o o o o o o o o o
o o o o o o
o o o o
o o o o o
o
o o
o o o o o o
o
c
c
c c
Canadian Cataloguing in Publication Data
c
Main entry under title:
(-
Diagenesis
c
(Geoscience Canada reprint series; 4) Includes bibliographical references. ISBN 0-919216-40-4
c
c c
1. Diagenesis.
2. Rocks, Sedimentary. I. Mcllreath , Ian A. II. Morrow, D.W. III. Geological Associat ion of Canada. IV. Series. QE571.D43 1990
552'.03
C90-090377-5
Cover Explanation: Realms of burial diagenesis (mesogenesis) are colour-coded in shades of yellow. Early diagenesis (eogenesis) is shown in purple and post-uplift diagenesis (telogenesis) is shown in green. Hot lower lithosphere marking the boundary between diagenesis and metamorphism is indicated in red. Directions of fluid movement are indicated by arrows . Black indicates open pore space in cavities or vugs . These open spaces are filled with diagenetic mineral cements during burial. Illustration Concept: David Morrow, Calga ry Graphics: Design Media Ltd., Toronto
Additional copies may beobtained bywriting to: Geological Associat ion of Canada Publications Department of Earth Sciences Memorial University of Newfoundland St. John 's, Newfoundland A1B 3X5 Canada
© 1990
ISBN 0-919216-40-4
Typesetting: Geological Association of Canada and Typeline, Mississauga. Ontario Printing: The Runge Press Ltd. Ottawa, Ontario
Autho rization to photocopy items for internal or personal use, or the internal or personal use of specific clients , granted by the Geological Associat ion of Canada for libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, prov ided that the base fee of $3.00 per copy is paid directly to CCC , 21 Congress Street, Salem , Massachusetts 01970, USA. 0·919216·40·4 $3 .00 + 0.00. The above perm ission does not extend to other kinds of copying , such as copy ing for general distr ibution, for advert ising or promotional purposes, for creating new collective works, or -for resale. For such copy ing , arrangements must be made with the publisher in advance of publ lcatlon .
is
L'Association geolog ique du Canada accorde I'autorisation de photocopier des documentspour un usage interne au personnel , ou pour I'utilisation interne ou personnelle de clients particuliers, aux bibliotheques et autres util isateurs inscrits au Copyright Clearance Center (CCC) Transactional Reporting Service, ala condition que Ie tar if de base de 3$ par copie soit paye directement au Cent re, 21 Congress Stree t, Salem, Massachusetts 01970, USA. 0·919216·40·4 $3 .00 + 0.00 . La presents permission ne s'applique pas a d'autres genres de reproduction, notamment la reproduction en vue d'une distribution generate, a des fins de publtcite au de promotion, pour la creation de nouveaux travaux collectifs ou pour la revente . Dans ces cas , iI faut prendre les dispos itions qu i s'imposent en cornmuniquant en advance avec I'ed iteur de l'Association.
c c c c c c c c c c c c c c c c c c c c c c C E l C C C C
C C C
( r:
CONTENTS
( r
r:
( (
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction I.A. Mcllreath and D. W Morrow
.
Limestones - Introduction N.P James and P W Choquette
.
Limestones - The Sea Floor Diagenetic Environment N.P James and PW Choquette Limestones - The Meteoric Diagenetic Environment N.P James and P W Choquette Limestones - The Burial Diagenetic Environment P W Choquette and N.P James
c
v
Dolomite - Part 1: The Chemistry of Dolomitization and Dolomite Precipitation D.W Morrow
@§)
@ 113
Dolomite - Part 2: Dolomitization Models and Ancient Dolostones . . 125 D.W Morrow
~~1P~~~~~e;~~d ·I i<.·L~~~~~t~i~' r
r:
r
Aspects of the Diagenesis of Coarse-Grained Siliciclastic Rocks .. I.E. Hutcheon
8 @
Catagenesis of Argillaceous Sedimentary Rocks .
177
Chemistry and Diagenesis of Organic Matter in Sediments and Fossil Fuels M.A. Barnes, WC. Barnes and R.M. Bustin
189
Determining Levels of Organic Diagenesis in Sediments and Fossil Fuels R.M. Bustin, M.A. Barnes and we. Barnes
205
Origin of Chert: Diagenesis of Biogenic Siliceous Sediments R. Hesse
227
A. E. Foscolos
Silica Diagenesis: Origin of Inorganic and Replacement Cherts .. .. 253 R. Hesse Early Diagenetic Pore Water/Sediment Interaction: Modern Offshore Basins R. Hesse
277
Fluid Geothermometry in Sedimentary Systems I.E. Hutcheon, H.L. Abercrombie and D. Gardner
317
Index
325
/
c Co r:
G C
o
( ( , r:
G
o o
c
o o o o r:
c o o
o o o o
v
PREFACE The articles that formed the Diagenesis series in Geoscience Canada were initiated by the senior editor, Ian A. Mcllreath, in the March 1982 issue of Geoscience Canada. The underlying philosophy of the Diagenesis series was similar to that of the Facies Models series edited by Roger G. Walker. Like the Facies Models series, the Diagenesis series has been directed toward the non-specialist, the "general purpose geologist" alluded to by Professor Walker in the preface to the Facies Models series, and toward research scientists interested in aspects of diagenesis outside their normal fields of specialization. The enormous success of the first review series indicated that other aspects of the geosciences, such as diagenesis , would benefit from this type of simplified approach. As with the Facies Models series, authors were asked to prepare reviews dealing with aspects of diagenesis that entailed a minimum of jargon ,but yet covered a signif icant part of the subject. Diagenesis itself is a natural associate of the subject of sedimentology, the main focus of the Facies Models series. An appreciation of both sedimentology and diagenesis are, of course, necessary for an overall comprehension of the origin of sedimentary rocks. In this sense, the Diagenesis review series is a natural complement to the Facies Models series. Twelve of the fifteen contributions within this volume originally were published in Geoscience Canada and have since been revised for publicat ion here. Three papers are new to this series and have been included to broaden the volume coverage on the subject of diagenesis . In this regard we extend our special thanks to Dr. R.J. Spencer of the University of Calgary for providing us, at relatively short notice , with a review paper dealing with the diagenesis of evaporites. This series emphasizes the role that diagenesis plays in the origin of common sedimentary rock types or, in other words, how diagenesis transforms unconsolidated sediments into Iithified sedimentary rocks. Most of the papers in this volume deal with specific sedimentary rock types and their diagenesis and_are segregated into groups according to general rock type so that carbonates and evaporites are grouped separately from siliciclastics. Diagenesis of organic material also is discussed separately and not as part of other rock types because of its widespread occurrence as a constituent in all rock types and because of its paramount economic importance both as coals and as fluid hydrocarbons.
Other new contributions to the original series have been provided by R. Hesse, and I.E. Hutcheon, H.J. Abercrombie and D. Gardner concerning Inorganic Chert and Fluid Geothermometry, respectively. We are grateful for these additional contributions. The emphasis of the latter paper is not on the origin of a particular rock type, but rather on an important aspect of the subsurface diagenetic environment. This paper was included to remind even the casual reader that diagenesis involves the study of processes that affect the physico-chemical characteristics of the diagenetic environment and to encourage the reader to look beyond the geologist's natural emphasis on the " end products " of diagenesis to the conditions and processes that formed these "end products". The paper by Hutcheon, Abercrombie and Gardner fulfils this need by showing how dissolution and precipitation of minerals in the subsurface is closely related to temperature and solution composition. Ian Mcllreath originated this series and invited one of the authors , David W. Morrow, to later join him as co-editor in order to facilitate the handling of what proved to be a review series of unusual length. The length of this series has also led to the decision not to include papers dealing primarily with techniques for the study of diagenesis. This has led to the exclusion of some papers that were part of the original Geoscience Canada series from the final Diagenesis volume and to the inclusion of some addit ional papers that were submitted after publication of the Geoscience Canada series. This is an unfortunate, but necessary, consequence of the decision to give priority to a more cohesive series and we extend our apprec iation to the authors of papers that have been excluded for their understanding with regard to this decision. Revised versions of some of these techniques papers have recently been included in other compilations (see Hutcheon , 1989). Monica Easton, as Managing Editor for Geoscience Canada, has been more than helpful in editing and in handling the manuscripts for the printer. We thank a succession of chairpersons (Glen Caldwell, John Kramers , Godfrey Nowlan and Bob Baragar) of the Geological Association of Canada publications committee for their help. Peter Russell designed the sketches that head each article. Also, thanks are due to Bob McNutt and Andrew Miall, the editors-in-chief of Geoscience Canada during the progress of this series. Finally, we extend our thanks to all authors and contributors of illustrations and photographs for their willingness to fulfil our requests and for their monumental patience in waiting for final publication. We hope their patience has been rewarded.
Ian A. Mcllreath and David W. Morrow Petro-Canada and the Institute of Sedimentary and Petroleum Geology, 1990
General Introduction L
11 01
(
( ( (
(
(
c
(
Diagenesis General Introduction David W. Morrow Institute of Sedimentary and Petroleum Geology 3303 - 33rd Street N. W Calgary, Alberta T2L 2A 7 Ian A. Mcllreath Petro-Canada P. O. Box 2844 Calgary, Alberta T2P 3E3 INTRODUCTION In this general introduction to a review of diagenesis, we want to give the reader a broad overview of the papers that are gathered together in this volume. Also, we aim to provide insight into what constitutes the scope of diagenesis and comment on some practical if not economic reasons why we should study diagenesis. Lastly, we hope to give the interested reader some indication as to what are the directions of modern research toward the study of diagenesis. DIAGENESIS - DEFINITION AND SCOPE The term "diagenesis", like the term "facies " (e.g., Walker, 1979), has undergone continual redefinition and discussion since its introduction to the geological literature by Von Guembel (1868). Larsen and Chilingar (1979), in their comprehensive review of diagenesis in all its aspects, provide an excellent historical summary of the term "diagenesis" as it applies to the origin of sedimentary rocks. Broadly speaking, as Larsen and Chilingar (1979) have stated, "diagenesis can be defined as the changes whi ur in the character and com osition of sed iments~ ginning from the moment of de osi' and lasting until the resultin materials roc s ~eJI1QY..e.d.J.ntQ the...r.e.al!J:L..Dl ~9J: ~_ " alJd includes ro-
cesses such as compaction, cementat io n, solut Lo~lJ.:Ill.adlated minera) recrystallization and r.eRlacement, subsurface_solutiQo:IeJiching,.coaliflcatic;>..D.. aiidttle degradation of organic material and generation of h drocarbon5.""Some au ors, part icularly Russian workers (e.g., Strakhov, 1956), differentiate between the period of initial lithification of unconsolidated sediment and later changes that affect already lith ified rock. Strakhov (1953, 1956) restr icted the use of the term " diagenesis" to apply only to the initial lithification stage and applied the term " catagenesis" or "metagenesis" to any and all subsequent changes that affect sedimentary rocks. However, in Norti ' America and Europe, the term " d i ag e ~ esis" is taken to embrace all Chang~ that occur in sediments and sedimen tary rocks up to the time when they enter the realm of metamorphism (Larsen and Chilingar, 1979; see also Foscolos, this volume, p. 177-187). .;;!.. The transition between diagenesis ~nd metamorphism has not been as rigorously defined as the transition between sediment deposition and subsequent diagenesis. As discussed by Larsen and Chilingar (1979), it is impossible to determine a unique combination of pressure and temperature that could be taken to characterize the transition between diagenesis and metamorphism because of the greatly contrasting degrees of mineralogic stabilities that characterize different rock types. For example, evaporites recrystallize at temperatures and pressures that have little affect on quartz sandstones. Most workers would probably concur with Fyfe at al. (1958)that diagenesis involves minor changes in the rock matrix and "crystallization of cement minerals in the hitherto open pores" whereas metamorphism occurs where "the coarse crystal grains are also extensively involved in reaction so that the rock becomes substantially recrystallized". In essence, therefore, the classification of a particular rock as metamorphic indicates that the original sedimentary components are no longer discernible as might be inferred from the etymology of the word "metamorphosis". Although it is not possible to assign unique ranges to the pressure and temperature conditions characteristic of subsurface diagenetic environments, it
! ",,, ~
is possible to assert, in a general way, that the realm of diagenesis is character ized by pressures of 0.1 MPa to 10MPa(1.0 bar to 1kb) and by temperatures ranging from 0° to 300°C (see Bates and Jackson , 1987). This corr sponds to a maximum burial depth of about 10 to 15 kilometres. In other words, diagenesis occurs under the pressure-temperature conditions characteristic of the earth 's outer crust and the earth's surface. One natural thermodynamically defined interface that might be considered to place an absolute operational upper limit on the realm of diagenesis coincides with the critical points of aqueous solutions involved in diagenesis. This is the curvilinear plane in pressure - temperature - composition space across which liquid solutions and their associated vapour phases merge to become single fluid phases. Little is known about the effects of supercritical fluids on sedimentary rocks except to say that recrystallization is likely to be much more rapid and complete under the ir influence than that experienced by rocks containing subcritical solutions. One problem with this type of demarcat ion of the diagenesislmetamorphism boundary is that even for pure water the critical temperature of 374°C is considerably greater than that generally assumed to coincide with the upper temperature limit of the diagenetic realm. This natural thermodynamic demarcation is even less realistic for the much higher critical temperatures and pressures that apply to the saline aqueous soutions (Crawford, 1981, fig. 4.12) that predominate in the subsurface (Dickey, 1969). Consequently, the pressure-temperature boundary separating diagenetic from metamorphic conditions in the subsurface occurs at considerably lower temperatures and pressures than those which characterize the critical points of subsurface fluids. In other words, diagenesis of rocks proceeds only under the influence of subcritical fluids; liquids and their associated vapour phases. Important consequences of the subcritical chemical behaviour of diagenetic fluids include the fractionation of gases, such as carbon dioxide, hydrogen SUlphide, methane and other light hydrocarbons between the gaseous and liquid phases (i. e., dissolved gases). For example, the existence of a
c 2
Diagenesis
Geoscience Canada Reprint Series 4
c (
large subcritical H20-C02 unmixing solvus (Todheide and Franck, 1963) controls the composition of the CO2 bearing diagenetic solutions that are most important for the creation and destruction of subsurface porosity through their participation in equilibria involving carbonate minerals (see Hutcheon , this volume, p. 165-176). In recognition of the fundamentally different processes that affect sedimentary rocks in a variety of diagenetically distinctive near-surface and deeper subsurface environments many modern workers have adopted classification schemes that subdivide these settings into two or three diagenetic realms. Recently, the tripartite classification of Choquette and Pray (1970) has gained in popular usage (e.g., Schmidt and Macdonald, 1979). In this classification of diagenetic realms, eogenesis is defined as those diagenetic changes that occur at or near the
[email protected] SUrface where the interstitial solutiollS_ are still in communication with the overlying water mass. After burial has effectively sealed the pore-filling solutions from the overlying water mass , the enclosing sediments are considered to r ~h ave passed into the mesogenetic dia~genetic realm of the dge~ubsurface. Most sediments remain in the mesogenetic regime throughout most of their diagenetic history. Diagenesis that occurs under the direct influeMfLQf. r?i\eteoric solutions following uplift and \.:::::./eroslon IS conl:1!gne.cLto..tbe..telogeaetic diagenetic .J..Ei9.i.m_e.... Other, similar schemes that separate early diagenesis from later diagenesis during burial and from diagenesis after uplift and erosion have been proposed (Fairbridge, 1967). Organic geochemists tend to use the term catagenesis in place of "mesogenesis" to indicate the diagenetic changes that affect organic material during burial (see Bustin and Barnes, this volume, p. 000-000). Diagenesis for most ancient sedimentary sequences involved the repeated exposure of these sequences to diagenetic realms as these sequences underwent cycles of subsidence and uplift. Generally, however, the imprint of the original eogenetic and mesogenetic history is preserved because of the large porosity reduction and lithification that occurs during the initial episodeof burial diagenesis. This severely limits the extent of subsequent mineral pre-
(b
cipitation. Exceptions to this rule occur where secondary porosity is developed with the opportunity for further significant porosity occlusion by diagenetic minerals precipitated during later burial episodes . The effects of telogenesis can, of course, be repeated many times and are particularly important in the diagenesis of soluble carbonates and evaporites. Telogenesis can totally erase the effects of earlier diagenetic events in these soluble rock types. Conversely, telogenesis may have only a minor effect on insoluble siliciclastic rocks and on organic material. THE DIAGENETIC CYCLE PROCESSES, CONTROLS AND PRODUCTS Classifications, such as that of Choquette and Pray (1970), are useful in that they communicate to the reader important information in a cursory manner and the reader is forced to understand the rationale that underlies these classificat ions. However, all classifications have a drawback in that they are generalizations. If a classification scheme deals with purely descriptive attributes, such as classifications of sedimentary rock types, then it pre-
sents a quantitative, or at least semiquantitative, approximation of objectively measureable parameters such as mineralogic composition and texture. In contrast, the classification of Choquette and Pray (1970) and other similar schemes primarily are inferential and are based on the inferred origin or genesis of diagenetic features observed in sedimentary rocks. It is precisely these features and phenomena about which our knowledge is very in complete. Rocks which have very different diagenetic histories can be quite similar in mineralogy and texture. Dolomites (or dolostones) are perhaps the prime example of a rock type for which classifications such as that of Choquette and Pray (1970) can not be used with confidence even though dolostones themselves are considered by most workers to be, almost invariably, completely diagenetic in origin (Larsen and Chilingar, 1979). Consequently, the objective of the reviews gathered together in this series of articles lies in demonstrating the diversity of processes through which sediments are diagenetically altered and Iithified rather than to categorize them rigorously within classification schemes.
THE DIAGENETIC CYCLE
.
DIAGENETIC PROCESSES ROCK/FLUID INTERACTION
~
DECREAS.... THERIIIOGYNAMIC 8RADIENTS
INCREASING THERMODYNAMIC GRADIENTS
c c c c c c
c c c
c
SEDIMENTS
\
c c c c c c c c c
,r------,~ DIAGENETIC PRODUCTS ROCK AND PORE SPACE WITH CONTAIMED FLlHDS
+
ROCK NON-POROUS HtGHL Y-CEMENTED
Figure 1 A schematic representation of burial diagenesIs. The commonly observed correlation of porosity reduction with burial depth is accompanied by repeated departures of the rock/ water system from thermodynamic equilibrium. These departures result in mineral precip itation or dissolution until the pore solution becomes saturated with respect to potential mineral precipitates. The natural evolution of pore fluids in the direction of increasing concentration or salinity with increas ing depth of burial favours mineral precipitation and porosity reduction during burial diagenesis . However, many processes can interrupt or reverse this trend and cause the retention or secondary development of porosity in the subsurface that is necessary for the emplacement of hydrocarbon accumulations and many mineral deposits.
c c c c
c c c c c C E \.
C C C C l
'-<.
(, (
General Introduction
3
(
c
o (
c ( (
o
o o o o
o
CO
o
o o o
o
o o o o o o c (
l r
"~
Commonly, sediments undergo multiple stages of diagenesis in the subsurface reflecting a succession of episodes of cementation, dissolution and recrystallization. In a very generalized manner, these episodes may be viewed as part of an overall diagenetic cycle (Figure 1) involving cyclical departures of the rock/fluid system from thermodynamic equilibrium. These departures from equilibrium are followed by rock/fluid interaction that may be manifested as the precipitation of mineral cements, as pore space creation and enlargement by dissolution or by fluidmediated mineral replacement reactions. These processes all reflect the return to thermodynamic equilibrium of the rock/fluid system. This type of generalization emphasizes the great variety of subsurface chemical processes that cause dissolution and precipitation in response to departures from thermodynamic equilibrium. However, diagenesis itself is often represented in terms of unidirectional trends of porosity reduction through physical compaction, pressure solution and mineral precipitation that commonly reflect the initial cycle of burial of a sediment which may have endured multiple episodes of burial and uplift. Diagenesis during burial is almost universally accompanied by a trend of increasing pore water salinity that commonly, though not invariably, is linked to a progressive approach to saturation with respect to dissolved minerals (Hanor, 1979). Consequently, there is an overall tendency for precipitation of mineral cements to continue throughout the burial history of a sedimentary rock resulting in a rock that is highly cemented and containing little porosity (Figure 1). Reversals to the general trend of porosity reduction occur if major mineral components of the rock become undersaturated during diagenesis. The generation of secondary porosity in carbonates by subsurface carbonate mineral dissolution during burial (Schmidt and MacDonald, 1979) is an excellent example of this type of reversal in which cementation is followed by dissolution and porosity enhancement leading to the development of subsurface hydrocarbon reservoirs. Many examples of hydrocarbon reservoir facies and ore deposit host facies exist because of porosity enhancement beneath uncon-
formities. The immense Fateh Field of the United Arab Emirates, with over on billion barrels of ultimate recoverable oil reserves, is an outstanding example of reservoir enhancement by secondary porosity development during the circulation of low salinity meteoric ground water beneath an unconformable s face (Jordan et aI., 1985). It can readily be appreciated then that the occurrence of hydrocarbon accumulations of economic size is linked to the preservation of original or primary porosity, or to the creation of secondary porosity throughout the burial range favourable for the maturation of hydrocarbons (see Barnes et a/., this volume, p. 189-204). This is true also for the occurrence of many economic metal deposits which, in addition, depend upon the generation of subsurface solution salinities compatible with the formation of soluble chloride metal complexes favourable for the transport of metals in solution to precipitation sites in porous rock (Hanor, 1979). In other words, under normal conditions, burialdependent processes of hydrocarbon generation and metal concentration in subsurface fluids are countered by the progressive disappearance of porosity and hence a reduced opportunity for the entrapment of hydrocarbons or formation of ore deposits. Atypical events are necessary for the generation of enhanced subsurface porosity although there is now some evidence to suggest that organic maturation itself is responsible, in some cases, for the development of "aecondary porosity t1:1"Ol 19b tb~ generation of organic acids during early bur ial diagenesis (e.g., Surdam and Crossey, 1985). Commonly, however, undersaturation with respect to diagenetic minerals in subsurface fluids and consequent porosity enhancement through mineral dissolution is accomplished by mixing with less saline meteoric solutions. This can occur where deeply buried sediments are uplifted to shallower depths so that the sedimentary pile can be infiltrated by fresh or brackish ground water. An excellent example of meteoric ground water invasion that accompanied post-burial uplift is afforded by the Mesozoic and Tertiary foreland basin sediments of western Canada that were uplifted and subaerially exposed during the Tertiary Laramide Orogeny (Hitchon, 1984). The release of interlayer water and water of
crystallization from clay minerals during burial-induced clay diagenesis also has the potential to cause mineral undersaturation in pore solutions and may play a significant role in the development of subsurface "overpressured" zones (see Foscolos, this volume, p. 177-187).
SEDIMENTARY ROCKS AND FLUIDS - DIAGENETIC END PRODUCTS The primary focus of the articles presented in this volume is on the end products of diagenesis - the rocks themselves and the processes that led to the development of individual rock types. The sequence of articles here is different from their original order of appearance in the Geoscience Canada series. Here, there is an overall subdivision between sedimentary rocks that originated as chemical sediments and those that are siliceous or siliciclastic. Papers concerned with organic material are included in the latter part of the volume because of their emphasis on the maturation of organic material in siliciclastic sediments. The first four papers by James and Choquette deal with the diagenesis of limestones from their initial deposition through deep burial diagenesis and subsequent exposure to meteoric waters. They point out that diagenesis in limestones is profoundly influenced by the initial sediment mineralogy, whether it is dominantly magnesian calcite or aragonite. They outline the intriguing research that has increasingly indicated that shallow marine carbonate sedimentation and associated synsedimentary cementation has not always been dominated by aragonite. Early Paleozoic seas may have been t.dominated by calcite, rather than arag-LV 'f onite precipitation. The role of climate as a major control on meteoric diagenesis (i.e., telogenesis) and the discussion of pressure-solution effects during burial diagenesisare emphasized. Pressure-solution, both intergranular and along stylolites, is identified as the most active process in porosity reduction and as a major contributor of cement during deep burial diagenesis. An abundance of textural, chemical and isotopic evidence is utilized to compare modern with ancient examples of carbonate diagenesis or to simply infer diagenetic processes or conditions. Among these criteria, oath-
4
odoluminescence of cement fabrics, although commonly used to infer precipitational conditions during the cementation of ancient limestones, appears to us to lack documentation in modern examples of diagenesis. Some key research, in our estimation, could be done through the systematic investigation and documentation of the cathodoluminescent character of modern or near-modern calcite cements from a variety of diagenetic environments. This would place inferences concerning the significance of cathodoluminescent cement stratigraphy in ancient limestones on a firmer footing. The role of organic maturation toward the generation of secondary porosity and late stage cementation is described in the paper dealing with deep burial diagenesis. It is evident that research into the interaction of organic material with carbonates during diagenesis has barely begun. Two papers by Morrow discuss dolomitization, one of the more elusive diagenetic gremlins . The first paper deals with what is known concerning the chemistry of dolomitization and dolomite precipitation. The influence of a variety of kinetic inhibiting factors affecting dolomite precipitation at low temperature is discussed and the recent perception that the presence of sulphate in solution strongly inhibits the precipitation of dolomite, even at the high temperatures typical of deep burial, is assessed. The following paper deals with the origin of ancient dolostones and a wide variety of models for dolomitization in synsedimentary to deep burial settings are described. Emphasis is given to mass balance considerations in assessing the potential relevance of particular models for dolomitization. Recent research concerning the origins of dolomite crystal textures and their potential application as diagenetic geothermometers are discussed along with the recent application of fluid inclusion studies to the problem of the origin of dolomite. Spencer's contribution concerning the diagenesis of evaporites rounds out the first half of the volume series. Textural studies and criteria for the recognition of successive episodes of evaporite diagenesis are emphasized. The long neglected importance of diagenetic cementation, inferred from the degree of porosity reduction observed in buried evaporites and ancient evaporites,
Diagenesis
is addressed through the development of criteria for the identification of cementation and recrystallization textures. The presence of certain " high temperature salts" in evaporites may be useful as minimum temperature geothermometers and, along with fluid inclusions, provide evidence as to temperatures and pressures that affected these evaporites during diagenesis. The second half of the volume, dealing with siliciclastic and siliceous rocks and the diagenesis of organic material, begins with Hutcheon's paper on the diagenesis of coarse-grained siliciclastics. Burial diagenetic reactions, such as feldspar albitization and reactions between clay minerals and carbonates occur in many siliciclastic sequences and are shown to have a profound effect on the chemistry of pore solutions and on the potential for porosity reduction through reaction-related cementation. Some of these reactions generate CO2 in addition to CO2 generated by organic maturation. This CO2 can contribute to the development of secondary porosity or to additional carbonate cementation and porosity reduction depending on whether carbonate equilibria or non-carbonate equilibria control the hydrogen ion activity (pH) of diagenetic pore solutions. Hydrocarbon recovery efficiencies are related to the presence of porefilling diagenetic clay and carbonate minerals. Diagenes is of fine-grained argillaceous sediments is outlined by Foscolos. Here, the emphasis is on the effects of the temperature controlled, dominantly isochemical clay reactions by wh ich swelling clays (smectites) evolve to the non-swelling iIIitic clays that characterize most ancient basinal shales. These changes are accompanied by the sporadic release of water from clay minerals to the pore solution and may contribute to the development of overpressured zones in shale sequences and, possibly, to hydrocarbon migration. The dissolution of feldspars generates amorphous hydrous gels of silica, aluminum and iron that can participate in later stages of clay diagenesis, such as the growth of diagenetic chlorite in pore spaces. Diagenesis of organic material is reviewed in the follOWing two papers by Barnes , Bustin and Barnes. In their first paper, they outline the transformation of initial biologic materials, such as pro-
Geoscience Canada Reprint Series 4
teins, carbohydrates, lipids and lignin into hydrocarbons. Early diagenesis ends with the synthesis of " kerogen", the ubiquitous source material for hydrocarbons . Unlike many other diagenetic processes, maturation of hydrocarbons is irreversible and strictly temperature dependent. Plots of the hydrogen 'to carbon ratio versus the oxygen to carbon ratio of kerogen (or bitumen) and coals on "van Krevelen" diagrams delineates the progressive thermal maturation of these residual organic solids during burial and generation of hydrocarbons of progressively shorter carbon chain lengths. Their second paper outlines the tremendous advances that have been made in the quantification of stages in the thermal maturation of kerogens and coal. They discuss vitrin ite reflectance and the othe r, less well documented optical methods , such as conodont colour alteration indices (CAl), palynomorph thermal alteration indices (TAl), and liptinite fluorescence that are commonly compared to an equivalent rank of vitrinite reflectance . Chemical methods for evaluating source rock potential and for characterizing maturation rank also are discussed. These include techniques for isolating biological markers or " geochemical fossils", such hopanes and steranes, from extractable organic matter (i.e., organic matter soluble in standard organic solvents). These markers provide information concerning the source rocks for par- • ticular hydrocarbon accumulations . Rock-EvaliP analysis is presented as a means for the routine estimation of both the organic source rock potential and the maturation rank of most sedimentary rocks in a manner that is most accessible to petroleum exploration ists unfamiliar with organic geochemistry. Modelling of thermal maturity in basins is also discussed using calculated timetemperature indices (TTl) based on subsidence histories. The origin of chert and siliceous sediments is discussed in two following papers by Hesse. In the first paper, he describes the origin and distribution of siliceous sed iments in the world's oceans emphasizing the important role of water depth in relation to the calcite compensation depth. Low temperature maturation of siliceous radiolarians and diatoms results in their transformation from opal-A to the opal-CT of porcel-
c c c c c c c c c c
c c c
c c c c c c c
c c
c c c c c c c c c c c e <--
C C
C C C C
C r: C c.,
r
o o c c c c ( (
o o o o o o o C l (
C
o
General Introduction
5
lanites and finally to chert composed dominantly of quartz. The shift in the d(101) peak position of opal-CT has some potential as a maturation indicator in sediments less than 100 million years old. The second paper emphasizes the various modes of siliceous replacements that sediments and rocks undergo. Commonly, such replacements occur under high pH conditions that favour silica precipitation. The final two papers of the series, by Hesse and Hutcheon, Abercrombie and Gardner depart from the main series theme emphasizing the "end products" of diagenesis in that they deal instead with diagenetic conditions and processes. Hesse shows how the interpretation of concentration profiles of pore water chemistry in modern marine basins can be used to infer rates of sedimentation and pore water advection with respect to ongoing sediment reactions. The role of organic material in controlling early sediment diagenesis because of the generation of carbon, nitrogen, sulphur and hydrogen as bacterial metabolites is emphasized. Hutcheon, Abercrombie and Gardner show how concentrations of ions, such as sodium, potassium and silica, in solution can be used as geothermometers in the presence of equilibrium mineral asemblages. Table 1 is meant to provide a generalized assessment of the relative degree of importance of individual diagenetic controls on the development of the diagenetic rock types that are described in this series. These controls are envisaged as primarily being external in nature and are imposed upon the rockor sediment-pore fluid system. For Table 1
example, externally imposed changes in fluid chemistry commonly exert a strong effect on the diagenesis of carbonate, clays and aluminosilicates and silica but has little effect on the diagenesis of organic material or evaporites. Organic material is commonly not affected because its diagenesis is dominated by thermal organic maturation whose reactions proceed relatively rapidly to form liquid hydrocarbons that tend to exclude other fluids from the pore space. Evaporites are little affected by compositional changes in pore solutions because dissolution of soluble evaporite minerals effectively buffers the solution composition against externally imposed solution composition changes. Temperature is, of course, an important external control on all diagenetic changes involving chemical reactions. Silica cementation has been singled out as an exception to this rule because of the greatly dominant role that the pH of the solution plays in silica diagenesis, particularly during early diagenesis, or eogenesis. The rate of fluid flow has a strong effect in most diagenetic situations, but generally not on the diagenesis of organic material. Similar reasoning may be applied to the other cited diagenetic controls in assessing their relative significance. This table is presented merely to give the reader an overview of the generalized diagenetic controls that apply to the diagenetic end product rock types discussed in this volume. DIRECTIONS OF MODERN RESEARCH It is difficult to summarize in a concise manner the directions of modern re-
search toward understanding diagenesis. Such summaries become dated very quickly with respect to such a rapidly evolving subject in which research efforts are proceeding in many different directions. With this caveat in mind, certain aspects of diagenesis can be identified as undergoing particularly intensive research at the present time. The final paper of this volume by Hutcheon, Abercrombie and Gardner highlights one important and very active modern research direction. This is the application of computer-based models of mineral and aqueous chemistry to simulate the thermodynamic state of diagenetic systems. Figure 2 is an example of a stability diagram generated by the PC-based program (PTA) of Brown et al. (1988). This diagram was used to infer the compositional changes of waters that participated in diagenesis of the Clearwater Formation at Cold Lake, Alberta, based on the petrographically determined sequenceof diagenetic minerals (Abercrombie, 1989). Programs such as PTA of Brown et al. (1988)generate phase diagrams in pressure - temperature - activity space for thermodynamic systems with userdefined components. Other programs, such as SOLMNEQ (Kharaka and Barnes, 1973) bridge the gap between the thermodynamic activities of components and their concentrations in aqueous systems to permit the calculation of mineral saturations from concentration data. Other programs, such as PHREEQE (Parkhurst et al., 1980), have been developed which permit the modelling of reaction pathways, which may be plotted on phase diagrams, and can even give estimates of the extent of
Dependence of diagenesis on major diagenetic controls. Diagenetic Controls
Diagenetic Products
l r
"
'---
Biologic Activity Strong
Time
Temperature
Pressure
Fluid Flow
Weak
Strong
Strong
Strong
Fluid Chemistry Strong
Evaporite Cementation and Recrystallization
Weak
Weak
Strong
Strong
Strong
Weak
Clay and Alumino-Silicate Diagenesis
Weak
Strong
Strong
Weak
Strong
Strong
Organic Material
Strong
Strong
Strong
Weak
Weak
Weak
Silica Cementation and Recrvstallization
Strong
Weak
Weak
Weak
Strong
Strong
Carbonate Cementation and Recrystallization
c Diagenesis
6
Geoscience Canada Reprint Series 4
c
c
mass transfer of chemical components between the solid phases and the solution during fluid-mediated chemical reactions . Ultimately, this approach will yield important quantitative estimates of the amount of porosity loss or gain during successive stages of diagenesis for a wide range of rock types . A closely related intensive research effort has recently been devoted toward an improved understanding of the interaction between organic matter and mineraI diagenesis. The production of organic acids and carbon dioxide as reaction products during the diagenesis of organic matter has been shown to exert a strong influence on the saturation state of carbonate minerals and the acidity of subsurface environments (Surdam and Crossey, 1985). Again the accuracy of the assessment of values for these solution parameters is dependent upon the development of accurate models of aqueous chemistry. One important aspect with regard to the improvement of these models concerns the acquisition of more accurate thermodynamic data for organic acids, such as acetic acid, which are abundant in subsurface environments. Recent advances in the understanding of early diagenesis of organic material have been made concerning the correlation of early diagenetic mineral reactions, such as the generation of pyrite by sulphidic diagenesis, with the degree of preservation and availability of organic material for diagenesis during later burial (Gautier, 1985). The diagenetic aspects of models for basin development have undergone definite changes under the impact of new ideas concerning the role of organic acids. For example, Surdam et aJ. (1989) have shown that variations in the amounts and types of organic material results in profound differences in the style of mineral diagenesis and in the evolution of porosity during burial diagenesis of many hydrocarbon-bearing siliciclastic sequences . Rapid recent advances have also occurred with respect to techniques that provideestimates of the relative and absoluteagesof subsurfacediageneticevents. Absolute ages for the precipitation of authigenic clay minerals and feldspars in siliciclastic sediments may be obtained from the K/Ar, 4OArJ39Ar, and K/Ca isotopic dating techniques and from fission track dating with greatly improved
c
c c
COMPUTER-GENERATED MINERAL PHASE DIAGRAM
(
c
12
c c
Analcime
10
c c c
C C C
Potassium Feldspar
C
C C C
C C C
o
C C C C
o
2
10
12
C
C C Figure 2 A computer-generated phase diagram that has been used to infer the evolution of fluid compositions during diagenesis of the Clearwater Formation in the Cold Lake heavy oil area, Alberta (Abercrombie, 1989). The succession of diagenetic minerals observed in this unit indicate initial deposition in brackish waters of mixed marine/river composition (stage I). Enrichment of sodium and potassium is reflected by the progression from mixed layer clays (lIa) to the maximum diagenetic assemblage, stage lib, comprising quartz-Kspar-analcime-iIIite(?). This was followed by precipitation of Kaolinite and isotopically distinct calcite reflecting Tertiary uplift of the Clearwater, an influx of low salinity meteoric fluids, and biodegradation of oil. This caused pore fluids to change composition along the path labelled 1/1.
C C C
c (
C
C (
C
C
/'
(
General Introduction
7
r
C G (
( ( ( (
c o o c
o o o
o o o o o (
r:
accuracy in recent years. Such advances greatly improve our ability to unravel the geochronological history of diagenetic cements and consequently our ability to understand the history of diagenesis and its application to basin analysis (Lundegarde, 1989). Fission track dating of heavy minerals, such as apatite and zircon, also gives important paleotemperature infor mat ion that spans the temperature range for oil generation (Naeser, 1986). Modern studies of diagenesis have also moved in the direction of semiquantitative to quantitative modelling of diagenetic processes from the scale of diagenetic micro-environments to a global scale. Chilingar and Wolf (1988) provide a broad overview of research concerning models for diagenesis, particularly as they apply to the origin of ore deposits and hydrocarbon reservoirs. It is evident that the study of diagenesis has expanded far beyond a purely descriptive stage involving merely the identification of minerals or rocks of diagenetic origin. This series has emphasized the contribut ion of diagenetic processes to the origin of specific rock types. The reader will appreciate, however, that diagenesis itself involves the interaction of many different processes in the lithosphere , biosphere and hydrosphere. The complexity of some of these processes may be glimpsed in the papers of this series. It is our hope that this information may be of assistance to the reader in pursuing his or her research interests or as an aid to exploration ists involved in the search for hydrocarbon and mineral resources. REFERENCES CITED Abercrombie, H.J., 1989, Water-rock interaction during diagenesis and thermal recovery, Cold Lake, Alberta, Ph.D . thesis, Department of Geology and Geophysics, The Unversity of Calgary, Calgary, Alberta , 183 p. Bates, RL. and Jackson, J.A., 1987, Glossary of Geology, Third Edition: American Geological Institute, 788 p. Brown, T.H., Berman, R.G. and Perkins, E.H., 1988, GEO-CALC: Software package for calculation and display of pressure-temperature-composition phase diagrams using an IBM or compatible personal computer: Computers and Geosciences, v. 14, p. 279-289.
Chilingar, G.V. and Wolf, K.H., 1988, eds., Diagenesis , I: Elsevier, Amsterdam , Developments in Sedimentology, v. 41, 591 p. Choquette, P.w. and Pray, L.C., 1970, Geologic nomenclature and classification of porosity in sedimentary carbonates: American Association of Petroleum Geologists Bulletin, v. 54, p. 207-250. Crawford, M.L., 1981, Phase equilibria in aqueous fluid inclusions, in Hollister, L.S. and Crawford , M.L. , eds., Short Course in Fluid Inclusions: Applications To Petrology: Mineralogical Association of Canada, Short Course Handbook Volume 6, p. 75·100. Dickey, P.A., 1969, Increasing concentration of subsurface brines with depth: Chemical Geology, v. 4, p. 361-370. Fairbridge, R.W., 1967, Phases of diagenesis and authigenesis , in Larsen, G. and Chilingar, G.V., eds., Diagenesis in Sediments: Elsevier, Amsterdam . Fyfe, W.S., Turner, F.J. and Verhoogen, J., 1958, Metamorphic reactions and metamorphic facies: Geological Society of America, Memoir 73, 259 p. Gautier, D.L., 1985, Interpretations of early diagenesis in ancient marine sediments, in Gautier, D.L., Kharaka, Y.F. and Surdam, R.C., eds., Relationship of Organic Material and Mineral Diagenesis: Society of Economic Paleontologists and Mineralog ists, Short Course No. 17, p.6·72. Hanor, J.S., 1979, The sedimentary genesis of hydrothermal fluids, in Barnes, H.L., ed., Hydrothermal Ore Deposits: John Wiley and Sons, New York, p. 137·169. Hitchon, B., 1984, Geothermal gradients, hydrodynamics, and hydrocarbon occurrences: American Association of Petroleum Geologists , Bulletin, v.68, p.713·743. Hutcheon, I.E., 1989,ed., Burial Diagenesis: Mineralogical Association of Canada , Short Course Handbook, v. 15, 409 p. Jordan, C.F., Jr., Connally, T.C., Jr. and Vest, H.A., 1982, Middle Cretaceous carbonates of the Mishrif Formation, Fateh Field, offshore Dubai, UAE., in Roehl, P.A. and Choquette , P.W. eds., Carbonate Petroleum Reservoirs: Springer-Verlag, New York, p. 425-442. Kharaka, Y.K. and Barnes, I. , 1973, SOLMNEQ: Solution-mineral equilibrium computations : Un ited States Department of Commerce, NTIS Report PB 215-899, Springfield, Virginia, 81 p. Larsen, G. and Chilingar, G.v., 1979, eds., Diagenesis in Sediments and Sedimentary Rocks: Elsev ier, Amsterdam, Developments in Sedimentology, v.25A, 579 p.
Lundegarde, P.D., 1989, Temporal reconstruction of sandstone diagenetic histories, in Hutcheon, I.E., ed., Burial Diagenesis : Mineralogical Association of Canada, Short Course Handbook, v. 15, p. 161-194. Naeser, N.D., 1986, Neogene thermal history of the northern Green River Basin, Wyoming-Evidence from fission track dating, in Gautier, D.L., ed., Roles of Organic Matter in Sediment Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication No. 38, p. 65-72. Parkhurst, D.L., Thorstenson, D.C. and Plummer, N.L., 1980, PHREEQE - a Computer Program for Geochemical Calculations: United States Geological Survey, Water-Resources Investigations 80-96 (revised and reprinted in 1987), 193 p. Schmidt, V. and McDonald, DA , 1979, The role of secondary porosity in the course of sandstone diagenesis, in Scholle, P.A. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication No. 26, p. 209·226. Strakhov, N.M., 1953, Diagenesis of sediments and its significance for sedimentary iron formation: Izvestia Akademia Nauk SSSR, Series of Geology, v. 5, p. 12-49. Strakhov, N.M., 1956,Towards knowledgeof diagenesis, in Questions of Mineralogy of Sedimentary Formations: L'vov Gos. Univ., L'vov. Surdam , R.C. and Cressey, L.J., 1985, Mechanisms of organic/inorganic interactions in sandstone/shale sequences, in Gautier, D.L., Kharaka, Y.F. and Surdam, RC. , eds., Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralogists, Short Course No. 17, p. 177-232. Surdam , RC., Dunn, T.L., Heasler,H.P. and MacGowan, D.B., 1989, Porosity evolution in sandstone/shale systems , in Hutcheon, I.E., ed., Burial Diagenesis, Mineralogical Association of Canada, Short Course Handbook, v. 15, p. 61-127. Todheide, K. and Franck, E.U., 1963, Das Zwe iphasengebeit und die kritische Kurve im system Kohlendioxid-Wasser bis zu Drucken von 3500 bar: Zeitschrift fur Physiks und Chernie, N.F., v.37, p.388-40t Von Guernbel, C.w., 1868, Geognostische Beschreibung des ostbayerischen Grenzgebirges, I-III, 700 p. Walker, R.G., 1979, ed., Facies Models: Geoscience Canada Reprint Series 1. 211 p.
8
Diagenesis
Geoscience Canada Reprint Series 4
SEM photomicrograph of sandstone in the Cretaceous Viking Formation ofthe Caroline Oil Field. Intergrown clusters of authigenic microquartz and chlorite (OC) are interspersed with authigenic pyrite (P) in the deeply buried, chert-rich Caroline sandstones. Chlorite is commonly associated with late-stage burial diagenesis of siliciclastic sediments. Photograph courtesy G.E. Rein!lon and A.E. Foscolos (Geological Survey of Canada).
c c c c c c ( c c c c c c c c c c c c c c c c c c c c c c c c c c
C
E C ( (
C
C (
C C
Limestones:
Introduction
9
Limestones: Introduction Noel P. James Department of Geological Sciences Queen's University Kingston, Ontario K7L 3N6
r:
Philip W. Choquette Department of Geological Sciences University of Colorado Boulder, Colorado 80309-250
o r r
INTRODUCTION Limestones are among the most capricious of rocks. Mostly born from the sea and generally biologic in origin, they possess an extremely sensitive record of past life and conditions on earth, yet because they are lime and so easily susceptible to chem ical change , this record is commonly blurred or rendered unreadable by diagenesis. Paradoxically, this same diagenesis is economically beneficial because the holes and voids created during alteration may be later filled by base metal precipitatesor hydrocarbons. So, to understand limestones for any purpose it is imperative to decipher the often complex series of processes that have modified their texture and composition through geologic time. The following articles are an overview and not an in-depth treatment of the topic ; they are intended as a guide to current understanding and a synthesis of concepts useful when interpreting the diagenetic history of limestones. Since most limestones are marine, the articles Y concentrate on these rocks. The papers were originally written between 1982 } and 1986 and have been revised and \,\t. updated for this volume.
, ' h:j.. ...... t--
1. /
,.-
~j
-;
-1k
/VCr X,'lA..j
'hV/Iet
~\"
LV'
,
l'\f'~1 Y '-'" v· ~ \ ' r , i"\J~ Q I).
j
~ ~ <:l ,J'....f
c-s«
* c)
® DEEP
BURIAL
(;'j~ I,
l\)nD~
Y-;,'
1\'" \ i
2.-0"~ (f,f)lAJ.. .0~ I-/t)
r . ;...
~ ./ ~
D'
tation, although not ubiquitous, is the most important diagenetic process on the modern sea floor. It is manifest in the f ormation of grains (i.e., ooids), the hardening of soft grains by intragranular precipitat ion (i.e., peloids), and the lithification of loose carbonate sediments to form sea floor limestone. In contrast, the main biological process is boring, by a variety of macro and microendoliths, who may destroy much depositional fabric . Both of these processes, precipitation and boring, take place on the shallow tropical sea floor. With increasing water depth, accompanying increases in pressure and decreases in temperature combine to result in some cases in ~ neomorphism ..and ultimately in complete dissolution. _ The Meteoric. If carbonate sediments are, through uplift, fluctuations in sea level or simply rapid accretion, exposed subaerially to percolating meteroic waters, dramatic changes take place. Diagenesis is driven by two reac- 0'1 tions, one between the carbonates , LV regardless of composition, and meteonc waters of differing temperature andlor ~~ Peo• called water-controlled meteoric V diagenesis and another between minerals of differing solubilities and meteoric water calledminerakontrolled meteoric diagenesis. Water-controlled alteration is reflected by the formation of caves and Other karst features, local precipitation of spelean carbonates ao.d_calccete-d~I-
'<
,,,)1 \; ~v
~
tion is surface and subsurface fluid composition. Waters most commonly in contact with carbonate rocks and sediments are marine, meteoric or deeE subsurface in origin ~With time, as carbonates are deposited, precipitated, buried, eroded, exposed and reburied, they interact with these fluids, each of which affects the sediments or rocks in a special way and leaves a unique diagenetic signature. Each region characterized by waters of different composition (Figure 1) can be thought of as a separate diagenetic environment (Purdy, 1968). In this synthesis we recognize three major diagenetic environments : (1) the sea floor and underlying shallow marine phreatic, characterized by marine waters, together with the strandline bathed in mixed marine and fresh waters; (2) the meteoric distinguished by the freshwater vadose and plireatic zones and the shallow phreatic region of mixed fresh and marine waters ; and (3) the deep burial, where pores are filled with waters that may once have been marine but have been moderately to drastically modified by burial diagenesis. Changes that take place in each of these settings is the topic of a separate article in the following series, but to place the whole spectrum in context we first summarize the main attributes of all the realms and outline the diagenetic changes that occur in each. The Sea Floor. Carbonate precipi-
..
J7;;';;::;t? .r~"
uO#'/ (~"
Figure 1 A sketch outlining the main diagenetic environments through which carbonate
DIAGENETIC ENVIRONMENTS sediments and rocks pass in geologic history. Depending upon hydrodynamics, configuration Since rock-water jnteractionJs the priof the aquifer and permeability, the freshwater phreatic waters may underlie only the exposed mary force driving carbonate diagenearea as illustrated here or may extend seaward beneath the shelf and underlie a thin shallow sis, a key tactonln the diagenetic equa- ,~t IV marine phreatic zone.
\..~( ~\ (btl I( I J.t M~'" ~(' '6' ~<,((}v c /)l/f()..rf~1J.0 I' \1lI' >\l ~ / \) II 1.:-......... lJY~i...,I \ \I\I"~V ' J
C~
--.-/
o
.
U
I)
(1'1
~ ~
"'.,
II 0
i
'I
~ U1u .
I
1 / _
U.lI\lvV~
I-(JI/Y
tiC " l
0'1
V'!f
{f.Jl.l/tl I l l , , , -( .,
~16'''r;j /1(/
"!~.-.:!.!!./7'oJ~
'j \
YI"'V 'I
-!.-_--
Diagenesis
10
Geoscience Canada Reprint Series 4
o
~
TECHNIQUES AND LITERATURE opment. Mineral-cQntrolled alteration burial environment for extended results in the transformation Qf aragperiods. NQt all carbonates , however, The indispensable source of information onite and magnesium-calcite (high are flushed by meteor lc waters . Nevon the topic of carbonate diagenesis is magnesium calcite) to calcite witfi negertheless, it is common to see some the text by Bathurst (1975) and the folIigible magnesium CQntent (low magnelowup article in 1980 . For a more general evidence of meteoric diagenesis, gensium calcite). Many components have erally because in shallow-marine settreatment of the subject the reader is their microstructure altered dramattings they accumulate rapidly to sea referred to the books by Friedman and ically during this process ot mineral level. Combined with subtle fluctuaSanders (1979), Blatt et al. (1980) and transformation, while others are distions in sea level, this commonly leads Flugel (1982). Good sourcesfor the idensolved completely leading tQ new pattification of components are the volto subaerial exposure and thus meteumes by Scholle (1978) and Horowitz terns of porosity and permeability. oric diagenesis, even if short lived. These changes are accompanied by In actual experience different diaand Potter (1971). For a parallel treatwidespread lithfficatlon through cegenetic processes vary markedly in ment of diagenesis with a somewhat different approaches the articles by ment precipitation. importance Irorn environment to enLongman (1980) and Harris et al. (1985) The processes and products of metevlronrnent (Table 1). Most sediments on oric diagenesis cannot be applied the modern sea floor are affected to are recommended. Papers treating speequally tQ carbonates ot all ages some degree by bloerosion, but cecific aspects of cementation and neomorphism, petrography and geochembecause ot changes in the biosphere mentat ion is facies specific and most istry, can be found in compilations by (varying skeletal mineralogy and terdeposits are unlithified. Once exposed restrial vegetation) and the atmosphere to percolatinq non-marine fluids, all Schneidermann and Harris (1985) , (varying PC0 2) through geQIQgic time. sediments suffer some dissolution Schroeder and Purser (1986) and James and Choquette (1988). toqether with variable cementation and The rates at which mineral transformaSince the changes that take place are .t ion, dissoiution, early cementation and neomorphism. The effect of organisms largely chemical, the study of carbonate the formation of karst features occur is localized to the soil zone with the
A&-
~
----
--
c c c c c c c c c c c c c c c c
c
c c c c c c c c c c c c c
c c
c C E C C (
(
C C C C
c./ (
Limestones
Introduction
11
C f
c
o r c C
o o o c c o (
,
(
l
6
SELECTED REFERENCES The literature on carbonate diagenesis is vast - the following texts and papers together, give the reader a working overview and introduction to the topic. Amieux, P., 1982, La cathodolominescence: methode d'etude sedimentologie des Carbonates: Bulletins des Centres Ressources Exploration Prod. Elf Aquitaine, v. 6, p. 457-483. An excellent overview of cathodoluminescence - as applied to carbonates, with good photographs. Bathurst, R.G.C., 1975, Carbonate Sediments and Their Diagenesis: Elsevier, New York, 658 p. The authoritative text on the subject of carbonate diagenesis - very readable with exhaustive scholarship. Any worker in the field should have this book. Bathurst, R.G.C., 1980, Lithification of carbonate sediments: Science Progress Oxford, v. 66, p. 451-471. An update on research in limestone diagenesis since 1975. Blatt, H., 1982, Sedimentary Petrology: WH. Freeman, San Francisco, 564 p. A good chapter reviewing carbonate diagenesis. Blatt, H., Middleton, G. and Murray, R., 1980,Origin of Sedimentary Rocks: Prentice-Hail, Englewood Cliffs, NJ, 782 p. One overview chapter on carbonate diagenesis. Bricker, O.P., 1971, Carbonate Cements: Johns Hopkins Press, Baltimore, MD, 376 p. Numerous short articles which together illustrate the spectrum of cements in carbonate rocks, modern and fossil. Flugel, E., 1982, Microfacies Analysis of Limestones: Springer-Verlag, New York, 633 p. Although this is a general text there are good reviews on diagenesis - and an excellent source of information on Mesozoic-Cenozoic carbonates. Folk, R.L., 1965, Some aspects of recrystallization in ancient limestones, in Pray, L.C. and Murray, R.C., eds., Dolomitization and Limestone Diagenesis, a symposium: Society of Economic Paleontologists and Mineralogists, Special Publication, No. 13, p. 14-48. A good introduction to a complex topic. Folk, R.L., 1973, Carbonate petrography in the post-sorbyan age, in Ginsburg, R.N., ed., Evolving Concepts in Sedimentology, Johns Hopkins Press, Baltimore, MD, p. 118-159. A thoughtful and readable essay on the petrography of carbonate diagenesis. Friedman, G.M. and Sanders, J.E., 1978, Principles of Sedimentology: John Wiley, New York, 792 p. A short section on diagenesis.
Friedman, G.M. and Ali, SA, 1981, Diagenesis of Carbonate Rocks: Cementporosity relationships: Society of Economic Paleontologists and Mineralogists, Reprint Series No. 10. A collection of 12 papers on carbonate diagenesis which have appeared in the Journal of Sedimentology Petrology. Fuchtbauer, H., 1969, ed., Lithification of carbonate sediments, parts 1 and 2: Sedimentology, v.12, p.7-159, and p. 163-222. These 2 volumes contain a suite of excellent papers on carbonate cementation. Harris, P.M., Kendall, C.G. St.C. and Lerche, I., 1985, Carbonate cementation-a brief review, in Schneidermann, N. and Harris, P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication No. 36, p. 79-96. A succinct overview of carbonate cements in different settings. Horowitz, A.S. and Potter, P.E., 1971, Introductory Petrography of Fossils: Springer-Verlag, New York, 302 p. Good photomicrographs of wide variety of limestones. Hudson, J.D., 1977, Stable Isotopes and Limestone Lithification: Journal of the Geological Society, v. 133, p. 637-660. An excellent overview and good place to start when consid~ring the application of stable isotope analysis to carbonates. James, N.P. and Choquette, P.W, 1988, Paleokarst: Springer-Verlag, New York, 421 p. A variety of different papers on all aspects of karst and paleokarst, ranging in age from Precambrian to modern. Longman, M.W., 1980, Carbonate diagenetic textures from nearshore diagenetic environments: American Association of Petroleum Geologist, Bulletin, v. 64, p.461-487. A good review of the processes and products from nearsurface diagenetic realms. Machel, H., 1985, Cathodoluminescence in calcite and dolomite and its chemical interpretation: Geoscience Canada, v. 12, p. 139-147. An excellent overview of the chemical considerations that should be taken into account when using cathodoluminescence. Meyers, W.J., 1974, Carbonate cement stratigraphy of the Lake Valley Formation (Mississippian) Sacramento Mountains, New Mexico: Journal of Sedimentary Petrology, v. 44, p. 837-861. A landmark paper introducing the concept of cement stratigraphy and illustrating the use of cathodoluminescence in carbonates.
Milliman, J.D., 1974, Marine carbonates: New York, Springer-Verlag, 375 p. Good review sections on techniques and diagenesis of modern carbonates. Purdy, E.G., 1968, Carbonate Diagenesis: An environment survey: Geologica romana, v. 7, p. 183-228. A beautifully presented, wide-ranging paper on diagenetic environments. Roeder, R.J., 1983, Carbonates: mineralogy and chemistry: Reviews in Mineralogy, v. 11, 394 p. A number of excellent short papers, with superb, pertinent reviews on crystal chemistry (R.J. Roeder and J.A. Speer), Magnesian calcites (F.T. Mackenzie et al.), Aragonite-calcite transition (W.O. Carlson), dissolution and precipitation (J. W. Morse) and trace elements and isotopes (J. Viezer). Roehl, P.O. and Choquette, P.W, 1985, Carbonate Petroleum Reservoirs: Springer-Verlag, New York, 622 p. A series of 35 papers on the geology of carbonate reservoirs, many of which deal extensively with diagenesis. Schneidermann, N. and Harris, P.M., 1985, Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication No. 36, 379 p. A collection of 21 papers on the petrographic and geochemical attributes of cements in all diagenetic environments. Scholle, P.A., 1978, A Color Illustrated Guide to Carbonate Rock Constituents, Textures, Cements and Porosities: American Association of Petroleum Geologists, Memoir 27, 241 p. An excellent reference while doing carbonate petrography with a good bibliography on techniques and classification. Schroeder, J.H. and Purser, B.H., 1986, Reef Diagenesis: Springer- Verlag, New York, 455 p. The biological and chemical aspects of diagenesis in modern and fossil reefs are treated in 21 papers.
Originally published in Geoscience Canada v. 10 Number 4 (December 1983) Revised 1988
12
Diagenesis
Geoscience Canada Reprint Series 4
c c ( c c c c c ( c c c c c c c c c c c c c c c c c c c c c c c c
c e (
Cements fabrics in ooid peloidal grainstone, Stubbs Cay, Turks and Caicos Islands, British Bahamas . Sample is from a Pleistocene exposed surface immediately below a Holocene, eolian, ooid sand dune. Early meniscus cement between grains is blocky, anhedral low magnesium calcite. Later cement, filling cavity, is randomly oriented rhomb ic . needle and bladed low magnesium calcite. Enlargement illustrates the complex arrangement of rhomb units within the later cement fabric. Thesub-structure of the bladed crystals is similar to that of "Iublinite ' ~ Subaerial-vadose environment. Photograph courtesy of A. Oldershaw (University of Calgary) .
c c c c c c
c (
Limestones -
The Sea-Floor Diagenetic Environment _ _
(
c o o ( o o (
o o c o o c o o o o o o o
(
,.
r
LJ-'VZO £:(/./ {n <./ 1'1 ~ b~aA'1~
Limestones - The Sea-Floor Diagenetic Environment Noel P. James Department of Geological Sciences Queen's University Kingston, Ontario K7L 3N6
Philip W. Choquette Department of Geological Sciences University of Colorado Boulder, Colorado 80309-250
J0J
jVVt l->.. J'l (.AI,",~o--
~;;-'2tA.-:.;:.;"
_
knowledge would be a powerful tool when attempt ing to predict trends of reservoir lithologies in subsurface carbonates . Carbonates lying on the sea floor are affected by a variety of chemical and biological processes. Precip itation of CaC0 3 predominates in shallow tropical marine environments, whereas dissolution is im ortant at nig latitudes and in deeB..'lolat~. The most important bIOlogical process is infestation of carbonate hosts by a s~%ctr~ of large ang small boring organisms Because geologists are so preoccupied with lithification, any pore-lining or pore-filling carbonate precipitate is commonly called "cement", whether it acts as a binding agent or fills a pore inside a grain. Since both carbonate genesis and diagenesis occur within the environment of deposition, often under identical conditions, the line between the two is often blurred. In this essay we arbitrarily consider all processes outside direct biomineralization as being diagenetic, even though some processes involve actual particle formation. Sedimentary particles and cements are either calcite or aragonite. The calcite may be enriched in magnesium up to about 18 mole% . Calcites are commonly referred to as calcite or lowmagnesium calcite or LMC_(less.tba0.2 mole% MgC0 3 ) and Mg-calcite or highmagnesium calcite or HMC (12-18 mole% MgC03 ) . While this has proven a useful separation in the past because most shallow water components are one or the other, discovery of precipitates in deeper waters with intermediate amounts of magnesium (2-12 mole% MgC03 ) has made the differentiation somewhat artificial. In the following pages, we first summarize our current understanding of the diagenesis of carbonate sediments on the modern sea floor, and then assess the applicability of these observations to the rock record.
INTRODUCTION Diagenesis commonly begins as soon as carbonate sediments are formed: tests of foraminifers settling onto the deep sea floor gradually dissolve; living bivalves are infested w ith boring sponges; cement precipitates in coral pores just below the living surface. These early changes are in many cases the most critical in the history of a limestone because they both modify depositional facies and direct the pathways of future diagenesis. For example - if crusts of lithified sediment sporadically develop on the surface of a carbonate sand shoal through precipitation of intergranular cement, a new hard substrate suitable for colonization by skeletal invertebrates is created where none existed before. This new community of organisms may in turn build reefs and so, in a very short time , a sand shoal facies has been replaced by a reef facies with no change in the environment except partial lithification ZONES OF SEA-FLOOR of the sediment. The effect of this local cementation will continue as these carDIAGENESIS bonate sands pass through subseThere are two realms of carbonate deposition in the modern ocean :' quent diagenetic environments - the unlithified and porous parts will be the, ( 1 ) shallow water, characterized ....Qy first flushed by new diagenetic fluids;~\l ra idl accumulatin sediments ca. the cemented, non-porous parts will be t 100 cm/1000 years) ~.hich form mainly affected more slowly. Thus, if the rela in warm tro u.o.a a d ~J,! bt ro ical envitive intensity of sea-floor diagenesis ronments; and (2 dee -w !!~ could be tied to depositional facies, this by slowly accumulating sedir:R9Ats-(ca.
~.
LlF" -,.{
~ II:s ) fc:) d r cr« /Q":;' ( 13 k\ c.~ Itr.'J..J' 'VI). ~~ s:- e.g. (r l/VIJ M t Q. ). • J
C~{o~o.
fJ flu ~::... n () k{!-t>"w et~ 10 cm/1000 years) from the fallout of calcareous plankton in the ocean worldwide. Sediments on the slopes of shelves and platforms are mixtures of shallow-waterand pelagic components. Whereas shallow-water deposits accumulate quickly and form impressively thick carbonate bodies, comparatively thin pelagic carbonates , because of their enormous extent covering more than a quarter of the Earth's surface, account for over two-thirds of the carbonate currently being extracted from seawater (Scholle et al., 1983). Current estimates indicate that between 75 and 95 percent of pelagic carbonate is subsequently dissolved (Broecker, 1974) at the sediment-water interface in the deep sea (Adelseck and Berger, 1975). In the perspective of geologictime it is important to recall that carbonate sediments made up of pelagic organisms are largely a Mesozoic and Cenozoic phenomenon and are insignificant in Paleozoic and Precambrian strata. Vertical Zonation. The general patterns of carbonate accumulation have been known for over a century (Murray and Renard, 1891; Murray and Hjort , 1912). The primary controls on carbonate precipitation and dissolut ion are Pea?and wa~r~m peratu re and pres- ~ lJi: sure. Although water pressure in~ses in a linear fashion with depth, water temperature does not. Seawater at low latitudes is temperature stratified and comprises three layers (Figure 1): a warm, homogeneous, near-surface mixed layer;!!J.e thermocline, a zone at rapid temperature decreasEU:l/ith deptb to around 5°C; and a deep-water zone where temperature is more or less constant to abyssal depths. Precise determination of the solubility of calcite and aragonite in seawater has proved to be one of the more elusive quests of low-temperaturegeochemists, and as Morse and Berner (1979, p. 531) point out, "after 100 years of study the major problem of determination of the saturation state of the deep ocean remains largely unresolved ". At present our understanding rests on the often conflicting results of laboratory exper imentation, mapping of bottom sediments, and field observations of particles hung in open containers at various depths in the ocean. These data have most recently been synthesized by Broecker and Takahashi (1978) and Morse and Berner (1979). For a
c 14
summary of carbonate dissolution and precipitation kinetics the reader is referred to the excellent review by Morse (1983). In general, near-surface tropical and subtro pical seawater appears to be supersaturated with re~m.e.Q.L1P-b.Q!b. aragonite and calcite to depths of about 500 metres in the Pacific and 2,000 metres in the Atlantic (Takahashi, 1975). Below this supersaturated water, sediments on the sea floor are exposed to more and more undersaturated water with increasing depth because of increasing pressure and, to a lesser ~temperature. Thus, sediments are subjeCtetl to Ilrster and faster dissolution rates until finall y a depth is reached where the rate of . so u Ion IS equal to the rate of sediment supglv from above. !3elow this depth, c alled the carbonate cO,JJJP.ensationoop th or ~(Figure 1), !ill-QaLcite..isdissolved. In regions where carbonate productivity and resulting sedimentation rates are high, such as those near the equator, the CCD is depressed. Oceanic Q.CD.for.calcite is now at about ~n the Pacific ~~d 5,500 m in the Atlantic. _and is sfiallowesCiii1he
ri9.Ltb..ea~ifjc.ancld~.e.p-estJRi~ii
~nt1c.
Variations of the CCD in the world oceans are discussed in detail by Kennett (1982). Hills that rise from abyssal depths up through the CCD can be pictured as "the snow-capped mountains of the deep sea" (Berger, 1975), their lower slopes below the CCD covered by red clays and siliceous ooze, and their tops above the CCD mantled with white carbonate sediment. Evidence for dissolution can be seen at depths considerably shallower than the CCD. This gives rise to the concept of tb!!JYAo.cUoe (Berger, 1968), or depth where evide[lC_e~of.conside.r.able.(selec tive) dissolution is first .e n c9M.r.tt~ d . The lysocline is in general 1,000 m shallower than the CCD and most dissolution takes place within this 1,000 m interval. Broecker and Takahashi (1978) have demonstrated that to within ± 200 m the depth of the calcite lysocline can be explained in terms of carbonate ion concentration in deep waters: the lysocline lies at greatest depth where waters with high carbonate ion concentrations extend to the bottom. It should be noted that there is often a difference between the "hydrographic
Diagenesis
lysocline", that zone determined byfield experiments in the water column, and the "sedimentary lysocline", marking an increase in the amount of dissolution on the sea floor (Berger, 1975). In fertile regions, near the continental margins in particular, the sedimentary lysocline may rise toward the surface and become diffuse. But what of sediments on the sea floor in the several thousand metres between the shallow saturated waters and the lysocline? Why, if water pressure is the main control , is there not a gradual but steady increase in dissolution with depth, and why do the particles that lie on the sea floor show only scant evidence of corrosion? The answer appears to be a complex interplay between rates of carbonate accumulation, kinds and sizes of sedimentry particles and, perhaps most important, the kinetics of dissolution (Morse and Berner, 1979; Morse, 1983). In addition, orthophosphate, present as strongly adsorbed coatings on grain surfaces, is an inhibitor of calcite and Mg-calcite (but not aragonite) dissolution (Berner and Morse, 1974). T~D is the deRth at whiJ:
"*
TEMPERATURE °c OF SEAWATER 0 0
,0
7
20
Geoscience Canada Reprint Series 4
lysocline and compensation depth are poorly known. A good practical measure for aragonite saturation is the dissolution behaviour of pteropods, small pelagic gastropods with an aragonite shell (Berger, 1978). In marginal seas with anti-estuarine circulation, such as the Mediterranean Sea, Red Sea and Persian Gulf, pteropods are common in sediments at all depths, indicating aragonite saturation of these waters. At low latitudes in the open ocean pteropods disappear below about 1,500 m with depth valuesshoaling toward the higher latitudes. On average, the aragonite com ensfltion..depthTACDTIl:igUl:eJ n~ shallower by3 kilometres than the CCO, llYt this maY~rY. As with the CCO, tile ACD is deeper in the Atlantic than ii1tlie Pacif ic because Atlantic waters are closer to being saturated for calcite and aragon ite due to different overall pattern S-.oLdeep_water_c~. In the northeast Pacific and eastern tropical Pacific aragonite dissolution begins right below the mixed layer (500 m) within an oxygen minimum layer. Zonation with Latitude. There is a much poorer understanding of latitude variations in shallow-water carbonate solubility. Obvious changes in carbonate sediments are complicated or masked by the effects of surface currents, terrigenous sedimentation and climate, as well as by the fact that outside the tropics many of the carbonates "10 CARBONATE IN SEDIMENT
30
0
50
100
0
I
ZONE OF PRECIPITATION
c\,\~'i..
",0
-t....'i..'if.
:n:
:l:
~2
x .....
Q. ILl
Cl 3
DEEP WATER
3
ll: ILl
..... c(
~ 4
5
ZONE OF DISSOLUTION AND PRECIPITATION
m:
ZONE OF ACTIVE DISSOLUTION
:nL
ZONE OF NO CARBONATE
4
5
Figure 1 Generalized diagrams illustrating the relative positions of calcite and aragonite solubilityprofiles in the modern tropical ocean and the variation in temperature with depth. The major zones of diagenesis are plotted to the right.
c
c c
c c c c c c c c c c c c c c c c c c c c c c c c c c
c c
c c e c ( c c c c c c
(
C'
Limestones - The Sea-Floor Diagenetic Environment
15
(
C.
are relict (Emery, 1968; Ginsburg and James, 1975). In general, rapid fixation of biogenic and inorganic carbonate in shallow water appears to be limited tothetropics and subtropics and rarely extends north or south of 30° latitude .Although acornplete spectrum of carbonate skeletons is produced in these environments,they are distinguished by abundant production of corals and green algae, both of which are aragonite . Lees and Buller (1972)have, as a result, designated this the Chlorozoan assemblage to differentiate it from the Foramol assemblage of temperate and polar seas.The principal contributors to the cooler water carbonate assemblages are molluscs, foraminifers and bryozoans with echinoids, barnacles and coralline algae often present. The bulk of these higher latl tude carbonates are Mg-ealcite or calcite with rare aragonite, and in mixedlayer mineralogies there are lower amounts of aragonite in colder climates (Milliman, 1974). Non-skeletal grains (ooids, peloids and lumps) and marine cements do not generally occur in these cooler waters and are restricted to the tropical zone, although Alexandersson (1974) has found aragonite and Mgcalcite cements inside coralline algae nodules in the North Sea where waters are undersaturated, suggesting strong biological influence. Even though rates of carbonate production are much lower in cooler latltudes than in the tropics, carbonates may locally constitute up to 50% of the sediment. Observing the high productivity and lack of terrigenous mater ial in such areas, however, the amount of carbonate is somewhat lower than might be predicted - suggesting removal of some portion (Milliman, 1974; p. 105). It is common to observe the effects of dissolut ion in both calcite and aragonite
o o o o o o o o o o o o o o o o o () o o o o o o o o 6 C
~
. : \...
~'\'\~
NORTH
70
:\'~
50
40
LATITUDE
30
20
10
0
SOUTH
10
20
30
40
50
60
70
mnTTfm~W~w'm7TI77J
\~
lCJ-
60
skeletons from latitudes above 40° . Although the actual removal of carbonate is thought to be mostly by physical abrasion aided by maceration, dissolution "opens up" the grains and aids in these processes. There is no clear evidence of dissolution on a large scale. Thus, in shallow-water environments north andsouth of the tropics conditions seem to be similar to those below the zone of saturation and above the lysocline in the deep tropical ocean. On balance we can roughly differentiate four zones in the modern ocean in which the conditions for carbonate diagenesis are the same (Figures 1 and 2). I. Zone of Precipitation. These are shallow tropical to subtropical settings in which a complete spectrum of skeletal and non-skeletal particles of aragonite, Mg-ealcite andlor calcite occur, as well as aragonite and Mg-calcite cements. The lower limit of this zone where seawater becomes undersaturated with respect to aragonite is about 1,000 m (varying between 500 m and 2,000 m depending on the ocean) in the tropics and shoals to the surface around 30 0N and 30 0S latitude . II. Zone of Dissolution/Precipitation. Shallowtemperatewatershelfsettings north and south of about 30° latitude to the polar regions, which are characterized by wholly biogenic carbonate accumulat ion (mainly Mgcalcite and calcite), no intergranular cementation, and minor dissolution. Deep-water tropical slope and basin settings from the base of Zone I to the calcite lysocline are likewise cnaracterized by local dissolution of allochthonous aragonite and ?Mg-calcite components, neomorphism and precipitation of calcite (LMC) cements. III. Zone of Active Dissolution. Deepsea settings between the calc ite lysocline and the CCD which may rise to
[7'
:x:
l-
n. w
o
4 -
5 -
ZONE
n
6 - ' - - - - - - -- - - - - - - - - - - - - -- - - '
Figure 2 Variations in the different zones of sea-floor diagenesis in the modern ocean.
near the surface in polar seas, are distinguished by wholly biogenic calcite accumulation showing abundant evidence of corrosion. IV. Zone of No Carbonate. Wholly deep-sea settings, below the CCD where no carbonate is accumulating. It should be stressed that this is a generalized zonation and appl ies chiefly to the open ocean and to associated shelf-slope and platform-slope systems. The major part of the following discussion on sea-floor diagenesis is focussed on Zones 1 and II. ZONE I: WARM SHALLOW WATER PLATFORM CARBONATES
Mineralogy of Precipitates. Sedimentary particles on \he modern shallow sea floor in tropical regionsconsist of aragonite or calcite or magnesium-enriched calcite (Mg-calcite). Although some organisms precipitate skeletons of calcite , only Mg-calciteand aragonite appear to precipitate directly from seawater here. It is still not clear what control the mineralogy and crystal habit of these minerals or why both Mg-calcite and aragonite are so common (see Bathurst, 1980,for summary). The complexities involved as well as the differences between experimental results and natural observations are discussed and debated by Given and Wilkinson (1985a,b) and Morse (1985). From a thermodynamic standpoint, calcite with from 2 to 7 mole% MgC03 is the preferred stable phase in seawater(Berner, 1975). It has been suggested that the presence of strongly hydrated Mg2+ ions absorbed onto the crystal surface (the Mg/Ca ratio in seawater is ca. 511) inhibits the growth of calcite, and that the Gibbs free energy needed to dehydrate the ions and form aragonite is lower (Bathurst, 1968). On the other hand, it is also known that calcite crystals may precipitate and absorb large quantities of Mg2+ as they grow (Berner, 1966). The Mg-calcites so formed are in non-exchange kinetic equilibrium with seawater. The main conclusion of much experimental work and observat ion of natural systems appears to be, as summarized by Mackenzie et al. (1983) and echoed by Given and Wilkinson (1985a,b), that the MgC0 3 contents of magnesium calcite precipitated from seawater, including organism skeletons and cements,
*
Diagenesis
16
are related to temperature and CO~' concentration of the water. The amount of MgC0 3 in the calcites increases with: (1) increasing temperature (Burton and Walter, 1987), (2) increasing carbonate ion concentration (Given and Wilkinson, 1985) and (3) increasing Mg/Ca ratio of the fluid (Mucci and Morse, 1983). It is also clear that the rates of fluid flow are important. Mg/Ca ratios do not seem to be sufficently variable in the modern ocean to account for observed trends in mineralogy. Since temperature and carbonate ion concentration are covarient in the ocean it is difficult to determine which is the most critical factor. The importance of temperature has been stressed by Burton and Walter (1987) who confirm that the growth rates of aragonite versus calcite increase with increasing temperature. Aragonite has about the same solubility as Mgcalcite with ca . 12 mole% MgC03 (WaIter and Morse, 1984; Walter, 1985). Seawater has been described as "a dilute electrolytic organ ic soup" so that prediction of any reactions that lead to precipitation must also take into account the large amount of organic material. That organics affect precipitation is clear, but their role is ambiguous (Mitterer and Cunningham, 1983). Inthe first place, all particles are coated with layers of organic molecules which inhibit nucleation, allowing seawater to maintain supersaturation (Suess, 1970, 1973). At the same time, organic matter extracted from ooids is similar in composition to that found in skeletons (Mitterer, 1971). These proteins and humic substances contain carboxyl groups with similar geometry and charge as the carbonate anion. Thus, there may be substitution by the anion for the organic group, and so crystals may be nucleated and grow by epitaxy. The presence of more 13C in marine cements than expected if they were completely inorganic precipitates (Milliman and Muller, 1977) also points to some sort of involvement by organics. ~ Although in general Mg-calcite appears to be the most widespread marine cement (James and Ginsburg, 1979), aragonite is common in waters of slightly elevated temperature and salinity. Finally, in ooids it appears that the degree of seawater agitation mayfavour the precipitation of one mineral over another (Davies et sl., 1978).
Petrography of Precipitates. Magnesium calcite (Figure 3). Magnesium-enriched calcite precipitates generally contain a range of between 12 and 19 mole% MgC03. The crystals (trigonal CaC0 3) precipitate either as tiny rhombs less than 4 micrometres across (micrite-size) or as fibres a few tens of micrometres long. Although iron can be accomodated in the calcite crystal lattice, microprobe analysis of these cements reveals no detectable Fe2 + . Probably th is is because there is very little iron in seawater and most precipitation takes place under oxidizing conditions where any iron present is the Fe3 + state, and the partition coefficient is close to one (Richter and Fuchtbauer, 1978). Because strontium with its relatively large ionic radius fits uneasily into the calcite lattice, Mgcalcite rarely contains contains more than a thousand ppm Sr2 + (Kinsman, 1969). Micrite-size crystals. This is the most common type of Mg-calcite cement occurring in awide variety of sediments . It ranges from thin rinds only tens of micrometres thick around particles and algal filaments to complete fillings of pore spaces. Although micrite is easily seen as cement when encircling particles or lining ooids, it is difficult to differentiate from lime mud if the pores are
c c c c c c c ( c c c c
Geoscience Canada Reprint Series 4
entirely filled. Thus, the addition of Mgcalcite micrite cement may, in the process of lithification, change the texture of a grainy sediment from a grainstone to packstone . Of interest here is the ubiquitous association of micro-peloids, ranging in sizefrom 20 to 100 mlcrornetres on average, with micrite cement. This is the structure grumuleuse of Cayeux (1935), a fabric of numerous clots or vague small peloids (aggregates of Mg-calcite rhombs), surrounded by micr ite and microspar. While these may well be sedimentary particles of uncertain origin, the persistent association has led Macintyre (1983) and Marshall (1985) to suggest that the peloids are the resutrof sp-ontaneous nucleation and preciRita·tion-ofMg-calcite in the water column Which- men settle out of suspension. Ttiis precipitation may be eIther i:lirector organically induced. Chafetz (1986) has found that the finely crystalline nucleus contains rods of bacteria , suggesting that the crystallites originated as a precipitatewithin and around active clumps of bacteria and that the vital activity of the bacteria influenced precipitation of the calcite . Fibrous crystals. This cement (Figure 3) is generally found as isopachous encrustations up to several hundred micrometres thick.The elongatecrystals
(
c c C
r
C C C C C C C C C C C C C C C C
SHALLOW MARINE CEMENTS MAGNESIUM CALCITE
ARAGONITE
MICRITE
Ye.,{o·, ~ ~
IlIl
11 \~ (
Ctl l-WV"·
OOp/)
MESH OF NEEDLES
c e
000"
c c
r'\
FIBROUS TO BLADED RINDS Figure 3
BOTRYOIDAL 10
-.r~( erA (D
.£'7 1-(e.' <. /)/c<-, /-
Different types of modern sea-troor carbonate cements. I 1t:lV\.L J rd.r
J~)
o
U,-"tn.f
""8
~ r:-r~.(A Ci..- 1_n L 1'oW)
C--l/ '
t.( V'
/'<-.0 II,
,
~ II-./'
/h-CJUl tJ,"""
'7 f' ·o /11,R
.I ~/
I
~
C
C
c
l (
c (
o o c o o o o o o o o o o c o o o
Limestones - The Sea-Floor Diagenetic Environment
flO C are either nucleated picket-fence style on the substrate or located at intervals, and grow as small spherulites which quickly interfere with one another as they develop. The spherulites are often arranged in tiers in the cement rind. Blocky or equant crystals. These small crystals, ranging in size from 20 to 50 micrometres, are not common as shallow marine precipitates, but do occur locally in both reefs and hardgrounds. Epitaxial crystals. Although particles of echinoderm skeletons, composed generally of a single crystal of Mgcalcite, are common elements in many modern carbonate sediments, the development of marine epitaxial Mgcalcite cement on these grains has not yet been reported. The closest occurrence is the presence of epitaxial crystallites developed on cores of echinoderms in ooid cortices (Marshall and Davies, 1975). Aragonite (Figure 3). Aragonite (orthorhombic CaCOa ) precipitates are either microcr ystalline or needles a few micrometres or tens of~icrometres long. Loreau (1982) separates these nannocrystals, from largest to smallest, into needles (generally cements), baguettes (generally in the ooid cortex) and nannograins, with increasing turbulence favouring needle formation and
o o
o o o
increasin g organics favouring nannoJ1,rains. The cements are reviewed and illustrated in a recent article by Sandberg (1983). Strontium isthe majortrace!J element in a'ragonite whose orthorhombic structure accomodates it easily and cement crystals usually carry about 8,000 to 9,000 ppm, as do most aragonite skeletons (Kinsman, 1969; Bathurst, 1975). The major exceptions are aragonite molluscs which contain between 4,000 and 1,000 ppm Sr2+ , The needles occur in a variety of different habits (Figure 3): (1) Isopachous rinds, particularly common in ooid and skeletal sands and as epitaxial growths on aragonite substrates such as corals, green algae and gastropods. (2) An intergranularmesh of crystals nucleated randomly on grain surfaces and growing in a variety of directions into a void until the void is completely filled by an irregular growth of crystals or crystallites. (3) Botryoids, which range in diameter from tens of micrometres to centimetres, sometimes developed in coarse carbonate sands but most spectacular in reef cavities. Isotopic Composition of Precipitates. When CaCOa is precipitated from seawater - whether in calcareous skeletons, as non-skeletal grains such as ooids, or as marine cements - it acquires characteristic chemical "signa-
+6
tures" in the form of 0180/160 and
~otoPic ratios as well as minor
cation (e.g., Sr2+, Mg2+, Na +) concentrations (Viezer, 1983). Many analyses have been done on carbonate sediments and on many of their more common constituents, such as ooids, calcareous green and red algae, corals, brachiopods, molluscs, lime mud, etc. (Figure 4). Milliman (1974), Bathurst (1975) and Morrison and Brand (1986) have summarized many of these isotopic data on Recent marine carbonate sediments including the pioneering work of Gross (1964). Though much of the spectrum of marine cements and cementation processes is poorly understood, enough isotopic data has been gathered to suggest some general relationships. ~ know most about cements being precipitated in modern sediments and reefs at low latitudes, in warm shallow-shelf and platform setti nqs. These cements have relatively heavy, 180-enriched oxygen-isotopic compositions of 0 to + 3%0 (Gross, 1964; Shinn, 1969; Milliman, 1974; James and Ginsburg, 1979), and their 13CS are heavier (ca. + 2 to + 5%0) than would be expected had they been precipitated in isotopic equilibrium with seawater ( + 2.0 to + 2.5%0 at 25°C; Lloyd, 1971). Specifically (Figure 5), relative to calcite equilibrium compositions, aragonite cements are enriched by 1.0to 1.4%0 013C and 1.5to 2,0%0 018 0 while magnesium calcites ,------,---,----,---,-----,
..
ARAGONITE
+4 o
-e o o
o
~ +2
0'
oCD
Seds.
z
a: cl
u
OI--------------i----------i~.-:.:--1-~~-_t
...---!
SEDIMENTS 8 SKELETONS, Recent, Shallow Warm Water
r:
8
17
-2
II \ \
/
,' ....... _,,'"
MAGNESIUM CALCITE
// / '
.,
,,///
/
-2
.,
-, OXYGEN
Figure 4 Cross-plot of carbon versus oxygen isotopic compositions for recent sediments and component particles, and for buried chalks. PDB-1is the isotopic standard in this diagram. Data are from Gross (1964), Anderson and Schneidermann (1973), Milliman and Muller (1977), Schlager and James (1978) and Rao and Green (1982). V fa..e n /L, (.
--
(
~ fit. (
I~
t..~tI.,!VV!J.:
'2
+3
('M)
Figure 5 Isotopic composition of aragonite and Mg-calcite cements in modern reefs from Bermuda, Enewetak, Bikini and Belize.(Modi~ fi d from Gonzalez and Lohmann, 1985). (<(
~.~
'df~
C O\.~·
Ill"~5yfl--f ..
C) It~ ~'/i Q M' .... ~6.J::O;......,.-.-r-rN~"""r,--.('PIJ-,~0 {h( y":;r'l 1 ,$'1 LI ~ t I
X/pV:"'J
i "0
I
of,
eQ
Diagenesis
18
ZONE I
• :
' :_ .0.
to
cipitates lining chambers and other pores inside skeletal elements (Alex· andersson, 1978; James et al., 1976) and as precipitates in micro- and macroborings. Thus, it is not unusual in the fossil record to find grains exhibiting good evidence of internal ssa -noor CaC03 precipitation, yet showing little or no indication that they were actually cemented together in the sea-floor environment. Between Particles. This type of precipitation is the least common in the sea-floor environment but the most important geologically because it is "rock-making" cement.Thuslimestones may be formed soon (tens to thousands ofyears)after deposition ifconditionsare right in the environmentof accumulation (Figure 6). The habitats of these intergranular cements are varied. (1) Platforms. The most desirable conditions for interparticle cementation appear to be water turbulence and sta· bilized sediment. Both conditions are met on the insides of reefs, especially at or near the platform margin (Ginsburg and Schroeder, 1973; James et al., 1976; Macintyre , 1977 ; Aissaoui et al., 1986) or on the windward sides of platform reefs (Marshall, 1986). Likewise, sand shoals near the platform edge are likely locales, but here most lithification takes place when the sediments are at rest inside the sand shoals or bound by micro-organisms in sheets (Dravis, 1979) or stromatolites (Dill et al., 1986). In contrast, muddy lagoonal sediments are rarely Iithified, probably because of sluggish water movement, continuous bioturbation and general reducing
Loreau, 1982), although Mg-calcite ooids (Marshall and Davies, 1975) and even bimineralic ooids (Land et al., 1979) do form locally. Although there is compelling evidence that the lime mud found in many shallow marine settings IS mostly biogenic in origin (Stockman et aI., 1967; Weumann and Land, 1975), some may be due to direct precip itati on. The clouds of aragonite crystallites or "whitings" seen in many shallow tropical EWvironments are commonly thou ght to be mud derived from benthic sources stirred up by feeding fiSh - yet in some cases they may be products of spontaneous precipitation. In the Persian Gulf, for example, there are few aragonitic ~Igae~ yet numerous whitin gs and mu ~ mud , and many fish . While Loreau (1982) suggests, on the basis of Sr2+ and Mg2+ contents, that both Persian Gulf and Bahamian aragonite needle" muds are likely inorganic precipitates , Ellis and Milliman (1985) found thattlTesuspended sediments in the Persian Gulf and..Bed-Sea...ar:.ELbl..ogenic and detrital. While most readers will agree that we are stretching the concept of sea-floor diagenesis by including these sediments, we have done so to place the whole topic in context. Style of Cementation. Inside grains and skeletons. This type of precipitate, found inside particles and/or skeletons, is the most widespread CaC0 3 "cement", occurring even in grains still rolling around on the sea floor. The cements occur between silt-size particles inside peloids, as pre-
(average 16 mole% MgC0 3) are enriched by 0 .3 to 0.4%0 <5 13C and 1.8 to 1.9%0 <5 180 (Gonzalez and Lohmann, 1985). Because of the consistent fractionation relative to predicted compositions it is clear that the cements do not represent equilibrium precipitates. Since the aragonitic thalli of calcareous green algae and the Mg-calcite of micrite rims produced by endolithic algae also have heavy <5 13Cs, dominantly of + 2 to + 6%0, several workers have proposed that these groups of algae are " involved" if not directly responsible for the origin of marine CaC0 3 cements (Lloyd, 1971; Schroeder, 1972; Moore, 1973; Milliman and Muller, 1974). Biologically mediated precipitation should result in isotopically depleted carbonates due to incorporation of metabolic CO2(Viser, 1983), yet cements depleted in 13C and 180 are not known from shallow marine environments. I/t...- In any event, marine cements in mod~ern warm/shallow environments can be clearly differentiated isotopically from Late Cenozoic meteoric vadose to shallow-phreatic cements, and the evidence is accumulating that these differences survive, though often much reduced, in the face of complex diagenetic histories spanning tens to hundreds of millions of years. Precipitated Particles. Perhaps the most important precipitates in terms of the amount of CaC0 3 removed from shallow seawater are ooids. They may have a cortex of tangentially or radially oriented crystallites. In the modern ocean the overwhelming majority are aragonite (Simone, 1980;
100'S
Geoscience Canada Reprint Series 4
; : : :. ;:
:
•
• •
•
~
. .. .
'0
.'1. ';,.
.....
. ..:,.....
ZONE
.'
Lysocl ine-
~.
, 0,
. ," , ', : . ,
:', ' :
-
CCD
no DEEP SEA Carbonate
I~~~~:::='ii=-~)~ Figure 6
extenSive hardoroundS
or
hardgrounds In upper
pi '" of sand and limestone blocks
port
--
ZONE IS[
•• • ••
Carbonate ....... ..;.' Sediment '" : "" "
ZONE ill -
SHALLDW MARGIN
DEEP BASIN
L-::.'-='-'-' sANDSHQAls
]I
-.-.. '.
.;:.•..: v-
1000'S of meters
,,',
~
;-'
hardQrounds
REEFS extenS ive cemen tat ion
IPLATFORM or
LAGOON
I
SAND SHEETS - sporadic lithification or extensive hardgrounds
SANDYBEACH beachrock
REEFS- sporadIC lithIfIcatIon MUDDY LAGOON -
no cementation
c c c c c c c c c c e
MUDDYTIDAL FLAT crusts, dolomIte
REEF MOUNDS
~
c c c c c c c c c c c c c c c c c c cc c c c c
eJlU!ns iwe
cementation
neomorphism
The locations of sea-floorprecipitation and neomorphism on a shallow carbonate platform and in adjacentdeep-water settings. In a/l . of these habitats, most sediments are unlithified.
c c l
('.
c. (
G
o o (
c c
c
r:
Limestones - The Sea-Floor Diagenet ic Environment
19
conditions . Strandline sands also are commonly cemented but this may notbe entirely due to reaction with seawater. Even though the majority of sediments on shallow carbonate platforms are not cemented, when viewed in total there appears to be a facies preference for precipitation. While this generalization may hold true for muddy shelves and platforms, cemented crusts or hardgrounds (Figure 7) are common in extensive sand sheets on parts of the Bahama Banks (Taft et a/., 1968), wide gently dipping shelves such as off the Trucial Coast in the Persian Gulf (Shinn, 1969),and partially enclosed embayments such as Shark Bay, Australia (Read, 1974). Lithification is most intense at the surface and decreases in intensity downward, generally ceasing within tens of centimetres below the sediment-water interface. As Wilkinson et al. (1982) point out in a discussion of fossil hardgrounds , the processes of lithification may be quite different between platform-margin reefs which have large Figure 7 An irregular submarine hardground (between 129 and 128 em on the tape) cavities and are subject to strong curdeveloped in Lower Cambrian ooid limestones of the Forteau Fromation, western Newfoundrents or waves moving large volumes of land . The photomicrograph illustrates the truncated ooids and cements at the hardground normal seawater, and non-reefal sands surface (arrow indicates location; scale bar 0.5 mm). where current velocities are reduced, pore volumes small and communication with surrounding waters limited. (2) Shallow enclosed basin. In relatively shallow basins such as the eastern Mediterranean and the Red Sea, where there is anti-estuarine circulation and warm waters extend to the seafloor, basin-floor sediments are locally cemented by Mg-calcite and occasionally aragonite to form crusts (Gevirtz and Friedman, 1969; Milliman and Muller, 1977). (3) Slope and deep basin. The oceanfacing slopes of shelves and platforms illustrate a range of Iithified sediments. The top of the slope, which generally comprises deep portions of the marginal reef and forereef, locally exhibits extensive cementation (James and ~ G i n s b u rg , 1979;Land and Moore, 1980). gementation seems to die opt with increasing water depth so that below, 300 m OI-SO ha.J:dgrQullds-give way first to nodules alliLt b~ILto unlithified ~e dime ot (James and Schlager, pers. Figure 8 A debr is flow from the Ordovician part of the Cow Head Group , a carbonate slope observ. , 1979; Mullins et al., 1980). deposit in western Newfoundland, illustrating the effects of multi-generation sea-floor cemenThese partly to completely lithified tation. The tabular clasts are fragmented thin-bedded lime mudstones that were lithitied.on the slope sediments are commonly subject slope before being eroded and incorporated into the flow. The boulders of tabular clast breccia to failure and become clasts in gravity to the left of the hammer and at upper right were debris flow deposits lithified on the sea floor flow deposits (Figure 8). before being incorporated into yet another flow. Photograph, M. Coniglio.
( Diagenesis
20
(
Geoscience Canada Reprint Series 4
( Hardgrounds are common off the western Bahama Banks where the sea floor is swept by the Gulf Stream. In depths of 600 00 or so, Iithified mounds (lithoherms) formed by numerous superimposed crusts and populated by deep-sea ahermatypic corals and crinoids rise 50 m or more above the bottom (Neuman net al., 1977). The cement in these structures is Mg-calcite. Below these depths (Figure 6) is the transition zone where water temperatures decrease rapidly in the thermocline and waters at the sea floor pass from Zone I (precipitation) to Zone II (dissolution and precipitation).
Neomorphism. On the shallow tropical sea floor, when aragonite and Mg-calcite are precipitated they appear to remain I r.ge.ly unchange untl a fected by waters of a radicallyaTferen character. Nevertheless, there are some reports of recrystallization: Mg-calcite foraminifers and coralline algae altering to aragonite (Murray, 1966; Kendall and Skipwith. 1969; Purdy, 1968); aragonite cements partially replacing aragonitic mollusc shells (Shinn, 1969); parts of aragonitic cements and skeletons changing to Mgcalcite (Alexandersson, 1972; Scherer, 1974). These occurrences seem to be rare but do point out again the subtletie-S. governing prec iRit at ion of different cac 0 3 phases and tell us that the posSibi lities fo r ne-o.mar.pfiiSii'CCti5-ex.lSf; although they ma~ not be co m!J.lQD.J.o. today's sbaIlQW.S.e.as._ ZONE II: TEMPERATE WATER SHELVES, TROPICAL CARBONATE SLOPES Perhaps the single most intriguing advance in our understanding of seafloor diagenesis since this article was first written (James and Choquette. 1983) has been the confirmation that fabric-selective dissolution, calcite cement precipitation and calcitization of metastable carbonates do take place in some cool Qcean water 89tliR§s. Coolwater environments occur in two locales; (1) in shallow temperate environments where carbonate sediments are dominated by the foramol assemblage (Mg-calcite and calcite); (2) in deepwater, along the margins ottropical carbonate platforms and on the tloors of intervening basins where shallow water sediments of the chlorozoan assemblage (araqonite and Mg-calcite)
have been redeposited as sediment gravity flows. Cementation in shallowwater temperate carbonates appears to be limited to prec ipitation within lntraparticle pores (Rao, 1981) and there are to date no reports of sea-floor neomorphism.ln contrast, in deep-water tropical settings where the sea floor lies within the thermocline or just below it, waters are close to or below aragon ite satu ration but supersaturated with respect to calcite. Thus, a situation is created similar to that in meteoric environments (see James and Choquette, this volume) in which the metastable carbonates have a high "diagenetic potential" to alter to calcite through dissolution and preclpitation . There are relatively few documented cases of this diagenesis, because of the great difficulty in studying such sediments in waters 100s to 1000s of metres deep. (1) Holocene fore-reef limestones off Jamaica appear to be undergoing local neomorphism (Land and Moore, 1980). (2) Pleistocene periplatform ooze in the Bahamas is lithified by dissolution and reprecipitation when exposed for long periods as vertical erosional valley walls in isolated troughs (Schlager and James, 1978) or buried in the shallow subsurface (less than 30 00) on ocean facing slopes (Mullins et et., 1985) . Metastable components in carbonate turbidites are also altered in a similar fashion (Gomberg and Bonatti, 1970). (3) Relict sediments on seamounts are altered when exposed and winnowed for IQngperiods (Milliman, 1966). (4) Cretaceous to Miocene shallow water limestones which have subsided into deep water by vertical platform accretion suffer dlssolutlon and calcite precipitation (Saller, 1985; FreemanLynts et aJ. , 1986).
Di~owtron.
j
Aragonite appears to be the only mi eral undergoing dlssolutton, resulting i partial to complete fossil molds.
Precipitation. All of the precipitates are calcite. In deep water Bahamian muds the crystals are blocky calcite, 2-4 micrometres in size, as cement QvergrQwing coccoliths, as druse in foraminifera chambers, calcispheres, and the molds of pteropod shells. Calcite also precipitates as epitaxial overgrowths on echinoid particles. In MiQcenereef carbonates intergranular pores, sponqe borings and aragQnitefossil molds are IQcally lined with crusts, 0.05 to 2.0 moo thick, of
C
equant, bladed and radiaxial fibrous calcite, both clear and inclusion rich. In Cretaceous limestones the cements are medium to coarsely crystalline (100-300 micrometres) equant spar. These calcites contain low to intermediate amounts of magnesium , in accordance with the experimental results of Fuchtbauer and Hard ie (1976) and BurtonandWalter(1986)which indi~e that calcites precipitated from waters of decreasing tem r ontarnpro essrve y less magnesium. The l inely crys talffiie calcite cements in Bahamian peri platform ooze contain 3-5 mole% MgC03 while the radiaxia calcite cements from Enewetak contain 5.0-11.6 mole% MgC03.
Neomorphism. While some aragonite pteropods are dissolved, others are partially miCj ritized. There is also selective alteration of Mg-calcite allochems, e.g., some foraminifera illustrate crystal enlargement.
r
C C C (
C
C C
I
Isotopic Compositions. While the 0'3C values for lithified Bahamian peri-platform ooze in the Tongue of the Ocean are roughly similar to the values from shallow water precipitates, the heavier oleo values clearly ind icate precipitation from cold ocean waters (Figure 4). The isotopic record of the radiaxial and equant calcites is, however, unclear. These are clearly marine precipitates but their isotopic compositions do not confirm precipitation from cold waters. i The importance of this deeper water neomorphism in the modern ocean is difficult to ~~ at present. It is localized to areas hich receive metastable sediments from platforms. In most of the deep ocean, where sediments are composed of calcite planktonic organisms, there is little "diagenetic potential". In the Paleozoic when there were virtually no planktonics and all deep-water carbonate sediments including lime mud came episodically from nearby platforms or shelves, the times between sedimentation events equid have been periods of recrystallization and cementation by calcite on the sea floor.
C C C C C C
C C C
C C C C C
C C C C C C (
c e "(
ZONES III AND IV: DEEP BASINAL ENVIRONMENTS For deep-water environments, numerous studies confirm that very little dissolution occurs as particles settle
c
c
(
c
c (
o C' o o o (
(
c (
c
,.
,..
or:
21
Limestones - The Sea-Floor Diagenetic Environment
through the water column; most carbonate is dissolved at the sediment-water interface. If the sea floor is overlain by welloxygenated water, as is the case in most modern oceans, then burrowing allows mixing of interstitial and overlying waters, and thus dissolution may take place some centimetres below the sediment surface. If euxini c ~nditions prevail and the ' \ ~verlying waters are CO2 rich, corrosion may take place on the sediment surface. Dissolution ceases once the carbonate is buried just below the surface, because an increase in alkalinity within the sediment is brought about by bacterial sulphate reduction (Berner, 1971). Carbonates may not suffer dissolution in euxinic basins or on slopes where the oxygenminimum zone impinges on the sea floor. Although corrosion surfaces have been reported in limestones of basins interpreted as rather restricted (e.g., Lindstrom, 1979), signs of dissolution beneath such surfaces appear to be extremely rare. In fact, under the right conditions carbonate precipitation will occur, the chemistry of which is discussed by Hesse (this volume). Summary. In summary, there is a clear zonation in diagenesis along the margins of tropical carbonate platforms facing deep water (Figure 6); parts of shallow water reefs and sand shoals are commonly lithified early by aragonite and Mg-calcite, as are the adjacent fore-reef sediments (Zone I); .lithification decreases down .§lQp.e, as the waters become cooler, grading from pavements to nodules to little cementation; in cold wate..!§., which are undersaturated with resRectto aragonite, dissolution of the metastable sediments and precipitation of calcite, resulting in cementation, may occur (Zone II); in very cold deep waters most (Zone III) or all (Zone IV)ofthe carbonate is dissolved.
::::J.t-
BIOLOGICAL ALTERATION Even though neomorphism sensu strictu is not a common process in shallow-water carbonates, sedimentary particles and rocks do undergo drastic alteration to micrite or mudstone through a combination of biological and chemical processes. The mechanism involves infestation of the carbonate host by one of a variety of boring or endolithic organisms and, after their
death, filling of their holes by fine sedi(Zankl and Schroeder, 1972;James and Ginsburg, 1979; Land and Moore, 1980). ment and/or cement. The end result, however, is the same: a Particle Alteration. On a microscale this process involves skeletal-rich rock is transformed by sevboring by endolithic algae or fungi into eral generations of alteration into a sand-sized particles. Algae, mostly muddy limestone and much of the ori9.!:.. blue-green and green, appear to be JiaL texture is lost. most effective in the upper part of the photic zone to depths of about 70 m STRANDLINE DIAGENESIS Diagenetic processes that characterize (Budd and Perkins, 1980). Fungi extend to depths of 500 m and more, whereas the sbaU~GfiiiOllQor iiIso preheterotrophic algae and bacteria occur dominate in the intertidal and supratidal to abyssal depths (Friedman et a/., zones.gHere the environment IS more complex: (1) instead of a simple rock1971). Infestation appears to be most water system the setting is three-phase, active (Kobluk and Risk, 1977) when grains are at rest in lagoonal or other onrock-water-air; (2) freshwater flow from land extends seaward out underneath shelf settings. the shore, with the water table at or just In shallow tropical environments below the strandline sediment surface; where precipitation is dominant (Zone I) and (3) the waters themselves are comthe vacated holes of endoliths are filled monly mixtures of marine and meteorip by Mg-calcite and/or aragonite cement fluids. __ £r() /),/.1- , W/VJ t::j"w47 (Margolis and Rex, 1971; AlexandersBeaches. ~ 10. (er.R.cll1~{ son, 1978).If the numerous generations of boring endoliths are restricted to the The most com ffl6~ifestation of grain margin, a rind of microcrystalline strandline precipitation is beachrocklayers of cemented beach calcarenite Mg-calcite or aragonite called a "micrite envelope" (Bathurst, 1966) develops. (Figure 9) that dip seaward at the same attitude as beach sediments and are The formation of these envelopes is composed of the same grains as the often critical to grain preservation in beach sand, which can be anything from later diagenetic environments. carbonate to quartz to volcanic lithoIf infestation is intense and prolonged clasts; good reviews are to be found in the entire grain may be transformed to microcrystalline Mg-calcite or aragStoddart and Cann (1965)and Milliman (1974). Some tropical. intertidal zones onite, with little trace left of the original are all beachrock, others have only scatstructure (Kendall and Skipwith, 1969). tered layers, and many are without it. This pervasive micritization is parStrata generally are localized to the ticularly common in grapestone lumps intertidal zone extending up into the (Bathurst, 1975) and in stromatolites zone of wave splash and disappearing (Logan, 1974). These "diagenetic pelseaward, landward, and with depth into oids" are in some instances difficult if unconsolidated sediment. The beachnot impossible to distinguish from fecal rock may be present as scattered nodpellets or Mg-calcite cement aggreules, isolated slab-like layers, or whole gates. As a result, it should be stressed thicknesses of beach sediment, generthat the term "peloid" applies to grains ally developing a few centimetres below that may have had very different origins the sediment surface. but look the same. The surface of the beachrock is comOutside the tropics and in deeper water monly pitted and bored and exhibits dis(Zone II)infestation is often just as intense solution basins or potholes. In crossbut the holes are not filled because of the section beachrock is well laminated and lack of precipitation and so the grains are the base of each layer is sharp. The rock gradually broken down into smaller partiis characteristically jointed into blocks cles (Alexandersson, 1979). and textured by cracks and channels. Limestone Alteration. Erosion of rock slabs and reworking durVoring also occurs on a macroscale with ing intense storms leads to conglomer~ 'sponges, bivalves and polychaete ates or ramparts of beachrock clasts. worms being the excavating organisms. The grain-size of the sediment ranges Particularly common in reefs and on from sand to boulders. hardground surfaces, the cavities creInduration varies from loosely ceated by these larger excavating inverteQrates may be filled with fine sand- to mented sand that can be disaggresilt-sized sedirrumt,and/or .ceme.riC:. gated by rubbing to well-lithified rock
--
-
22
that can be broken only with a hammer. Commonly the surface is hard and the inside soft. Cements are aragonite or Mg-calcite with the same spectrum of fabrics as on the sea floor, but microcrystalline aragonite cement is particularly common (see numerous articles in Bricker, 1971). In some localities these cements have a pendulous or mini-stalactite orientation, signifying vadose precipitation in an air-filled void. Beachrock forms at surprisingly rapid rates on the order of years to tens of years, so that today coins, bottles, skeletons and other artifacts are commonly found in beachrock. Most workers now seem to agree that precipitation of beachrock cement is from seawater by evaporation and CO 2 degassing in the intertidal zone during low tide (Bricker, 1971; Meyers, 1986), although the composition of the water may be modified greatly by meteoric mixing (Moore, 1983). Other alternatives are by degassing of CO 2-rich carbonate-saturated groundwaters which have migrated into shore areas (Hanor, 1978), or by biologically induced precipitation. Beachrock is particularly susceptible to alteration by biological alteration (James and Choquette, 1984). Aragonite crusts. Aragonite crusts up to 20 mm thick and called "coniatites" (Purser and Loreau, 1973) often coat beachrock, carbonate grains and exposed bedrock in the splash zone. These coatings, which look like grey to cream-coloured enamel paint, range from smooth encrustations to porous surfaces that resemble tufa or travertine. Beneath overhangs or in beachrock they have dripstone morphologies. These crusts appear to develop only when the substrate is impermeable. In the fossil record such crusts could easily be mistaken for ~ea-f1oor precipitates. ~ \Vv 1J'I'N~ V"'oj \ ~ Tidal Flats. \ Diagenesis of carbonate sediments on wide, muddy tidal flats is discussed in part in other articles in Geoscience Canada (James, 1984; Kendall, 1984; Morrow, this volume). The diagenetic processes are complex and involve not only CaC0 3 precipitation but also the formation of dolomite and evaporite minerals. In terms of limestone formation in particular, the same processes seen on
Diagenesis
beaches predominate, but the rate and intensity of peritidal diagenesis appear to be functions of climate . On humid tidal flats, as typified by the Bahamas, cemented crusts are relatively rare. Only a few centimetres thick (Figure 10), they are found just above the normal high-tide mark. In the inland marshes crusts are composed of algal filament molds encrusted with Mg-cal-
Geoscience Canada Reprint Series 4
cite. In the more seaward channel belt, crusts are peloidal sediment without filaments cemented by Mg-calcite micrite (and often containing dolomite) in which cracks and fenestrae are filled with acicular to bladed aragonite (Hardie, 1977). On arid tidal flats, such as those along the Trucial Coast of the Persian Gulf (Taylorand lIIing, 1969)and around
Figure 9 A group of geologists walking on the black, algal-infested surface of beachrock composed of ooid and grapestone particles cemented by aragonite on the shore of Cockroach Cay, Berry Islands, Bahamas. The white sand near the trees (right) and the sediments beneath the water (left) are unlithified.
c c c c c c c c
c c c c c c ( c c c: c c c c c c c c c c c c c c c
e (
Figure 10 Brecciated crusts composed of peloids and algal tubules, cemented by Mgcalcite and containing minor dolomite in the supratidal zone of the muddy tidal flats along the western side of Andros Island, Bahamas (handle of the dive knife is 10 em long).
c c c c c c c
Limestones -
The Sea-Floor Diagenetic Environment
Shark Bay, (Logan, 1974), cementation is more extensive and crusts extend over the whole tidal range . Cryptocrystalline aragonite is the most common cement, quickly lithifying intertidal stromatolites and algal-bound sed iment.ln Shark Bay,this grades seaward into acicular aragonite cementing hardgrounds offshore. In some instances, the aragonite cement forms pendant growths from the ceilings of farge voids and fenestrae. Perhaps most useful as signatures of shoreline precipitation are the largescale structures produced. Rapid intergranular precipitation at or near the sediment surface produces expansion fractures and arcuate to polygonal ridges and tepees. These broken crusts may in turn be reworked by storms to produce pavements of tabular clasts. Micritization of skeletal and oo id grains in many of these crusts is so intensive that peloidal sediment is in part depositional and in part diagenetic.
c o
r
Q ( ( (
o o
THE FOSSIL RECORD The products of sea-floor diagenesis can be recognized in carbonates of all ages. Almost all information, however, comes
C O
o I).
23
from the Phanerozoic, with little information on Precambrian limestones. Biological Alteration. Microborings. Endol ithic algae are known in carbonates as old as Precambrian (Campbell, 1982) and micritized grains are an integral part of most shallow-water limestone suites. Macroborings. Macroboring ichnofossils are known in limestones as old as Lower Cambrian (James et a/., 1977) but only one, Trypanites (a probable siphonculid worm trace), is found until the Middle Ordovician when reefs con taining large skeletons developed and there was a rapid diversification of endoliothictaxa(Kobluketal., 1978). Theborings of various small bivalves, sponges and echinoids are found in reefs , reefmounds and hardgrounds of later Paleozoic and particularly Mesozoic and Cenozoic age (Warme, 1975). Cementation. Confident recognition of sea-floor cementation and its spectrum of products is a critical part of any study in carbonate petrogenesis, yet two problems bedevil this seemingly simple task: (1) what precisely are the fabr ics of the aragonite and Mg-calcite cements
1,J!./ qt. o "~JI\
O~ ~D c " ,...",
..'
r
once they undergo neomorphism to calcite in another diagenetic environment, and (2) were aragonite and Mg-calcite always the normal precipitates on shallow carbonate platforms in the past? In our experience , marine cementation is not a common or widespread feature in ancient limestones. It is usually documented from fossil reefs (Figure 11), especially from reef-mounds (James, 1983, 1984) and reef blocks in fore-reef deposits (Figure 12), where marine cement may make up 50% or more of the rock volume . It is commonly inferred to be present from field evidence of hardgrounds yet rarely is demonstrated there on the petrographic level, possibly because micrite-size cements are the norm and cannot be resolved microscopically. Fibrous cements are rarely seen in marine-shelf gra instones but have been described from fossil beachrocks (Purser, 1969; Choquette and Steinen, 1980). Recognition of Sea-floor Lithification. Prior to any analysis of sea-floor cement fabrics the following criteria are useful in establishing early lithification on the sea floor (Figure 13): (1) If in bedded sediments, are the lithologic contacts knife-sharp, and if grains or cements are present are they truncated (e.g., Figure 7)? (2 Is the surface sta ined by iron or manganese salts or phosphate minerals? (3) Are there cobbles or pebbles of limestone or sand-grade lithoclasts similar in composition to the matrix, and if so are they encrusted with organisms? (4) Is the surface encrusted by organisms that require a solid substrate? (5) Is the rock surface bored and, in thin section , if cements are present, are they bored as well? (6) If in reefs, is the first-stage cement localized to the reefs? (7) If in reefs or hardgrounds, is the firststage cement interlayered with marine sediment in cavities (Figure 14)? (8) If in reefs, are there n ep.tu ~n ("J-IJ dykes? s:r...-rAIJI~ r.~('" 'tfI'(.pt 'I r
~
Figure 11 Photomicrograph of reef mound limestone composed of the ? phyllo id alga Palaeoaplysina (P)surroundedby several generations offibrous calcite submarine cement (C) (interpreted to have been Mg-calcite originally) and internal sediment (5); Lower Permian Nansen Formation, Ellesmere Island, NWT. (scale 0.5 em). Photograph, G. Davies.
.
H'.~
r.e.all)~
(9) In reef talus, are there clasts with first-stage cement or clasts of the cement (Figure 12)? (10) If in bedded sed iments, are expansion ridges and polygons present? While these features point to early lithification, many alone may be the result of a short-lived period of subaerial exposure.
24
Diagenesis
Geoscience Canada Reprint Series 4
c c c (
c c
c c
c c c c c c
Figure 12 Fore-reef debris from an Upper Devonian reef, Canning Basin, WesternAustralia. Theclast at centre is composed of reef limestone (R) (mostly the alga Renalcis and small stromatoporoids) and fibrous calcite (C). Thiscement is abundant in contempora neous reef deposits. Such clasts are often good evidence that the reef itself was subject to sea-floor cementation.
c c
Sediment overlying cement Clasts of cemented l imestone
Figure 13 Criteria for the recognition of see-tloor cementation.
Encrusting fossils Fe/Mn/P impregnated surface dork colour increase in amount of sea floor cement
r
Figure 14 Photomicrograph of i nterlayered marine sediment and radiaxial fibrous calc ite cement in laminar cavities from a Middle Ordovician reef mound at Meiklejohn Peak, Nevada. (scale 1 mm).
c c (; c c c c c c c c c c c c c c c e c c (
c ( (
(
r
Limestones - The Sea-Floor Diagenetic Environment
25
r
..
il!.,_""""_~
,,.r m lc robOnngs roun ded groins f ibrous or rnrcrocr ystcltine cement
(
Figure 15 Criteria for the recogn ition of fossil beachrock,
pilled and corroded surfaces
Figure 16 Photomicrograph of a bioclastic grainstone from the Ste. Genevieve Formation (Mississippian) Bridgeport Field , Illinois Basin , in which the first cement is a fringe of fibrous calc ite that is thickest on the undersides of grains (arrows) resembling small stalactites. Thisearly cementprobably developed while the sediment was beachrock . (scale 2 mm).
c r:
r
r»
-8 C)
0 .-
I
'
~ •
I
_ _ FASCICULAR ) OPTIC CALCITE
I'
curved
FIBROUS CALCITE
cleo~Dges
~
RAOIAXIAL FIBROUS CALCITE
Figure 17 Fabrics and morphologies of coarse marine cements . Fibrous calcite is generally interpreted as derived from calcite or Mg-calcite and spherulitic calcite from botryoidal aragonite.
'--'
,-
"
reloC~ raYs~
'-
SPHERULITI C CALCITE
PARAMORPHIC REPLACEMENT
square
-~
'spherulitic cetcue cement
SPAR-FILLED MOLDS
Diagenesis
26
In the fossil record, beachrock can be differentiated from sea-floor hardgrounds only with difficulty, butthis may be facilitated by the observation of features such as (1) associated edgewise conglomerates, (2) intertidal erosion features, (3) jointing, (4) keystonevugs, (5) meniscus or pendulous cements, and (6) associated facies (Read and Grover, 1977; Donaldson and Ricketts, 1979; Figures 15 and 16). Cement Petrography. A helpful guide is the petrographic nature of the cements themselves. Accumulated evidence from the fossil record suggests that most "fibrous" cements as well as many "spherulitic" or mammillary cements were marine originally, although they are also present in speleothems. Although microcrystalline calcite, or Mg-calcite and possibly aragonite, were precipitated from seawater in the past as they are today,the small size of the crystals, their similarity to carbonate mud and their susceptibility to rapid recrystallization all make their identification difficult (Mountjoy and Riding, 1981). Fibrous calcites. These cements (Figure 17) fall into two general groups: (1) fascicular-optic (Figure 18) cones of fibrous calcite with divergent optic axes (Kendall, 1977), and (2) radiaxialfibrous crystals (Figure 19) with consertal boundaries and convergent optic axes (Bathurst, 1959; Kendall and Tucker. 1973). While common in fossil reefs and reef mounds. radiaxial calcite
is almost unknown from modern settings, and so has long been an enigmatic cement fabric. These two types are not mutually exclusive and may grade laterally into one another along the same cementfringe.lt seems clear that these cements were originally precipitated as calcite (Sandberg, 1985). The problem is whether this calcite was low-magnesium calcite, and sois moreor less original, or whether it was highmagnesium calcite which subsequently altered to its present form . This is complicated because diagenesis appears to favour selective alteration of compositionally uniform Mg-calcite (reflecting differences in the original crystal structure) leading to calcite with highly variable magnesium contents (Videtich, 1985). On the one hand, calcite skeletons, known to have been HMC originally, now contain tiny rhombs of dolomite, 1-10 micrometres in size (Macqueen and Ghent, 1970). Since many, but not all, fibrous calcites also contain these dolomite inclusions they are likewise interpreted to have been HMC originally (Lohmann and Meyers, 1977). On the other hand, on the basis of petrography and geochemistryof Paleozoic and very young Cenozoic radiaxial fibrous calcites it seems that these cements precipitated as LMC and HMC (Kendall, 1985 ; Saller, 1986). The peculiar fabric of radiaxial fibrous calcite is thought to form by a process of asymmetric growth within calcite crystals undergoing split-growth (Kendall.
Geoscience Canada Reprint Series 4
1985). On balance, the wide variety of fabrics shown by these cements in the fossil record (James and Klappa, 1983), even when dolomitized (Tucker, 1983), and the wide range of magnesium contents in Miocene radiaxial fibrous calcites, suggest that these fibrous cements may have had a rangeof magnesium contents originally, depending upon the chemisty of the precipitating waters. Thus, those that were LMC have remained unaltered while those that were HMC have neomorphosed. These fibrous calcites are generally preferentially enriched in Mg2+ and have c5 13 C values consistent with precipitation from seawater (e.g., Davies, 1977; Walls et al., 1979; Given and Lohmann, 1985; Kerans et al., 1986). Stromatactis. An important structure in fossil carbonates, especially mud mounds, is stromatactis, loosely defined as irregular masses of fibrous calcite spar often displaying a flat base. These features originally were thought to be the recrystallized remains of a mound-forming organism, but since the work of Bathurst (1959) are now known to be sediment-floored cavities filled with fibrous , radiaxial calcite cement. Furthermore, Bathurst (1982) thinks that the whole structure is marine in origin and forms by a complex history of multigeneration sea-floor cementation and cavity formation . Epitaxial cement. In contrast to the modern sea floor, there seems to be abundant evidencefromthe fossil record
c c c c c
c c
c c c c
c c
c c c c c c c c c c c c c c c ~
Figure 18 Photomicrograph in partially polarized light of fascicular optic calcite cement from Lower Cambrian reef mounds in the Forteau Formation, southern Labrador. The numerous nested cones have the same irregular crystal boundaries as radiaxial fibrous calcite but lack the curved twins and have divergent optic axes. (scale 0.5 mm).
Figure 19 Photomicrograph in partiaffy polarized light of radiaxial fibrous calcite cements in the Middle Ordovician reef mound at Meikeljohn Peak, Nevada. The crystals characteristicaffy have concertal boundaries, curved twins, convergent optic axes and contain diverging subcrystals. (scale 0.5 mm).
c c c c c e c c c c c c c c
C'
C C C
o (
r (
C
o ( c
o
o o o o o o
o
o o o
o
Limestones - The Sea-Floor Diagenetic Environment
that syntaxial overgrowths of Mg-calcite cement on echinoderm particles were common phenomena in the shallow marine environment (Meyers, 1974; Lohmann and Meyers, 1977) . In some localit ies, marine epitaxial cements are also common on brachiopods, corals, stromatoporoids, calcispheres and ooids (Kerans et al., 1986). Scalenohedral cement. Clear columnar crystals which contain few inclusions and display unit extinction are locally common and interpreted as marine cements (Kerans et a/., 1986). Spherulitic calcite. There are numerous instances of spherulitic calcite (Figure 20) which mimic the geometry of Holocene aragonite sea-floor cements, especially in reefs. Like former aragonite skeletons, the fabric in these botryoids is preserved either as a mosaic of tiny crystals or as a paramorphic replacement (Assereto and Folk, 1980; Mazzullo, 1980) or even as chert (Grotzinger and Read, 1983). For a detailed discussion on styles of alteration see Sandberg (1985) and the following chapter on meteoric diagenesis. Actual aragonite cement crystallites preserved in calcite have been discovered in rocks as old as Pennsylvanian (Sandberg and Popp, 1981). These replaced aragon ite spherulites are sometimes still enriched in Sr and again have a e5 13 C consistent with marine precipitation . Microcrystalline calcite. Finely crystalline calcite is extremely difficult to interpret as a fossil cement (Mountjoy
27
and Riding, 1983). Lasemi and Sandberg (1982) have discovered that those micrites and microspars with apparently aragonite-dominated lime mud precursors have neomorphic calcite crystals which show pitted surfaces and/or relic aragonite inclusions in polished and etched sections . In contrast, micrite with apparent calcite-dominated precursors are characterized by finely crystalline < 4 micrometre textures, lack any inclus ions and have unpitted crystal surfaces . Isotopic Signature of Fossil Marine Cements. From a number of careful studies involving selectively sampled cements and other components of limestones (e.g., Choquette, 1968; Hudson, 1977; Dickson and Coleman , 1980; Brand and Veizer, 1981; Given and Lohmann, 1985; Lohmann, in press), it is clear that marine and other CaC0 3 cements in the fossil record have by and large preserved enough of their original signature to be distinguishable from one another (Figure 21). In general , the marine cements (isopachous, radiaxial fibrous, micritic and botryo idal) contain isotopically heavier carbon and oxygen
than near-surface meteoric or burialdiagenetic cements. Compared with co-existing skeletal debris believed to have been low-Mg calcite originally and thus relatively stable (e.g., brachiopods, bryozoans, trilob ites), marine cements have roughly the same e5 13 C but e5 18 0 which may be lighter or heavier. It now seems clear that these cements, along with all other CaC0 3 components of limestones, undergo varying degrees of geochemical alteration from slight to extensive (Brand and Veizer, 1980, 1981; Given and Lohmann , 1985) depending in part upon whether the majority of their diagenesis took place in closed, partly closed or open systems with respect to pore wate r. Given and Lohmann (1985) have argued , using both cathodoluminescence petrography and a large body of isotopic analyses, that closed-system diagenesis operates on a microscale to preserve the isotopic signatures of many marine cements. There is growing evidence that isotopic compositions of botryoidal (former aragonite) and fibrous calcite cements show systematic yariations through the Phanerozoic which seem to transcend
+6
..
+4
~
o
~ +2
z
o
o
lD
II:
e:t
o o o
o
ol-------------_.,....",...,.,=-.:-----t--------I
-2
( - ~1L,.2-....L..-+-.l.---,1,----l--!---l--+-.l.----!1,----l-~-....L..-+--.l----'·
l
-6
C. (
/"
-4
-2
+4
OXYGEN, 8 180 ("10.)
Figure 20 Photomicrograph of coalesced, strontium-rich calcite botryoids (interpreted to have been aragonite originally) from a reef mound in the Nansen Formation (Lower Permian) Ellesmere Island, NWT. (scale 2 mm). Photograph, G. Davies.
Figure 21 Cross-plot of carbon versus oxygen isotopic compositions of Holocene and ancient fibrous and spherulitic (botryo idal) carbonate cements. The values shown are the " heaviest" compos itions from covar iate populations of data, believed to approximate most closely the original marine-isotopic compositions of the cements (method of Given and Lohmann, 1985). The data are from James and Ginsburg (1979), Moldavanyi and Lohmann (1982), Marshall and Ashton (1980), Given and Lohmann (1983), Davies and Krause (1975), Lohmann (1983) , MeyersandLohmann(1983), Walls etal.(1979), MattesandMountjoy(1980), Cercone and Lohmann (1983), Sears and Lucia (1980), Ross et al. (1975), and James and Klappa (1983), listed in order of increasing age of the stratigraphic units involved.
J 11C) -.
; . ? '" • !.'''-''''''v- ';;~ L~C'r-v-I (in\' )lQ!J..;a.--
Diagenesis
28
regional differences in diagenetic history (Figures 21 and 22). These variations are consistent overall with the variability of isotopic compositions through Phanerozoic time for whole-rock limestones (e.g., Veizer and Hoefs, 1976) and for brachiopods (Pigott, 1981). Isotopic variations ofthis magnitude can be explained in a number of ways, including differences in degree of diagenetic alteration and re-equilibration (e.g., Dickson and Coleman, 1980). There is a grOWing basis for believing that the oxygen isotopic composition of seawater itself has varied since Precambrian time (Brand and Veizer, 1981; Popp et al., 1981; Lohmann, 1988). Differences of up to 5.5%0 in 0180 between present-day and past seawater have been postulated by Brand and Veizer (1981), and differences in 013C may have been nearly as great (Figure 22). THE PRESENT IS NOT ALWAYS THE KEY TO THE PAST One of the guiding tenets of carbonate petrology is uniformitarianism . especially the concept that, as in present times, aragonite and Mg-calcite were the most common carbonates forming on shallow sea floors ofthe past. This axiom has recently been challenged, and it now seems quite likely that an alternative is possible, i.e., thatthere were times when calcite was the normal sea-floor precipitate (see Sandberg, 1983 and an excellent review by Wilkinson, 1982). Twenty-five years ago, Lowenstam (1963) pointed out that the composition of marine skeletons apparently had changed during the Phanerozoic: Paleozoic seas were dominated by calcite-secreting invertebrates, Cenozoic and modern seas were characterized by organisms which secreted aragonite and Mg-calcite skeletons, and Mesozoic seas contained roughly equal numbers of each. In the last few years, on the basis of new data from fossil skeletons (Wilkinson, 1979; Pigott, 1981), ooids (Sandberg, 1975; Pigott and Mackenzie, 1979; Wilkinson et al., 1984), cements (Folk, 1974; Wilkinson, 1982; Sandberg, 1983), and carbonate muds (Folk, 1974; Sandberg, 1975), it appears that much carbonate in the fossil record was deposited or precipitated as calcite. Milliken and Pigott (1977) coined the term "calcite seas" to differentiate these periods from times similar to the "aragonite seas" of today.
Geoscience Canada Reprint Series 4
patterns in the lithosphere. Sea level is high and shallow seas vast because of extension in the length and width of the mid-ocean ridge system; the "submergent mode" of Wilkinson et al. (1985). Aragonite Seas. Theatmosphereisin the "icehouse" mode with lower Pco, because of the reduced volcanism due to sluggish plate movement and aggregated continents. Sea level is low and shallow seas reduced to narrow strips along continental margins because the mid-ocean ridge system has dropped; the "emergent mode" of Wilkinson et al. (1985). The systematic variations in isotopic composition of brachiopods (Veizer et et., 1986; Poppetal., 1986), fossil marine cements (Figures 20 and 21) and carbonates in general (Veizer and Hoefs, 1976) tend to support this hypothesis since "aragonite seas" and "calcite seas" should be characterized by heavier and lighter 013C, respectively. Since marine carbonates show a linear relationship between 013 and 0180 (Milliman and Muller, 1974), we would expect
An original model of linear change from calcite seas in the early and middle Paleozoic to later aragonite seas has now been replaced by a non-linear model tied to changes in atmospheric PC0 2 (Mackenzie and Pigott, 1981; Pigott, 1981; Sandberg and Popp, 1981; Sandberg, 1983). This most recent model has more universal appeal as it is also related to world-wide sea-level fluctuations (e.g., Vail et ei., 1977) and global tectonics. Fischer (1981) usesthe term "greenhouse mode" to label times when the atmosphere contained more CO 2 than it does today and so the surface of the earth was warmer, and the term "icehouse" to characterize times like the present, with lower CO 2 , cooler temperatures and continental glaciation. Integrating these concepts, it is postulated that in the past shallowwater carbonate deposition took place under one of the following conditions (Figure 22): Calcite Seas. The atmosphere is in the "greenhouse" mode with high PC0 2 because of increased volcanism resulting from rapid and complex convective
e
EUSTATIC MARINE
bleo s 8
CEMENTS
~p C. 0
RISING
-
SEA LEVEL--FALLING
0/00
+8
W
o
o
T
W
(f)
:::l
0
K
I Z
100
W
w II
w
(f)
:::l
0
I
W
~
w (f) :::l
0
I Z
W W
0:
(9
i
II
ARAGONITE SEAS
R
~ C
200
400
500
c c c c c c
c c c c c ( c c c c c c c c
c c E
C C C
c e c.
£
pe;
(
C 300
0
S 0
c c
C
J
0:
(9
c c
600
Figure 22 (Right) The possible distribution of "calcite" and "aragonite" seas in geologic history. Modified from Fischer (1981) and Wilkinson (1982). (Left) Variations in oxygen and carbon isotopic compositions of marine cements through Phanerozoic time. The data are from the sources acknowledged in Figure 21 (caption).
c
c ( c
(
c'
c: c (, o C o ( o ( (
o ( o o o o o o o o G
o o o
o (
o o o
Limestones - The Sea-Floor Diagenetic Environment
that radiaxial fibrous , botryoidal and other marine CaC0 3 cements also should have had original isotopic signatures which were heavier during times of "aragonite seas" and lighter during times of "calcite seas", This is exactly the relationship we appear to be seeing in Figure 22, where, from groups of analyses the heaviest (most nearly original) isotopic values have been plotted versus geologic time, The departure of the curves for 0'80 and 0'3C in late Carboniferous through Triassic time implies a diagenetic and/or original seawater temperature effect But the general relationship suggests some consistent non-linear trends in seawater composition through the Phanerozoic which show a close tie with first-order sea-level excursions and temperature variations , The question of whether seawater isotopic composition has, indeed, varied significantly and systematically through time is critically important in any attempts to interpret the origins of carbonate cements and carbonate rocks using geochemical compositions, If this is the case then the discussion of carbonate precipitation on the modern sea floor outlined in this essay applies mainly to specific times in the past : (1) late Precambrian and Cambrian, (2) Pennsylvanian to Triassic, possibly Jurassic and (3) Cenozoic - the " emergent" modes. During the early and middle Paleozoic and much of the Mesozoic abiotic carbonate precipitation in shallow seas would have been mostly calcite. Another impliction is that clear, rhombic calcite cement, generally thought to be characteristic of precipitation from fresh water (see the next article in this series) is not diagnostic alone as evidence of exposure to meteoric fluids (Given and Wilkinson , 1985), and may be a sea-floor precipitate (Wilkinson et al., 1982).
SUMMARY
c
Taken as a whole, marine sea-floor diagenesis is not common in either modern or ancient carbonate sediments. Cementation, however, is of major importance in certain facies, namely along platform and reef margins, on sand shoals and in the strandline. Early, shallow water lithification of platform and reef margins is probably one of the main reasons for the structural integrity
of these features . There is a clear zonation present on reef-rimmed platforms with cementat ion decreasing into the lagoon, onto the platform and downslope into deep water. If the lower parts of the slope are in waters undersaturated with respect to aragonite, then lithification may take place in deep cold waters by dissolution and reprecipitation. Such azonation is less apparent on open platforms and ramps, where extensive lithification of sand sheets to form hardgrounds is common. The major prerequisites for early seafloor cementation appear to be an oxidizing environment, a stable substrate and good water exchange. Recognition of submarine cements in the fossil record depends upon a combination of field, petrographic and geochemical techniques , and even then their presence is easily inferred but proven with difficulty. . The present is but a general guide to sea-floor dia enesis in the ast. Th~ world ocean appears to have oscillated between periods when calcite was th main precipitate and periods when, Iik today, aragonite and Mg-calcite we dominant Geochemical interpretations of fossil precipitates are further complicated by the fact that the isotopic composition of seawater seems to have varied with time. The implications of sea-floor cementation are profound in the creation of new environments, maintenance of large carbonate structures and the preservation of depositional textures and fabrics. ~xtensive intergranular precipitation, however, reduces dramatica y the depositiona porosl yan permeabilit of s ecific aCles ren erm them unsuitable as reservoirs for hydrocarbons a e metals.
ACKNOWLEDGEMENTS This series of articles is an outgrowth of a graduate-level course in carbonate petrogenesis given at Memorial University and a carbonate training seminar presented to Marathon Oil Company. We thank all those willing and unwilling participants who helped shape our ideas . K.C. Lohmann generously allowed us to use some of his unpublished data on the isotopic composition of fossil cements, thus enabling us to present a much more coherent picture than would otherwise have been possible. This article was originally critically
29
read by R.G.C. Bathurst, N. Chow, M. Coniglio, P. Sandberg and R.K. Stevens, to all of whom we owe special thanks. N.P. James acknowledges the Natural Sciences and Engineering Research Council of Canada for ongoing financial support
REFERENCES General References and Reviews
Bathurst , R.G.C., 1975, Carbonate Sediments and Their Diagenesis: Elsevier, North Holland, 658 p. Chapter 9 and the updated appendix are a good introduction to this topic. Bathurst, R.G.C., 1980, Lithification of carbonate sediments: Science Progress Oxford , v. 66, p. 451-471. An excellent update to the material covered in the 1975 text. Bathurst , R.G.C., 1986, Carbonate diagenesis and reservo ir development: conservation, destruction and creation of pores , in Carbonate Depos it ional Environments; Part5: Diagenesis 1, Colorado School of Mines Quarterly, v. 81, p.1-24. An essay focussed on porosity highlighting the importance of early diagenesis. Br icker, O .P., 1971, ed . , Carbonate Cements: The Johns Hopkins University Press, Baltimore, MD, p. 201-236. Theearly work on sea-floor diagenesis is well documented in 7 important papers. Hudson, J.D., 1977, Stable isotopes and limestone lithification: Journal of the Geological Society, v. 133, p. 637-660. An excellent overview of stable isotopes, carbonates and seawater. James, N.P.and Ginsburg, R.N., 1979, The seaward margin of Belize barrier and atoll reefs: International Association of Sedimentologists, Special Publication 3,199 p. A detailed, well-illustrated documentation of early diagenesis along a modern platform margin setting. Milliman, J.D. and Muller, J., 1977, Characteristics and genesis of shallow-water and deep-water limestone , in Anderson, N.R. and Malahoff, A., eds., The Fate of Fossil Fuel CO2 in the Oceans: Plenum Press, New York, p. 655-673. A summary of the attributes of marine cements. Schneidermann, N. and Harris, P.M., 1985, eds., Carbonate Cements: Society of Econom ic Mineralogists and Paleontologists, Special Publication No. 36, 379 p. A collection of papers on modern and ancient carbonate cements, a good source of current information on the topic.
30
Schroeder, J.H. and Purser, B.H ., 1986, eds., Reef Diagenesis: Springer-Verlag, New York, 455 p. A series of articles devoted entirely to the diagenesis of reefs , modern and ancient, including a review of current thinking on the topic. Zones of Sea-floor Diagenesis Adelseck, C.G. and Berger, WH., 1975, On the dissolution of planktonic foraminifera and associated microfossils during settling and on the sea floor, in Sliter, W.V., Be, AW.H. and Berger, WH ., eds. , Dissolution of Deep Sea Carbonates: Cushman Foundation Foraminifera Resources, Special Publication 13, p.70-81. Berger, WH., 1975, Deep-sea sedimentation, in Burk, C.A. and Drake , C.L., eds., The Geology of Continental Margins : Springer-Verlag, New York, p.213-243. Berger, WH ., 1978, Deep-sea carbonate: pteropod distribution and the aragonite compensation depth: Deep Sea Research, v. 25, p. 447-452. Berner, A.A., 1966, Diagenesis of carbonate sediments: interaction of magnesium in sea water with mineral grains: Science, v. 153, p. 188-191. Berner, A.A. and Morse, J.W, 1974,Dissolution kinetics of calcium carbonate in sea water IV, theory of calcite dissolution : American Journal of Science, v.274, p.l08-134. Broecker, WS. and Takahashi , T., 1978, The relationship between lysocline depth and in situ carbonate ion concentration: Deep Sea Research, v. 25, p. 65-95. Fuchtbauer, H. and Hardie, L.A., 1976, Experimentally determined homogeneous distribution coefficients for precipitated magnesium calcites: application to marine carbonate cements: Geological Society of America, Abstracts with Program, p. 877. Ginsburg, R.N . and James, N.P., 1975, Holocene carbonate sediments of continental shelves, in Burk, C.A. and Drake, C.L., eds., The Geology of Continental Margins: Springer-Verlag, New York, p. 137-157. Hsu, K.J. and Jenkyns, H.C., 1974, eds. , Pelagicsedimentson land and under the sea: International Association of Sedimentologists, Special Publication 1, 778 p. Kennett, J .P., 1982, Marine Geology : Prentice-Hall, Englewood Cliffs, 813 p. Lees, A. and Buller, A.T., 1972, Modern temperate-water and warm-water shelf carbonates contrasted: Marine Geology, v. 13, p. 67-73. Milliman, J.D., 1974, Marine Carbonates: Springer-Verlag, New York, 375 p.
Diagenesis
Morse , J.W and Berner, R.A., 1979, Chemistry of calcium carbonate in the deep oceans, in Jenne, E.A., ed., Chemical Modelling in Aqueous Systems: Amer ican Chemical Society, Washington , p.499-535 . Morse, J.W., 1983, The kinetics of calcium carbonate dissolution and precipitation, in Reeder, R.J., ed., Carbonates, their mineralogy and chemistry: Reviews in Mineralogy, v. 11, p. 227-264. Richter, D.K. and Fuchtbauer, H., 1978, Ferroan calcite replacement indicates former magnesian calcite skeletons: Sedimentology, v. 25, p. 843-860. Scholle, P.A.,Arthur, M.A. and Ekdale, A.A., 1983, Pelagic environment, in Scholle, P.A., Bebout, D.G. and Moore, C.H o, eds ., Carbonate Depositional Environments: American Association of Petroleum Geologists, Memoir 33, p. 619692. Sliter, W.V., Be, AW.H . and Berger, WH ., 1975, eds., Dissolution of Deep Sea Carbonates: Cushman Foundation Foraminifera Research, Special Publication 13, 159 p. Takahashi, T., 1975,Carbonate chemistry of seawater and the calcite compensation depths in the oceans, in Sliter, W.v., Be, A.W.H. and Berger, W.H., eds ., Dissolution of Deep Sea Carbonates: Cushman Foundation Foraminifera Resources, Special Publication 13, p. 11-26.
Modern Carbonate Sediments Zone I: Warm Water, Shallow Tropical Environments
Mineralogy of Precipitates Bathurst, A.G .C., 1968, Precipitation of ooids and other aragonite fabrics in warm seas, in Muller, G. and Friedman, G.M., eds., Recent Developments in Carbonate Sedimentology in Central Europe: Springer, Berlin, p. 1-10. Berner, R.A., 1975, The role of magnesium in the crystal growth of calcite and aragonite from seawater: Geochimica et Cosmochimica Acta, v. 39, p. 489-504. Berner, R.A., 1980, Early Diagenesis: a theoretical approach: Princeton University Press, 241 p. Burton, E.A. and Walter, L.M., 1987, Relative precipitation rates of aragonite and Mg calcite from seawater: Temperature or carbonate ion control: Geology, v. 15, p. 111-114. Davies, P.J., Bubela, B. and Ferguson, J., 1978, The formation of ooids: Sedimentology, v. 25, p. 703-731. Folk, R.L., 1974, Natural history of crystalline calcium carbonate: effect of magnesium content and salinity: Journal of Sedimentary Petrology, v. 44, p. 40-53 .
Geoscience Canada Reprint Series 4
Given, R.K. and Wilkinson, B.H ., 1985a, Kinetic control of morphology, composition and mineralogy of abiotic sedimentary carbonates: Journal of Sedi· mentary Petrology, v. 55, p. 109-119. Given, R.K. and Wilkinson, S.H ., 1985b, Reply - Kinetic control of morphology, composition and mineralogy of abiotic sedimentary carbonates : Journal of Sedimentary Petrology, v. 55, p. 921-926. Loreau, J.P., 1982, Sediments aragonitiquesetleurgenese: Memoirdu Museum National D'Histoire Naturelle, New Series Tome XLVII, 483 p. Kinsman, D.J.J ., 1969, Interpretation of Sr +2concentrations in carbonate minerals and rocks: Journal of Sedimentary Petrology, v. 39, p. 489-508. Morse, J.W., 1985, Discussion - Kinetic control of morphology, composition and mineralogy of abiotic sedimentary carbonates: Journal of Sedimentary Petrology, v. 55, p. 919-921. Mackenzie, F.T., Bischoff, W.D., Bishop, F.C., Loijens, M., Schoonmaker, J. and Wollast, A., 1983, Magnesian calcites : low temperature occurrence, solUbility and solid solution behaviour, in Reeder, A.J ., ed., Carbonates, their mineralogy and chemistry: Mineralogical Society of America, Reviews in Mineralogy, v. 11, p.97-144. Mitterer, R.M., 1971, Influence of natural organic matter on CaC03 precipitation, in Bricker, O.P., ed. , Carbonate Cements: The Johns Hopkins University Press, Baltimore, MD, p. 252-258. Mitterer, R.M . and Cunningham, R., Jr., 1985, The interactions of natural organic matter with grain surfaces: implications for calcium carbonate precipitation, in Schneidermann , N. and Harris, P.M., eds ., Carbonate Cements: Society of Economic Mineralogists and Paleontologists, Special Publication No . 36, p.17-32. Mucci, A. and Morse , J.W., 1983, The incorporation of Mg2+ and Sr 2+ into calcite overgrowths: influences of growth rate and solution composition: Geochimica et Cosmochimica Acta, v. 47,p. 217-233. Suess, E., 1970, Interaction of organic compounds with calcium carbonate -I : associated phenomena and geochemical implications: Geochimica et Cosmochimica Acta, v. 34, p. 157-168. Suess, E., 1973,Interaction of organic compounds with calcium carbonate - II: organocarbonate association in recent sediments : Geochimica et Cosmochi mica Acta, v. 37, p. 2435-2447. Walter, L.M., 1985,Relative reactivity of skeletal carbonates during dissolution: implications for diagenes is, in Schneidermann, N. and Harris, P.M., eds ., Carbonate Cements: Society of Economic Mineralogists and Paleontologists, Special Publication No. 36, p. 3-16.
c c c c c c c c ( c ( c c c c c c c t; c c (
c
c c c c ( c c c c c c
e (
c
c (
C l
C (
(
(
Limestones -
31
The Sea-Floor Diagenetic Environment
c C'
o o o ( (
( (
o o o a o o o o r
o
o o
o o
"
Walter, L.M. and Morse, J.w., 1984, Magne· sium calcite stabilities : are-evaluation : Geochimica et Cosmochim ica Acta , v. 48, p. 1059·1069. Petrography of Precipitates
Chafetz, H.S. , 1986, Marine peloids: a product of bacterially induced prec ipitation of ca lc ite: Journal of Sed imentary Petrology, v. 56, p. 812-817. Macintyre, I.G. , 1985, Submarine cementsthe peloidal question , in Schneider· mann, N. and Harris , P.M., eds ., Carbonate Cements : Society of Econom ic Mineralogists and Paleontologists, Special Publication No. 36, p, 109·116. Marshall , J.F., 1983, Submarine cementation in a high-energy platform reef: One Tree Reef, southern Great Barrier Reef : Journal of Sedimentary Petrology, v. 53, p. 1133-1149. Sandberg, P., 1985, Aragonite cements and their occurrence in ancient limestone , in Schneidermann, N. and Harris, P.M., eds. , Carbonate Cements : Society of Economic Mineralog ists and Paleon tologists , Special Publ ication No. 36 , p.33-58. Isotopic Geochemistry of Precipitates
Gonzalez, L.A. and Lohmann, K.C., 1985, Carbon and oxygen isotopic compostt ion of Holocene reefal carbonates : Geology, v. 13, p. 811-814. Gross, M.G., 1964, Variations in the 18 0 / 160 and 13C/'2C ratios of diagenetically altered limestones in the Bermuda Islands: Journal of Geology. v.72 , p. 170-194. Keith, M.L. and Weber, J.N., 1965, Carbon and oxygen isotopic compositon of selected limestones and foss ils: Geochimica et Cosmochimica Acta, v. 40, p. 1095-1108 . Lloyd , A.M., 1971, Some observations on recent sediment alteration ("micritization") and the possible role of algae in submarine cementation, in Br icker. D.P., ed. , Carbonate Cements: The Johns Hopk ins University Press. Baltimore, MD, p. 72-79. Moore, C.H., 1973, Intertidal carbonate cementation , Grand Cayman , West Indies: Journal of Sedimentary Petrology, v. 43, p. 591-602. Morrison, J. and Brand, U., 1986, Geochemistry of Recent marine invertebrates: Geoscience Canada, v. 13, p. 237·254. Schroeder, J.H ., 1972, Calcified filaments of an endolithic algae in recent Bermuda reefs : Neues Jahrbuch fUr Geologie, Palaeontology Montash v. 1, p. 16-33.
Viezer, J., 1983, Trace elements and isotopes in sedimentary carbonates , in Reeder, R.J., ed., Carbonates, the ir mineralogy and chemistry: Mineralogical Society of America, Reviews in Mineralogy, v. 11, p. 265-300 . Precipitated Partlc/es
Ellis, J.P. and Milliman , J.D., 1985, Calcium carbonate suspended in Arabian Gulf and Red Sea waters: biogenic and detrital , not " chemogenic" : Journal of Sedimentary Petrology, v. 55, p. 805-808. Land, L.S ., Behrens. E.W. and Frishman , SA, 1979 , The ooids of Baffin Bay Texas: Journal of Sedimentary Petrology, v. 49 , p. 1269-1278. Marshall , J.F. and Davies , P.J., 1975, Highmagnesium calcite ooids from the Great Barrier reef : Journal of Sedimentary Petrology, v. 45, p. 285-291. Neumann , A.C. and Land, L.S., 1975, Lime mud deposition and calcareous algae in the Bight of Abaco, Bahamas: a budge t: Journal of Sedimentary Petrology, v. 45, p.763-786. Simone , L., 1980, Ooids: a review: Earth Sciences Reviews, v. 16, p.319-355 . Stockman, K.W., Ginsburg, R.N. and Shinn , E.A ., 1967, The production of lime mud by algae in south Flor ida: Journal of Sedimentary Petrology, v.37, p.633-648. Cementation of Sand Shoal, Reef and Fore-reef Sediments
Aissaoui, D.M, Bu igues, D. and Purser, B.H., 1986, Model of reef diagenesis: Mururoa Atoll, French Polynesia , in Schroeder, J.H . and Purser, B.H., eds., Reef Diagenesis: Springer-Verlag. New York, p. 27-52. Dill, R.F., Shinn , E.A., Jones, A.T., Kelly, K. and Ste inen, A.P.• 1986. Giant subtidal stromatolites forming in normal salinity waters : Nature, v. 324, p. 55-58. Dravis, J., 1979, Rapid and widespread generation of recent oolitic hardgrounds on the high energy Bahamian Platform, Eleuthera Bank, Bahamas: Journal of Sedimentary Petrology, v. 49, p. 195-209. Ginsburg, R.N. and Schroeder, J., 1973, Growth and submarine fossilization of algal cup reefs, Bermuda: Sedimentology, v. 20 , p. 575-614. James , N.P., Ginsburg, R.N., Marszalek, D.S. and Choquette, P.w., 1976, Facies and fabric spec ificity of early subsea cements in shallow Belize (British Honduras) reefs: Journal of Sedimentary Petrology, v. 46, p. 523-544. Land, L.S. and Goreau, T.F., 1970, Submarine lith ification of Jamaican reefs: Journal of Sedimentary Petrology, v. 40, p.457-461.
Land, L.S. and Moore , C.H., 1980, lithification, micritization and syndepositional diagenes is of biolithites on the Jama ican Island slope : Journal of Sedimentary Petrology, v. 40, p. 357-371. Marshall , R.J., 1986, Regional distribution of submarine cements with in an epicon tinental reef system : Central Great Barrier Reef , Australia, in Schroeder , J.H. and Purser, B.H., eds ., Reef Diagenes is: Springer-Verlag, New York, p. 8-26. Macintyre , I.G., 1977, Distribution of submar ine cements in a modern Caribbean fringing reef , Galeta Point , Panama : Journal of Sedimentary Pstrolcqy, v. 47, p.503-516. Pierson, B.J. and Shinn, E.A., 1985 , Cement distribution and carbonate minerai stabilization in Pleistocene limestones of Hogsty reef, Bahamas , in Schne idermann , N. and Harris , P.M .. eds., Carbonate Cements : Society of Economic Mineralogists and Paleontologists , Special Publicat ion No. 36, p. 153-169 . Read, J.F., 1974, Carbonate bank and wavebuilt platform sedimentation, Edel Province, Shark Bay. Western Australia, in Logan, B.W., et al., eds., Evolution and diagenesis of Quaternary carbonate sequences. Shark Bay, Western Australia : American Assoc iat ion of Petroleum Geolog ists . Memoir 22, p. 1-61. Shinn , E.A., 1969, Submarine lithification of Holocene carbonate sediments in the Persian Gulf: Sedimentology, v. 12, p. 109-144. Taft, W.H., Arr ington , F., Hainov itz, A., Macdonald , C. and Woolheater, C., 1968, Lith ification of modern carbonate sediments at Yellow Bank, Bahamas : Bulletin of Marine Science, v. 18, p. 762828. Cementation on Carbonate Slopes and in Shallow Basins
Gevirtz, J.L. and Friedman, G.M., 1966, Deep sea carbonate sediments of the Red Sea and their implication on marine lithification : Journal of Sedimentary Petrology, v. 36, p. 143-151. Mullins, H.T.A., Neumann, C., Wilber, R.J. and Boardmann, M.R., 1980, Nodular carbonate sediment on Bahamian slopes: possible precursors to nodular limestones: Journal of Sedimentary Petrology, v. 50, p. 117-133. Neumann, A.C., Kofoed , J.w. and Keller, G.H., 1977, Lithoherms in the straits of Florida: Geology, v. 5, p. 4-10. Neomorphism
Kendall, C.G.C. and Skipwith, P.A., 1969, Holocene shallow-water carbonate and evaporite sediments of Khor alBazam, Trucial Coast, Southwest Persian Gulf :
32
American Association of Petroleum Geologists, Bulletin, v. 53, p. 841-869 . Murray, J.w., 1966, The foraminifera of the Persian Gulf. 4: Khor al Bazam: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 2, p. 153-169. Purdy, E.G., 1968, Carbonate diagenesis : an environment survey : Geologica Romana, v. 7, p. 183-228. Scherer, M., 1974, Submarine recrystallization of a coral skeleton in a Holocene Bahamian reef: Geology, v. 3, p. 499500. Zone II: Temperate water, shallow shelves and deep water tropical slopes and basins
Alexandersson, T., 1974, Carbonate cementation in coralline algal nodules in the Skagerrak, North Sea: biochemical precipitation in undersaturated waters: Journal of Sedimentary Petrology, v. 44, p.7-26. Alexandersson, T., 1978, Destructive diagenesis of carbonate sediments in the Eastern Skagerrak, North Sea: Geology, v. 6, p. 324-327. Alexandersson, T., 1979, Marine maceration of skeletal carbonates in the Skagerrak North Sea: Sedimentology, v. 26, p.845-852. Freeman-Lynde, R.P., Whitley, K.F. and Lohmann , K.C., 1986, Deep-marine origin of equant spar cements in Bahamian escarpment limestones: Journal of Sedimentary Petrology, v. 56, p. 799-811. Gomberg, D.N. and Bonatti, E., 1970,Highmagnesium calcite : leach ing of magnesium in the deep sea: Science, v. 168, p. 1451-1453. Milliman, J.D., 1966, Submarine lithification of carbonate sediments: Science, v. 153, p. 994-997. Mullins, H.T., Wise, S.W., Jr. , Gardulski, A.F., Hinchey, E.J., Masters, P.M. and Siegel, D.I., 1985, Shallow subsurface diagenesis of Pleistocene periplatform ooze: northern Bahamas: Sedimentology, v. 32, p. 473-494 . Nelson, C.S. and Lawrence, M.F., 1984, M~thane-derived high·Mg calcite submarine cement in Holocene nodules from the Frazer Delta, British Columbia, Canada: Sedimentology, v. 31, p. 645-655. Rao, C.P., 1981, Cementation in cold-water bryozoan sand, Tasmania, Australia: Marine Geology, v. 40, p. m23-m33. Rao, C.P. and Green, D.C., 1983, Oxygen and carbon-isotope composition of cold shallow-water carbonates of Tasmania, Australia: Marine Geology, v.53 , p. 117-129. Saller, A.H., 1986, Radiaxial calcite in Lower Miocene strata , subsurface Enewetak Atoll: Journal of Sedimentary Petrology, v. 56, p. 743-762.
Diagenesis
Schlager, W. and James, N.P., 1978, Low magnesium calcite limestones forming at the deep sea floor, Tongue of the Ocean, Bahamas: Sedimentology, v. 15, p.675-702. Biological Alteration
Alexandersson, T., 1978, Micr itization of carbonate particles: processes of precipitation and dissolution in modern shallow-water sediments : Bulletin of Geologists Institution , University Uppsala , New Series 3, p. 201-236. Bathurst, R.G.C., 1966, Boring algae, micrite envelopes and lithification of molluscan biosparites: Geological Journal, v. 5, p. 15-32. Bromley, R.G. , 1978, Bioerosion of Bermuda reefs: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 23 , p. 169-197. Budd, D.A. and Perkins, R.D., 1980, Bathymetric zonation and paleoecological significance of microboring in Puerto Rican shelf and slope sediments : Journal of Sedimentary Petrology, v. 50, p.881-905. Friedman, G .M., Gebelein, C.D. and Sanders,J .E., 1971, Micrite envelopes of carbonate grains are not exclusively of photosynthetic algal origin : Sedimentology, v. 16, p. 89-96. Kobluk, D.R. and Risk, M.J., 1977, Rate and nature of infestation of a carbonate substratum by a boring alga: Journal of Experimental Marine Biology and Ecology, v. 27, p. 107-115. Margolis, S. and Rex, R.W., 1971, Endolithic algae and micrite envelope formation in Bahamian oolites are revealed by scanning electron microscopy: Geological Society of America, BUlletin, v.82, p.843-852. Neumann , A.C., 1965, Processes of recent carbonate sedimentation in Harrington Sound, Bermuda: Bulletin of Marine Science, v. 15, p. 987-1035. Zankl, H. and Schroeder, J.H., 1972, Interaction of genetic processes in Holocene reefs off North Eleuthera Island, Bahamas : Geologische Rundschau, v. 61, p. 520-541. Strandline Diagenesis
Hanor, J.S ., 1978, Precipitation of beachrock cements: mixing of marine and meteoric waters vs . CO 2 degassing : Journal of Sedimentary Petrolgy, v. 48, p.489-501 . Hardie, L.A ., 1977, Algal structures in cemented crusts and their environmental significance, in Hard ie, L.A. , ed., Sedimentation on the Modern Carbonate Tidal Flats of Northwest Andros Island, Bahamas: Johns Hopkins Press, Baltimore , MD, p. 159-178.
Geoscience Canada Reprint Series 4
Jones, B. and Goodbody, Q.H., 1984, Biological alteration of beachrock on Grand Cayman Island, British West Indies: Bulletin of Canadian Petroleum Geology, v. 32, p. 201-215. Logan, B.W., 1974, Inventory of diagenesis in Holocene - Recent carbonate sediments, Shark Bay, Western Australia, in Logan, BW., et al. , Evolution and Diagenesis of Quaternary Carbonate Sequences, Shark Bay, Western Australia : American Association of Petroleum Geolog ists, Memoir 22, p. 195·250. Meyers, J.H., 1987, Marine vadose beachrock cementat ion by cryptocrystalline magnesian calcite- Maui, Hawaii:Journal of Sedimentary Petrology, v. 57, p. 558570. Purser, B.R. and Loreau, J.P., 1973, Aragonitic, suprat idal encrustations on the Trucial Coast, Persian Gulf, in Purser, B.H., ed., The Persian Gulf: SpringerVerlag, p. 343-377. Stoddart, D.R. and Cann , J.R., 1965, The nature and origin of beachrock: Journal of Sedimentary Petrology, v.35, p.243-247. Taylor, J.D .M . and IIling, L.V., 1969, Holocene intertidal calcium carbonate cementation , Qatar, Persian Gulf: Sedimentology, v. 12, p. 69-109.
The Fossil Record Biological Alteration
Bromley, R.G., 1975, Trace fossils at omission surfaces , in Frey, R.W., ed., Study of Trace Fossils: Springer-Verlag, New York, p. 399-229. Campbell, S.E., 1982, Precambr ian endoliths discovered: Nature, v.229, p.429-431. James, N.P., Kobluk, D.R. and Pemberton, S.G., 1977, The oldest macroborers : Lower Cambrian of Labrador : Science, v. 197, p. 980·983 . Kobluk, D.R., James, N.P. and Pemberton, S.G., 1978, Initial diversification of macroboring ichnofossils and exploitation of the macroboring niche in the lower Paleozoic : Paleobiology, v. 4, p. 163-170. Warme, J.E., 1975, Borings as trace fossils and the processes of bioeros ion, in Frey, R.W., ed., Study of Trace Fossils: Springer-Verlag , New York, p. 181-229. Petrography of Sea-floor and Shallow Marine Burial Cements
Bathurst, R.G.C ., 1959, The cavernous structure of some Mississippian stromatactis reefs in Lancashire, England: Journal of Geology, v. 67, p. 506-521. Bathurst, R .G .C., 1982, Genes is of stromatactis cavities between submarine crusts in Paleozoic carbonate mud buildups : Journal of the Geological Society, v. 139, p. 165-181.
c c c c c c c c c c ( c c c c c c c c c c c c c c c c c c c c c c e c c c ( (
C (
C
Limestones -
The Sea-Floor Diagenetic Environment
33
r
o c () r
c o o
o o o o o o
o
r
rl
V
o o o
o r:
o
o o
Bromley, R.G., 1978, Hardground diagene sis, in Fairbridge , R.W. and Bourgeo is, J., eds ., The Encyclopedia of Sedimentology: Dowden, Hutch inson and Ross, Stroudsburg, PA, p. 397-400 . Davies , G.R. and Krause , H.R., 1975, Carbon and oxygen isotopic composition of Late Paleozoic carbonates - Canadian Arctic Archipelago - prelim inary results and interpretations: Geological Survey of Canada, Pape r 75-18 , p.215-220 . Davies , G.R., 1977, Former magnesian calcite and aragon ite submarine cements in Upper Paleozoic reefs of the Canadian Arct ic: a summary: Geology, v. 5, p. 11-15. Grotzinger, J.P. and Read , J .F., 1983, Evidence for primary aragonite precipitat ion , lower Proterozo ic (1.9 Ga) Rocknest Dolomite, Wopmay Orogen , northwest Canada : Geology, v. 11, p.710-713. James , N.P. and Klappa, C.F., 1983, Petrogenesis of Early Cambrian reef li mestones, Labrador, Canada : Journal of Sed imentary Petrology, v. 53, p. 1051· 1096. Kendall , A.C., 1977, Fasc icu lar-opt ic calcite : a replacement of bundled acicular carbonate cements: Journal of Sedimentary Petrology, v. 47, p. 1056-1062. Kendall , A.C., 1985, Radiaxial fibrous calcite: a reappra isal, in Schneidermann, N. and Harris, P.M., eds., Carbonate Cements: Society of Economic Mineralogists and Paleontologists , Spec ial Publication No. 36, p. 59-78. Kendall , A.C. and Tucker, M.E., 1973, Radiaxial fibrous calcite: a replacement after ac icular ca rbonate: Sedimentology, v. 20 , p. 365-389. Lindstrom, M., 1979, Diagenesis of Lower Ordovician hardgrounds in Sweden : Geologica et Palaeontologica, v. 13, p.9-30. Lasem i, Z. and Sandberg, P.A., 1984, Transformation of aragon ite-dom inated lime muds to microcrystalline limestones: Geology, v. 12, p. 420-423 . Lohmann, K.C. and Meyers, w.J., 1977, Microdolomite inclusions in cloudy prismat ic calcites - a proposed cr iter ion for former high magnesium calc ites: Journal of Sedimentary Petrology, v.47, p. 1078-1088. Mazzullo, S.J., 1980, Calc ite pseudospar replac ive of mar ine acicular aragon ite and implications for aragonite cement d iagenesis : Journal of Sedimentary Petrology, v. 50, p. 409-423. Macqueen, R.W. and Ghent , E.D., 1970, Electron microprobe study of magnesium distribution in some Mississippian ech inodern limestones from Western Canada: Canad ian Journal of Earth Sciences, v. 7, p. 1308-1317.
Mountjoy, EW. and Riding , R., 1981, Foreslope stromatoporoid-renalcid bioherm w ith evidence of early cementation , Devonian Ancient Wall reef complex, Rocky Mounta ins: Sedimentology, v. 28, p. 299-321. Richter, D.K. and FGchtbauer, H., 1978, Ferroan calcite replacement indicates former magne sian calcite skeletons: Sedimentology, v. 25, p. 843-861. Sandberg, P., 1985, Aragon ite cements and their occurrence in ancient limestone , in Schneidermann , N. and Harris , P.M. , eds ., Carbonate Cements: Society of Economic Mineralogists and Paleon tologists, Special Publication No. 36, p.33-58. Sandberg, P.A. and Popp, B.N., 1981, Pennsylvanian aragonite from southeastern Kansas - environmental and diagenetic implications (abstract): American Association of Petroleum Geolog ists , Bulletin, v. 65 , p. 985 . Stoakes, FA, 1980, Nature and control of shale basin fill and its effect on reef growth and termination : Upper Devonian Duvernay and Ireton Format ions of Alberta, Canada: Bulletin of Canadian Petroleum Geology, v. 28, p. 345-411. Tucker, M.E., 1983, Diagenesis, geochemistry, and or igin of a Precambrian dolo mite: the Beck Spring Dolomite of eastern California: Journal of Sedimentary Petrology, v. 53, p. 1097-1119. Videtich, P.E., 1985, Electron microprobe study of Mg distribu tion in recent Mg calcites and recrystallized equivalents from the Pleistocene and Tertiary: Journal of Sed imentary Petrology, v. 55, p.421-429. Walls, R.A., Mountjoy, E.W. and Fritz, P., 1979, Isotopic composition and diagenetic history of carbonate cements in Devonian Golden Spike reef, Alberta, Canada: Geological Society of America, Bulletin, v. 90 , p. 963-982 . Wilkinson, B.H., Janecke, S.V. and Brett, C .E ., 1982, Low-magnesian calcite marine cement in Middle Ordovician hardgrounds from Kirkfield, Ontario: Journal of Sedimentary Petrology, v. 52, p.47-59. Strandline Diagenesis
Asseretto , R. and Kendall , G.C., 1977, Nature, or igin, and class ification of peritidal teepee structures and related breccias: Sedimentology, v. 24, p. 153-210. Asseretto, R. and Folk, R.L. , 1980, Diagenet ic fabrics of aragon ite calcite and dolomite in an ancient peritidal-spelean environment: Triassic Calcare Rosso , Lombard ia , Italy: Journal of Sedimentary Petrology, v. 50, p. 371·395.
Choquette, P.w. and Steinen, R.P., 1980, Mississippian non-supratidal dolomite, Ste . Genev ieve Li mestone, Ill inois Basin : evidence for mixed-water dolomitization, in Zenger, D.H., Dunham , J.B. and Ethington , R.L., eds., Concepts and Models of Dolomit ization: Society of Econom ic Paleontologists and Mineralogists, Special Publication No. 28, p. 163-196. Donaldson, J.A . and Ricketts, B.D., 1979, Beachrock in Proterozo ic dolostone of the Belcher Islands, Northwest Territories , Canada: Journal of Sedimentary Petrology, v. 49, p. 1287-1294. Purser, B .H ., 1969 , Syn-sed imentary marine lithification of Middle-Jurassic limestones in the Paris Basin : Sedimentology, v. 12, p. 205-230. Read, J.F. and Grover, G.A., 1977, Scalloped and planar erosion surfaces, Middle Ordovician limestones, Virginia : analogues of Holocene exposed karst or tidal rock platforms: Journal of Sedimentary Petrology, v. 47, p. 956-972. Isotopic Geochemistry of Precipitates
Brand, U. and Veizer, J., 1981, Chemical diagenesis of a multi-component carbonate system - 2: stable isotpes: Journal of Sed imentary Petrology, v.51, p.987-999. Choquette, P.w., 1968, Marine diagenesis of shallow marine lime-mud sediments : insights from 0'80 and O'3C data : Science, v. 161, p. 1130-1132. Dickson, J.A.D. and Coleman , M.L., 1980, Changes in carbon and oxygen isotope composition during limestone diagene sis: Sedimentology, v. 27, p. 107·118. Davies, G.R. and Krouse, H.R., 1975, Carbon and oxygen isotopic composition of Late Paleozoic calc ite cements , Canadian Arct ic archipelago - preliminary results and interpretation: Geological Society of Canada , Pape r 75 -1 B, p.215-220. Given, R.K. and Lohmann , K.C., 1985, Derivation of the original isotopic composition of Permian marine cements : Journal of Sed imentary Petrology, v. 55, p.430-439 . Kerans, C., Hurley, N.F. and Playford, P.E., 1986, Marine diagenesis in Devonian reef complexes of the Canning Basin , western Australia, in Schroeder, J.H. and Purser, B.H., eds ., Reef Diagenesis: Springer-Verlag, New York, p. 357-380. Lohmann, K.C., 1983, Diagenetic history of carbonate reservo irs - integra tion of petrogrpahic and geochemical tech niques , in New Ideas and Methods for Exploration for Carbonate Reservoirs, Amer ican Assoc iat ion of Petroleum Geolog ists , Short Course Notes , Dallas, TX, Apr il 16-17, 1983.
34
Diagenesis
Lohmann, K.C., in press, Covariant carbon and oxygen isotopic trends in calcitized marine cements - estimation of marine isotopic signatures:Journal of Sedimentary Petrology, in press. Marshall , J.D. and Ashton , M., 1980, Isotopic and trace element evidence for submarine lithification of hardgrounds in the Jurass ic of Eastern England : Sedimentology, v. 27, p. 271-289. Mattes, B.W.and Mountjoy, E.W., 1980,Burial dolomitization ofthe Upper Devonian Miette buildup, Jasper National Park, Alberta, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists , Special Publication No. 28 , p.259-297. Popp, B.N., Sandberg , P.A. and Anderson, T.E, 1981, Coord inated textural isotopic and elemental analysis of constituents in some Middle Devonian limestones (abstract): Geological Society of America, Abstracts with Programs, p. 531. Ross, R.J., Jr., Jaanusson, V. and Friedman, I., 1975, Lithology and origin of Middle Ordovician calcareous mudmound of Meiklejohn Peak, Southern Nevada: United States Geological Survey, Professional Paper 871, 48 p. Sears, S.D. and Lucia, EJ., 1980, Dolomitization of northern Michigan Niagara reefs by brine refluxion and freshwater/ seawater mixing , in Zenger, D.H., Dunham, J.B. and Eth ington, R.L., eds ., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication No. 28, p. 215·237. Veizer, J. and Hoefs, J., 1976, The nature of 180 /180 and 13C/12C secular trends in sedimentary rocks : Geochimica et Cosmochimica Acta, v. 40, p. 1387-1395. Walls, R.A., Mountjoy, E.W. and Fritz, P. , 1979, Isotopic composition and diagenetic history of carbonate cements in Devonian Golden Spike reef, Alberta, Canada: Geological Society of America, Bulletin, v. 90, p. 963-982. Temporal Variations In the Mineralogy and Isotopic Composition of Sea-floor
Precipitates Fischer, A.G., 1981, Climatic oscillations in the biosphere, in Nitecki, M.H ., ed., Biotic Crises in Ecological and Evolutionary Time: Academic Press , New York, p. 103-133. Lowenstam, H., 1963, Biologic problems relating to the composition and diagenesis of sediments, in Donnelly, T.W., ed., The Earth Sciences: University of Chicago Press, Chicago, p. 137-195.
vV \\1
sy ---
\/)/
Lohmann, K.C., 1988, Geochemical patterns of meteoric diagenetic systems and their application to studies of paleokarst, in James, N.P. and Choquette, P.W., eds., Paleokarst: Springer-Verlag, New York, p. 58-80 . Mackenzie, ET. and Pigott, J .D., 1981, Tectonic controls of phanerozoic sedimentary rock cycling : Journal of the Geological Society of London, v. 138,p. 183-196. Milliken, K.L. and Pigott, J.D ., 1977, Variation of oceanic Mg/Ca ratio through time - implications for the calcite sea: Geological Society of America, South-Central Meeting , Abstracts, p. 64-65. Pigott, J.D., 1981, Global tectonic control of secular variations in Phanerozoic sedimentary rock/ocean/atmospheric chemistry (abstract) : American Association of Petroleum Geologists, Bulletin , v. 65, p. 963. Pigott, J.D. and Mackenzie, F.r., 1979, A signature of paleo-ocean and atmospheric chemistry: Geological Society of America, Abstracts with Program , p.495-496. Popp, B.N., Anderson, T.F. and Sandberg, P.A., 1986, Brachiopods as indicators of original isotopic conpos itions in some Paleozoic limestones : Geological Society of America, Bulletin, v.97, p. 1262-1268. Sandberg , P.A., 1975, New interpretations of Great Salt Lake ooids and nonskeletal carbonate mineralogy: Sedimentology, v. 22, p. 497-537. Sandberg, P.A., 1983,An oscillating trend in Phanerozoic non-skeletal carbonate mineralogy: Nature, v. 305, p. 19-22. Vail, P.R., Mitchum, J.R. and Thompson, S., 11I,1977, Seismic stratigraphy and global changes in sea-level, part 4, in Peyton, C.E., ed., Seismic Stratigraphy: American Assoc iation of Petroleum Geologists , Memoir 26, p. 83-97. Veizer,J., Fritz, p. and Jones, B., 1986,Geochemistry of brachiopods: oxygen and carbon isotopic records of Paleozoic oceans : Geochimica et Cosmochimica Acta, v. 50, p. 1679-1696. Wilkinson, B.H., 1979, Biomineralization, paleoceanography and the evolution of calcareous marine organ isms: Geology, v. 7, p. 524-527. Wilkinson, B.H., 1982, Cyclic cratonic carbonates and phanerozo ic calcite seas: Journal of Geological Education , v. 30, p.189-203. Wilkinson, B.H., Bucynski, C. and Owen, R.M., 1984, Implications from Upper Pennsylvan ian calcite-aragonite ooids of southeastern Kansas: Journal of Sed imentary Petrology, v.53, p.932-947.
Geoscience Canada Reprint Series 4
Wilkinson, B.H., Owen, R.M. and Carroll, A.R., 1985, Submarine geothermal weathering , global eustacy and carbonate polymorphism i n Phanerozoic marine oolites: Journal of Sedimentary Petrology, v. 55, p. 171-183. Other References Cited In Text Alexanderson , T.F. and Schneidermann, N., 1973,Stable isotope relationships in pelagic limestones from the central Caribbean : Leg 15 Deep Sea Drilling Project , in Edgar, N.T., Saunders, J.B., et el. , eds., Initial Reports of the Deep Sea Drill ing Project, Vol. XV, p. 795-803. Berner, R.A., 1971, Principles of Chemical Sedimentology: McGraw-Hili, New York, 240 p. Broecker, W.S., 1974, Chemical Oceanography : Harcourt, Brace and Jovanovitch, New York, 214 p. Cayeux, L., 1935, Les Roches Sedimentaires de France: Roches carbonatses : Masson, Paris, 436 p. James, N.P., 1983, Reef environment, in Scholle, P.A., Bebout, D.G. and Moore, C.H ., eds ., Carbonate Depositional Environments: American Association of Petroleum Geologists , Memoir 33 , p.345·440. James, N.P., 1984, Shallowing-upward sequences in carbonates , in Walker, °R.G., ed., Facies Models, Second Edition : Geological Association of Canada, Geoscience Canada Reprint Series 1, p.213-228 . James, N.P., 1984, Reefs, in Walker, R.G., ed., Facies Models, Second Edition : Geological Association of Canada, Geoscience Canada Reprint Ser ies 1, p.229-244. Kendall , A.C., 1978, Evaporites, in Walker, R.G., ed., Facies Models, Second Edition : Geological Associat ion of Canada, Geoscience Canada Reprint Series 1, p.259-296. Murray, J.L. and Renard, A.F., 1891, Report on deep sea deposits based on the specimens collected during the voyage of H.M.S. Challenger in the years 1872-1876, 525 p. [Reprinted by Johnson Reprint, London, 1965. J Murray, J. and Hjort, J., 1912, The Depths of the Oceans: Macmillan , London. 831 p.
c c c
c c
c c c (
c
c
c c c c c c c c:
c c c
c
c c c c c c
c c
c c
e (
Originally published in Geoscience Canada v. 10 Number 4 (December 1983) Revised 1988
c c
c
c ( (
C
Limestones - The Meteoric Diagenetic Environment
~The meteoric diagenetic env iron/~ent is one with many apparent contra-
r:
o
c
c o r (
c
Limestones - The Meteoric Diagenetic Environment Noel P. James Department of Geological Sciences Queen's University Kingston, Ontario K7L 3N6 Philip W. Choquette Department of Geological Sciences University of Colorado Boulder, Colorado 80309-250
INTRODUCTION Carbonate sediments, born in the sea, remain largely unaltered while bathed in their embryonic fluids of high ionic strength. With time, however, the composition ofthe pore waters will inevitably change , and if exposed to circulating ground water, the sediments may undergo profound and geologically rapid diagenesis (Figure 1).
dictions. For example, the main process is carbonate dissolution , yet here is where soft sediment is transformed into hard limestone, the same waters that precipitate CaC0 3 when mixed may dissolve CaC0 3 ; while spectacular caves are being excavated, delicate fossils are being preserved. The environment has been studied more intensively than any other diagenetic realm, yet we are really only reasonably confident about processes and products that occur above the water table. The wide variety of limestones that result from meteoric diagenesis often beginning with the same sediment - are the end result of several reactions, governed by intrinsic and extrinsic factors (Figure 2) that may proceed at different rates for varying lengths of time. The most important intrinsic factor is original mineralogy: sediments composed of several CaC0 3 minerals , dominated by aragonite and magnesium-rich calcite (Mg-calcite), which are metastable with respect to fresh water, alter most; those sediments composed only of various calcite minerals alter least. Grain size control the rate at which these components change; fine-grained sediments and finely crystalline particles, because of their large ratio of surface area to particle size, change most rapidly, whereas
r
35
large skeletons of the same mineralogy change more slowly. Individual grains will also alter at diffe rent rates, depending upon the crystal size and relative amount of organic material. Finally, porosity and permeability of the sediment, which govern the rate at which fluids flow through and the length of time they are in contact with the sediment or rock, affect the speed at which changes take place. Perhaps the most "cruclal extrinsic factor is climate; in areas of extensive rainfall and warm temperatures diagenesis proceeds quickly and with great vigour, whereas in warm arid regions, because of little water, sediments are often unchanged for long periods, or are virtually unaffected by subaerial exposure. As the dissolving power of fresh water is largely controlled by biologically produced CO2 , the amount of large vegetation is also important. Finally time is critical: the longer a sediment is exposed in the meteoric environment, the more intensive the changes will be. THE METEORIC SETTING As outlined in the introduction to this series on limestone diagenesis, the meteoric environment can be subdivided in most general terms into vadose and phreatic settiogs.(Figure 3). ~he water table marks the top of the phreatic zone and represents the level at which hydrostatic pressure is equal to atmospheric pressure. The most common aquifers are unconfined, or open to the atmosphere, and the hydrostatic level is within the water-bearing unit. There are two types of gravity. induced flow in this situat ion (White, 1969). Under conditions of diffuse flow METEORIC DIAGENESIS SETTINGS -. C ;l"" fi ll ~ ~ : ~ : a / .' .
CONTROLS INTRINSIC MINERALOGY GRAIN SIZE POROSITY" PERMEABlUTY
EXTRINSIC CUIoIATE
.~
•
.,'-"<>.
ll";~I ..
ZONE
~ .
WATER TABLE . ~
VEGETATION
PHREATIC
TIME
ZONE
I
~
'r ..XING ZONE Figure 1 Theshore of Bimini, The Bahamas. The rock on the right is carbonate sediment that has been cemented by exposure to meteoric waters over the last several thousand years.
Figure 2 The principal controls on, and settings of, meteoric diagenesis.
Diagenesis
36
c c
Geoscience Canada Reprint Series 4
c
there is a well-defined water table and water movement obeys or nearly obeys Darcy's law. The sediment or rock has relatively homogeneous porosity and permeability characteristics, karst landforms are rare and subdued on a generally low-relief surface , small cavities are somewhat randomly arranged and flow is relatively deep. In contrast, under free flow conditions water moves through integrated conduits which func tion like a series of three-dimensional subterranean rivers. The water table is discontinuous or difficult to recognize at all, but there is nevertheless a continuous potentiometric surface. Diffuse flow aquifers with a well-defined water table are typ ically developed in newly exposed young carbonate sediments, but with time and diagenesis aquifers evolve into the free flow type characterized by conduit flow and a discontinuous water table. In this paper the focus is on diffuse flow aquifers. The other entirely different type of flow is confined. The aquifers are bounded by aquitards (impermeable layers), and are not connected vertically to the atmosphere. Waterflow is by pressure and will rise above the level of the aquifer inwells (artesian flow). Confined aquifers are particularly important as conduits by which fresh water can flow deep into sedimentary basins and considerable distances out beneath the continental shelf.
Unconfined aquifers characterized by diffuse flow are most usefully divided into the vadose zone, above the water table; the active, shallow, freshwater phreatic zone or lenticularphreatic zone (Jakucs, 1977), below the water table; and the sublenticular mixing zone, also referred to as the transition, mixing, diffusion or dispersion zone. The general relationships of these settings are shown in Figure 3. The Vadose Zone. Rainwater, meltwater or runoff enters a carbonate sequence either directly on bare rock or through a calcrete or soil. The zone of infiltration (Figure 3) is a realm of complex carbonate-water interaction. Although water may pass directly into the rock, in most situations there is a soil or protosoil on top of bedrock. Biological activity in the soil not only alters the composition of the fresh water but, in many cases, induces carbonate precipitation. This zone is subject also to evaporation or evapotranspiration, especially in the tropics, so that some of the downward percolating waters are drawn back up to the surface, again leading to carbonate precipitation. The underlying and more extensive zone of gravity percolation is characterized by pores variably filled with water andlor air or organically produced gas. Under conditions of vadose seepage water trickles through the
limestone via a network of tiny fractures or small pores. In contrast, vadose flow (Thrailkill, 1968) is the relatively rapid movement of water downward via joints, large fissures and sinkholes, ofte ~ 1.,":/ directly to the water table. ~""~ The Wafer Table. Dl<;S1:>W . ~ ,,,IT' As will be discussed later;-CUie water table is'a surface of great inportance in carbonate diagenesis. In older, prePleistocene limestones its former presence and position can often only be recognized approximately if at all because of long and complex diagenetic histories . Some criteria that can be helpful in delineating paleowater tables are presented later in this article. On relatively small , present-day limestone islands the water table is usually very gently bowed upward toward the island interior, as a rule rising less than a metre per kilometre (Figures 3 and 4). Beneath widespread carbonate platforms and mainland carbonate plains underlain by flat-lying strata the water table appears to have almost imperceptible relief (Figure 3). For example, two of the better known karst terrains of North America - the northern Yucatan and the southern peninsula of Florida (Back and Hanshaw, 1970; Back et et., 1976) regions of about 50,000 km2 - have watertable elevations which reach maxima of only a few metres and mostly less than 20 metres, respectively.
e
c
c c (
c C
c
C C C C
C C
C C C C C
C C C
THE METEORIC
C C C C
ENVIRONMENT
~
C
C C C SHALLOW PHREATIC LENTICULAR ZONE (FRESH WATER) ~
~
MIXING ZONE (BRACKISH WATER)
~
.>
?
r :
..
-: :"::::\::'2 -: >:MA~INE' ::
-
e
(.S~~ ~.~~~
c
: . : ...'.: . .::- ·.: · · .
. .
. ~ .. ..... :-:: ::.:.::::.-:: ::::.-:::.:..: .:.:.:-: . ..
::: ::::::::::::::::::::::::.:::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::.:.: ::::::::::.::::::.::::::::.:.::: ::::. :: :.::
c
:. '.'
PHREATIC ~ .
ZONE
<.".::'
-:.:.:-:.:-:-:-:..:.::-:-:-:-:.:.: -:-:-:".::-:-:..::":-: DEEP PHREATIC ZONE .:-:-:.:":.:-:..:.::..:.::..::..:.:.::...: : : : : : . : : :." ...":-. :-:- :-:-:"....: :::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::: .(~~l:~~. ~~:r.~~). .. .. . . . . . . . .... ... .. . .. . . . . . . . . . . ... .. .. .. . . ... . .. ... .. . . . . . . . . . . . . ... . .. . . . . . . . .. . .. . ... . . .. . . . .. . .. ..... . ... . .-. . . . . .. - .
::::::::::::::::.::::::: ::::::::::::::::::::': :::::" .::':::::::::::'.::::.: '::::::::.::
c C \.
,
Figure 3 A sketch outlining the main zones of the meteoric diagenetic environment.
C
C C C
(
~
C
p.
C
tJ'
C' ('
G-
r C C C
C (
o C-
O
o o o o o o c o a o o o o o o o o c
c
,f'.
Q~J fhJ~ kJ«I-!?Vv~ ;nC-rr~ .. :::;6171) L,.-. l r- ~ 0 - l .;;) ;:'V'v""'.- J.
-
~9' O l!.imestones "f The Meteoric Diagenetic Environment
(oJ"'
~~
...
fv'~t..fo"~-l~ (ODD~ clJ'~ > w~ en s: 6a ~~ ;.-v-- F"f7.s:k
f-
The r Jnticular Zane. This zone has as its lower boundary either an aquiclude or a water mass of different composition (Figure 4). A common situation today is one in which the freshwater lens "floats", iceberg style, balancing on a column of denser marine water beneath it. Early observers (DuCommun, 1828; Ghyben, 1888-89; Herzberg, 1901, cited by Fetter, 1980) noted in unconfined coastal aquifers that the depth to which the lenticular zone extends below sea level is about forty times the height of the water table above sea level. This relationship is commonly called the GhybenHerzberg principle (e.g., Todd, 1980). The precise configuration of the lens is dependent also upon, amongst other things, the permeability of the rock and rate of ground-water recharge (Buddemeir and Oberdorfer, 1986). Regard-
less, the important point is that only minorelevation above sea level can lead to deep penetration of meteoric water and attendant diagenesis. Inthiszonewater, which fills the available pore spaces, is actively moving. Except in regions of relatively high elevation, water movement in the lenticular zone is largely horizontal or subhorizontal (Figures 3 and 4) and directed toward the local baselevel of erosion which may be a spring, river or other inland water body, or the ocean. If the aquifer is very permeable and there is dynamic water movement in the lenticular zone, the interface may be deeper and displaced seaward toward the coast compared to what would be predicted by the Ghyben-Herzberg relationship, because the marine water beneath the lens is actively moving rather than static, as assumed for the
SMALL SAND CAY OR OCEANIC ATOLL
~ mm;:-~2 ~~@» '~; ~ ....... ... :: :;:..;: :.:: : : : : : : : : :: : :: : " .. .. .. ,
,
• •
I
•
. _. . __ ._-- - - _.. -- ... - .. .
'
...... .
,
LARGE CARBONATE PLATFORM
TECTONICALLY UPLIFTED TERRANE
~~~
,
l,I
CO~1l <;:kx-l
t ~
Figure 4 A sketch showing different scales and configurations of freshwater lenses .
...... ,'"
~y~-'~~~'\tJt!'~ (J" ~
C/V'-
-
•
....... ~ I
I'Yr.f t..
/Al 0
37
relationship. In strongly dynamic systems, "tongues" of fresh water may extend many tens of kilometres seaward beneath the continental shelf in confined aquifers (Manheim, 1967; Johnson, 1983). The Transition or Mixing Zone. The brackish water zone at the base of the freshwater lens is of great geologic as well as practical interest. It has been suspected since the early 1970s that this zone may be a site where earlY dolomitization take.s.place (see Morrow, this volume). Processes that occur in the transition zone are still poorly understood, but its hydrologic and chemical characteristics are becoming better known through studies such as those on which this brief discussion is largely based: Bear and Todd(1960); Back and Hanshaw (1970); Back et al. (1976); Vacher (1974); Plummer et al. (1976); Wigley and Plummer (1976); and Back et al. (1979, 1984, 1986). In general, the transition zone is a product of physical and diffusive mixing. It is thickest in more permeable strata where these mixing processes are more effective, and also it isthickest toward the coast because of both the higher velocities of ground-water flow near coastal discharge points and the influx of seawater drawn shoreward and upward in the shallow subsurface to replace the discharge. Salinity profiles in the transition zone - as well as profiles of water velocity, concentrations of single and multiple ion species and such water-dependent mass properties as electrical conductivity - have the form of an s-shaped curve (Figure 5). Because flow velocity and salinity are inversely related, profiles of the two essentially mirror one another; therefore, flow will be faster and waters more dilute in the upper part of the transition zone. The importance of these relationships here is that both conditions should have a bearing on whether carbonate will dissolve or precipitate. It is important to be aware that, although the transition zone is often portrayed as being quite thin compared to the freshwater zone above it, the zone can be quite thick in permeable carbonates of oceanic islands. Figure 5 illustrates the variations in transition zone thickness that Vacher(1974) found in the main freshwater lens on Bermuda, where the transition zone is
/~
Diagenesis
38
Geoscience Canada Reprint Series 4
E
W
UJ I-
~
~
\
-l
CD
et
I-
\
0:
Figure 5 A hydrological case study : map of the water table and profiles of the transition (mixed-water brackish) zone in the main (Devonshire) lens on the main island of Bermuda . (Mod ified from Vacher, 1974).
c
WATER
Page l Fm
c
(low permeab il i ty)
c
\
-c
\
15
~
!.-- <,'00% _ _
/ 1
-,
o
-,
-l UJ
CD
20L-
:x:
m
lll. UJ
o
o
50
100
c c c c c c c
./ /
......
./
1 Belmont Fm . '(h igh permeab il ity) ---l.--....::........:
--=-_--'
HYDROGEOLOGIC CROSS SECT ION WITH RELATIVE SALINITY CONTOURS THRU TRANSITION ZONE
C C
C C C......
TRANSITION ZONE PROFILES IN 5 WELLS
C C
REL. SALINITY, %
C IK m
V1lJ
Pagel Fm
C
C C C C
March 1974 ' . ,
AVG . ELEVATION OF WATER TABLE
C C
C C
®
C e-;
C~~~:'~N <. \~'t~~\ ,~,\~~ ~ \~\ ~,~~\\ ~ \;\~ \\:~\
(j)
SIMPLE CORROSION
~
WATER TABLE
~~ ~
\\\
;X
~~ \ \ \~
;---7"
\ \\
~
PRECIPITATION (C02 LOSS) ~ DEGASSING
L..----'~
FRESH WATER
.s>
.>
C C C C C
E MARINE WATER
l_
C
C MIXED WATER Figure 6 A sketch illustrating the areas of dissolu tion (corrosion) and prec ipitat ion of carbonate dur ing the exposure of calc ite limestone to percolating meteoric waters ; surface features are sketched in Figures 12 and 14, detail of square in Figure 21.
C C (.
<--(
t: (
Limestones - The Meteoric Diagenetic Environment
c c r o c o o o C~
\
o o o o o c o o o o o o o o o o o o o (
G
c c " ,-.
often as thick as or thicker than the freshwater lens itself (as defined by the 1% relative salinity contour) . To a close approximation the Ghyben-Herzberg relationship, which assumes no mixing of waters, would apply in effect to the part of the lens bounded by the 50% relative salinity surface and the water table (Vacher, 1978). Deep Phreatic. Water in the deeper phreat ic zone may be slowly moving or almost stagnant ground water, or in the case of young exposed carbonate platforms or islands, it may be seawater. An important result of the dominance of vertical, gravity-controlled seepage and flow in the vadose zone, compared with horizontal water movement in the phreatic zone, is that large solution cavities formed in the two zones have different elongation , commonly vertical in the vadose zone, but mostly horizontal along and for a short distance below the water table. Caves that clearly have formed in brackish-water transition zones have been documented along the eastern Yucatan coast (Back et el., 1979), a setting where one would expect them to be elongate parallel to the main flow direction, as suggested in Figure 3. Where dissolution in the phreatic zone is strongly influenced by flow rates, transition-zone caves may tend to be larger and more numerous toward the coast where flow rates are particularly high. Changes in base level due to tectonic disturbances or world-wide sea level changes must inevitably shift the positions and spatial relationships between the major hydrologic zones of the meteoric environment. DISSOLUTION AND PRECIPITATION It is well kn that limestone d' ...§pl\l en ex osed to fres Although the processes are easily observed and the general theory is well understood, some of the principles governing the dissolution and precipitation of carbonate are still poorly known. This is because we must deal with a multicomponent system simultaneously involving all three natural phases gas, liquid and solid. An additional complicating factor is that in perhaps the more important case, the transformation of sediment into limestone, we must also deal with a suite of several different minerals. Trying to understand all the
processes which are happening at once is much like learning to juggle: having finally mastered the art of keeping three objects in the air, you are handed several more and told to put on a good show! The reaction between carbonate minerals and meteoric waters in general is most simply expressed by the equation CaC03 + H20 + CO2 = Ca2 + + 2HCO; . The exact mechanisms of the CaC0 3 - CO2 - H20 chemical system are complex, involving a series of reversible and mutually interdependent reactions, all proceeding at different rates and each regulated by different equilibrium constraints. Recent reviews can be found in Bathurst (1975), Plummer et al. (1979), Bogli (1980), and Drever (1982). Under natural conditions this equilibrium reaction is represented by (1) dissolution of carbonate, (2) oversaturat ion and (3) precipitation of new carbonate. Diagenetic processes would be relatively simple if this reaction were all that was involved, but dissolution and precipitation rates are often slowed down or halted altogether by kinetic factors. Kinetics are particularly important in slowing down precipitation and thus maintaining supersaturation. Dissolution (Corrosion). Dissolution and precipitation of CaC0 3 are controlled in most natural situations by the flux of CO2 in and out ofthe water. Addition of CO2 by any means will drive the reaction to the right, resulting in dissolution. The simplest way is to increase pressure or decrease temperature. Kinetic effects do not seem to be significant. In the meteoric environment, dissolution (corrosion) is brought about in several ways (Figure 6), some of which are quite unexpected. The reader is referred to excellent treatments of th is topic by Thrailkill (1968) and Bogli (1980). Simple corrosion. The amount of CO2 dissolved in water open to the air depends upon the part ial pressure of CO2 (PC0 2 ) in the air at the air/water interface. Calcite is soluble to the extent of about 12 to 15 ppm in pure water (depending upon the temperature). Rainwater, however, is in equilibrium with atmospheric CO2 , which averages about 0.03% by volume (PC0 2 about 10- 3 .5 atm) and so is in reality an extremely weak acid. Thus simple cor-
39
rosion is the dissolution brought about by rainfall on bare rock surfaces (Figure 6). Biogenic corrosion. Air in the soil zone has a significantly higher Pco 2 than the atmosphere because of plant respiration and decay of organic matter. As rainwater percolates through the soil, addition of biogenic CO2 increases the CO2 content , typ ically to an equivalent PC0 2 of about 10 - 1.0 atm. Values of 1-2% are common, but in some poorly ventilated tropical soils may be as high as 20-25% (PC0 2 = 1O- 0 .7 atm). Consequently, waters emerging from below the soil are very "aggressive " chemically and can rapidly dissolve orders of magnitude more.carbonate than could ordinary rainwater. Mixing corrosion. Runnels (1969) has emphasized that mixing of two solutions which are both saturated with respect to a given mineral may lead to a mixture which is either supersaturated or undersaturated depending upon the character of the original solutions and on the minerals in question . These nonlinear relations can result in extensive subsurface limestone alteration . Mixing of waters in the subsurface can be expected to bea common phenomenon in all carbonate terranes. In the simplest cases this may involve (1) mixing of different meteoric waters, (2) mixing of meteoric and deeper subsurface br ines, and (3) mix ing of meteoric waters and seawater. When carbonate ground waters are mixed, the resulting saturation state depends upon (1) the PC0 2, (2) the temperature, (3) the ionic strength (salinity), (4) the degree of calcite saturation, and (5) the pH of the end member solutions prior to mixing. Chemically, the simplest type of mixing occurs when two bodies of meteoric water, isolated from contact with the atmosphere and each at equilibrium but with different CO 2 contents , come together (Figure 6). The resulting mixture will lie somewhere along the straight dashed line joining A and B in Figure 7, depending upon the proportions of each.The composition will move up and to the left until the saturation line is reached, with dissolution occurring in the process. This has been called "mixing corrosion " by many authors, but Wigley and Plummer (1976) prefer to call it the PC0 2 effect, because
I
Geoscience Canada Reprint Series 4
Diagenesis
40
there are other types of mixing corrosion as outlined below. This process is most effective when each of the waters mixed is saturated with respect to the calcite. If one of the waters is supersaturated, then the effect is diminished or negligible, as can be seen in Figure 7 by joining waters of compositions A and C. In natural systems this mixing can be envisaged as the mixing of high Pco 2 vadose water with low Pco 2 phreatic water. The effect is much reduced, however, if the vadose water is in equilibrium with a low PC0 2 (e.g., vadose flow), the phreatic water is in equilibrium with a high PC0 2 (because of slow degassing), or if the vadose seepage or ground water is supersaturated with respect to calcite. A similar corrosion effect may be achieved by mixing waters of different temperature. This can occur when vadose seepage is cooled at the water table. If the system is closed to PC0 2, then the effect is minimal; but if it is open to Pco 2 , changes as small as 1°C can be significant. In reality, however, in well-
tions . As might be expected, mixing of waters with varying pH may also lead to corrosionwith the degreeof undersaturation increasing with decreasing pH. Undersaturationcan also resultfrom a decrease in Ca2 + concentration in surface streams during times of high flow, which then descend to the water table by vadose flow and mix with ground water. Hydrostatic corrosion. As the hydrostatic head increases in the phreatic zone, so pressure increases, and thus fluids are able to dissolve more carbonate . This process was thought by many European karst workers to be at least as important as mixing corrosion in the subsurface (Jakucs, 1977). Oversaturation and Precipitation. Under this topic we must consider two reactions: (1) a first-order reaction in which precipitation occurs because of Pco 2 variations in the meteoric water, here called water-controlled precipitation, andlor (2) a second-order reaction in which precipitation is brought about by differences in CaC0 3 mineral solubility, here called mineral-controlled precipitation. .
ventilated caves this is more than balanced by precipitation from waters supersaturated with respect to calcite. When otherwise similarly saturated solutions with differing ionic strengths are mixed, appreciable undersaturation may result. In natural systems, because waters of differing salinity generally have unlike PC0 2 and pH values and are at different saturation states when mixed, the results may bequite variable; a spectrum of expected situations has been synthesized by Plummer (1975). Although the precise curves for any given pair of variables are different, the results follow the same general trend (Figure 8), showing that as seawater mixes with carbonate-bearing ground water the mixtures initially become increasingly undersaturated with respect to calcite. With continuing addition of seawater undersaturation decreases, so that mixtures containing large amou nts of seawater are oversaturated. Once in the field of oversaturatlon, precipitation should result but is generally inhibited kinetically by high Mg2+, PO~- and SO~- concentra-
)J 1
I
OVERSATURATION
/'VI
20
.;,\\
J,/.fr[. v[v.
{I\","
I' /1''1
J'/ Q'v'"
w· t-v(.I
>< 1.0 III Q
z z
oj: e( ~
~
Undiluted
0.8
Seawater~
0.6
OVERSATURATION Calcite-Saturated / Fresh Groundwater
l-
e(
tn w -0.2
I-
(j -0.4 ..J
-e (J 10
15
20
25
30 4
.
~.-.""'--'--'
-0.6 ::· :: :: :: ::: : :: :: :: :: : UNDERSATURATION ::. -0.8
5
· .· . . .
o
. .
.
20
. .
.
.
.
40
.
.
.
. . . .
60
.
. . . . .
80
.
..
100
% SEA WATER IN MIXTURE
EQUILBRIUM ~02(atm)X 1 0
Figure 7 A diagram illustrating the non-linear relationship between equilibrium PC0 2 and dissolved CaC03- The mixing of waters such as A andB in any proportion will result in undersaturation. The mixing of waters such as A and C will lead mostly to oversaturation. (After Thrailkill, 1968).
c c
c c c c c (
c
c c c c c c c
c c c c c c c c c
c
0.4
0.2
c c
Figure 8 A generalized diagram based on families of curves from Yucatan, Bermuda and Florida illustrating the effect on calcite saturation in groundwater by mixing waters of different ionic strength, in this case seawater and meteoric groundwater. The curve is a spectrum ofblends representing the mixing of seawater with a groundwater that is exactly saturated with respect to calcite.
c c c c c c e L C
C C C
C C C
( (
Limestones -
The Meteoric Diagenetic Environment
41
c.
c ( o c (
o (
c. ( o o o o o
o O· o o o o
o o o o o c r.
c
Water-controlled precipitation. Precipitation of calcite (low magnesium calc ite, i.e., calci te with less .~ mole% MgC03 ) can occur only under conditions of oversaturation. From the earlier equatl ori it can be seen that one way in which this may be achieved is by removing some CO2 from the system, most simply by heating the water, decreasing the pressure, or evaporation. Nearthe top ofthevadose zonethis occurs by either evaporation or loss of CO2, Assimilating plants also can take their CO2 from the water, causing precipitation. Vadose waters that are at equilibrium with a high PC0 2 degas and precipitate carbonate when emerging into a cave or the atmosphere with a lower PC0 2' If the CO 2 content of the water remains constant as in the phreatic zone, for example, where it is isolated from contact with the atmosphere, then theoretically the subsurface waters should be saturated with respect to calcite. But they should not be oversaturated, because there are calcite nuclei everywhere in the surrounding limestone available for precipitation. In addition, from the previous discussion of dissolution it is clear that if two waters saturated with respect to calcite are mixed, dissolution and not precipitat ion will occur, and the same situation seems to be true for seawater-freshwater mixing . Under natural conditions, however, ground waters are commonly supersaturated. Once the waters are saturated, calcite will precipitate if it can, but is often prevented from doing so by various inhibiting factors. This aspect is poorly understood and several things, such as kinetics of crystal growth, flux of CO2 and inhibition by other ions, serve to illustrate the problems . It is well known, for example, that more energy is needed to grow a crystal than to dissolve it, and so the rate of crystal growth is slow relative to dissolution . Also, the rate at which CO2 gas dissolves in water is very fast, but the rate at which a solution evolves CO2 to the atmosphere is very slow, and the rate of calcite growth is slower still. Finally, ions such as Mg2+, SO~ - and PO~ - are known to inhibit both dissolution and/or precipitation . If precipitation is prevented, then the saturation state will continue to rise, with the solution becoming progressively oversaturated, until a thermo-
dynamic drive is reached that is sufficient to overcome the kinetic problem. As rates are commonly the main problem given sufficient time precipitation will occur. At our present level of understanding , it seems that the phreatic lens and transition zones in calcite-only systems are zones of dissolution, and if, cementation does take place there, the process by which it does so is poorly understood. An exception might occur near the shoreline where CO2 evasion, induced by tidal pumping, is important (Hanor, 1978). Mineral-controlled precipitation. Aragonite, magnesium-calcite (HMC) and calcite (LMC) are the CaC0 3 minerals that make up most shallow-water carbonate sediments. The processes of dissolution described above are equally applicable to these minerals, but precipitation is another matter. Because we are now dealing with a group of minerals with differing solubilities, the main process is not one of calcite dissolution, but instead (1) dissolution of the more soluble phases (aragonite and magnesium-calcite), (2) resultant oversaturation with respect to calcite, and (3) precipitation of new calcite.
CaCO a
calcite ~ Ca 2 + + CO~aragonite Mg-calcite
The important point here is that, in this process, calcite is never dissolved because oversaturation is achieved and maintained by dissolution of aragonite, and to a lesser degree by magnesium-calcite. In addition, this process leads to a wholesale change in mineralogy, from metastable carbonates to calcite. Aragonite and magnesium-calcite are stable when bathed in warm seawaterbecause it was from this seawater, with its high ionic strength , that they were precipitated. All three minerals, however, are soluble in fresh water, but to differing degrees. Specifically, the solubility of magnesium-calcite progress ively increases with increasing magnesium content, as the calcite lattice becomes more and more distorted . Aragonite is also more soluble than calcite and has the same solubility as magnesium-calcite with about 12 mole% MgC03(Walterand Morse, 1984).Thus,
in relative terms, magnesium-calcite components with more than 12 mole% MgC0 3 should be the most soluble and so alter the most rapidly, followed by aragonite and magnesium calcites with about 12 mole% MgCOa and then magnesium calcites with less magnesium, the least soluble phase of all being calcite with virtually no magnesium . In addition, many of these components, because they are either biogenic or are precipitated from seawater and have variable microstructures, relatively high trace cation contents as well as organic and fluid inclusions, are slightly more soluble than pure minerals. In certain situations grain-microstructural complex ity can override thermodynamic restraints (Walter and Morse, 1985). In the Pleistocene Miami Oolite for, example, aragonite ooids have dissolved but magnesium-calcite foraminifera are unchanged (Evansand Ginsburg, 1987). All of the CaC0 3 mineral transformations take place by dissolution of one mineral and precipitation of calcite. What is important in this process is that, unlike the simple calc ite-water interactions described previously, large amounts of water are not necessary. Consider, for example, the aragonitecalcite reaction, which can be envisaged as occurring in a beaker of fresh water (Figure 9-1) in which two carbonate minerals, a piece of aragonite and a piece of calcite, are immersed. Immediately after immersion (Figure 9-2) both the calcite and the aragonite begin to dissolve until the water is saturated. As noted in the earlier section , this reaction is almost instantaneous. However, since calcite is less soluble than aragonite, the water becomes saturated with respect to calcite first, but is still undersaturated with respect to aragonite, so the aragonite continues to dissolve. But now (Figure 9-3) the water is oversaturated with respect to calcite, and as there is a calcite crystal present, CaC0 3 will precipitate as calcite crystal cement. With this precipitation the solution is now undersaturated with respect to aragonite again, and so more aragonite dissolves. Thus the process continues, with aragonite continuing to dissolve but never contributing enough CaC03 to reach saturation because of continuing calcite precipitation. Eventually (Figure 9-4) all the aragonite is gone and the process stops, the final product
c
42
Diagenesis
Geoscience Canada Reprint Series 4
c (
being an original calcite crystal or grain surrounded by newly precipitated calcite. This process, in which dissolution is followed by precipitation of a secondary carbonate phase, is referred to by some as incongruent dissolution (Turner
et al., 1986). Quite clearly the same reactions could begin all over again if a new aragonite crystal were introduced. Theoretically, the same beaker of water could be used to alter large amounts of aragonite and magnesium-calcite to calcite. In reality, what would happen is that the ions from the aragonite and magnesium-calcite that do not fit into the new precipitated calcite, especially Mg2+ and Sr2+, would become so concentrated in the water as to inhibit the precipitation of calcite. In the intermediate stages, though, because of the rising trace-element concentration in the water and selective absorption on the calcite, these elements would be partitioned into the new calcite in small but ever-increasing amounts. An analogous situation exists in nature. Waters at the recharge end of a freshwater aquifer have low trace element concentrations, but by the time they reach the distal, discharge point, because they have taken part in so many reactions along their path, they have relatively high trace element concentrations. Calcites precipitated in the distal parts of the aquifer will therefore have higher trace element concentrations compared to those precipitated in the proximal part (Kinsman, 1969). An important point that must be remembered in all of this is that once the aragonite and magnesium-calcite have all dissolved to produce new calcite, the process stops. The intrinsic driving force for calcite precipitation
-
throughoutthesystem isgone, and now, in a monomineralic all-calcite system, precipitation is localized and largely water-controlled. Kinetics also are important in this system. Under natural conditions, if the freshwater diagenetic alteration of aragonite and magnesiumcalcite to calcite were governed solely by equilibrium processes, then ground waters would never be very oversaturated with respect to calcite, whereas in reality they are. Also, if the kinetics of the processes were rapid relative to the age of the ground waters, all water should be in equilibrium with calcite, but it is not. In most instances, the waters are oversaturated with respect to calcite and just undersaturated with respect to aragonite (Plummer et al., 1976). Thus non-equilibrium processes such as dissolution, precipitation and crystal growth as well as the flux of CO2 in and out of the ground water are again important. For example, it has been estimated that the rate of aragonite dissolution in "fresh" meteoric water is 100 times faster than the rate of calcite precipitation (Schmalz, 1967). Thus it is important to remember that waters in the meteoric environment, as long as they are still in contact with metastable carbonates, may be several times oversaturated with respect to calcite, not because the nuclei are not there to precipitate on, but because the processes are slow in the time frame of water movement. In summary, the mineral-controlled processes involve dissolution of aragonite and magnesium-calcite and precipitation of calcite. This new calcite, called diagenetic calcite (dLMC) by some workers, can be precipitated in open voids ascement between grains or
within grains, orcan be precipitated on a microscale inside a particle, replacing the original aragonite or magnesium calcite. Styles of Diagenesis. From the foregoing, it is clear that meteoric diagenesis is driven by two rockwater reactions, one involving congruent dissolution and generally watercontrolled alteration, and the other incongruent dissolution which is minerai-controlled. Congruent dissolution means that the entire solid dissolves and generally no solid phase is precipitated atthe site of dissolution.lncongruent dissolution means that part of the solid dissolves to leave behind a solid phase different in composition from the original, and the reaction is usually accompanied by precipitation of a new phase (Drever, 1982; Turner etaI., 1986). The new phase maydiffer in crystal form from the original but havegenerally similar composition, like dLMC compared with dissolved aragonite, or maydiffer in composition but essentially not in crystal form, like dLMC compared with precursor HMC, or may have both new crystal form and new composition, like dolomite compared with HMC or aragonite.
DH 0 2 (1)
r
~
r:
Ca++ fS r ++@ i8£HC0
(2)
3
~ tIl
O~HCO= 3 (3)
c
c c c
c
c
c c c c c c; c c c c
c c c c c
~ ca::,
c c c
c c c c c c
DISSOLUTION & PRECIPITATION ~
~~ ~
c
(
r:
ca
c c
H ,
Sr++ HCO= 3
(4)
Figure 9 A sketch illustrating the simultaneous dissolution of aragonite and precipitation of calcite in a beaker of distilled water.
Figure 10 Modern hard laminated caliche
C
(e) developed on the upper part of Pleisto-
cene reef limestone, Barbados. (Scale in em).
c C l
C
( (
Limestones - The Meteoric Diagenetic Environment
43
C r:
(
o c o
() ( (
c () o
c o
o
o
o o
In a suite of metastable minerals, either water-controlled alteration (congruent dissolution) or mineraI-controlled alteration (incongruent dissolut ion and dLMC precipitation) may be operative, depending mainly on the amount and flow rate of meteor ic water through the system . The difference between these processes is well illustrated in the vadose zone. Water coming into contact with metastable aragon ite and high-magnesium calcite directly beneath the soil profile will dissolve all species until it is saturated, but as this solut ion percolates downward, alteration will occur because of differing minerai solubilities, not changes in Peo z' But if these waters should subsequently emerge into a cave open to the atmosphere, then precipitation due to degassing and not different solubilities will take place. In contrast, if sediments in the vadose zone are entirely calcite, then dissolution beneath the soil and precipitation in the cave will still occur, but no mineralogical changes will take place in the bulk of the vadose zone. Congruent dissolution takes place in both situations, in response to flowthrough of water in large enough volumes to prevent oversaturation for calcite at the sites of dissolution. Watercontrolled alteration under conditions of this sort might be thought of as congruent diagenesis . Still using the vadose zone as example, incongruent dissolution and precipitation of dLMC will take place where
the sediments are still aragonite and high-magnesium calcite, provided that the flow of meteoric water is slow enough that solubility differences can make the precipitation operative. The alteration is mineral-controlled and might in fact be viewed as incongruent diagenesis. Diagenesis of this kind may often be dominant in zones of relatively slow diffuse flow. Because the flux of water through the system is such a dominant control, it seems quite possible that both mineraland water-controlled alteration (incongruent and congruent diagenesis) take place simultaneously and side by side in some sequences. For example, in a newly exposed sequence of little-cemented aragonitic and high-magnesium calcite sediments , relatively slowmoving diffuse flow may occur in most places, fostering incongruent dissolution and mineral-controlled alteration , but higher volume channel flow may develop locally along incipient joints or root molds, such that the dissolution there is congruent and the diagenesis is water-controlled. In mineral-controlled, incongruent diagenesis, because each kind of grain is different mineralogically, dissolution and other forms of alteration will tend to be " fabric selective" (Choquette and Pray, 1970). In water-controlled, congruent diagenesis, reactions and reaction rates are governed by differences in crystal size, grain size, amount of adsorbed organic matter, porosity and
c
c
(
l'
permeab ility, or the presence or absence of fissures and cracks . Most changes will not be governed by the fabric of the limestones and so will be cross-cutting or "non-fabric selective", In the following sections , we first outline the processes of water-controlled (congruent) diagenesis because they are applicable to virtually all carbonates in relatively high-flow systems. Next,we describe processes that occur during mineral-controlled (incongruent) diagenesis because they appear to be widespread in limestones containing metastable suites of CaCOs minerals where there is relatively slow, diffuse flow of meteoric water. PART 1 - WATER-CONTROLLED METEORIC DIAGENESIS
The Rock-Air Interface. The limestone surface in the meteoric environment , whether open to the air or beneath a soil cover, is the first carbonate to come in contact with acidic waters. Limestone surfaces forming the upper part of the zone of infiltration are usually altered intensively and develop features that are easily discerned in outcrops, cores and thin section and so are valuable indicators of subaerial exposure. Esteban and Klappa (1983) havesynthesized what was previously a very diverse and widespread literature concerning these structures, and much of the following section is condensed from their article. They concluded that there are really two end-member diagenetic facies here: (1) the caliche facies and (2) the surface karst facies . These diagenetic facies are not mutually exclusive: karst and caliche may coexist at anyone time and overlap in any one area. Whereas karst can develop under all climatic conditions, caliche is attribu table to a generally semi-arid climatic reg imen . Before outlining the attributes of these two facies, it should be stressed that many subaerial exposure surfaces may not contain any diagenetic features, so the absence of such features in the rock record does not necessarily mean that the limestone was never subaerially exposed. CALICHE
I'
,
Figure 11
Ancient caliche (C - between arrows) illustrating good alveolar texture , developed in the Carboniferous Newman Limestone, Kentucky. (Scale in em).
Caliche (or calcrete or duricrust) is that carbonate-Iithified port ion of the soil profile developed commonly, but not exclusively, on carbonate sediments and rocks (Figures 10 and 11). Because caliches are accretionary and thus
44
commonly preserved in the rock record, they have been intensively studied over the last twenty years. Caliche zones range from centimetres to metres in thickness and may be complex in the extreme, with part of the profile resulting from alteration of the underlying limestone, part from precipitation of new calcite and much from a combination of both sets of processes. Characterized by such features as irregular, often laminated crusts, rh izoconcretions, diagenetic peloids, ooids and pisoids, breccias, clotted micritic and chalky carbonate, leached and vuggy porosity and microborings and iron oxide, these horizons commonly stand out in a sedimentary sequence. Because their features are similar to some formed in the marine environment, however, they are often overlooked (James, 1972; Read, 1976). Lithologies. As with other soils, modern caliche is vertically zoned and upward is composed of four rock types: (1) massive chalky carbonate, (2) nodular and crumbly carbonate, (3) irregular plates and sheets, and (4) a compact crust or hardpan. The position and development of these lithologies in a vertical sequence is highly variable (Figure 12),but the massive chalky carbonate most commonly is found at the base, grading downward into underlying rocks or sediment. The caliche hardpan, usually cream to brown in colour, may range from 1 mm to over a metre in thickness, is composed of microcrystalline to cryptocrystalline calcite, and ranges from structureless to horizontally laminated. Thicker hardpan generally is brecciated and exhibits dissolution and precipitation features, pisoids and rhizoconcretions. The upper surface ofthe hardpan in semi-arid climates commonly is colonized by lichens which may cause textural and fabric changes as deep as 2 cm below the surface, producing spongy, microscopic or micritic layers composed of organic-rich and organicpoor mm laminations. Thus, the fabric of the hardpan may resemble that of a stromatolite and, in fact, Klappa (1979) has called these structures "lichen stromatolites". Criteria that can be used to differentiate calcrete crusts from marine stromatolites have been outlined by Read (1976). Beneath the hardpan or, if the hardpan is not developed, directly beneath
Diagenesis
soil cover, there is a series of thin, commonlyfriable,horizontaltosub-horizontal plates or sheets which are separated from one another but join and bifurcate laterally. This zone, which generally grades down into nodular calcrete, may be several metres in thickness. Caliche nodules (glaebules of soil terminology) range from silt to pebble size and from spherical to irregular or cylindrical in shape, and may be isolated or coalesced in arrangement. Thesestructures, which are commonly concentrically laminated, have been called caliche ooids, pseudo-ooids, oollths, peloids, pellets, pelletoids, pisolites , coated particles, and, more recently, have been included by Peryt (1983) under the term vadoids. Chalkycaliche is often well developed and may be up to one metre in thickness, but is generally absent in areas where the rocks have high initial porosity. The white- to cream-coloured uncemented carbonate is composed of siltsized calcite grains, commonly microspar, with scattered nodules . The transition zone down into the host carbonate is characterized by strong evidence of in-place alteration and replacement of the original carbonate. Fossils, for example, are often preserved but embedded in calcified host material, or coated with calcite laminations. If developed on bedded carbonate , alteration takes place preferentially along bedding and joint planes. This basal zone may be absent or metres in thickness. Because roots may penetrate any of the zones, rhizoconcretions - features formed by the precipitation around or
Geoscience Canada Reprint Series 4
replacement of roots by calcite in the form of root molds, casts of tubules , concretions or actual petrification are a diagnostic fabric of the caliche profile. Roots both contribute to the formation of platy calcrete and leadto brecciation . This brecciation, which also may be due to expansive crystallization , results in teepee and pseudo-anticline structures. Included in many calcrete profiles are black pebbles. The blackening may be due either to trapped organics within crystals (Ward et al., 1970), metallic oxides precipitated on fungal hyphae (Esteban and Klappa, 1983) or fires (Shinn and Lidz, 1988). Petrography. The petrographic appearance of caliche is best described as messy. The wide variety of precipitates are reviewed by Chafetz et al. (1985). Predominant fabric is a clotted peloidal micrite with microspar-filled channels and cracks. Grains and fragments of limestone are separated from thesurroundingmatrixbycircumgranular cracks due to shrinkage and expansion . Fragments of original limestone , or individual grains if a relatively young carbonate sediment, may be coated with micrite laminations, which in turn may be connected by thin laminations bridging grains. A complete spectrum of partly replaced (micrit ized) particles to completely altered grains may be visible. Replacement of some grains and not others commonly results in a "floating texture" of particles in a micrite matrix. Voids in the caliche profile often display an alveolar texture (Figure 11),or a
.. ~
CALICHE PROFILE
~?~Y
-
-
-
~
c
c
c c c c c ( c c c c c c c C E C C
C C
i
0
~~~~~~~. LIMESTONE OR SEDIMENT
VARIATIONS
l(}\tJ'
HARDPAN PLATES & CRUSTS - - -
o~
c c c c c c c c c c c c
nm !
Figure 12 Sketch of a caliche (calcrete) profile showing all the elements (left) and some of the observed variations from various modern examples (right).
c e c c ( c (
c c
c
r:
Limestones - The Meteoric Diagenetic Environment
45
r: r:
o r:
o
o o o
o o o
o o o r:
....... r
"'o""
o o ...... r
network of anastomosing micrite walls 100-150 JLm across inside cylindrical to irregular holes. Some open voids may be floored with geopetal crystal silt (Dunham, 1969) which filters into cavities from the chalky zone . In other instances, the rock has a more vermicular texture, or dense networks of micrite tubules and rods in a micr ite matrix yielding a spaghetti-like texture. In addition to micrite, needle fibres of calcite a few microns to tens of microns long are common . They may be a randomly oriented mesh (the " Iublinite" of soil science) and occur in voids within the hardpan; they may form platy horizons or nodules; or they may be tangential with the needle-shaped crystals arranged in a band. Another distinctive caliche precipitate is Microcodium or elongated to petal-shaped calcite prisms or ellipsoids or bell-shaped clusters which Klappa (1978) proved to be calcified mycorrhizae (soil fungi and cortical cells of higher plant roots). Variations. Depending upon the length of exposure and climate, the caliche profile will change with time. Following the development of weathered detritus and a protosoil by the colonization by lower plants (lichen, fungi, algae, bacteria), the caliche profile becomes differentiated into two separate horizons; water percolation and plant roots generate the upper transition zone; precipitation of calcite forms the chalk zone. As accumulation of CaCOs continues, a point is reached where soilforming organisms can no longer maintain viability so that soil-forming processes decrease and cementation and hardpan formation increase, fossilizing the profile. Finally, the profile is so Iithified that it is just another limestone and the process begins again, a protosoil forms and plant roots penetrate the hardpan, leading to brecciation and rhyzoconcretions. Seminiuk and Searle (1985)differentiate between vadose or rhizoconcretionary calcrete and ground-water calcrete (massive, mottled, laminar) in the zone of capillary rise just above the water table. As with the other features described in this article, there is a gradation from the wholly meteoric formation of caliche inland to a transition zone along the shoreline. It is common to find the hardest crusts developed in the supratidal
zone and just above because there the sea spray supplies carbonate to the crust , enabling precipitation to continue at a more rapid rate; as would be expected , the calcite there has a higher MgCOs content (James, 1972). In some areas, such as the shoreline along the southern Persian Gulf where virtually the only source of water is sea spray, caliche crusts and pisolites develop but are composed of Mg-calcite and aragonite (Scholle and Kinsman, 1974). In summary, the caliche zone is a diagenetic soil profile which involves alteration of original limestone or sediment and precipitation of new carbonate, both of which are modified or partly controlled by plants of various types. This assemblage, which changes character with time , is probably the most obscure and confusing of all diagenetic zones. Calcrete may be locally transitional into terra fossa and similar soils (Goldhammer and Elmore, 1984; Wright and Wilson, 1987). Detailed geolog ical study ofthese featu res isjust beginning . Ancient caliche. There are few welldocumented examples of caliche in the rock record (Estebanand Klappa, 1983), and none from the Early Paleozoic. SURFACE KARST Surface karst or exokarst, because it is a dissolution and not an accretionary phenomenon , is difficult to recognize in the rock record. In vertical section, it is represented only by irregular bedding contacts, and so only on well-exposed bedding planes can the style be determined with any confidence. Paleokarst has only recently been recognized by sedimentologists (James and Choquette , 1988),although there is awealth of information about modern karst in the geomorphological literature (Jennings, 1971; Sweeting , 1972; Jakucs, 1977; Ritter, 1978).Karst occurs in all modern climates, is the result of normal and biogenic corrosion, and is slow under arctic conditions but extremely rapid in the tropics. The process of dissolution takes place both on bare rock and beneath soil, but generates somewhat different features under the two conditions . Corrosion is especially active beneath soil and intensive in the humid tropics because of high temperature and increased vegetal litter and soil. Although biogenic corrosion by carbonic acid is the most important pro-
cess beneath tropical soils, dissolution duetoother acids, such asfulvic,crenic, sulphuric and nitr ic, is also significant, but unlike the reactions associated with carbonic acid, many of these reactions are irreversible. The style of development also depends a great deal on the porosity and permeability of the limestone itself, with surface features developed best on well-lithified limestones. Regardless, it is clear that climate is a major controlling factor in the development of karst. Surface alteration at high latitudes or altitudes is restricted to slowly developed karst. In temperate or Mediterranean-type climates, karst and calcrete are common, but their development is seasonal. In general, deserts display poor karst and in semi-arid regions caliche is common. But, in regions of high rainfall, karst forms a fantastic array of towers, jagged ridges, hogbacks, canyons and sinkholes which, together with extensive vegetation, make such regions pract ically impenetrable to surface travel and poorly known. Even though surface karst is most intense in the humid tropics in areas of substantial relief, in the past most carbonate platforms were exposed as vast flat plains. just above sea level with little relief. Such a platform today in the Nullarbor Plain in southern Australia (oneof the world's largest karsts, ca. 100,000 km 2 ) which, because of the semi-arid climate, is heavily mantled with calcrete, exhibits minimal solution sculpture and contains less than a score of caves that reach to the water table and relatively few shallow caves (Jennings, 1971). Other level karst plateaus, such as the Yucatan Peninsula (Stringfield and LeGrand, 1974) or Bahama Banks, because they are in a rainy climate,look like Swiss cheese from the air,with innumerable dolines or sinkholes (Figure 13). On the ground, the limestone is fluted or pitted by solution sculpture and dotted with small, solution basins (Bourrouilh, 1974). Surface karst features , from a practical standpoint, can be subdivided into small-scale solution sculpture or features smaller than a metre in size which can be seen in outcrop-size exposures, and large-scale features tens of metres to kilometres in size which, although sometimes seen in outcrop, are better thought of as karst landforms.
( 46
Solution Sculpture. These dissolution or corrosion features, developed either beneath a soil or on bare rock, are generally called "karren" in the English and German literature, or "Iapies" in the French and eastern European literature. These features have a wide variety of shapes, and as B6gli (1980, p. 53) succinctly states, " the multiplicity of possible karren forms makes a morphological system endless". Nevertheless, there are a number of common, recurring shapes which are outlined below and diagrammed in Figure 14.As ageneral rule, the sculptures tend to be sharp and jagged on exposed rock, but rounded if below a soil cover. (a) Small karren: In temperate climates bare limestone surfaces are generally smooth (although the larger-scale sculptures may be sharp edged), but, in tropical climates or where there is rainsplash and mist, the rock surface is textured by tiny, slightly elongate , cuplike pits generally less than 3 cm in diameter called "cockling", which intersect at knife-sharp edges and give the rock a crinkly or cindery appearance (Figure 15). In tropical regions, a similar texture of black-coated , jagged pinnacles, marked by delicate, lacy dissection, has been called "phytokarst" (Folk et al., 1973) because of the intensive activity of endolithic microflora in its formation. These smallest of features can form very quickly, in as little as four years. They are especially characteristic of limestones with good intergranular porosity and are well developed in the intertidal and supratidal zones. (b) Medium-sized karren : These slightly larger features on bare rock develop in the form of solution runnels and solution pans. Two of the most distinctive forms are solution flutes (rillenkarren) or razor-sharp finely chiselled runnels (Figure 16) and solutiongrooves (rin nenkarren) or slightly larger runnels with rounded furrows, both of which develop on slopes of wellcemented hard limestones. All of these forms develop quickly and may reach lengths of metres in the tropics . These may be straight on steep surfaces or meandering on flattish surfaces. On more horizontal surfaces,hee/-printkarren (frittenkarren), in the form of crescentic steps, are more common.
Diagenesis
Solution basins (kamenitza or kamenica) are small, flat-bottomed pans or basins often exhibiting overhanging edges which characteristically develop on horizontal limestone sur-
Geoscience Canada Reprint Series 4
faces (Figure 17). They can develop as rainwater collects in small depressions, beneath humus patches or in the intertidal zone through the combined action of mixing corrosion and bioerosion.
C C C ( ( (
(
C C C (
(
C C (
C
C C C Figure 13 Areal photograph of large sinkholes, several hundred metres in diameter, developed in Pleistocene limestone, Andros Island, TheBahamas, and now filled with water or vegetation.
SURFACE SOLUTION SCULPTURE (KARREN) FLUTES
HEEL-PRINTS
GROOVES
( ( ( ( (
C C C ( (
C BASINS OR PANS
CLiNTS & GRIKES
ROOTS
-~-
( ( (
E
C (
(
C Figure 14 Sketches of different types of surface solution sculpture .
(
C (
C
( r: Limestones - The Meteoric Diagenetic Environment
r
47
o
r/
( (
o r. o o c o o o o o o o o o o o o
Figure 15 Syngenetic phytokarst characterized by sharp irregular cockling and numerous irregular pores in Pleistocene limestone, Andros Island, The Bahamas. (Arrow points to hammer for scale).
Figure 16 Ancient rillenkarst (Late Devonian - Early Mississippian) developed on Ordovician limestone, western Newfoundland (scale in em). The sculpted runnels are partially to completely filled with light grey Mississippian limestone.
o o
o
o o o r
,. Figure 17 Modern solution pans (kamenitza), developed on Pleistocene limestones, Bermuda and partially filled with modern carbonate sand. (Scale bar in 2 cm divisions).
48
(c) Large karren: Once covered by soil, increased corrosion tends to smooth all edges and points, remove small karren forms and widen and deepen pre-existing features (Figure 18). The general term rundkarren is applied to these features, which range from wavekarren or wavy surfaces like corrugated tin in temperate climates to cavernous karren in the tropics. Forms that seem to develop best under these conditions are clints and grikes (f1achkarren) or flat-topped blocks (clints) bounded on all sides by solution-widened joints (grikes) which are rectangular to diamond-shape in plan. While most workers (e.g., Sweeting, 1972; Bogli, 1980) think that these develop beneath soil, Purdy (1974) suggests that they may also develop on bare rock surfaces. At joint intersections, cylindrical pits several metres deep may be produced, which may evolve into solution pipes or potholes. As the grikes widen with time, the surface passes through a phase referred to as kluftkarren to a stage of solution pinnacles (spitzkarren) or assemblies of upward-pointing pyramidal or projectile-shaped bodies of rock separated by interconnected clefts or basins. While these pinnacles may develop beneath soil-cover inland, they also form on bare rock and on coastal carbonate rocks in the intertidal zone and higher. In addition to the soil, the roots of plants and trees in the tropics etch tortuous single and branching chan nels in the rock (hohlkarren) that may reach depths of 25 metres. Locally, the channels may criss-cross the limestone so densely as to turn it into something resembling a sponge, in places with more than 75% pore space. In four to ten years, for example, a hairline crack can be turned into a rock channel of arm or thigh thickness (Jennings, 1972). In Pleistocene carbonates, the complex interrelationship between karst and caliche is well demonstrated by common pipes which owe their origin to root corrosion but are lined by calcrete and filled with rhyzoconcretions, formed by organically-induced precipitation. Karst Landforms. The most widespread of these larger features are dolines or sinkholes which may be funnelto bowl- to flat, dish-shaped structures, metres to kilometres in diameter and up to 100 metres deep (Figure 13). They owe their origin to dissolution beneath a
Diagenesis
soil cover or subsurface dissolution and either gradual subsidence or collapse. Breccias that fill collapse dolines are characterized by mixed soils and speleothems together with the v-shape of fallen overlying beds. The other large-scale features, such as cenotes, cockpits, mogotes, uvalas, karst valleys and poljes, are more in the realm ~f geomorphology, and excellent summaries of these landscapes can be found in general texts 'o n karst (e.g., Jennings, 1971; Sweeting, 1972; Jakucs, 1977; Bogli, 1980). SUBSURFACE KARST Many characteristics of dissoiutional cave systems in carbonate rocks are discussed in detail by Ford (1988). The Vadose Zone. The zone of inflitration is dominated by processes of intensive physicochemical and biological corrosion related to intensive organic activity (Esteban and Klappa, 1983)::Vertical.caves- are ' pronounced .here. iand collapse breccias may be common. Precipitates are usually fine grained (moonmilk) and/or globulites (popcorn). As the CO 2-rich acidic waters emerge from below the soil they dissolve carbonate quickly, become saturated, and lose their aggressiven!3ss within a few decimetres of. the soil -zone. Once the waters -are saturated -in -an -all-calclte carbonate host rock, no further disso-
Geoscience Canada Reprint Series 4
lution or precipitation takes place. Consequently, this zone is relatively "inactive". Reaction does take place, however, if the limestone grains are extremely small and so reactive, or if mixing of vadose seepage and vadose flow occurs, or if large cavities open to the atmosphere are encountered, or if there are changes in temperature. Conversely (Figure 19), if the sediments are composed of a mixture of aragonite, magnesium-calcite and calcite, each with a different solubility, then this is a zone of intensive activity (see part 2 of this discussion). Under conditions of vadose flow, the system is effectively open to CO2 and the aggressiveness of the waters after initial dissolution is governed by the diffusion rate of CO2 across the waterair interface. If the channels are large, rapidly flowing water may be aggressive to depths of 100 metres or so (Thrailkill, -1968). Dissolution also occurs, when - vadose seepage and vadose flow meet, by mixing corrosion. Many caves found in the vadose zone are due to dissolution at perched water tables or are relic water-table caves abandoned when the water level dropped. It is clear that some caves, however, particularly those strongly elongated vertically, form in the vadose zone itself, either to survive - unchanqed orto be modified by phreatic processes.
c c ( c c c c c c c c c c c c c c c c c c c c c c c c C
E C C C C
c e c c c
c Figure 18 Ancient ;undkarst developed on Ordovician limestone: the depressions are filled by light grey Mississippian limestone (arrows), western Newfoundland. (Scale bar in 2 em divisions).
c c
c c
r:
Limestones -
o o o
The Meteoric Diagenetic Environment
The Water Table. This realm, including the base of the vadose zone, the water-table surface and the upper part of the phreatic zone , is one of intense chemical activity, perhaps more intense than anywhere else in the shallow subsurface.
o o
o o o
Dissolution. Most dissolution occurs at or below the water table, in the phreatic part of this realm . (a) Processes: If the system is entirely subterranean, closed to outside air, then this region where vadose and phreatic waters mix is a level of intense
METEORIC DIAGENESIS EFFECT OF ORIGINAL CARBONATE MINERALOGY
---I
o
o o o
49
Mg-CALCITE ARAGONITE CALCITE
CALCITE
I g c w
NO ALTERATION
(I)
>
t
EXTENSIVE ALTERATION
wcN
DISSOLUTION & PRECIPITATION WCA WCA~ ~~:r-<~~!(.::..-(-W-C-A-)-1WATER TABLE f---:(::-M-C--A"'-)--='!.4~.Yl~~ MINOR DISSOLUTION NON-FABRIC ) ( "SELECTIVE
EXTENSIVE DISS. & PRECIP.---r--,· ( FABRIC & )
~~~cM~~C
: : ",::':".".
"/=!-- DISSOLUTION (W C A)
o o
o o o
NO ALTERATION
DISSOLUTION (W C A) NO ALTERATION
Figure 19 A sketch illustrating the different diagenetic processes that affect a carbonate composed 0.' calcite (sediment or limestone). versus one composed of a mixture of magnesium-calcite. aragonite and calcite. Water-controlled diagenesis (WCA) is driven by evasion and Invasion of CO2 while mineral-controlled diagenesis (MCA) is driven by differences In CaC03 mineral solubility.
o o o e o
Figure 20 Large dissolution cavities (ca . 1.5 m in diameter) excavated into Permian carbonates, Carlsbad Caverns. New Mexico - note the lack of decoration.
corrosion. It is here that many caves are formed. Phreatic water movement is generally horizontal and the origin of the waters may be allochthonous, from outside the local area . Since the level of the water table varies seasonally, the water table may rise into the vadose zone during times of flood, resulting in a whole zone of caves at this level. (b) Features: There area wide variety of dissolution features, such as passages, chan nels and shafts (Figure 20), which can be related to both the type of water flow and lithology (Bagli, 1980; Ford, 1988). Smaller cave karren are superimposed on the larger cavities. The walls and ceilings may be smooth or pockmarked with corrosion pockets, dimples and pits, or textured by rills and grooves similar to surface karst and especially common where vadose waters emerge. Pendants of relic limestone in the form of cones of rock may hang down from the ceiling or protrude up from the floor. Scallops or crescentic, shell-shaped dissolution features can form on the surface where a stream flows over limestone, but are particularly common in caves. The crests in these features lie transverse to flow and the steeper slopes face downstream. Precipitation. Most precipitation, but not all, takes place in the meteoric part of this realm, in the upper parts of caves and cavities above the water table. (a) Processes: Regardless of the mineralogy of the sediment or rock, if the system is ventilated with atmospheric air, the reverse occurs. High Peo, vadose waters emerging at the top of water-table caves equilibrate with low-Peo, atmospheric air, degas and precipitate carbonate. Exactly the same phenomenon occurs as springs emerging from a hillside precipitate travertine (Julia, 1983; Chafetz and Folk, 1984). Flowing waters in the cave are also at equilibrium with atmospheric air, and so precipitation only occurs because of vanattons in temperature and turbulence. It is a paradox that once caves are open and can be visited by geOlogists, the processes which formed them largely have ceased, and they are in the process of being decorated by various precipitates (Thrailkill, 1976; Moore and Sullivan, 1978).
c Diagenesis
50
(b) Features: Subterranean precipitates can be subdivided into two types: (1)calcareous tufa - soft porous rock to which plants contribute significantly,and (2) calcareous sinter (speleothems) nonporous, crystalline, hard limestone. The most common tufa is moonmilk, which is composed of microscopic carbonate crystallites or lublinite (random needle crystals); although plastic and containing from 35% to 75% water when precipitated, upon drying it turns into a powder. Plants playa significant role in the precipitation by removing CO2 from the water, and the resulting crystallites are enmeshed in a net of filaments from bacteria, actinomycetes and algae. Sinter is precipitated either from thin films of water flowing over the rock or in pools of water. Precipitation from water films results in crystals that grow outward from the rock surface for only a short distance, with growth terminated at the air-water interface (Figure 21). As a result, the outer surface of the calcite is smooth, with scalenohedral terminations never developing. This style of precipitation is typical of the vadose part of the cave, and forms flowstone and dripstone (Figure 22). Dripstone ranges from draperies to "soda-straws" to stalagmites and stalactites to helectites. Deposition starts from a drop of water on the ceiling, with precipitation beginning at the area of attachment. Since CO2 is given off from the surface of the drop, precipitation starts there, causing a ring of calcite. Ring is added to ring to form a tubular structure - the "soda straw". Water flowing as a film over the exterior surface of the soda straw deposits a second type of calcite which constructs the familiar conical or candle-shaped form . This outer coating is most commonly composed of elongate calcite crystals which, in sections cut normal to the stalactite long axis, have their long and optic axes radially disposed. Some of the growth surfaces are discernible by bands which contain abundant impurities. In an effort to reconcile the growth of large crystals with a complex internal fabric from a thin film of water, Kendall and Broughton (1978)suggest that precipitation takes place initially from thin water films in the form of tiny crystallites which grow syntaxially. The columnar crystals form by a process of syntaxial coalescence immediately behind the
growing surface. This two-stage process leadstofrettedcrystal boundariesand not to straight ones, as might be expected from competitivegrowth of large crystals. Those crystals in speleothems with paraaxialboundariesareinterpretedto bedue to precipitation during episodes of cave flooding. The variety of fabrics are reviewed and illustrated by Chafetz et a/. (1985) and Gonzalez and Lohmann (1988). In addition to these deposits in the vadose part of the cave, bars of sinter called rimstone form on the cave floor, where water from a pool flows over an obstruction as a thin film. When it flows over this rim, there is an increase in the loss of CO2 and calcite is precipitated. Most precipitation in pools or beneath the water table in the phreatic part ofthe zone, however, is different. Because there is no bounding surface, crystals grow out into the water equally in all directions as true crystals with scalenohedral terminations (Figure 21). The largest crystals grow in the quietest water; the more the water flows, the more numerous the calcite nuclei are and the closer the calcite crystals stand, and in rapidly moving water, rounded surfaces like cauliflower, composed of innumerable tiny crystals, can form. The seed crystals for precipitation are
Geoscience Canada Reprint Series 4
thought to form at the water surface because of CO2 diffusion into the air. These may sink or may aggregate into small rafts. Cave pearls (concentrically laminated pisoids) form in pools of moving water or splash pools. Plant life in caves has important geological implications. As caves lack light, heterotrophic bacteria (including actinomycetes or mold-like filamentous bacteria), algae and fungi as well as chemo-autotrophic bacteria are the most common forms . In addition to their contribution to the formation of moonmilk, these organisms result in the precipitation of other minerals. The black, sooty coating on cavewalls and pebbles is a variety of manganese minerals whose precipitation is aided by specialized bacteria. On the other hand, chemo-autotrophic iron bacteria, which obtain N2 from the air and carbon from iron carbonates, liberate ferrous iron, which is oxidized to give a red colour to cave sediments. The Lenticular Zone. Carbonate-water interaction in this shallow phreatic setting is poorly understood because it is inaccessible, and conclusions must be drawn from theoretical calculations and laboratory experimentation (Plummer, 1975) and data from wells (Back et a/., 1979).
) DRI~STONE HELECTITES 6 b FLOWSTONE
0 ~
CAVE PEARLS SCALLOPS COLLAPSE BRECCIA Figure 21
c c c
c
c
c c ~
c c c
c
c c c c c
/
/ -.:..
c c c c c
c
SC~LLOPS
----~..,;;;:~
c c c
c c
) .
c c
A sketch illustrating the styles of precipitates (sinter) formed above and below the water table in a cave open to the atmosphere.
c c c e (
c
c c c c c c
G C
Limestones - The Meteoric Diagenetic Environment
51
-, r:
r
r
c ( c C' ,.'
o r
Dissolution. There is a brackish zone of mixing at the base of the freshwater lens. Theoretical calculations and observations, especially near the shoreline, from Yucatan (Back et al., 1979, 1986; Hanshaw and Back, 1980) and Bermuda (Palmer, 1984) indicate that this mixing zone between fresh and marine waters is dominantly one of corrosion and porosity formation. It is significant to note that corrosion becomes more intensive when fresh and hypersaline waters mix (Plummer, 1975). Since the saturation state is dependent upon so many factors, however, in some areas of the Florida aquifer mixing does not result in undersaturation. In areas far removed from the effect of seawater, Jakucs (1977)has noted that caves are common at the base of the lenticular zone (see Vernon, 1969, for examples in the Florida aquifer) and has ascribed their formation to hydrostatic corrosion, but some of this may be due to mixing of fresh waters from the upper phreat ic and more stagnant and saline waters from the lower phreatic. Precipitation. There is considerable confusion about calcite precipitation in the lenticular zone. From the above, it appears that corrosion is the dominant process in calcite limestones . Hanor (1978) has proposed several processes to promote degassing and so precipitation near the area of outflow, but there does not seem to be much suggestion of widespread prec ipitation .
SYNGENETIC KARST The rock-air interface in metastable carbonates is particularly susceptible to alteration because of the generally porous nature of the sediments. The sediment surface hardens almost instantaneously as a thin, millimetrethick crust of altered grains cemented by calcite. This grey crust also contains algae and fungi and develops within months to years in quarries and new road cuts. The porous natureofthesediments also aids in the rapid and deep development of caliche, particularly in semi-arid climates. In spite of the often poorly consolidated nature of metastable sediments during early phases of meteoric diagenesis, karst processes are ongoing. The products have been called syngenetic karst (Jenn ings , 1971) because they develop during lithification. Not all karst features develop, however. Solution sculpture is minor, except where calcrete is exposed directly to rainwater. Solution pipes are common, espec ially those associated with roots, but the familiar solution features along joint intersections are missing . Caves are well developed, generally due to lateral dissolution at the water table and minor vertical dissolution. Because excavation accompanies lithification and the surrounding poorly lithified sediments are inherently weak, roof collapse dominates more of their history than is characteristic in consolidated lime-
o o o r
Figure 22 Dripstone curtains about 1 m in length decorated by cave popcorn, Carlsbad Caverns, New Mexico - arrows point to water droplets.
stone and is a distinctive trait of these early caves. Once welliithified into calcite limestone, the full specturm of karst features may develop. PALEOKARST Numerous examples of ancient karst, ranging in age from Precambrianto Cretaceous, are documented in the compilation by James and Choquette (1988). Choquette and James (1988), recognize3typesof paleokarst: (1) depositional paleokarst, formed at the top of metrescale shallowing-upward sequences, (2) local paleokarst, generatedwhenpart of a carbonate shelf or platform is exposed by local tectonism or small drops in sea level, and (3) interregional paleokarst, related to major eustatictectonic events and commonly defining sequence boundaries. Since karst is a dissolution phenomenon, a major problem is recognizing when dissolution occurred (Wright, 1982), i.e. , whether the features observed were formed soon after deposition, or are the result of present processes, or were formed at some unknown time. Particularly troublesome is intrastratal or subjacent corrosion, which forms in the subsurface along lithologic boundaries and creates features that resemble surface karst and could be mistaken for such. The surest way of confirming that the feature is fossil is to find skeletons or calcretes cemented onto, or borings penetrating into, the paleosurface. Other criteria that might be useful in certain casesare suggested by Read and Grover (1977) and Wright (1982). On a smaller scale, there is a problem in the differentiation of paleokarst surfaces and stylolites (Walkden, 1974). PART 2 - MINERAL-CONTROLLED METEORIC DIAGENESIS This diagenesis can be seen going on today or, at least, inferred to havetaken place recently in Holocene and Pleistocene carbonates in numerous warm, sunny places, and so is exceptionally well documented. The reader is referred to excellent recent reviews by Bathurst (1980), Longman (1980) and Fliigel (1982) for additional information . The Process. The commonstagesbywhich metastable carbonatecomponentsalterto calciteare now well known (Friedman, 1964; Land, 1967; Purdy, 1968; Matthews, 1974). Soon
52
after meteoric waters begin percolating through a sediment , small elongate or equidimensional calcite crystals precipitate as cement on grain surfaces. They grow equally on magnesium-calcite or aragonite particles and earlier submarine cements, giving the sediment some coherence . The next step, which occurs as more cement is being precipitated , is the start of alteration of magnesium-calcite components to calcite bydissolution of Mg-calcite and reprecipitation of calcite on a microscale (Turner et al. , 1986).By now, most particles are coated with a rind of calcite cement which may constitute up to onequarter of the rock volume. It is at this stage that aragonite alteration usually begins, either dissolving to form molds or altering to calcite across an alteration front with much textural preservation. Dissolution of aragonite produces copious quantities of CaCOa to be precipitated locally as cement both inside and between particles and in new holes created by aragonite dissolution. By the time all aragonite has gone , the sediment has generally been transformed into hard calc ite limestone (Figure 23). In this system, it seems that almost all the cement needed for lithification can be derived locally from aragonite dissolution. So, in spite of wholesale fabric rearrangement, there may be only minor net loss of carbonate and just a change in the style of porosity, from intergranular to moldic.
Diagenesis
dence of a limestone in different parts of the meteoric diagenetic environment. Many examples of these cements are illustrated in Bricker (1971), Halley and Harris (1979), and Jones et a/. (1984); their main attributes are outlined below. Vadose cements. These precipitates grow from void walls into poresthat mayat different times be completelyfilled
with either water or air or both,with water just wetting the grain surfaces. This, and the fact thatvadosepercolationof wateris not uniform, results in cements that typically show quite irregular distribution . Inoutcrops or subsurface cores of Pleistocene limestones it is common to see relatively well cemented areas only millimetres away from areas that are
SHALLOW DEEPER VADOSE
I
SHALLOW PHREATIC
I
MIXED-WATER TRANSITION
I
I
I
I
I
c c
c c c c c
CALCITE CEMENT IN MACROPORES I
I I
c
en zt-
c
c~
a:: CJ ARAGONITE I ."':".~.::f::L''....L.LCL.!.J.W-'~~. I CJ W (GRAINS & MARINE . .. . CALCITE CEMENT IN MOLDS
! :: CE~~~~) a .'. . ~
»:'<'::'::- ':'.:.'.:'.:.':':..:.''.'-:. ':::
. , .:" :"1'" .: : ~ I .. ' '. ' LOW-Mg CALCrtE (NEOMORPHIC) ' ' . ' Mg-CALCIT~ . . . I • • • • • . • ' . . . . . . . ." • • : • • • . • . '" _
c
.: .'
.
~ CIll ~~I~~. ~ ~r'~.I~~ . ~!-~.'. :'.'.... '.: '.'.'.'.' .:.,:'.'.:..:'. :.'.1: . -:: . ' '. '. '.':
- - - BURIAL-DIAGENETIC STAGE ----~ Figure 23 Idealized diagenetic evolution of an aragonite and magnesium-calcite sediment as it passes through meteoric diagenetic zones , in a terra in of low to moderate rainfall . Inferred changes in the trans ition at the base of the lenticular zone are based on calcite saturation relationships suggested in Figure Z
(", d\. -o o--.I-0v METEORIC CEMENTS
Cementation. By the term "cement" we mean here the growth of precipitated calcite into original pore space, either between grains or inside grains, or into holes newly created by the dissolution of aragonite. Techniques for differentiating cement from neomorphic or altered pre-existing precipitates are outlined by Bathurst (1975, p. 416-439) and Dickson (1983). Among the more useful criteria, although neither is foolproof, are (1) increase in crystal size away from the substrate and (2) development of sealenohedral terminations directed away from the nucleation surface or substrate. Even these criteria are best checked in fossil limestones by staining and cathodoluminescence for the possibility that the crystals are neomorphic (e.g., Bathurst, 1983). Vadose and phreatic cements generally have differing morphologies and so can provide a useful guide to the resi-
Geoscience Canada Reprint Series 4
c c c c c c c c c
VADOSE ZONE
PHREATIC ZONE
~
MENISCUS
,
ISOPACHOUS
./)
VADOSE ~~ SILT
~~ ' ). r
STALAGTITIC
~~IQ.A,Jb"
EPITAXIAL
t .:'- . \ .. /
MICRITE ENVELOPE
Figure 24 The different types of cement prec ipitated in the vadose (left) andphreatic (right) parts of the meteoric diagenetic environment. Epitaxial cements on echinoderm particles may be precipitated in either zone.
c c c
c
c
c c c c
c c c c c e l (
C (
C C C
( ( r
Limestones - The Meteoric Diagenetic Environment
uncemented. Although some cements may be in layers of equal thickness around pore walls (precipitated when the pores are filled with water), more commonly cements reflect growth outward to capillary water-air interfaces of
(
partly filled pores and so have meniscus or pendant shapes (Figure 24). Crystals in these settings rarely have good terminations (Figure 25). Meniscus and pendant cements are excellent criteria for vadose precipitation, but
53
if cementation is prolonged and pores are filled, these characteristic shapes are usually lost. In our experience, epitaxial overgrowth cements also develop here, particularly on echinoderm grains and some benthic foraminifer tests. It also seems that some early cements in this zone become detached from pore ceilings and fall to the bottom of the pores forming geopetallayers of vadose crystal silt (Dunham , 1969). Phreatic cements. Pores in the phreatic zone, unlike those in the vadose, are always filled with water,and so crystals can grow unimpeded except by intercrystalline competition. Thus cement rinds formed here, at least inthe shallow lenticular phreatic, are either well developed isopachouslayers of calcite crystals around pore walls (Figure 24) or blocky calcite (Figure 26). As a general rule, crystals are somewhat larger than in the vadose zone and epitaxial growth on echinoderm particles is more rapid. These cement morphologies are, unfortunately, not restricted to this zone and form in deeper burial and cold water marine settings. So shape alone is not a reliable criterion. Staining. A useful technique for differentiating separate stages of cement precipitation and determining cement Figure 25 Pore-rounding cements, probably of vadose origin, in ooid grainstone. The morphologies is staining with an acid arrows show location of meniscus-style cement borderingpartly filled interparticle pores. 5te . solution of potassium ferricyanide, Genevieve Formation, Bridgeport field, southeastern Illinois. (Width of photograph is 0.5 mm). which imparts a blue stain to iron-rich calcites (Dickson, 1966; Lindholm and Finkleman, 1972). Since the vadose zone is an air-water system, conditions there are generally oxidizing, so that any iron is in the ferric state (Fe3 +) and can not be incorporated into calcite. As a result, vadose calcite cements are ironpoor. On the other hand,phreatic waters are often reducing, and so iron is in the ferrous state (Fe2 +) and can be incorporated easily into the calcite lattice , with the result that phreatic cements are commonly slightly iron-rich. There are, however, some caveats to this concept: first, a source of iron must be present; --. second, perched water tables are common in the vadose zone; third, not all phreatic waters are reducing; and fourth, many ferroan calcite (and dolomite) cements form in the deep subsurface, below the lenticular zone. Consequently, the interpretation of the stain should be used with caution, remembering that the lack of an iron-rich calcite Figure 26 Partially to completely dissolved ooids with some aragonite remaining which are cement does not preclude the possurrounded by medium crystalline phreatic calcite cement. (Photomicrograph, partially polarized light, is 2.5 mm across). sibility that it is phreatic in origin.
c
Diagenesis
54
Cement stratigraphy. Identification of stages of cement precipitation can be a powerful tool in ascertaining the sequence of diagenetic environments through which a limestone has passed (Figure 27). In some areas, such stages have been found to be systematically arranged and correlatable over tens of kilometres through tens of metres of limestones (Meyers, 1978; Meyers and Lohmann, 1984). This "cement stratigraphy" (Meyers, 1974), also discussed in Choquette and James, this volume, can be best unravelled by using a combination of staining and cathodoluminescence. Alteration of Magnesium-Calcite Components. On the modern sea floor, particles and cements are composed of micritic and fibrous magnesium-calcite crystallites (Folk and Land, 1975). As outlined above, magnesium-calcite, because of its generally greater solubility in dilute waters, begins to alter to calcite before aragonite, very soon after percolating waters have entered the sediment. Because the change does not involve a modification of the original crystal habit, textural changes that accompany alteration to calcite are invisible with the light microscope (Figure 28). There is, however, dissolution-reprecipitation, and under the SEM it is clear that micrometre-size crystallites undergo slight enlargement (Towe and Hemleben, 1975; Oti and Muller, 1984). Regardless, the precise crystal orientation of crystals is preserved, which is especially evident in echinoderm tests . Skeletal magnesium-calcite is heterogeneous, divided into domains having different levels of MgC0 3 (Moberly, 1970; Blake and Peacor, 1981; Turner et et., 1986). Although the replacement process is not precisely known, magnesium calcites do dissolve incongruently, probably reflecting the greater solubility of the MgC03 phases (Schroeder, 1969). Thus, it is probable that the MgC03-rich domains dissolve first and calcite is simultaneously precipitated. Under the SEM the new calcite can be seen to have grown epitaxi ally on adjacent calcite surfaces, preserving the original crystallographic orientation (Benson and Matthews, 1971). Consequently, the resultant calcite component is partly original and partly diagenetic. Most particles, however, are somewhat porous, and it
appears that even before the skeleton loses much MgC03 , calcite cement is precipitated into these small intragranular pores. A potentially useful technique for differentiating between original magnesium-calcite and calcite components in the fossil record, since both display excellent microstructure, has been suggested by Richter and FOchtbauer (1978). Since calcite precipitated from
Geoscience Canada Reprint Series 4
seawater contains negligible iron, grains with excellent fabric preservation but composed of iron-rich calcites must, at some stage, have altered in contact with phreatic or deep subsurface waters. As calcite is stable and should not alter, these iron-rich calcite grains must have been magnesium-ealcite originally.The converse, however,is not true, as magnesium-calcite grains altered in the vadose zone would also
CEMENT STRATIGRAPHY
c c c
c c c c
c c c c
c
c c
c
c c
c
1. 2. 3. Figure 27
MARINE VADOSE DEEP PHREATIC
1. BEACHROCK 2. VADOSE 3. DEEP BURIAL
A sketch illustrating t wo different examples of cement stratigraphy.
c c c c c c c
c c c
c
c
c (
c e
c c (
Figure 28 Pleistocene limestone (Bermuda) in which the Mg-calc ite grains (Foramin iferaleft, lower right, centre) are now calcite while the mollusc (centre) is still aragonite: the isopachous cement rind is probably phreatic. (Photomicrograph, partially polarized light, is 2.0 mm across).
C l.
C C
C
c
c
Limestones - The Meteoric Diagenetic Environment
55
(' (
o
C)
( (
c (
(
o c o o o o
o
o o o o
be iron-free calcite. In add ition, some aragonites calcitize with considerable iron incorporation (Sandberg and Hudson, 1983), but the replacement textures are coarsely crystalline. Alteration of Aragonite Components. Aragonite is volumetrically more abundant than magnesium-calcite in most modern shallow-water, low-latitude carbonate sediments (e.g., most ooids, gastropods, codiacean algae , dasyclad algae and corals are aragon ite). Particle and cement habits range from microcrystalline to platy to acicular to fibrous, though most modern marine cement maybe acicular. Unlike magnesium-calcite, the change from aragonite to calcite involves a wholesale crystallographic change from orthorhombic to hexagonal and the resultant process is one of complete dissolution of aragonite and precipitation of calcite. Aragon ite dissolution is controlled by the microfabric of the particle, skeleton or cement. It begins in areas with the highest organic content or smallest crystal size, then proceeds to etch along intercrystalline boundaries commonly yielding a "chalky" aragonite,
and finally d issolves the crystals altogether (James, 1974). This dissolution may begin at one point and expand outward or it may proceed from numerous starting points. From a geological viewpoint, it is the way in which the subsequent calcite precipitation takes place that is important (Sandberg, 1965), and two styles are common (Figure 29): (1) Macroscale: In some cases the whole component - grain, fossil or cement - dissolves away, leaving a hole that mayor may not be filled with calcite cement (Figure 26). This is probablythe single most important process creating small-scale porosity in the meteoric environment. The mold may be filled quickly by cement in the meteoric zone or may remain open during burial to be filled much later by deep subsurface cements, brines or even hydrocarbons . A note of caution! In our experience, magnesium-calcite grains in the meteoric environment do not generally dissolve to form molds (but see Schroeder, 1979, for such an example). Recent studies of pre-Pleistocene limestones,
o o
o o o o o o
/Jtr,~ _0
vi"
""""~/p.um.e-
DISSOLVED
~
FILLED
STAYS OPEN
JffY' 6' t
CALCITIZED
~
PHREATIC
VADOSE
i-
-!-
-!-
-!-
MOLD
CEMENT-RLLED MOLD
POOR PRESERVATION
GOOD PRESERVATION
G
8 r
MACROSCALE
MlCROSCALE
ALTERATION
ALTERATION
Figure 29 A sketch illustrating the different ways in which an original aragonite skeleton may be changed during meteoric diagenesis. If dissolved, a moldis formed which may later be filled by cement. If calcitized, some relics of shell structure (often including aragonite relics) remain in the replacement calcite. The best preservation generally occurs in the vadose zone.
however,suggest that underdeep burial cond itions an d/or later fltiSfilng by waters in ne~s.ur.fac.e-emlimen1s....-. some original low-magnesium calcite~ may dissolve selectively, leaving molds (Donath et al., 1980 ; James and Klappa, 1983; Sandberg, 1983). This may be due to subtle differences in crystal size, contained organicmatter ordomainsof unreplaced calciteof low,but notlowest, magnesium content (e.g., 4-10 mole%). (2) Microscale: In other examples, dissolution of aragonite takes place simultaneously with precipitation of calcite on the opposite side of a water film nannometres to micrometres in thickness (Wardlaw et al., 1978), and the basic structure of the original component is partially retained (Figure 30) that is, the resulting "calcitized " aragonite is generally a mosaic of calcite crystals which cross-cut original fabric, with the former structure outlined by relics of organic matter and other insoluble material (James, 1974; Pingitore, 1976). This process generally is called neomorphism. In some instances, replacement is incomplete and tiny crystallites of aragonite remain as relics entombed in the new calcite (Sandberg et al., 1973; Sandberg and Hudson, 1983). Also, because of the fine scale and the microenvironment in which alteration takes place, the concentration ofSr2+ mayquickly rise inthis water film and so be partitioned back into the new calcite in relatively large amounts (rarely, up to almost 1%; Davies, 1977). Because fine-scale replacement resuits in calcitized components that clearly resemble their aragonitic precursors, interpretation of original aragonite mineralogy in ancient limestones is often difficult. Sandberg (1983) suggests a range of criteria that may be used with increasing levels of confidence to confirm original aragonite mineralogy (Table 1). It appears possible that this process may operate in slightly different ways in the vadose and phreatic zones (Pingitore , 1976). Aragonitic components altered in the vadose zone tend to have fabric-selective mosaics with relatively small crystal sizes, show excellent replacement, and contain some organic residue. The resulting calcites have a relatively low Mg2 + and high Sr2+ content. Aragonite components that have changed in the phreatic zone reportedly show poor preservation in coarse,
c 56
Diagenesis
Geoscience Canada Reprint Series 4
c
c
C C C C ('
C C (
C C C
C C
C C C C C C C C Figure 30 A Pleistocene gastropod in cross-section (Strombus gigas) wh ich is partially aragon ite andpartially altered to calcite (C); the lower part of the photograph is enlargedatthe leftillustrating excellentreplacement of aragonite (A) bycalcite (C) with the or iginal fabric now entombed in new calcite as relic organ ic material and inclusions.
C C C
Table 1
C C C
Aragonite Replacement Criteria
C
1. PRESERVATION OF ARAGONITE a. with aragonite relics 2. COARSE" CALCITE MOSAIC , IRREGULAR CROSS-CUTIING STRUCTURE (AS SHOWN BY ORGANIC RELICS OR FLUID INCLUSIONS)
b. no aragonite relics but high Sr ' {
c. no aragonite relics, and low Sr '
3. CEMENT-FILLED MOLDS " relative to original aragonite n.b. 2b, 2c and 3 are strengthened for cements if external morphology shows large square-ended crystals.
~
<-
iii
C
::::i
<: ::::i w II:
o Z
u;
c
e
-c
(
~
\.
~
Criteria, in order of increasing reliability, for determining original aragonitic nature of calcitized components (after Sandberg, 1983). Although useful alone, they are best utilized in conjunction with observations on the whole spectrum of preservation styles within a single limestone.
C C (
C (
c (
Limestones - The Meteoric Diagenetic Environment
57
c
c o o o ( (
o o c o o o c o C G
o o o o o c. C. (
l (
0
cross-cutting mosaics and have relatively high Mg2+ contents and low Sr2+ values. These differences are thought to result from alteration across a film of water micrometres thick in the vadose zone but across a chalky zone in the phreatic zone. In the phreatic, because the skeleton is surrounded by water, calcite continues to grow into a void, whereas in the vadose zone the voids may be filled with air. These changes imply that alteration in the vadose zone can be viewed as somewhat closed, whereas phreatic calcitization takes place in a more open system. This process is also viewed as occurring in a "two-water system" (Pingitore, 1982), with (1) a small amount of slow-moving water at the dissolution- reprecipitation interface which exchanges with (2) a large volume of faster-moving water percolating or flowing through the rock or sediment. Water at the interface can be considered an almost closed system which is, in turn, connected by diffusion to a large open reservoir. The alteration of aragonite skeletons to calcite also results in changed petrophysical characteristics. In corals, for example, calcitization leads to an overall decrease in porosity, mainly because of precipitation of calcite cement into pores, but an increase in permeability because of changes in pore-aperture size (Pittman, 1974). Alteration of Calcite Components. Biogenic components composed of calcite, such as brachiopods, some bryozoans, planktonic foramininfers and layers in some molluscs, appear to remain unaltered under most meteoric conditions. It should be remembered, however, that these particles are not pure crystalline calcite, but are composed of calcite, organic templates, usually small crystallites, and numerous micropores. Thus, they are more soluble than the new diagenetic calcite. So, once all aragonite and Mg-calcite components have altered to calcite, the limestone is composed of original calcite particles and less soluble diagenetic calcite components. A second stage of alteration, affecting only these original calcite elements may then take place in deeper parts of the meteoric environment (AI-Aasam and Veizer, 1986a, b) or in the burial setting (Choquette and James, this volume).
Importance of Grain Size. Most of the processes and products outlined above have been worked out from Pleistocene and Holocene calcarenites because of their convenient grain size and pore diameters. Likewise, grainy rocks should be looked at first in ancient limestones to establish the general diagenetic sequence. When considering the full spectrum of carbonate sediments, however, it is clear that grain size plays an important role. Mudstones, for example, because of their small grain size, alter much more rapidly than do contemporaneous calcarenites. Even in calcarenites, the finer grained layers are commonly cemented first, probably because the finer grained layers hold water longer by capillary action (Evans and Ginsburg, 1987).Such preferential cementation of calcarenites is important because subsequent water movement is preferential through channels and voids around the cemented layers, leading to rapid porosity enhancement and beginnings of karstification. The alteration of polymineralic muds in the meteoric environment is difficult to study because oftheirsmall grain size and propensity for rapid alteration. Recent studies utilizing SEM (Steinen, 1978,1982;Lasemi and Sandberg, 1984) indicate that the same processes occur as in calcarenites. Plio-Pleistocene mudstones formed from aragonite-rich muds are composed of both micrite and microspar (sensu Folk, 1965) and appear to be the end result of concurrent aragonite calcitization and porefilling cementation. The neomorphic calcite exhibits aragonite pits and relics on polished and etched surfaces and the rocks have been called ADP (aragonite-dominated precursor) mudstones by Lasemi and Sandberg (1984). The cement regions are commonly composed of several crystals in which each continues into adjacent relic-rich neomorphic calcite as minute examples of "cross-cutting mosaics" observed by Pingitore (1976)in calcitized corals. This process transforms carbonate mud with 50% to 70%. porosity into mudstone with commonly less than 5% porosity. These mudstones contrast with COP (calcite-dominated precursor) mudstones, which are mostly micrite composed of crystals with no relics or pits, and are from more ancient limestones.
The importance of grain size or crystal size is well illustrated in the common phenomenon of micrite envelopes (Bathurst, 1966;Alexandersson, 1972). On the sea floor, the outsides of carbonate particles, such as bivalve shells, commonly are bored byendolithic algae whose vacated microborings are filled by microcrystalline aragonite or magnesium-calcite cement, forming an "envelope" of finely crystalline carbonate. When these grains come in contact with meteoric waters the small crystals inthe envelope, because of their tiny size, alter quickly to calcite (more quickly if magnesium-calcite than if aragonite), before the aragonite of the bivalve begins to dissolve. So for a time the bivalve is still aragonite while the envelope is calcite. Eventually the shell may be leached away and only a thin microcrystalline rind is left outlining the bivalve mold.This mold isthen filled with calcite cement (Figure 24). If it were not for the difference in grain size, mineralogy, or relative resistance to dissolution (if both aragonite), and thus different alteration rates, all would have been dissolved together and there would be no record of the presence of bivalve shells. At the other end of the spectrum, large whole aragonitic fossils (gastropods, bivalves, corals) alter more slowly than sand-size aragonite particles, and it is not uncommon to see partly altered corals surrounded by well-cemented, but porous, calcite limestone containing molds of smaller aragonite grains. Diagenesis in Vadose versus Phreatic Settings. The style of and relative rates of sedi-. ment alteration in the vadose and phreatic zones have been studied on relatively large oceanic islands mantled by Pleistocene carbonates (Bermuda: Land, 1970; Plummer et al., 1975; Jamaica: Land, 1973b; Barbados: Steinen and Matthews, 1973). These results can be used in a general way to approximate conditions on larger carbonate platforms. Studies on small Holocene sand cays (e.g., Joulters Cay: Halley and Harris, 1979) probably better reflect conditions of ephemeral exposure of small portions of platforms. Two examples of vadose versus phreatic alteration are outlined in Figures 31 and 32.
Diagenesis
58
Suspicion that sediments continuously in contact with meteoric waters in the phreatic zone alter more quickly than those in the vadose zone, where water flow is ephemeral, is borne out by studies on the larger islands. Waters in these large phreatic lenses not only percolate down from the vadose zone, but also arrive at the water table directly from lakes, swamps, rivers, and via vadose flow through fissures. Because the phreatic waters are from different sources, their chemistries and mineral saturations show considerable temporal variation. In some areas, it is estimated that the bulk of the water in the phreatic lens is not from vadose seepage but from vadose flow or lakes, rivers and bogs. Thus, the phreatic waters should be of different composition than the overlying vadose waters and, because of mixing, the water table should be a level of considerable chemical activity. Sediments composed of metastable carbonates seem to alter rapidly in the phreatic zone. Those areas studied to date display both cementation and dissolution that result in extensive lithification and in abundant aragonitic fossil molds as well as calcitized skeletons. There is no clear information about where in the phreatic zone grains alter most quickly or in what manner, or whether all parts are affected equally. However, Longman (1980) has outlined a hypothetical sequence of alteration for this zone, which remains to be tested. In any case , because of these differences a common situation in Pleistocene carbonates is for sediments in the vadose zone to be poorly lithified and still mainly metastable, while the same sediments in the phreatic zone are calcite limestone, with a striking mineralogical discontinuity developed at or near the water table. It is not clear whether dissolution or precipitation will prevail in the phreatic zone during these early stages of diagenesis because the processes depend upon so many factors. If, for example, vadose waters several times oversaturated with respect to calcite because of abundant dissolution of aragonite are mixed with phreatic waters that are just saturated with respect to calcite, and PC0 2 levels in the waters are not too different, then cementation can occur at least adjacent to the water table . If, on the other hand, PC0 2 levels are very
different, then mixing is likely to result in dissolution, as in a wholly calcite system. Meteoric diagenesis on small islands or sand cays with only local aquifers appears somewhat different. Diagenesis throughout is very rapid in the best studied example, Joulters Cay (Halley and Harris, 1979), but this may be due partly to the high aragonite content of the sediments (almost all ooids). An important difference from large islands is that all the waters pass downward through the vadose zone before they reach the water table and there is
PETROGRAPHY
Geoscience Canada Reprint Series 4
little vadose flow, only seepage. Although sediments are only in the initial stages of alteration, several trends are nonetheless evident. The main point is that on small islands the rates of diagenesis above and below the water table do not appear to be much different . In both cementation is patchy, but oolite is more friable below the water table than above because precipitated calcite is a true "cement" in the vadose zone, forming an "onion skin" on the grains and concentrating as a meniscus at grain contacts . In contrast, calcite in the phreatic zone is precipitated as
GEOCHEMISTRY
CAUCHE
0/00
MARINE LEVELS
o
+
I
Mg++-I
I:
VADOSE
Sr+++--
I I
I
I
I
I I I I PHREATIC
I I I I
I I I I MARINE
Figure 31 A sketch illustrating a hypothetical core through an ooid grainstone shoal exposed to meteoric diagenesis under semi-arid conditions, resulting in a caliche profile. Sediments below the mixing zone are saturated with marine pore waters and essentially uncemented. The geochemical curves are based on results one might obtain from "bulk" analyses. The limestone was originally composed of aragonite ooids, hence the high initial Sr2+ and low Mg2+ concentrations. Stabilization is essentially complete in the freshwater phreatic zone. A Pa/eoexposure Surface - caliche crust, partial conversion of ooids to calcite micrite, needle and other calcite cements, gravity effects, vadoids. B Vadose zone - conversion of ooids to calcite with mostly good preservation, meniscus and gravity calcite cements. C Freshwater Phreatic Zone - extensive dissolution (fabric selective). calcitization, isopachous calcite cement. o Mixing Zone - preservation of aragonite ooids, poor induration, some rhombs of low to high magensium-calcite.
c c c c
c c c (
c c c c c c c c c c c c c c c c c c c c c c c c c e (
c (
C l l
C C C
c c
c c o (J (
( ( (
( (
o o o o o o 0_ o o (
Limestones - The Meteoric Diagenetic Environment
scattered rhombs all over the grain surfaces and not necessarily at grain contacts . The most intensively cemented zone is right at the water table . Cementation decreases with depth in the phreatic zone as the proportion of seawater increases. Interestingly,calculations suggest that at the currentrate all the sediment will be stabilized 16 calcite within 10,000 to 20,000 ye~rs on this small carbonate
PETROGRAPHY
59
sand cay, whereas stabilization will probably take some 100,000 to 1,000 ,000 years on Bermuda and along the coastline of South Australia (Reeckmann and Gill, 1981). The Freshwater-Seawater Mixing Zone. The possibil ity that dolomite may form in miXing zones has provoked considerable interest in the hydrology, chemistry and mineralogy of such zones (see Morrow, th is volume) .
GEOCHEMISTRY MARINE LEVELS
o +
/
/
® S
/
I
/
~I
Mg++ , VADOSE
I I
/:
""
I
I
l{}.1J I ~ h?-f ~ I I/v e:r Id--v I /
(-1: I PHREATIC
o
I
I
Cementat ion of limestones by CaC03 in mixing zones may not be extensive. In very young Holocene sediments (ca. 1,000 years old) beneath south Joulters Cay on Northeastern Great Bahama Bank, for example, there is virtually no cementation at all below the freshwater phreatic lens. Minor amounts of low-Mg calcite cement have been found in paleo-mixing zones on Barbados (Benson, 1973; Steinen, 1974), northern Jamaica (Land, 1973a, b), and a coastal area of Grand Cayman Island (Moore, 1973). Coastal mixing zones are, in a sense, involved in the formation of beachrock (discussion in James and Choquette , this volume), but CO 2 degassing in these zones, rather than mixing effects , may be mainly responsible for CaC0 3 cements there (Hanor, 1978). There is considerable evidence that dissolution is important in mixing zones. Non-fabric selective dissolution occurs along some 150kilometres of the northeastern Yucatan coast, apparently in a coastal mixing zone (Back at al., 1978, 1979, 1984), producing a striking scalloped morphology and related collapse features (Figure 33). Mixing corrosion
I I
: I I
I
MARINE
o l l
Figure 32 A sketch illustrating a hypothetical core through a reef limestone exposed to meteoric diagenesis in a warm , moderately humid climate . Subaerial exposure produced a corroded fractured surface karst. Sediments below the mixing zone are saturated with marine pore waters. The geochemical curves are based on results one might obtain from "bulk " analyses . The limestone was originally composed of coralline algae (A, magnesium-calcite), bivalves and corals (8 and C, aragonite). Stabilization is mostly complete in the freshwater phreatic and the offset of the 5' 3C profile at the water table is due mostly to precipitation of diagenetic calcite cement. A Pa/eoexposure Surface - dissolution and fracturing, geopetal vadose sediments, good skeletal preservation. B Vadose Zone - mostly good preservation though skeletal aragonite is calc itized, minor development of molds and matrix vugs. C Freshwater Phreatic Zone - extensive dissolution , prolonged flushing, non-fabr ic selective voids, some calc itization and calc ite cement precipitation (black = porosity) . D Mixing Zone - not exposed to meteoric waters, aragonitic fossils may be calcitized, magnesium content of coralline algae high in lower part, dissolution likely in upper part (dilute pore waters).
Figure 33 An oblique areal photograph of the coastline at Caleta XelHa, Quintana Roo, Yucatan Peninsula Mexico. The irregular coastline is the result of dissolution of Pleistocene limestone along joints by brackish groundwaters which are undersaturated because of mixing fresh water and seawater. The waters come to the surface in this locality at the irregular area about 200 m across marked by the arrow.
(
60
on a large scale has probably occurred along the base of the vast freshwater lens in parts of the south Florida aquifer (Vernon , 1969), and distributions of caves in limestones beneath large oceanic islands, such as Bermuda, suggest that brackish waters may have had a part in their origin (Palmer, 1984) . Where CO 2 is not a major factor and fresh water and seawater mix according to the error function relationship described earlier, it seems likely that dissolution is favoured in the upper, more dilute reaches of the mixing zone (see Figure 3), as might be expected from Figure 8, and that CaCOa precipitation (and dolomite formation?), if significant, should take place in the lower reaches of this zone. Where calcite is the cement precipitated, in metastable limestones it might be expected to become more Mg-rich downward in the mixing zone. This Mgrich fingerprint should not be preserved long or deeply into the sublenticular phreatic realm, however, unless by such features as microdolomite inclusions (Meyers and Lohmann, 1978). It is attractive to interpret as products of mixing-zone precipitation those calcite cements which have these inclusions, are clearly not synsedimentary marine in origin , and are spatially associated with identifiable phreatic cements. Meyers and Lohmann (1984) argue persuasively for precipitation of these inclusion-rich cements from regionally extensive transition zones, and it is true that modern brackish mixed-water zones with comparable distributions
Diagenesis
and geographic extent have been documented in some widespread continental-coastal aquifers (e.g., the South Florida aquifer: Back and Hanshaw, 1970). The curve in Figure 8, illustrating the effect of mixing fresh and seawater, is plotted for calcite only. Other minerals show the same effect, but have different saturation indices. Thus, there is a portion of the curve (or percentage of seawater in the mix) where waters are saturated or oversaturated with respect to calcite, yet undersaturated with respect to aragonite - a situation which would lead to aragonite alteration. Along parts of the Yucatan coast, for example, the critical mixture appears to be between 50% and 60% seawater (Back et aI., 1986). It is likely that the rock record will be found to contain more examples of diagenesis and cement stratigraphies imprinted by local and regional mixing zones. GEOCHEMICAL ASPECTS OF METEORIC DIAGENESIS Minor Elements. Dissolution of aragonite (at whatever scale from micro to macro) releases strontium (Sr2+) and other less abundant trace cations to the surrounding pore waters ; incongruent disssolution of Mg-calcite releases Mg2+ . The result in the carbonates is a net decrease in both Sr2+ and Mg2+ either with accompanying passage through the vadose and freshwater phreaticzones, or during continued residence in either zone (Figures 16 and
Geoscience Canada Reprint Series 4
17). ThislossinSr2+ and Mg2+ is apparently reversed with passagethrough the mixing zone (Wagner, 1983) , which appear to be particularly well defined by downward increases in Mg2+ content. Below the water table, beyond the reach of oxygenated water, metals that can exist in lower oxidation states such asMn 2+ andF&+ become available for incorporation into diagenetic calcite (C02+ and Ni2+ also may playa role here) , result ing in an overall net increase in these elements with progressive burial Meyers and Lohmann (1984). The level of cathodoluminescence (Cl) varies sharply among growth zones in crystals of carbonate cement (see Machel, 1985, for detailed discussion). As illustrated in Figure 34 , vadose cements show little or no Cl because vadose waters normally are well oxygenated, so that neither Fe2+ or Mn2+ exist. Below the water table oxygenation and Eh decrease rapidly, so in the shallow lenticular phreatic zone Cl is both present and highly variable, but is essentially nil (dark) in the deeper phreatic zone. Variability can best be attributed to fluctuations in supply of activator ions (mainly Mn2+) and suppressor ions (mainly F& +) and to variations in water flux in response to either seasonal or longer cycles; these fluctuations are greatest in near-surface hydrologic regimes, particularly, the vadose. Phreatic-zone Cl, first identified by Meyers (1974) in a landmark study, has since been described in many parts of the Phanerozoic (e.g.,
C C C C C C C C (
C C C C C C C C C C C C C C
C C C
E l C C
C
c e (
(
C Figure 34 (Left) two stages of cementation in an Early Cambrian grainstone: V, vadose pendulous cement; P, phreatic blocky calcite (plane polarized light). (Right) cathodoluminescence, vadose cement has irregular trace element content, phreatic cement is two-staged; dark and bright luminescent zones, probably shallow phreatic; P 2 , dull luminescence, probably deep phreatic. (Photo width is 5 mm).
p,.
(
\-
C C C
c
c
Limestones - The Meteoric Diagenetic Environment
61
C'
c.
o o ( c ( c c c c o o o o o
Grover and Read, 1983; James and Klappa, 1983 ; Meyers and Lohmann, 1984). Cathodoluminescence is a powerful tool for deciphering of tencomplex cementation histories, particularly where CL zones can be calibrated with minor-cation compositional data . Mussman et al. (1988), for example, have utilized these techniques to interpret the depth to which meteroic fluids were important diagenetic agents below the Middle Ordovician, postKnox interregional unconformity in the Appalachians.
Although the tempo of research on minor elements in carbonate diagenesis is increasing (see discussions by Brand and Veizer, 1980 ; and Lohmann, 1983, 1988), our understanding of the behav iour of these elements even in the near-surface meteoric environment is still fragmentary and evolving . In general terms, however, the profiles in Figures 31 and 32 appear to be reasonable. Stable Isotopes. Considerably more is understood about the distribution and diagenetic "behaviour" in carbonates of stable carbon
+6 AVG 0 0 OF I CALC ITE PPT'D IN OPEN METEORIC' PORE-WATER SYST I INCREASING METEORIC EFFECTS 18
®
+4
+2
Q If
MARINE SEDIMENTS
t;~-,I4ilIjj_~~,
AVG"O I3C MARINE SEDS
~w z~
o
w~ ::Eo::
OW
~~ ~ ,14J
w~
ss s
< ~~
-2
~-~
~I
© C
-4
s s» ~
(J)~
~g
<---
,..g/B ,..
0::
~
c>z
'
z-
iil4J~
-W
(J)u
-6
~~CJ
Ww
0:::;) u....J
~u.
~o
§iil4J
o'
o~~
~'
Z , .........1.1'1
OCJ~
-8 PLEISTOCENE COMPOSITIONS
-10
@
BARBADOS
@
BERMUDA
-12 -10
-8
CALCITE PRECIPITATED IN CALCITE & SOIL PROFILES
o
+4
l Figure 35 A diagram showing isotopic compositions and trends in metastable carbonates where mineraJ-controlled alteration takes place in a meteoric system capped by a caliche-soil profile. The covariant trend (A) has as end member marine sediments (seawater) and 13C_ enrichedpore waters; diagenetic calcites precipitatedin a marine-meteoric mixing zone might have compositions along such a trend. If no caliche-soil profile existed, alteration of marine sediments might follow the trend shown by the top arrow gently down to the left (B). The long vertical trend (C) would apply to rock-water interaction beneath a caliche-soil profile in the shallow vadose, where 613C becomes heavier downward as "rock" CO~- overpowers soil COz while 6180remains the same, as shown in Figures 31 and 32. The data fields are from Gross (1964) andAllan and Matthews (1977, 1982). The trends are in part based on interpretations by Hudson (1977), Lohmann (1982, 1983), and Magaritz (1983).
and oxygen isotopes within the nearsurface meteoric. At present, our knowledge of carbonate-ion isotopic compositions diminishes with increasing depth below subaerial exposure surfaces, even in Holocene and Pleistocene carbonates. The profiles in Figures 31 and 32 typify fairly well the kinds of stratigraphic distributions of carbonate 51 3C and 51 80 in young Pleistocene sequences. In these profiles, the patterns are emphasized rather than absolute values, and it is clear that with variations in absolute composition these patterns are valid for many, if not most, limestones SUbjected to near-surface meteoric diagenesis at least as far back in geologic time as Late Paleozoic. The profiles are based on a series of recent studies (e.g., Magaritz, 1975, 1983; Allan and Matthews, 1977, 1982; Salomons et ei. , 1978; Videtich and Matthews, 1980; Wagner, 1983). The general subject of 51 3C and 5180 isotopic compositions and isotope geochemistry has been ably summarized by others (e.g., Bathurst, 1975; Hudson, 1977; Veizer, 1983) and a lucid view of principles is provided by Lohmann (1983, 1988). With sound petrographic work, it is possible to interpret isotopic signatures of meteoric diagenesis with a high degree of accuracy, provided that one can get an estimate of the original marine 513 C and 5180 of a limestone or limestone sequence prior to alteration. This can be done, as numerous studies (e.g., Lohmann , 1984; Given and Lohmann, 1985) have convincingly demonstrated, by taking the "heaviest" composit ions in a linear, covariant series of values determ ined for those marine components that show the least evidence of alteration. Such components might be well-preserved brachiopods (calcite originally) or marine radiaxial cement, or originally aragonitic rudists in which critera from Table 1 indicate good preservation following Sandberg (1983). The covariant trend approach, illustrated in Figure 35, has been used by Lohmann (1983) and others to estimate orig inal marine CaC0 3 compositions with apparent accuracy. These graphs were presented in the first of this servies of re views (see James and Choquette, this volume).The need for a " marine baseline" is critical in older Paleozoic limestones - and possibly Cretaceous limestones -
62
where original 01ac and 01eO of marine sediments and cements appear to have been several per mille "lighter" than in the present-day ocean (James and Choquette, 1983,Figs. 20 and 21). Contrary to earlier ideas that this isotopic "lightness" was the result of more prolonged diagenetic alteration in older Pleistocene limestones, it now appears that real, non-linear variations in the composition of seawater through Phanerozoic time are represented. The range, and more significantly the pattern, of carbon and oxygen isotope compositions found in many Cenozoic limestones in shallow meteoric settings - where limits of present-day or former vadosezones, phreatic lenses and even mixing zones can be determined with increasing confidence - also occur in many Phanerozoic limestones. Ranges in fossil limestones are usually smaller because of the damping effects of later burial cements or other forms of alteration, but patterns are similar. The general tendency in stratigraphic profiles through young metastable carbonate sequences in near-surface meteoric environments is for carbonisotope compositions to vary considerably down through the vadose, particularly within short distances (metres or less)of subaerial exposure surfaces; then to stabilize and remain relatively constant through the shallow phreatic; and finally to change through the mixing zone toward marine or nearly marine values. Oxygen-isotope compositions show little variation through the vadose and shallow phreatic except in the mixing zone where they become heavier, enriched in 1eO because of preferential loss of 160 by surface evaporation. Lohmann (1988) has called this invariant 01eO value, when preserved in carbonate precipitates, the meteoric calcite line, which is likely to be unique for each sequence and locality studied. These general relationships are shown in Figures 31 and 32. Although the relationships were recognized first in Pleistocene sequences, they are commonly preserved in Phanerozoic limestones (Allan and Matthews, 1982)where they may have value for signalling the presence of ancient subaerial surfaces. In the CaCOa-H 20-C02 system carbon comes almost wholly from the original carbonate and CO2 gas. Carbonisotope variability in near-surface vadose zones is due mainly to the influ-
Diagenesis
ence of isotopically "light" (1aC-poor) carbon. Calcites precipitated in caliche and soil profiles - in hardpans, nodules and associated vadoids - are typically -7%0 to -13%0 in Recent and Pleistocene profiles, changing downward within a few centimetres to metres at most to values of -2%0 to -4%0, which are typical for the rest of the vadose. The "light" compositions reflect contributions of CO2 from decaying organic matter to vadose pore waters seeping downward or being drawn upward to evaporation throughout the caliche-soil profiles (for discussion of these effects see Rightmire and Hanshaw, 1973;Hudson, 1977; Allan and Matthews, 1982; Lohmann, 1983, 1988). Deeper in the vadose, dissolved CO2 in pore water rapidly equilibrates isotopically with the marine carbonate sediment through which it is percolating. Speleothems, however, do not have the same isotopic signature as vadose cements, generally because of kinetic isotopic fractionation during rapid degassing. Both 01ac and 01eO are enriched (Hendy, 1971; Lohmann, 1988; Gonzalez and Lohmann, 1988). Oxygen in the CaCOa-H 20-C02 system can come from the minerals, the water and lor the gas, but in reality contributions from dissolved CO2 are miniscule compared to contributions from the water and from host sediment or limestone. Thus oxygen-isotope com-
Geoscience Canada Reprint Series 4
positions through the meteoric are much less variable than carbon-isotope values. The 01eO values of carbonates in caliche and soil profiles are much like those of carbonates in the rest of the vadose zone and the lenticular phreatic (except for the evaporation effect mentioned earlier), generally -2%0 to -6%0 in young Cenozoic carbonates. Covariant trends of 01ac and 01eO should also be displayed by CaCOa cements precipitated from mixtures of marine and fresh meteoric waters, as in a mixing zone. The "end members" can be identified most clearly if they are original marine CaCOa and carbonate precipitated in a caliche or soil profile (Figure 35). In progressively more altered carbonates below profiles of this kind (Figures 31 and 32), one often sees an inverted "J" or "L" shaped trend, as pointed out by Lohmann (1982, 1983), Magaritz (1983) and AI-Aasam and Veizer(1986).This kind oftrend on cross plots is a result of the extremely "light", 1aC-impoverished CO 2 supplied to near-surface pore waters by decaying organic matter, as noted earlier, and the differences in rates at which pore waters equilibrate isotopically with the host CaCO a sediments and rocks through which they pass. Where organic matter did not exist or was very sparse at the surface, so that a caliche or soil profile did not develop, a "J"-shaped trend will not be expected to develop.
Figure 36 Hard Pleistocene reef limestones overlying softOligocene chalks, Barbados. The Pleistocene limestone hardened quickly in the meteoric environment because it was composed of metastable minerals. The chalk remained soft because it was originally, and still is, composed of calcite.
c c c c
c
c
c c:
c
c c c c c c ( c c c c c c c c
c c c c c c c c c e c ( c c c c c c
c
c c co (
c
c (
C' (
o (
o (
o o o o c. (
Limestones -
The Meteoric Diagenetic Environment
DIAGENESIS OF STABLE CARBONATE (CALCITE) SEDIMENTS It is clear that the extensive alteration and cementation of metastable carbonate sed iments is driven by the co-ex istence of different minerals with different solubilities. But what about carbonate sediments that are mostly calcite? We cannot, unfortunately, make a direct comparison because shallow-water Neogene carbonate sediments are usually a mixture of minerals, but we can lookat deep-water chalks (CretaceousRecent) to gain some insights. These sediments are , or were , composed almost entirely of calcite skeletal elements originally, with relatively few aragonite bioclasts. Exposures of Cretaceous chalk are widespread in northwest Europe. Where these sediments have not been affected by deep burial or percolation of hot waters they are extremely soft, with over 40% porosity, and have altered relatively little despite 70 to 90 m.y. of shallow burial and prolonged exposure. The chief alteration is the loss of organic matter, aragonite and opaline silica. Calcite fossils are excep-
Table 2
63
tionally well preserved. There is some calcite cement, precipitated mostly as epitaxial overgrowths on coccoliths and echinoids. This minor cement is likely derived from the very smallest, supersoluble particles . Compared with the metastable suites just described, these sediments are virtually unaltered! The contrast is probably nowhere better displayed than on the island of Barbados where soft Oligocene chalks directly underlie hard, Pleistocene reef limestones and, even though the permeability of the chalk is much less than that of the reef limestone, both have been exposed to percolating fresh waters for nearly the same length of time, about 500 ,000 years (Figure 36). It is important , however, to note that in spite of the fact that the chalks are soft, they commonly exhibit deep surficial weathering and solution cavities, to an extent reminiscent of calcite limestones. This striking difference becomes extremely important if the composition of carbonate 'sediments in the past was at times different from those of today, and dominated by calcite constituents (see discussion of sea-floor diagenesis
Some features helpful in identifying former vadose and phreatic diagenetic zones. VADOSE OR PHREATIC
RELIABILITY
MEGASCOPIC
(
Calcrete, soil-zone features, root casts, oftencomplex travertine-filled fractures
V
Speleothems
V
Stratified/laminated sediment in larger cavities (vadose sediment often inclined)
P
2
Trains of horizontally elongate large solution cavities (may cross-cut structural dip)
P
2
Doninantly vertical solution cavities
P
2
MICROSCOPIC Preserved needle-fibres, flower spar, fungal sheath & related near-surface cements
V
Meniscus cement
V
Gravity cements (micro-stalactite, -stalagmite)
V
Isopachous cements"
P
2
• 1 is more reliable than 2. •• May be difficultor impossible to distinguish from isopachous cements of submarine, sea-floororigin (see James and Choquette, 1983).
by James and Choquette, this volume). If meteoric diagenesis, or more properly, the lack of meteoric diagenesis of chalk, is any guide, then the subaerial exposure of sediments in the lower and middle Paleozoic, for example, should leave little discernible evidence in the form of diagnostic cement, signs of fabric-selective dissolution, or geochemical signatures. Since there were likely some aragonitic components in these calcite sediments, there was probably a little cementation , but very little. Instead, the signature would be more in the form of surficial karst, etc. These conclusions have yet to be tested against the rock record. CRITERIA FOR THE RECOGNITION OF THE PRODUCTS OF METEORIC DIAGENESIS In summary, it is often possible to reconstruct whether a given stratigraphic unit has resided in a vadose or phreatic zone (or both) , either from field relationships, from petrographic features seen at scales ranging from hand specimen to the light and scanning-electron microscopes or, in some cases from geochemical information. Careful petrographic observation, aided by analysis of cathodoluminescent zonation and cement chemistry, can make possible detailed reconstructions of residence in a succession of diagenetic environments. Many of the petrographiccriteria have been distinguished in very young Holocene and Pleistocene carbonates, but are commonly preserved in the older stratigraphic record. Large-scale features of karst terrains are being applied with increasing success to the fossil . record (James and Choquette, 1988) and a great deal is now known about the kinds, if not always the origins, of features formed in the vadose zone. The criteria outlined in Table 2 are often helpful in both distinguishing and identifying the interface or transition zone between vadose and phreatic zones in the fossil record . Many of these criteria are quite reliable. Others are less so, either because they may be produced in both zones near the water table, or because they may originate in other depositional-diagenetic settings , such as beachrock where gravity cements are well known. The complexities of overprinting in this realm are well illustrated in numerous studies (Heckel, 1983; Roehl
Diagenesis
64
and Choquette, 1985).Atthesame time, we now understand the main processes and products well enough so that forward modelling of meteoric diagenesis in specific settings is being proposed (Matthews and Frohlich, 1987). THE EFFECTS OF CLIMATE Since water and temperature are the two most important extrinsic factors in the processes of meteoric diagenesis, climate - particularly the availability of meteoric water - influences both the intensity and rate of carbonate alteration (Figure 37). The effects are strongest in the vadose zone. Under hot , arid desert conditions both karst and calcrete may develop, butslowly and intheformofthin profiles ; likewise, alteration in the vadose zone is slow. Thus, carbonates exposed for extended periods under these conditions mayshow little or no effect of meteoric diagenesis. In the phreatic zone, alteration may be more extensive. Water flow in the phreatic is likely to be slow because of limited recharge . Cementation may be limited and porosity and permeability high. There are few case studies, however, and so clear information is lacking . If the climate is warm and semi-arid, caliche should be both thick and extensive, with surface karst also present locally. Alteration in the vadose zone is relatively rapid and is dominantly incongruent, if the mineralogies are metastable. Alteration is most generally fabric-selective, and replacement of original aragonite textures is good. Mineralogical changes have been discussed earlier and diagrammed in Figure 23. Geochemical changes, also discussed earlier and outlined in part in Figures 31 and 32, seem to follow the general direction shown , with slight abrupt offsets at the water table if the phreatic and vadose zones behave as somewhat separate systems (Wagner, 1983). Changes are rapid in the vadose and very subdued or essentially nil in the phreatic. Caves are rare and small , and movement of ground water in the vadose is mostly by vadose seepage. Under warm and rainy conditions (subtropical to tropical), although modern diagenetic changes are difficult to determine because of extensivevegetation and rugged karst topography, surface and subsurface karst should be extensive, dissolution would prevail
over precipitation , and both fabricselective and non fabric-select ive fabrics should occur. Caves and solutionenlarged fractures should be very extensive (Figure 37), stabilization to calcite limestone rapid, and vadose flow dominant over seepage. Congruent meteoric diagenesis is dominant.
Geoscience Canada Reprint Series 4
Geochemical profiles in metastable limestone beneath a karsted surface, outlined in part in Figure 32, ideally may resemble those in the right portion of Figure 38. Because of high-volume flow in the phreatic zone, substantial effects can be seen in compositional profiles. For the 0'3C and 0'80 profiles in the
CLiMATE & METEORIC DIAGENESIS
c
CALCITE
PHREATIC
c
CALCITE
Figure 37 A diagram illustrating the effects of different climates on a series of metastable carbonates of the same age and composition. The geochemical profiles in Figure 38 apply largely to semi-arid and wet (high rainfall) situatio?s.t\l,J ,J\.v{·
.Y\
SEMI ARID
EJ"')
P ~l'
y9 .J'\~~ /:
HUMID
depletion ~(--~) enrichment
~===:::::::=====::::=~~r=:~~~::::t~UBAERIA;:;r-;-~ -...!:6~18;0~;1~3:===:;~=~;~~~u~~:=1 EXPOSURE SURFACE VADOSE WATER r - --,l---,l--~l----j TABLE - 1 1 \ 1 1
1
I
I I
I
PHREATIC
11<)
r I
I I 1 I
,I-- ,.L---!-,--+---'t----l
I 1
\
\
\ \
1:3/ \
/ )<
\
TOP OF MIXING ZONE
'\ '\
c c c c c c c c c
b-=- :P~/-~~ 7'-l
depletion ~(--~) enrichme
c
c
VADOSE WATER TABLE
c
c
SEMI-ARID
ARID
c c c c
I I I I
I I
-,
\
\
\ \
Figure 38 Stratigraphic patterns of geochemical profiles in a sequence of initially aragonite and magnesium-calcite limestones beneath a subaerial exposure surface, under arid or semiarid conditions (left) and humid conditions (right). Open arrows show dominant directions of groundwater movement. Offsets in isotopic composition, cation concentrations and cathodolum inescence (Lum, mainly reflecting Mn z + if FeZ+ is not excessive) occur at the water table because the vadose and phreatic zones behave as more or less separate hydrologic systems , particularly in the humid setting where water is greater in the phreatic and water chemistry is affected by other recharge sources as well. In the ertd-semi-eria setting there is upward enrichment in Sr z + , MgZ + , and 18 0 in the shallow vadose due in part to upward movement of vadose pore waters in response to high surface evaporation " wicking effect': Profiles for the phreatic and mixing zone are similar in both situations. The profiles are somewhat modified from Wagner(1983). They were developed from data from boreholes in the Pleistocene of Barbados (see text) and from profiles of various ancient stratigraphic sequences beneath subaerial exposure surfaces (Allan and Matthews, 1982; Wagner, 1983).
c c c c c c c c c c c c c c e (
c
c c c c c c
c
C' C C C C.
C C C
C ( (
o o o (
o o
o r:
C l
o
C (' (I
Limestones - The Meteoric Diagenetic Environment
phreatic there may be an offset toward heavier compositions if cementation by calcite has been slight , or offset toward lighter values if meteoric calc ite cementation has been extensive . Dissolution of aragonite and magnesium alteration of Mg-calcite may be so extensive in the vadose zone that greatly enriched pore waters lead to enrichment of Sr2+ and Mg2+ in diagenetic calcite. In summary, in warm climates the supply of meteoric water is important in determining the rate, intensity and even direction of diagenetic alteration. Variations in rainfall may be substantial across the larger oceanic islands that rise some hundreds of metres above sea levelbecause of rain shadow effects and are generally even greater across large land masses. LARGE-SCALE VARIATIONS THROUGH GEOLOGIC TIME It is common practice to extrapolate the processes and products of meteoric diagenesis observed today to the fossil record. As emphasized by James and Choquette (this volume) , however, there have been changes in the nature of the atmosphere and biosphere through geologic time which may have profoundly influenced the style of meteoric diagenesis (Figure 39). Changes in the Atmosphere. It now seems probably that there have been variations in atmospheric PC02 through the Phanerozoic. The atmosphere seems to have oscillated between a "greenhouse" mode when SEDIMENTARY MINERALOGY
o
T
100
K
200
J 'Ii
(
(,
6
PC02 levels were higher than those today, and an "icehouse" mode when they were more like the modern atmosphere (Fischer, 1981). During the greenhouse mode, meteoric waters , with increased PC02, would have been more aggressive than they are today. How this would be reflected in the rock record is difficult to assess. In areas of thin soil cover, or at times when soils were poorly developed, increased simple corrosion would lead to more extensive surface karst. Intuitively, one feels that there should be more extensive karst development during these times, yet most extens ive dissolution occurs beneath a soil and not on bare rock. So the effect should be most noticeable in areas where there is little soil development, or at times when soils were poorly developed. Changes in the Biosphere. Terrestrial: Weathering effects and the generation of CO2-charged waters are heavily dependent upon the presence and evolution of higher plants. Although land plant fossils are known from the early Paleozoic, theywere not a significant part of the terrestrial landscape until Devonian time, but were widespread by the early Mississippian . Thus, even though soils may have existed in the early and middle Paleozoic, they would have been close to the protosoils of today, dominated by algae and lichens , and unlike the complex biogenic profiles in true soils. Marine: Of equal importance is the mineralogy of carbonate sediments LANO PLANTS
CUMATIC EPISODES
METEORIC OIAGB4ESIS
ANGIOSPERMS
ICE HOUSE
1
0
r
C
400
D S
500
0 C
r
800
P-C
GYMNOSPERMS
HOUSE
LOWER VASCULAR PLANTS
ICEHOUSE
ALGAE
,,
,
GREEN HOUSE
T K
GREEN
I 300
65
2
J 'Ii ~
3
4
• LICHENS
C 0 S
0
C ICE HOUSE
5
PC
Figure 39 A plot of variations in non-skeletal mineralogy (James and Choquette, 1983; Sandberg, 1983), skeletal mineralogy (approximated from Lowenstam, 1963; Wilkinson, 1979), terrestrial vegetation and atmospheric PC02 (Fischer, 1981) throughout Phanerozoic time. Numbers at right indicate approximate intervals when the combined effects of these variables probably resulted in different styles of meteoric, particularly vadose, diagenesis. ,-
that were available to be altered. As pointed out by Lowenstam (1963) and Wilkinson (1979), the mineralogical composition of shallow-water skeletons has varied through geologic t ime. Although the trends are not yet firmly established, first-order variations seem clear (Figure 39). In general, sediments on modern and Cenozoic shallow platforms are or were mostly aragonite and magnesium-calcite, whereas those in similar Mesozoic settings were about equally aragonite and calcite (either high-magnesium and low-magnesium). In contrast, sediments in the middle and early Paleozoic were mostly calcite (except perhaps for the earliest Cambrian). The history of the mineralogy of non-skeletal particles seems more complex (Sandberg, 1983), and varied in an oscillating fashion. Because the style of diagenesis is partly dependent upon sedimentary mineralogy, those sediments with abundant aragonite will have undergone the most extensive alteration and have been cemented most in the meteoric environment. Conversely, those predominantly calcite and Mg-calcite sediments will be little affected and illustrate features because of water-controlled alteration. These differences are well demonstrated byanyone who has marvelled at the superb preservation of early Paleozoic fossils, yet struggled to discern the nature ofthe numerous recrystallized and partially dissolved skeletons in Cenozoic limestones (Figure 40). It often seems that the older the limestone, the better the preservation! Synthesis: By integrating these variables in a general way,wecan speculate on potential variations in diagenesis. The Phanerozoic could be subdivided into five periods when the effects of meteoric diagenesis were probably different (Figure 39). 1. OLIGOCENE - PRESENT. Since conditions were similar to those today and in the recent past, the present meteoric environment is a good guide to rocks of this age. 2 . LATE TRIASSIC - EOCENE. A period when sediments had less aragonite, atmospheric CO2 was higher and there were no terrestrial grasses, although plant cover in general was extensive. The spectrum of textures and fabrics describedin this article shouldbe expected for rocks of this age, with perhaps more obvious surface karst.
c
66
3. LATE DEVONIAN/EARLY CARBONIFEROUS - MID TRIASSIC. During these times, the atmospheric PC0 2 was similar to that of today, plant
cover was potentially extensive, and sedimentary mineralogy was roughly the same as in younger rocks (less skeletal aragonite). Both surface karst and calcretes probably developed commonly at this time, with somewhat less cementation and dissolution and so less fabric-selective diagenesis than in younger rocks. 4.MID CAMBRIAN-MID DEVONIAN. A time quite different from the recent past: atmospheric PC0 2 was relatively high. There were no true land plants, so soils were poorly developed. Carbonate sediments were composed mostly of calcite. The effects of meteoricdiagenesis at this time were probably minimal and confined to surface and subsurface karst, non-fabric-selective alteration, little cementation, little or no caliche and much fabric retention. 5. LATE PRECAMBRIAN? - EARLY CAMBRIAN. Scant evidence suggests that the atmosphere was like that of today in terms of Pco 2 and sedimentary mineralogy was like that in the middle Paleozoic, but that there were poor soils. Again much karst, and little or no caliche were developed, but some fabric-selective alteration. Cementation would have been the most significant results of meteoric diagenesis.
Diagenesis
It must be remembered that this scheme is a generalization, a background against which to consider meteoric diagenesis, realizing that local conditions may completely overprint the general trends. Conversely, where lack of alteration or absence of meteoric diagenetic features has in the past been interpreted as lack of exposure, it is also clear that exposure may have taken place without any significant signature. SUMMARY Carbonate sediments commonly are exposed to percolating meteoric waters at various times during their history. Since they are mostly deposited in shallow waters, small changes in sea level, whatever the cause, will result in subaerial exposure early in their history, either immediately after deposition or after shallow burial. Once they are exposed, a predictable hydrologic regimen is established, with only minor surficial exposure resulting in relatively deep percolation of fresh water and accompanying diagenesis. Although the occurrence of meteoric diagenesis can be established from textural and fabric criteria, pinpointing specific zones of alteration is difficult because, with time and fluctuating sea level, the zones migrate through the rock. Stages of alteration are thereby superimposed or early stages of diagenesis are obliterated by later ones.
Geoscience Canada Reprint Series 4
Although many reactions take place in this setting, in terms, of carbonate diagenesis two are critical: (1) a firstorder reaction, water-controlled (congruent) meteoric diagenesis, controlling dissolution and precipitation in general by invasion and evasion of CO2 from meteoric waters; all CaC0 3 minerals including calcite are dissolved, but only calcite is precipitated; (2) a secondorder reaction, mineral-controlled (incongruent) meteroric diagenesis, which affects minerals of differing solubilities; aragonite and Mg-calcite, but not calcite, are dissolved and new diagenetic calcite is precipitated. Both reactions occur concurrently when metastable carbonates are flushed by meteoric waters, but once the minerals haveall changed to calcite, only the firstorder reaction continues and much of the meteoric environment becomesdiagenetically inactive. Regardless of which reaction is prevalent and what carbonate minerals are present, there is a net loss of carbonate from the system. This is best reflected in a variety of karst-related features. Atthe sametime, soil-related processes form a variety of very distinctive caliche carbonates which, because they are accretionary, are preserved and are extremely useful as criteria of subaerial exposure. If the exposed sediments are composed of metastable carbonates, a specific, predictable series of changes takes place, and lithification is achieved
c c c
c c c c c c c c c c c c c (
c
c c c c c c c c c c c c c c e c
(
C l Figure 40 (Left) Ordovician limestone with all the fossils, trilobites (T), brachiopods (B), and bryozoans (C), preserved because they were originally calcite. (Right) Pleistocene limestone illustrating porosity developed by dissolving stick-corals (C) and gastropods (G).
C C. (
C
( (
Limestones - The Meteoric Diagenetic Environment
67
c C'
o CJ C' (
c ( (
o o o o o o c
o o o o o o ( c C l
C l
by precipitation of cements with distinctive structures which are preserved in ancient limestones . Concomitantly, because the process of alteration involves dissolution and precipitat ion, there is important cation and isotopic exchange with meteoricwaters,leading generally to lower Mg2+ and Sr2+ contents and higherFe2+ , Ca2 + and Mn2 + , as well as depletion in both 513C and /) 1 8 0 , compared with the original sediments. The intensity of meteoric diagenesis is dependent not only upon where in the environment changes have take place, but also upon the residence time in any particular zone . Climate is also an important factor in vadose alteration . The processes and products of meteoric diagenesis described in this article cannot be applied equally to carbonates of all ages. Changes in the biosphere (vary ing skeletal mineralogy and changes in terrestrial vegetation) and the atmosphere (varying Peo.) through geologic time will have had a profound , but as yet uncertain, effect on the style and intensity of freshwater diagenesis . Finally, when examining any limestone outcrop it should be remembered that this rock is now in the meteoric environment and many of the fabrics and textures present may be the result of very recent processes!
ACKNOWLEDGEMENTS We thank many colleagues who offered suggestions and guidance during the preparation of this manuscript: W. Back, R.G.C. Bathurst , N. Chow, M. Coniglio, D.H. Craig, A. Desrochers , J.A.D . Dickson, D. Haywick, J. Kennard, K.C. Lohmann, R. Reeder, P. Sandberg, L. Vacher, P. Wagner and B.H. Wilkinson . N.P. James acknowledges the Natural Sciences and Engineering Research Council of Canada for ongoing financial support.
REFERENCES General References These articles and books address various aspects of meteroic diagenesis and together cover most of the topics discussed in this paper . Allan, J.A. and Matthews, A.K., 1982, Isotope signatures associated with early meteoric diagenesis : Sedimentology, v. 29, p. 797-817. A good discussion of the theory behind and examples of isotop ic variations during meteoric diagenes is. Bathurst, A.G.C ., 1975, Carbonate Sediments and their Diagenesis: Elsevier Scientific Publishing Co., Amsterdam , 658 p. Bathurst, R.G.C., 1980, Lith ification of carbonate sediments: Science Prog ress Oxford, v. 66, p. 451-471. These two references are the best syntheses of information concerning the alteration of metastable carbonates in the meteoric environment. Bathurst, A.G.C., 1986, Carbonate diagenesis and reservoir development: conservation , destruction and creation of pores : Colorado School of Mines Quarterly, v. 81, No.4, p. 1-24. A succinct review of meteoric and other diagenesis , focussed on the evolution of porosity. Bagl i, J., 1980, Karst Hydrology and Physical Speleo logy: Springer-Verlag, Berlin, 285 p. A text devoted to the processes of karst formation: a good single source outlining basic concepts of rock-water interactions. Bricker, O.P. , 1970, ed ., Carbonate Cements: The Johns Hopkins Press, Baltimore, MD, 376 p. This book conta ins a suite of papers on meteoric diagenesis with excellent illustrations of different types of meteoric cements. Chafetz, H.S., Wilkinson , B.H. and Love, K.M., 1985, Morphology and composition of non-marine cements in near-surface settings, in Schne idermann, N. and Harr is, P.M., eds. , Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Spec ial Publication No. 36 , p. 337-348 . A good review of some of the cements precipitated in the meteoric environment, with excellent photographs. Choquette, PW. and Pray, L.C., 1970, Geological nomenclature and classificat ion of porosity in sedimentary carbonates: American Association of Petroleum Geologists, Bullletin, v. 54, p. 207-250. An outline of the basic concepts of fabricselective and non-fabric selective alteration of carbonates.
Esteban, M. and Klappa , C.F., 1983, Subaerial exposure, in Scholle, P.A., Bebout , D.G. and Moore , C.H., eds., Carbonate Depositional Environments: American Associat ion of Petroleum Geolog ists, Memoir No. 33 , p. 1-54. A beautifully illustrated, up to date reference synthesizing much of our understanding of surficial diagenesis . Folk, A.L., 1973, Carbonate petrography in the post -Sorbyian age, in Ginsburg, A.N., ed., Evolving Concepts in Sedimentology : Johns Hopkins University, Studies in Geology, v. 21, p. 118-158. A succinct essay presenting, amongst other things, an outline of the processes and products during subaerial diagenesis of metastable carbonates. Halley, A.B. and Harr is, P.M., 1979, Freshwater cementation of a 1,000 year old oolite: Journal of Sedimentary Petrology, v. 49, p. 969-988 . One of the few studies to integrate petrographic, geochemical and hydrographic data in the study of the early diagenesis of metastable carbonates. James , N.P. and Choquette , P.W., 1988, eds., Paleokarst: Springer-Verlag, New York, 421 p. A series of 18 papers covering all major aspects of karst and paleokarst, examples range in age from Precambrian to modern. Longman, M.W., 1980, Carbonate diagenetic textures from near-surface diagenetic environments: American Association of Petroleum Geologists, Bulletin, v. 63, p. 461-487. A useful summary of the processes and products of meteoric diagenesis of metastable carbonates, a good place to begin. Matthews, A.K., 1974, A process approach to the diagenesis of reefs and reef-associated limestones , in Laporte, L.F., ed., Aeefs in Time and Space: Society of Economic Paleontolog ists and Mineralogists, Special Publication No. 18, p.234-256. A useful paper describing many of the principles of meteoric diagenesis using the Pleistocene of Barbados as an example. Matthews , R.K., 1987, Eustatic controls on near-surface carbonate diagenesis: Colorado School Mines Quarterly, v. 82, No.1, p. 17-39. Outline of the dynamics of meteoric diagenesis with several examples to illustrate the principles. Plummer, L.N., 1975, Mixing of sea water with calcium carbonate groundwater, in Whitten, E.T.H., ed., Quantitative Studies in the Geological Sciences: Geological Society of American , Memoir 142, p.219-236 . A discussion of the chemical basis for many of the concepts of freshwater-seawater mixing.
c
68
Plummer, L.N., Vacher, H.L., Mackenzie, F.T., Bacher, O.P. and Land, L.S., 1976, Hydrogeochemistry of Bermuda: A case history of groundwater diagenesis of biocalcarenites: Geological Society of American, Bulletin , v. 87, p. 1301-1316. An integrated study of groundwater geochemistry of an island where many of the principles of metastable carbonate diegenesis were worked out. Sweeting , M.M ., 1972, Karst Landforms : Macmillan Publishing Co. Ltd., London , 362 p. A classic text discussion of the complete spectrum of karst features. Thrailkill,J ., 1968,Chemical and hydrologic factors in the excavation of limestone caves: Geological Society of America, Bullet in, v. 79, p. 19-46. A very readable outline of the principles of carbonate dissolution and cave formation. Thraikill, J., 1976, Speleothems, in Walter, M.R., ed., Stromatolites: Elsev ier Publishing Co., Amsterdam, p. 73-86. A well-illustrated paper describing the different types of sinter.
Hydrology of Carbonate Terranes Bear, J., 1979, Hydraulics of Groundwater: McGraw-Hili Inc., New York, 569 p. Bear,J. and Todd, O.K., 1960,The transition zone between fresh and salt waters in coastal aquifers: Water Resources CenterContribution No. 29, Hydrau lic Laboratory, University of California, 156 p. Back, Wand Hanshaw, B.B., 1970, Comparison of chemical hydrogeology of the carbonate peninsulas of Florida and Yucatan: Journal of Hydrology, v. 10, p.330-368. Back, W. and LaMoreaux, P.E., 1983, eds., V.T. Stringfield Symposium - Processes in Karst Hydrology: Journal of Hydrology, v. 61, p. 1-355. Buddemeier, R.W. and Oberdorfer, J.A. , 1986, Internal hydrology and geochemistry of coral reefs and atoll islands: key to diagenetic var iations, in Schroeder, J.H. and Purser, B.H., eds., Reef Diagenesis: Springer-Verlag , New York, p.91 -111. Davis, S.N. and DeWiest, R.J.M., 1966, Hydrogeology: John Wiley and Sons, New York, 463 p. Fetter, C.W., Jr., 1980, Applied Hydrogeology: Merrill and Company, Columbus , Ohio, 488 p. Johnson , R.H., 1983, The saltwater-freshwater interface in the Tertiary limestone aquifer, southeastern Atlantic continental shelf of the U.S.A.: Journal of Hydrology, v. 61, p. 239-251. Kohout, FA , 1960, Cyclic flow of salt water in the Biscayne aquifer of southeastern Florida: Journal of Geophysical Research, v. 65, p. 2133-2141.
Diagenesis
Manhe im, F.T. , 1967, Evidence for sub marine discharge of water on the Atlantic continental slope of the southern United States and suggestions for further research : New York Academy of Sc ience , Transact i o ns, v. 29 , p.839-853 . Smith, 0.1., Atkinson , T.C. and Drew, D.P., 1976, The hydrology of limestone terranes, in Ford, T.D. and Cull ingford, C.H.D., eds., The Science of Speleology: Academic Press Inc., New York, 593 p. Todd, D.K., 1980, Groundwater Hydrology: John Wiley and Sons, New York, 535 p. Vacher, H.L., 1974, Groundwater hydrology of Bermuda : Government of Bermuda, Public Works Department, Hamilton, 87 p. Vacher, H.L., 1978, Hydrology of Bermudasignificance of an across-the-island variation in permeability: Journal of Hydrology, v. 39, p. 207-226. Wheatcraft, S.W and Buddemeir, R.W., 1981, Atoll island hydrology: Groundwater, v. 19, p. 311-320. White, WB., 1969, Conceptual models for carbonate aquifers: Groundwater, v. 7, p. 15-21.
Dissolution and Precipitation (see also sections on geochemistry and freshwater-seawater mixing) Badiozamani , K., 1973,The dorag dolomitization model- application to the middle Ordovician of Wisconsin: Journal of Sedimentary Petrology, v.43 , p.965-984. Drever,J.I., 1982,The Geochemistry of Natural Waters: Prentice-Hall, Englewood Cliffs, NJ, 388 p. Folk, R.L., 1974, The natural history of crystalline calc ium carbonate - effect of magnesium content and salinity: Journal of Sed imentary Petrology, v. 44, p.40-54. Folk, R.L. and Land, L.S., 1975, The MgfCa ratio and salinity: two controls over crystallization of dolomite: American Associat ion of Petroleum Geologists, Bulletin , v. 59, p. 60-69. Haner, J.S., 1978,Precipitation of beachrock cements - mixing of marine and meteoric waters vs. CO2-degassing: Journal of Sedimentary Petrology,v. 48, p. 489-501. Kinsman, D.J.J., 1969, Interpretation of Sr2 + concentrations in carbonate minerals and rocks: Journal of Sedimentary Petrology, v. 39, p. 486-508 . Plummer, L.N ., Wigley, T.M.L. and Parkhurst, D.L., 1979, Critical review of the kinetics of calcite dissolution and precipitation, in Gould, R.F., ed., Chemical Modeling in Aqueous Systems: American Chem ical Society, Symposium Series No. 93, p. 537-577.
Geoscience Canada Reprint Series 4
Runnells , D.D., 1969, Diagenesis of chemical sediments and the mixing of natural waters: Journal of Sedimentary Petrology, v. 39, p. 1188-1201. Schmalz, R.F., 1967, Kinetics and diagenesis in carbonate sediments: Journal of Sedimentary Petrology, v. 37, p. 60-68. Walter, L.M. and Morse, J,W., 1984, Magnesian calcite stabilities : a reevaluation: Geochimica et Cosmochimica Acta , v. 48, p. 1059-1069. Walter, L.M and Morse, J.W., 1985, The dissolution kinetics of shallow marine carbonates in seawater: a laboratory study : Geoch imica et Cosmochimica Acta , v. 49, p. 1503-1513.
Modern Caliche Coniglio, M. and Harrison, R.S ., 1983, Holocene and Pleistocene caliche from Big Pine Key Florida: Bulletin of Canadian Petroleum Geology, v. 31, p. 3-14 . Harrison, R.S., 1977, Caliche profiles, indicators of near-surface subaerial diagenesis Barbados, West Indies: Bulletin of Canadian Petroleum Geology, v. 25, p.123-173. James , N.P., 1972, Holocene and Pleistocene calcareous crust (caliche) profiles ; criteria for subaerial exposure: Journal of Sedimentary Petrology, v.42, p.817-836 . Klappa , C.F., 1978, Biolithogenesis of Microcodium - elucidation: Sedimentology, v. 25, p. 489-522. Klappa , C.F., 1979, Lichen stromatolites criterion for subaerial exposure and a mechanism for the formation of lamina r calcretes (caliche): Journal of Sedimentary Petrology, v. 49, p. 387·400. Peryt, T.M., 1983,Coated Grains: SpringerVerlag, Berlin , 655 p. Read, J .F., 1974, Calcrete depos its and Quaternary sediments , Edel Province , Shark Bay, Western Australia, in Logan, B.W., ed., Evolution and Diagenesis of Quaternary Sequences , Shark Bay, Western Australia : American Association of Petroleum Geologists, Memoir 22, p. 250-280. Read, J.F.,1976,Calcretes and theirdistinction from stromatolites, in Walter, M.R., ed. , Stromatolites: Elsevier, Amsterdam, p. 55-71. Reeves, C.C., 1976,Caliche - origin , classification, morphology and uses : Estacado Books, Lubbock, TX, 233 p. Robbin , D.M. and Stipp, J.J., 1979,Depositional rate of laminated soilstone crusts , Florida Keys: Journal of Sedimentary Petrology, v. 49, p. 175-181. Shinn , E.A. and Lidz, B.H., 1988,Blackened limestone pebbles: fire at subaerial unconformities, in James, N.P. and Choquette, P.W, eds., Paleokarst: SpringerVerlag, New York, p. 117-131.
c ( c c c ( c C t.....
C C
C C C C
C C C
C C C C C C C
C C C E C C
C C
c
e \~ (
C C C (
c
(
c:
Limestones - The Meteoric Diagenetic Environment
69
r:
c r
c
c C' c (
(. ( (
o o (
o
o o o o o o o c c.
C' 6 C. ( r '-..
""-.. r:
Scholle , P.A. and Kinsman , D.J.J., 1974, Aragon ite and high magnesium calc ite caliche from the Pers ian Gulf - a modern analog forthe Perm ian of Texas and New Mexico: Journal of Sedimentary Petrology, v. 44, p. 904-916. Ward WC ., Folk , R.L. and Wilson , J.L., 1970, Blackening of eolinite and caliche adjacent to saline lakes Isla Mujeres, Quintana Roo Mexico: Journal of Sedimentary Petrology, v. 40, p. 548-555. Watts , N.L., 1980, Quaternary pedogenic calcretes from the Kalahari (southern Africa) mineralogy genesis and diagen esis: Sedimentology, v. 27, p. 661-686.
Ancient Soils and Caliche Adams, A.E., 1980, Calcrete profiles in the Eyam Limestone (Carboniferous) of Derbyshire petrology and regional sign ificance : Sedimentology, v. 27,p. 651-660 . Bernouilli , D. and Wagner, C.W , 1971, Subaerial diagenesis and fossil caliche in the Calcare Massiccio Formation (Lower Jurassic central Appen ines , Italy) : Neues Jahrbuch fOrGeologie und Palacntoloqie, Abhandlungen, v. 138, p.135-149. Esteban, M., 1976, Vadose pisolite and caliche: American Association of Petroleum Geologists, Bulletin , v. 60, p . 20482057. Goldhammer, R.K. and Elmore, R.D., 1984, Paleosols capp ing regressive carbonate cycles in the Pennsylvanian Black Prince Limestone, Arizona: Journal of Sedimentary Petrology, v. 54, p . 11241137. Harrison, R.S. and Ste inen, R.P., 1978,Subaerial crusts, caliche profiles and breccia horizons, comparison of some Holocene and Misissippian exposure surfaces Barbados and Kentucky: Geolog ical Society of America, Bulletin, v. 89, p. 385-396 . Walls, R.A ., Harris, W.B. and Nunam , WE., 1975, Calcareous crust (caliche) profiles and early subaerial exposure of Carboniferous carbonates, northeastern Kentucky: Sedimentology, v. 22, p. 417·440. Wright, V.P., 1984, The significance of needle fibre calcite in a Lower Carboniferous paleosol: Geological Journal, v. 19, p.23-32. Wright, V.P., 1986, ed., Paleosols: their recognition and interpretation : Blackwell Scientific Publications, London. Wright, V.P. and Wilson, R.C.L., 1987, A terra rossa-like paleosol complex from the Upper Jurass ic of Portugal: Sedimentology, v. 34, p. 259-274.
Karst in General Bagli, J., 1980, Karst Hydrology and Physical Speleology: Springer-Verlag, Berlin, 285 p. Jakucs, L., 1977, Morphogenetics of Karst Regions: John Wiley and Sons, New York, 284 p. James, N.P. and Choquette, P.W, 1988, eds., Paleokarst: Springer-Verlag, New York, 421 p. Jennings , J.N., 1971, Karst: The M.I.T. Press , Cambridge, MA, 252 p. Ritter, D.F., 1978, Process Geomorphology : WC. Brown , Dubuque, Iowa, p. 465-511 . Sweeting, M.M., 1972, Karst Landforms : Macmillan Publishing Co. Ltd., London, 362 p.
Surface Karst Allen , J.R.L., 1982 , Sedimentary Structures: their character and physical basis: Elsevier, Amsterdam, p. 222-251. Bourrouihl, F., 1974, Donnees geomorphologiques sur la reg ion de Fresh Creek, lie d'Andros (Bahama): Marine Geology, v. 16, p. 213-235. Chafetz, H.S. and Folk, R.L., 1984, Travertines : depositional morphology and the bacterially constructed constitutents : Journal of Sedimentary Petrology, v. 54, p.289-317. Folk, R.L., Roberts, H.H. and Moore, C.H., 1973, Black phytokarst from Hell Cayman Islands: Geological Society of America, Bulletin , v. 87, p. 2351-2360. Julia, R., 1983, Travertines , in Scholle, P.A., Bebout, D.G. and Moore , C.H ., eds ., Carbonate Depositional Environments: American Association of Petroleum Geologists, Memoir 33, p. 64-73. Purdy, E.G., 1974, Reef configurations , cause and effect, in Laporte, L.F., ed., Reefs in Time and Space : Soc iety of Economic Paleontologists and Mineralog ists , Special Publication No . 18, p.9-77. Str ingfield, Y.T. and LeGrand , H.E., 1974, Karst hydrology of northern Yucatan Peninsula, Mex ico , In Weldie , A.E., ed., Field Seminar on water and carbonate rocks of the Yucatan Peninsula, Mexico: New Orleans Geological Soc iety, p.26-44.
Subsurface Karst Back, W., Hanshaw, B.B ., Pyle , T.E ., Plummer, L.N. and Weidie , A.E., 1979, Geochemical significance of groundwater discharge and carbonate solut ion to the formation of Caleta Xel Ha, Quintana Roo, Mexico: Water Resources Research, v. 15, p. 1531-1535.
Back, W., Hanshaw, B.B. and Van Oriel, J.N., 1984, Role of groundwater in shaping the eastern coast line of the Yucatan Peninsula, Mexico, in LaFleur, R.G., ed., Groundwater as a Geomorphic Agent: Allen & Unwin lnc ., Winchester MA, p. 157-172. Ford , D.C. and Ewers , R.O., 1978, The development of limestone cave systems in the dimensions of length and depth : Canad ian Journal of Earth Sciences, v. 15, p. 1783-1799. Ford, D.C., 1988, Characteristics of dlssolutional cave systems in carbonate rocks , in James, N.P. and Choquette , P.W , eds ., Paleokarst: Springer-Verlag, New York, p. 25-57. Hanshaw, B.B. and Back, W., 1980, Chemical mass-wasting of the northern Yucatan Peninsula by groundwater dissolution : Geology, v. 8, p. 222-224. Kendall , A.C. and Broughton, P.L., 1978, Origin of fabrics in speleothems composed of columnar calcite crystals: Journal of Sedimentary Petrology, v. 48 , p.519-538. Moore, G.W. and Sullivan, G.N., 1978, Speleology: Cave Books , St. Louis, Missouri , 150 p. Palmer, A.N., 1984, Geomorphic interpretation of karst features, in LaFleur, R.G ., ed. , Groundwater as a Geomorph icAgent:Allen &Unwin, Winchester, MA, p. 175-209. Plummer, L.N. , 1975, Mixing of seawater with calcium carbonate groundwater, in Whitten , E.T.H., ed., Quantitative Studies in the Geological Sciences : Geolog ical Society of Amer ica, Memoir 142, p.219-236. Thrailkill , J., 1968, Chemical and hydrolog ic factors in the excavation of limestone caves: Geological Society of America, Bulletin , v. 79, p. 19-16. Thrailkill, J., 1976, Speleothems, in Walter, M.R., ed., Stromatolites: Elsevier PubIishing Co ., Amsterdam, p. 73-86 . Vernon, P.O ., 1969, The geology and hydrology associated with a zone of high permeab ility (Boulder Zone) in Flor ida : Society of Mining Engineers, Preprint 69-AG-12: Geochimica et Cosmochim ica Acta, v. 40, p. 989-995 . Wigley, T.M.L. and Plummer, N.L., 1976, Mixing of carbonate waters: Geochimica et Cosmochimica Acta, v. 40, p. 989995.
Paleokarst Cherns, L., 1982, Palaeokarst, tidal erosion surfaces and stromatolites in the Silurian Eke Formation of Gotland, Sweden: Sedimentology, v. 29, p. 819-833. James, N.P. and Choquette, P.W., 1988, eds., Paleokarst: Springer-Verlag, New York,421 p.
70
Kobluk, D.A., Pemberton, S.G., Karolyi, M. and Risk, M.J., 1977, The Silurian-Devo nian disconformity in southern Ontario: Bulletin of Canadian Petroleum Geology, v. 25, p. 1157-1186. Read, J.E and Grover, G.A., Jr., 1977, Scalloped and planar erosion surfaces: Middle Ordovician limestone, Virginia: analogues of Holocene exposed karst ortidal rock platforms: Journal of Sedimentary Petrology, v. 47, p. 956-972. Walkden, G.M., 1974, Palaeokarstic surfaces in upper Visean (Carboniferous) limestones of the Derbyshire block, England: Journal of Sedimentary Petrology, v. 44, p. 1232-1247. Wright , V.P., 1982, The recognition and interpretation of paleokarsts, two examples from the lower Carboniferous of south Wales : Journal of Sedimentary Petrology, v. 52, p. 83-94.
Diagenesis
Friedman, G.M. , 1964, Early diagenesis and lithification in carbonate sediments : Journal of Sedimentary Petrology, v. 34, p.777-813 . FIOgel, E., 1982, Microfacies Analysis of Limestones: Springer-Verlag, Berlin, 633 p. Gavish, E. and Friedman, G.M., 1969, Progressive diagenesis in Quaternary to late Tertiary carbonate sediment sequence and time scale : Journal of Sedimentary Petrology, v. 39, p. 9801006. Harrison, A.S., 1975, Porosity in Pleistocene grainstones from Barbados, some preliminary observations: Bulletin of Canadian Petroleum Geology, v. 23, p.383-392. Land, L.S., 1967, Diagenesis of skeletal carbonates : Journal of Sedimentary Petrology, v. 37, p. 914-930. Purdy, E.G., 1968, Carbonate diagenesis an environmental survey: Geologica Romana, v. 7, p. 183-228. Thorstenson, D.C., Mackenzie, F.T. and Ristvet , B.L., 1972, Experimental vadose and phreatic cementation of skeletal carbonate sand: Journal of Sed imentary Petrology, v. 42 , p. 162167.
Dickson, J.A.D., 1966, Carbonate identification and genesis as revealed by staining: Journal of Sedimentary Petrology, v. 36, p. 491-505. Dickson, J.A.D., 1983,Graphical modelling of crysta l aggregates and its relevance to cement diagnosis: Royal Society of London , Philosophical Transactions , v. A309, p. 465-502 . Dunham, A.J., 1969, Early vadose silt in Townsend Mound (reef), New Mexico, in Friedman , G.M. , ed., Depositional environments in carbonate rocks: asymposium: Society of Economic Paleontologists and Mineralogists, Spec ial Publication No. 14, p. 182-192. Evans, C.E. and Ginsburg, R.N., 1987, Fabric-selective diagenesis in the late Pleistocene Miami Limestone: Journal of Sedimentary Petrology, v. 57, p.311-318. Folk, A.L. and Land, L.S., 1975, Mg/Ca ratio and salinity: two controls over the crystallization of dolomite: American Associat ion of Petroleum Geologists, Bulletin, v. 59, p. 60-68. Jones, B., Lockhart, E.B. and Squair, C., 1984, Phreatic and vadose cements in the Tertiary Bluff Formation of Grand Cayman Island , British West Indies: Bulletin of Canadian Petroleum Geology, v. 32, p. 382-397. Lindholm, A.C. and Finkleman , A.B., 1972, Calcite staining: semiquantitative determination offerrous iron: Journal of Sedimentary Petrology, v. 42, p. 239-242. Meyers, W.J., 1974,Carbonate cement stratigraphy of the Lake Valley Formation (Mississippian) Sacramento Mountains , New Mexico: Journal of Sedimentary Petrology, v. 44, p. 837-861, Meyers, W.J., 1978, Carbonate cements their regional distribution and interpretation in Mississippian limestones of southwestern New Mexico: Sedimentology, v. 25, p. 371-400. Oldershaw, A.E. and Scottin, T.P., 1967, The source of ferroan and non-ferroan calcite cements in the Halkin and Wenlock Limestones: Geological Journal, v. 5, p.309-320. Schroeder, J.H. , 1973, Submarine and vadose cements in Pleistocene Bermuda reef rock: Sedimentary Geology, v. 10, p. 179-204.
Cementation
Magnesium-Calcite Alteration
Bathurst, A.G.C. , 1983, Neomorphic spar versus cement in some Jurassic grainstones : significance for evaluation of porosity evolution .and compaction: Journal ofthe Geological Society, v. 140, p.229-237. Bricker, O.P., 1971, ed., Carbonate Cements: The Johns Hopkins Press, Baltimore, MD, 376 p.
Benson, L.V. and Matthews, R.K. , 1971, Electron microprobe studies of magnes ium distribution in carbonate cements and recrystallized skeletal grainstones from the Pleistocene of Barbados , West Indies: Journal of Sedimentary Petrology, v. 41, p. 1018-1025.
Mineral-controlled Meteoric Diagenesis Overview
Geoscience Canada Reprint Series 4
Blake, D.E and Peacor, D.R., 1981, Biomineralization in crino id echinoderms: characte rization of crinoid skeletal elements using TEM and STEM analysis: Scanning Electron Microscopy III , p, 321-328. Macqueen, R.W., Ghent, E.D. and Davies, G.R., 1974, Magnesium distribution in living and fossil specimens of the echinoid Peronella lesceuric Agassiz, Shark Bay, Western Australia : Journal of Sedimentary Petrology, v. 44, p. 60-69. Moberly, R., 1970, Microprobe study of diagenesis in calcareous algae: Sedimentology, v. 14, p. 113-123. Plummer, L.N. and Mackenzie , ET., 1974, Predicting mineral solubility from rate data , application to the dissolut ion of magnesium calcites: American Journal of Science, v. 274, p. 61-83. Richter, D.K. and FOchtbauer,H., 1978,Ferroan calcite replacement indicates former magnesian calcite skeletons : Sedimentology, v. 25, p. 843-861, Schroeder, J.H. , 1969, Experimental dissolution of calcium, magnesium and strontium from recent biogenic carbonates: a model of diagenesis : Journal of Sedimentary Petrology, v.39 , p. 1057-1073. Towe, K.M. and Hemleben, C., 1976, Diagenesis of magnesian calcite evidence from miliolacean foraminifera: Geology, v. 4, p. 333-339 . Turner, J.V.,Anderson, T.E , Sandberg, P.A. and Goldstein, S.J., 1986, Isotopic, chemical and textural relations during the experimental alteration of biogenic high-magnesian calcite : Geochim ica et Cosmochimica Acta, v. 50, p. 495-506. Aragonite Alteration
Assereto , R. and Folk, R.L., 1980, Diagenetic fabrics of aragonite, calcite and dolomite in an ancient peritidal-spelean environment Triassic Calcare Rosso, Lombardia, Italy: Journal of Sed imentary Petrology, v. 50, p. 371-395. Bathurst, R.G.C., 1964,The replacement of aragonite by calcite in the molluscan shell wall, in Imbrie, J. and Newell, N.D., eds ., Approaches to Paleoecology : John Wiley and Sons, New York , p.357-376. Carlson, W.D., 1983, The polymorphs of CaC0 3 and the aragonite-calcite transformation , in Reeder, R.J., ed., Carbonates : mineralogy and chem istry : Reviews in Mineralogy, v. 11, p. 191-226. Davies, G.A. , 1977, Former magnesian calcite and aragonite submarine cements in Upper Paleozoic reefs of the Canadian arct ic - a summary : Geology, v. 5, p. 11-15.
c c c c c c c ( (-
C C
C C C C C
C C C C
C C C C C C C
C C C
C C C
e c l. ( <-
(
C C C
( (
Limestones - The Meteoric Diagenetic Environment
71
(
(
c
o ( (
c (
c CO
o o o C) o o o o (.I
o o o o
o o o
o (
l r
Donath, FA, Carozzi, A.V., Fruth, L.S., Jr. and Rich, D.W , 1980, Oomold ic poros ity experimentally developed in Miss issipp ian ool it ic limestone: Journal of Sed imentary Petrology, v.50 , p. 1249-1260. James , N.P., 1974, Diagenesis of sclera etinian corals in the subaerial vadose environment: Journal of Paleontology, v. 48, p. 785-799. Martin, G.D., Wilkinson, B.H. and Lohmann, K.C., 1986, The role of skeletal porosity in aragonite neomorphism - Strombus and Montastrea from the Pleistocene Key Largo Limestone, Florida: Journal of Sedimentary Petrology, v. 56, p. 194-203. Mazzullo, S.J., 1980, Calcite pseudospar replacive of marine acicular aragonite and implications for aragonite cement diagenesis: Journal of Sed imentary Petrology, v. 50, p. 409-423. Pingitore, N.E., 1976, Vadose and phreatic diagenesis: processes,products and their recognition in corals: Journal of Sedimentary Petrology, v. 46, p. 785-1006. Pingitore, N.E., 1982, The role of diffus ion during carbonate diagenesis: Journal of Sedimentary Petrology, v. 52, p. 27-41. Pittman , E.D., 1974, Porosity and permeability changes during diagenesis of Pleistocene corals , Barbados West Indies: Geological Society of America , Bulletin, v. 85, p. 1811-1820. Sandberg, P.A., 1983,An oscillating trend in Phaneorozoic non-skeletal carbonate mineralogy: Nature, v. 305, p. 19-22. Sandberg, P.A., 1985, Aragonite cements and their occurrence in ancient limestones , in Schneidermann , N. and Harr is, P.M., eds., Carbonate Cements : Society of Economic Paleontologists and Mineralogists, Special Publication No. 36, p. 33-58. Sandberg, P.A. and Hudson, J.D., 1983, Aragonite relic preservation in Jurass ic calcite-replaced bivalves : Sedimentology, v. 30, p. 879-892 . Sandberg, P.A., Schneidermann , N. and Winder, S.J ., 1973, Aragonite ultrastructure relics in calcite-replaced Pleistocene skeletons : Nature, v. 245, p.133-134. Schroeder, J.H ., 1979, Carbonate diagene sis in Quaternary beachrock of Uyombo, Kenya: sequences of processes and coexistence of heterogen ic products : Geolog ische Rundschau , v. 68, p. 894919. Wardlaw, N., Oldershaw, A. and Stout, M., 1978,Transformation of aragonite to calcite in a marine gastropod: Canadian Journal of Earth Sciences, v. 15, p. 1861-1866.
Fine-Grained Sediments Alexandersson , T., 1972, Micri tization of carbonate part icles : processes of precipitation and dissolution in modern shallow water marine sediments : Bulletin of Geological Institute , Uppsala , v. 3, p. 201-236. Bathurst , R.G.C., 1966, Boring algae , micrite envelopes and lith ification of molluscan biosparites: Geological Journal, v. 5. p. 15-32. Bathurst, R.G.C., 1971, Problems of lithification in carbonate muds: Geologists' Assoc iation , Proceed ings , v.81 , p.429-440. Harris, P.M., Halley, R.B. and Lucas, K.J., 1979, Endolith microborings and their preservation in Holocene-Pleistocene (Bahama-Flordia) ooids : Geology, v. 7, p.216-220 . Folk, R.L., 1975, Some aspects ofrecrystallization in ancient limestones, in Pray, L.C. and Murray, R.C., eds., Dolomitization and Limestone Diagenesis : Society of Economic Paleontologists and Mineralogists, Special Publication No. 13, p.14-48. Lasemi, Z. and Sandberg, P.A., 1984,Transformation of aragonite-dominated lime muds into microcrystalline limestones : Geology, v. 12, p. 420-423. Steinen, R.P., 1978, On the diagenesis of lime mud : scanning electron microscope observations of surficial material from Barbados, WI.: Journal of Sedimentary Petrology, v. 48, p. 1139-1148. Steinen, R.P., 1982, SEM observations on the replacement of Bahamian aragonitic mud by calcite : Geology, v. 10, p.471-475 . Wiggins, W.D., 1986, Geochemical signatures in carbonate matr ix and their relation to deposition and diagenesis , Pennsylvanian Marble Falls Limestone , central Texas: Journal of Sedimentary Petrology, v. 56, p. 771-783. Vadose versus Phreatic Diagenesis Benson, L.V., 1974, Transformation of a polyphase sedimentary assemblage into a single phase rock - a chemical approach : Journal of Sedimentary Petrology, v. 44, p. 123-136. Buchbinder, L.G . and Friedman , G.M ., 1980, Vadose, phreatic and marine diagenesis of Pleistocene-Holocene carbonates in a borehole · Mediterranean coast of Israel: Journal of Sedimentary Petrology, v. 50, p. 395-409. Land, L.S., 1970, Phreatic versus vadose meteoric diagenesis of limestones: evidence from a fossil water table: Sedimentology, v. 14, p. 175-185.
Reeckmann , S.A. and Gill, E.D., 1981, Rates of vadose diagenesis in Quaternary dune and shallow marine calcaren ites, Warrnambool, Victoria, Australia: Sedimentary Geology, v. 30, p. 157-172. Steinen , R.P. and Matthews , R.K., 1973, Phreatic vs. vadose diagenesis - stratigraphy and mineralogy of a cored borehole on Barbados , WI.: Journal of Sedimentary Petrology, v. 43, p. 1012-1020. Steinen, R.P., 1974, Phreatic and vadose diagenetic modification of Pleistocene limestone, petrographic observations from the subsurface of Barbados, West Indies: American Associat ion of Petroleum Geolog ists, Bullet in , v. 58, p. 1008-1024. Videtich, R.P. and Matthews, R.K., 1980, Origin of discontinuity surfaces in limestones: isotopic and petrographic data, Pleistocene of Barbados, West Indies: Journal of Sedimentary Petrology, v. 50, p.971-981. Diagenesis in the Freshwater-Seawater Mixing Zone Back, W, Hanshaw, B.B., Pyle, T.E. and Weidie, A.E., 1976, Hydrologic study of Caleta Xel Ha, Quintana Roo, Mexico, in Weidie, A.E. and Ward, WC ., eds. , Yucatan Guidebook : New Orleans Geological Society, New Orleans, LA, p.244-258. Back, W., Hanshaw, B.B. , Pyle, T.E., Plummer, L.N. and Weidie, A.E., 1979, Geochem ical sign ificance of groundwater discharge and carbonate solut ion to the formation of Caleta Xel Ha, Quintana Roo, Mex ico : Water Resources Research, v. 14, p. 1521-1535. Back, W.,Hanshaw, B.B., Herman, J.S. and Van Driel, J.N., 1986, Differential dissolut ion of a Pleistocene reef in the ground-water mixing zone of coastal Yucatan, Mexico : Geology, v.14 , p. 137-140. Beach, D.K., 1982, Depositional and diagenetic history of Pliocene-Pleistocene ca rbonates of northwestern Great Bahama Bank, evolution of a carbonate platform, Unpublished Ph.D. thes is , University of Miami, Coral Gables, FL, 447 p. Benson, L.V., 1973, Electron microscope stud ies of carbonates, Unpublished Ph.D. thesis, Brown University, Providence , RI, 180 p. Hanshaw, B.B., Back, W. and Deike, R.G., 1971, A geochemical hypothesis for dolomitization by groundwater: Economic Geology, v. 66, p. 710-724. Harr iss, W.H., 1971, Groundwater - carbonate rock chemical interactions: Barbados, West Indies: Unpublished Ph.D. thesis, Brown University, Providence, RI, 348 p.
72
Diagenesis
Geoscience Canada Reprint Series 4
( Land, L.S., 1973a, Holocene meteor ic dolomit ization of Pleistocene limestones, North Jamaica: Sedimentology, v. 20, p.411-422. Land, L.S., 1973b, Contempo raneous dolomitization of Middle Pleistocene reefs by meteoric groundwater, north Jamaica: Bulletin of Marine Sc ience, v.23, p.64-92. Major, R.P., 1984, The Midway Atoll coral cap-meteoric diagenesis , amplitude of sealevel fluctuations, and dolomitization, Unpublished Ph.D.thesis, Brown University, Providence, RI, 133 p. Meyers , w». and Lohmann , K.C. , 1978, Microdolomite-rich syntaxial cements : proposed meteoric-marine mixing zone phreatic cements from Mississippian limestones, New Mexico: Journal of Sedimentary Petrology, v. 48, p. 475488. Moore, C.H., 1973, Intertidal carbonate cementation, Grand Cayman, West Indies: Journal of Sedimentary Petrology, v. 43, p. 591-602. Pierson , B., 1983, Cyclic sedimentation , limestone diagenesis and dolomitization in UpperCenozoiccarbonatesofthe southwestern Bahamas, Unpublished Ph.D. thesis, University of Miami , Coral Gables, FL, 295 p. Ristvet, B.L., 1971, The progressive diagenetic history of Bermuda : Bermuda Biological Station for Research Special Publication 9, p. 118-157. Steinen, R.P., Matthews, R.K. and Sealy, H.A., 1978, Temporal variation in geometry and chemistry of the freshwater phreatic lens - the coastal aquifer of Christ Church, Barbados , West Indies: Journal of Sedimentary Petrology, v. 48, p.733-743. Wagner, P.O., 1983, Geochemical characterizations of meteoric diagenesis in limestone - development and applications. Part III. Mixing Zone diagenesis, Unpublished Ph.D. thesis, Brown University, Providence, RI, p. 123-146. Geochemical Aspects of Meteoric Diagenesis
Al-Aasrn, I.S. and Veizer, J., 1986a, Diagenetic stabilization of aragonite and LowMg calcite - I. trace elements in rudists : Journal of Sedimentary Petrology, v. 56, p. 138-152. AI-Asam, I.S. and Veizer, J., 1986b, Diagenetic stabilization of aragonite and LowMg calcite -II. stable isotopes in rudists : Journal of Sedimentary Petrology, v. 56, p.763-770. Allan, J.R . and Matthews, R.K., 1977, Carbon and oxygen isotopes as diagenetic and stratigraphic tools - data from surface and subsurface of Barbados, West Indies: Geology, v. 5, p. 16-20.
Brand, U. and Veizer, J., 1980, Chemical diagenesis of a multicomponent carbonate system -1. Traceelements: Journal of Sed imentary Petrology, v. 50, p. 1219-1236. Brand, U. and Veizer, J., 1981, Chemical diagenesis of a multicomponent carbonate system - 2. Stable isotopes: Jour na l of Sed imentary Petrology, v. 51, p.987-997. Choquette, P.w" 1978, Recent calcareous crusts (caliche) on Isla Mujeres, Quintana Roo, Mexico - insights from isotopic composition, in Ward, W.C. and Weidie, A.E., eds., Carbonate Rocks and Hydrology of the Yucatan Peninsula, Mexico: New Orleans Geological Society, [Guidebook for Field Trip 2], New Orleans, LA , p. 171-190. Dansgaard, W., 1964, Stable isotopes in precip itation : Tullus, v. 16, p. 436-468 . Galimov, E.M., 1966,Carbon isotopes of soil CO2 : Geochemistry International, v. 3, p.889-897. Galimov, E.M., 1967, Isotopic composition of carbon in gases of the crust: Geological Review, v. 11, p. 1092-1104. Gonzalez , L.A. and Lohmann, K.C., 1988, Controls on mineralogy and composition of spelean carbonates: Carlsbad Caverns, New Mexico, in James, N.P. and Choquette, P.w" eds., Paleokarst: Springer-Verlag, New York, p. 81-101. Gross, M.G., 1964,Variations in the 18 0 / 16 0 and 13C/12C ratios of diagenetically altered limestones in the Bermuda Islands: Journal of Geology, v. 72, p.170-194. Given, R.K. and Lohmann, K.C., 1985,Derivation of the original isotopic compos ition of Permian marine cements : Journal of Sedimentary Petrology, v. 55, p.430-439. Given, R.K. and Lohmann , K.C., 1986, Isotopic evidence of the early meteoric diagenesis of the reef facies, Permian reef complex, west Texas and New Mexico: Journal of Sedimentary Petrology, v. 56, p. 183-193. Hendy, C.H., 1971, The isotopic geochemistry of speleothems - 1. The calculation of the effects of different modes of formation on the isotopic compos ition of speleothems and their applicability as paleoclimatic indicators: Geochimica et Cosmochimica Acta, v. 35, p. 801-824. Hoefs, J., 1973, Stable Isotope Geochemistry: Springer-Verlag, Berl in, 140 p. Hudson , J.D., 1975, Carbon isotopes and limestone cement: Geology, v.3 , p. 19-22. Hudson , J.D., 1977, Stable isotopes and limestone lith ification: Journal of the Geological Society, v. 133, p. 637-660.
Lohmann, K.C., 1982, Inverted " J" carbon and oxygen isotopic trends - a criterion for shallow meteoric phreatic diagenesis (abstract): Geological Society of America, Abstracts with Program, 1982, p.548. Lohmann, K.C., 1983, Unravelling the diagenetic history of carbonate reservoirs, in Wilson , J.L. , Wilkinson, B.H. and Lohmann, K.C., eds., New Ideas and Methods for Exploration for Carbonate Reservoirs - notes for a short course: Dallas Geological Society. Lohmann, K.C., 1988, Geochemical patterns of meteoric diagenetic systems and their application to studies of paleokarst, in James, N.P. and Choquette, P.w" eds., Paleokarst: Springer-Verlag, New York, p. 58-80 . Magaritz, M., 1975, Sparitization of pelleted limestone - a case study of carbon and oxygen isotope composit ion: Journal of Sedimentary Petrology, v. 45, p. 599603. Magar itz, M., 1983, Carbon and oxygen composition of Recent and ancient coated grains, in Peryt, T.M., ed., Coated Gra ins: Springer-Verlag , Heidelberg, p.27-37. Magaritz, M., Kaufman, A . and Yaalon, D.H., 1981, Calcium carbonate nodules in soils - 18 0 /160 and 13C/12C ratios and 14C content : Geoderma, v.25, p. 1157-1172. Meyers, W.J. and James, A.T., 1978,Stable isotopes of cherts and carbonate cements in the Lake Valley Formation (Mississippian) , Sacramento Mountains, New Mexico: Sedimentology, v. 25, p. 105-124. Meyers, w'J. and Lohmann, K.C., 1984,Isotope geochemistry of regionally extensive calcite cement zones and marine components in Mississippian limestones , New Mexico, in Schneidermann, N. and Harris, P.M.,eds. , Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication, No. 36, p. 223-240. Pearson , F. and Hanshaw, B.B ., 1970, Sources of dissolved carbonate species in groundwater and their effects on carbon-14 dating: International Atom ic Energy Association, Isotope Hydrology Proceedings Symposium, p. 271-286. Rightmire, C.T. and Hanshaw, B.B., 1973, Relationship between the carbon-isotope composition of soil CO 2 and dissolved carbonate species in groundwater: Water Resource Research, v. 9, p.958-967. Salomons , w" Goudie, A. and Mook, w.G. , 1978, Isotopic composition of calcrete deposits from Europe, Africa and India: Earth Science Processes, v. 3, p. 43-57.
C (
C C
c
c c c c
c c
c c (
c
c
c c
c c c c c c c c c c ( e c ( (
C C
C C C
c~
( .
Limestones - The Meteoric Diagenetic Environment
73
( (~
r
o ( o r. ( (
c
o o C.
o o
o o
c
o o o o o '-'
CJ
o C
8 r:
Veizer,J., 1983. Chemical diagenesis of carbonates : theory and application of trace element technique , in Stable Isotopes in Sedimentary Geology: Society of Economic Paleontologists and Mineralogists, Short Course No. 10, p. 3-1-3-100. Veizer, J. and Hoefs, J., 1976, The nature of '80/'60 and '3C/'2C secular trends in sedimentary carbonate rocks: Geochimica et Cosmochimica Acta, v. 40, p. 1387-1395. Videtich, P.E. and Matthews, R.K., 1980, Origin of discontinuity surfaces in limestones - isotopic and petrographic data, Pleistocene of Barbados , West Indies: Journal of Sedimentary Petrology, v. 50, p.971-980. Wagner, P.D. and Matthews, R.K., 1982, Porosity preservation in the Upper Smackover (Jurassic) carbonate grainstone . Walker Creek fields , Arkansas, response of paleophreatic lenses to burial processes: Journal of Sedimentary Petrology, v. 52. p. 3-18. Diagenesis of Calcite Sediments Hsu, K.J. and Jenkyns, H.C., 1974, eds., Pelagic Sediments : on Land and under the Sea: International Association of Sedimentologists, Special Publication No.1, 447 p. Scholle, P.A., 1977, Chalk diagenesis and its relation to petroleum exploration: Oil from chalks, a modern miracle?: American Association of Petroleum Geologists, Bulletin, v. 61, p. 982-1010.
Examples of Meteoric Diagenesis in Ancient Limestones Grover, G.A., Jr. and Read, J.F., 1978, Fenestral and associated vadose diagenetic fabrics oftidal flat carbonates, middle Ordovician, New Market Limestone, southwestern Virgin ia: Journal of Sedimentary Petrology, v. 48, p. 453-473. Grover, G.A., Jr. and Read, J.F., 1983, Regional cathodoluminescent patterns, middle Ordovician ramp carbonates, Virgin ia: American Association of Petroleum Geologists , Bulletin , v.67, p. 1275-1303. Jacka, A.D . and Brand, J.P., 1977, Biofacies and development and differential occlusion of porosity in a Lower Cretaceous (Edwards) reef: Journal of Sedimentary Petrology, v. 47, p. 366-382 . James, N .P. and Klappa , C.F., 1983 . Petrogenesis of Early Cambrian reef limestones: Journal of Sedimentary Petrology, v. 53, p, 1051·1096. Harris, P.M., 1983,ed., Carbonate buildupsa core workshop: Society of Economic Paleontologists and Mineralogists, Core Workshop Number 4, 461 p. Heckel, P.H., 1983, Diagenetic model for carbonate rocks in midcontinent Pennsylvanian eustatic cyclothems: Journal of Sedimentary Petrology, v.53, p.733-761. Matthews, R.K. and Frohlich, C., 1987, Forward modeling of bank-margin carbonate diagenesis : Geology, v. 15, p.673-676.
Meyers. w.J ., 1978. Carbonate cements thei r regional distribution and interpretat ion in Mississippian limestones of southwestern New Mexico: Sedimentology, v. 25, p. 371-400. Roehl, P.O. and Choquette , P.w., 1985, eds., Carbonate Petroleum Reservoirs: Springer-Verlag, New York, 622 p. Other References Fischer, A.G., 1981, Climate oscillations in the biosphere, in Nitecki, M.H., ed., Biotic Crises in Ecological and Evolutionary Time , Academic Press , New York, p. 103-133. Lowenstam, H., 1963, Biologic problems relating to the composition and diagenesis of sediments, in Donnelly, T.w., ed., The Earth Sciences: University of Chicago Press, Chicago, p. 137-195. Wilkinson , B.H., 1979, Biomineralization , paleo-oceanography and the evolution of calcareous marine organisms: Geology, v. 7, p. 524-527.
Originally published in Geoscience Canada v. 11 Number 4 (December 1984) Revised 1988
( 74
Diagenesis
Geoscience Canada Reprint Series 4
c c (
(
c C C
c: '-,.
Scalenohedral calcite crystals and micrite with etch patterns following a carbonated-brine flooding experiment in the laboratory. Note that calcite crystal faces are etched in a rhombic staircase fashion and that crystal edges have become rounded. Small scattered particulate material coating the sample was released from the rock during the reservoir flooding experiment. Photograph courtesy of F. Krause and S. Sayegh (University of Calgary and Petroleum Recovery Institute, respectively).
c c c c c c c c c c c c c c c c c c c c c c c (
e c (
C l (
( r ~
U'
.s \
o~:. Limestones - The Burial Diagenetic Environment
o () (
r
.~
o') ...}-I o(
Limestones - The Burial Diagenetic Environment Philip W. Choquette Department of Geological Sciences University of Colorado Boulder, Colorado 80309-250 Noel P. James Department of Geological Sciences Queen's University Kingston, Ontario KlL 3N6 INTRODUCTION Carbonate sediments, or the limestones into which they may have evolved, once buried below the reach of sea-floor processes or below the meteoric environment and the sub-meteoric mixing zone, enter with progressive burial the "deep-phreatic" or subsurface burial-diagenetic realm (Figure 1). It is in this realm that, unless exhumed by tectonic uplift or lowered sea level, the great majority of sedimentary carbonates must inevitably spend most of their geologic existence. It is here, too, that the majority of carbonate sediments which escaped sea-floor lithification or prolonged meteoric diagenesis undergo most of the changes that lead to their lithification and reduction of porosity. The importance of deep-burial diagenesis has been recognized more widely in the last fifteen years or so, in considerable part because of the publication of a major synthesis on carbonate diagenesis by Bathurst (1971). Since that time, aspects of the subject have been examined in a number of general or specific studies, perhaps most notably the thoughtful works by Bathurst (1975, 1980a,b, 1984, 1985, 1987), Schlanger and Douglas (1974), Garrison (1981), Shinn and Robbin (1983), and Scholle and Halley (1985). During this same period, the development and refinement of instruments and
rV7D).J~(
_, . .--'_
~r/
->: r
•
_
techniques for the study of cathodoluminescence, ultraviolet fluorescence, stable-isotope, rare-earth and minor-element compositions, and microscopic fluid inclusions have made possible the acquisition of much new information about the petrologic, mineralogic and chemical changes that take place in the deep-burial realm. The fundamental importance of diagenesis in this largely occult realm in shaping the porosity, mineralogy, basic fabrics and numerous other features of carbonates is now clear, but many aspects remain to be clarified. The quest for a basic understanding and a general theory of diagenesis in the deep-burial setting has only begun. THE DEEP-BURIAL SETTING Definitions. Burial diagenesis can be defined in general as any change or collection of changes that takes place below the zone of near-surface diagenesis and above the realm of low-grade metamorphism (Figure 1). These limits can vary from one sedimentary basin to another or even within a single basin. By "near-surface" we mean at or close to the sea floor or in the meteoric environment, within the reach of surface-related processes (Choquette and Pray, 1970). The sub-lenticular mixing zone (see James and Choquette, 1984) is rightly viewed as part of the meteoric, although its nature and vertical as well as geographic extent are poorly known in all but a few sub-tropical limestone islands and a few aquifers in Cenozoic and Mesozoic carbonates. Defined in this way, burial diagenesis is mainly a consequence of overburden due to burial beneat h younger strata, and of resulting, ~ h ang es ' n ~ogy, pore-water chemIstr, ressure, and temperature. In eneral, it may begin at depths of a very few metres to a few hundreds of metres beneath subaerial surfaces, and a metre or so to very few tens of metres beneath sea-floor surfaces. The burialdiagenetic realm would include, for example, the mesogenetic or "deeper burial" realm of Choquette and Pray (1970); the "compactional" and "thermobaric" subsurface regimes suggested by Galloway (1985) for sandstone-shale sequences; the "deepphreatic" realm of many workers; and for geologists in the petroleum industry the "subsurface" in general. The term phreatic is not very useful because it
0/ /0
~ 0-
1.. r
% 75
r'YOo,S.! 0) N~ q,. '1 \ ~ MI ~Jb;
L/VI C connotes the zon/ IMsfttPr~iJJnd
water-filled pores anywhere below the water table. Differences from near-surface processes. The dominant processes, ambient conditions, space and time distributions, and principal products of diagenesis in deep-burial settings are quite different from those in near-surface settings. In the sea-floor environment, cementation by aragonite and Mg-calcite is the main inorganic agent of lithification (James and Choquette, 1983)and is highly facies-specific. Lithification in this setting is virtually instantaneous geologically, probably involving time intervals on the order of 101 to 104 years. In the meteoric environment, the main processes of lithification involve the precipitation of calcite, impelled either by dissolution of aragonite and microdissolution of Mg-calcite (forms of mineral-controlled alteration as discussed by James and Choquette, 1984, and this volume), or by more wholesale dissolution and precipitation of calcite (water-controlled alteration). Because of the incongruent dissolution of metastable aragonite and Mg-calcite particles and the accompanying precipitation of calcite, mineral-controlled alteration redistributes porosity from inter- and intraparticle sediment pores to new moldic pores, often with little net change in percent porosity (Harrison, 1975). Water-controlled alteration, on the other hand, creates new, larger, non-selective voids or adds CaC0 3 and reduces porosity. Meteoric-realm diagenesis must involve time intervals which approximate the duration of subaerial exposure at unconformities; in Plio-Pleistocene sequences on modern platforms glacio-eustatic unconformities involve time intervals on the order of 103 to 105 years (e.g., Steinen and Matthews, 1973; Enos and Perkins, 1979; Beach, 1982); older Phanerozoic platforms contain regional and interregional unconformities representing time intervals up to 106 to 107 years (Sando, 1974, 1985; Schlee, 1984). In the deep-burial environment, far more complex arrays of processes are operative (Table 1). Dominant among them are physical (mechanical) compaction and dewatering, chemical compaction (pressure-solution), cementation, subsurface dolomitization, alteration of hydrous minerals to anhydrous ones,
c 76
Geoscience Canada Reprint Series 4
Diagenesis
(
C
C C --
MIXING ZONE
BURIAL DIAGENETIC DOMAINS
-
SHALLOW MARINE , PERITIDAL
" -,
AND ASSOCIATED SEDIMENTS
pore waters of rnult ip l e orig ins
---=--=-
<,
+
::::,
,:::..-........
~
--" '"" -----_
<,
....
DEEP BURIAL
~
SHALLOW BURIAL
. . . ~--
--.. _-"'......, -
~
--
. . . . ·1 · · ···
...?
.
METAMORPHIC DOMAIN
<,
, '>'
----
BASINAL SEDIMENTS
marine-derived pore waters
DEEP BURIAL
Figure 1 A sketch outlining two general domains of deep-burial diagenesis. The nature and depth range of the boundary with the metamorphic realm are poorly known. The depth is likely to range from a kilometre or less to several kilometres or more.
thermally driven mineral stabilization reactions, and alteration and maturation of organic matter with attendant production of organic acids and hydrocarbons. An apparently common longterm trend with burial is progressive reduction of porosity by compaction and cementation , at increasing temperatures and pressures, in the presence of pore waters that may become increasingly saline. Important changes may occur over burial-history time spans of 106-108 years, although the rates of some modifications, such as mechanical compaction , will commonly fall off rapidly with increasing depths of burial. Factors Influencing Deep-Burial Diagenesis. Intrinsic factors. Among the most important intrinsic factors (Table 2) are the mineralogy and the amount and nature of the organic fraction . A sediment that retains aragonite and Mgcalcite as it goes into the deep-burial environment will have more "diagenetic potential " (Schlanger and Douglas, 1974) to generate calcite cement through dissolution and reprecipitation than will a sediment composed only of calcite. We know relatively little about the mineralogy of platform limestones as they leave the meteoric and submeteoric mixing zones and enter the deep-burial realm. Scholle and Halley (1985) have argued that many platform limestones altered in the meteoric zone are calcite and therefore have relatively low diagenetic potential like marine
C C
c c c
c c
c c c c c
General processes and products in the deep-burial domain.
Table 1
Products
Processes Physical compaction
Reduced thickness , porosity and permeability; re-oriented and commonly broken allochems ; compressed textures and structures
Chemical compaction
Reduced thickness, porosity and permeability; stylolites and other pressure-solution structures; ions for new carbonate cement
Cementation
Mosaicldrusy to very coarse or poikilotopic calcite and saddle dolomite
Burial dolomitization
Anhedral-crystalline dolomite, generally rather coarse
Mineral reactions
Smectite to illite Gypsum to anhydr ite Opal A and CT to quartz Fe hydroxides to Fe oxides
Biochemical alteration of organic matter
Kerogen and organic acids
Thermochemical alteration of kerogen
Hydrocarbons and CO2
Burial dissolution of Ca carbonate and sulphate
Solution porosity
Table 2
Principal controls in deep-burial diagenesis.
Intrinsic
Extrinsic
• Mineralogy • Grain size and texture • Organic matter, kerogen and hydrocarbons
• Temperature
• Early cement or dolomitization • Pore-water chemistry • Porosity and permeability
• Pressure
~
c c
c c c
c c c c e c c c c e c c (
• Time
(
• Fluid Flux
(
(
G (
Limestones - The Burial Diagenetic Environment
c C' c c (
(
c J
(
o o o o o o
o o o o o
e o o o r:
~ (
chalks . However, calc ite cement of meteoric origin is rarely homogeneous either isotopically or elementally, and therefore must have at least some diagenetic potential (L.S . Land, pers. comm., 1986). In any case, many studies of Recent and Plio-Pleistocene limestones have shown that mineralogic stabilization in meteoric water from aragonite and Mg-calcite to lowMg calcite can take place quickly, after a few metres to tens of metres of burial (e.g., Steinen and Matthews, 1973; Steinen, 1974; Beach, 1982). In another kind of " mineralogical" effect, limestones that contain at least a few percent clay minerals or unstable organic matter (in addition to being fine grained) are more susceptible than pure limestones to physical compaction, chemical compaction (pressuresolution), and even spaced cleavage due to tectonically induced pressuresolution. Furthermore, it is widely thought that the amount of clay and lor organic matter may be important. Whether clay and silt content is less than or more than about 10% appears influential in determining whether pressure-solution produces, respectively, sutured stylolites or non-sutured seams and fitted fabrics (R.G.C. Bathurst, pers . comm., 1986). Grain size and texture influence the rates at which CaCO s components change and sediments compact. Finegrained and poorly sorted sediments undergo neomorphic and solutionprecipitation reactions more rapidly because of their larger reactive surface areas; they also compact more readily than coarse sediments, in marine pore waters at least, because of their greater pore water content and potential for particle rearrangement and reorientation . One consequence of this is that some sediments that were deposited as lime muds undergo so much physical compaction of their fine-grained fraction that they become "diagenetic" wackestone or packstone (Shinn and Robbin, 1983). Early-precipitated cement or early replacement dolomite formed at or not far below the depositional interface will increase the bearing strength of a sediment so that it compacts less readily, at least at relatively shallow depths km?). Finally, the porosity and permeability of a sediment or rock, which govern
«2
both the volume of contained pore fluid and the rate at which fluids are transmitted, affect the speed at which diagenetic reactions take place. In Holocene sediments, porosity is generally greater in fine-grained sediments such as lime muds than in coarser carbonate sands (Enos and Sawatsky, 1981). In the geologic record, the highest known limestone porosities (up to 80% with very low permeabilities often below 1 md) are found in some Cretaceous and Danian chalks (Scholle, 1977). Permeability in limestones with interparticle porosity, however - other things being equal is directly related to grain size, so that in general the coarser-grainedcarbonates will transmit fluids in larger volumes and at higher flow rates. Extrinsic factors. Among the many extrinsic factors affecting burial diagenesis (Table 2) it appears that pressure, temperature, pore-water chemistry, and, in some settings, liquid hydrocarbons are most important. Pressureof at least three kinds is applied to sedimentary sequences being buried: (1) lithostatic pressure, transmitted through the solid framework; (2) hydrostatic pressure, transmitted only through the pore-system "water column"; and what Bathurst (1975) called (3) linear or directed pressure. The first two of these are induced by gravity. The third is really of tectonic origin and has been recognized increasingly in recent years as a major cause of spaced cleavage in tectonized carbonates that yield by brittle fracture. Typical hydrostatic and lithostatic pressure gradients, shown in Figure 2A, give an idea of the pressure ranges ambient over the range of subsurface depths that have been sampled thus far by petroleum exploration drilling. Pore pressure determines the effective stress on a sediment under overburden load, because the effective or net stress is the Iithostatic pressure (or total load) less the pore pressure (Terzaghi, 1940). Most subsurface strata at temperatures greater than about 100°C have pore pressures which are higher than "normal hydrostatic" and would lie in the stippled area of Figure 2A. Usually the differences between Iithostatic and porefluid pressure are small, but under various conditions the pore-fluid pressures can become abnormally high, in which case the strata are said to be
77
overpressured or "geopressured " and the pore-fluid column may support part or all of the overburden load. Abnormally high pore-fluid pressure can be an important inhibiting factor in burial diagenesis. It can actually " prop up" original pore spaces - either independently or in concert with early nearsurface cementation - thereby retarding or preventing physical compaction. It alsocan slow or shut off the export and import of pore water and prevent any changes in the partial pressure of CO2 and other dissolved gases, thereby creating a pore-fluid system which is both overpressured and static, in which both pressure-solutionand cementation may be sharply retarded. Feazel and Schatzinger (1985) also have discussed recentlythe retarding efects of high pore pressure. Many processes seem able to cause overpressuring of sedimentary rock pore-fluid systems, by trapping their fluids. Rapid sedimentation, as in marine parts of major delta systems, can bury sediments more quickly than these can expel connate water. Confinement of porous carbonates beneath stratiform permeability barriers such as compacting shales or evaporites, well-indurated hardgrounds, or limestones having swarms of laterally extensive stylolites, all can cause overpressuring. So also can some diagenetic reactions involving clay, sulfate and silica minerals as discussed next under "Temperature", since these reactions add water to the system. The generation of gases such as CO2 , methane and other hydrocarbons, and H2S at sites and in strata where the gases cannot escape may cause overpressuring. The abrupt release of overpressure has been proposed as a mechanism for producing certain breccias apparently by dilation, for example in thin-bedded,fine-grained carbonates of basinal origin in the Miocene Monterey Formation of California (e.g., Secor, 1965; Roehl, 1981). Our understanding of, and ability to predict, the realms and effects of abnormally high pore-fluid pressure in burial diagenesis of carbonates still are primitive. A very few studies haveshownthe importance, however, of abnormal pressures as co-agents (with early emplacement of hydrocarbons) in causing the retention of unusually high porosities in some Cretaceous and early Tertiary chalk reservoirs in the central North
Diagenesis
78
Sea (Scholle, 1977; van den Bark and Thomas, 1981; Feazel et el., 1985). Temperature appears to play at least three roles in burial diagenesis and may be of considerable importance. (1) The solubility of calcite and other carbonates - unlike that of quartz - is "retrograde" in water having dissolved CO2 and decreases with increasing temperature (Helgeson, 1969). The decrease is linear up to temperatures as high as 400°C and is substantial , on the order of two orders of magnitude at anyone pressure (Bathurst, 1971). This relationship, although complicated by the effects of increasing pressure, indicates that with increasing temperature alone, calcite cement (and by inference saddle dolomite) will be more likely to precipitate. (2) Increasing burial brings about a series of temperature- and pressuredependent mineral reactions. Some of these release water to the surrounding pore fluids and result in minerals that have higher specific gravities and occupy less space than their precursors. These reactions include the conversion of gypsum to anhydrite at around 1000 m (Murray, 1964; Kendall, 1984), and the conversion in some oceanic sediments of opal-A to opal-CT and finally quartz within a depth range of about 450 to 2000 m (Isaacs, 1981) which may correspond to temperatures of 40-100 oC. These reactions involve changes in crystal-lattice morphology and release water of crystallization. Other reactions which release water along with various cations are the conversion of smectite to mixed-layer clays and finally to illite commencing at around 2000 m and 60°C in the US Gulf Coast (Burst, 1969; Perry and Hower, 1970; Hower et a/., 1976; Boles and Franks, 1979),and the dehydration of various iron hydroxides such as limonite to yield hematite and free water. Still other reactions affecting carbonates, which take place in sandstones, include the conversion of calcic plagioclase to albite releasing Ca 2 + and other divalent cations such as Sr2 + which have been used as radiogenic tracers of burial reactions (e.g., Steuber and Pushkar, 1983; Moore, 1985; Woronick and Land, 1985). (3) Increasing temperature together with burial reaction time causes the conversion of sedimentary organic matter from unstable to more stable forms
cementation, and probably some bur ial dolom itization and dolomite neomorphism (xenotopic dolomite of Gregg and Sibley, 1984) also take place. The range of temperatures in sedimentary basins can be great, depending on the thickness of the sedimentary fill and on the geothermal gradients (Figure 2B), which in turn depend on ambient heat flow and on thermal conductivity of the sediment. Gradients in most sedimentary basins are in the range of 15-35°C per kilometre of depth and 25°C/km is an often-eited normal value (Wood and Hewett, 1984). Gradients can change sharplywith uplift and passage of a sedimentarypackageinto a nearer-surface hydrologic regime fed by cooler meteoric waters,as has happened in the Western Canada basin (Hitchon, 1969a,b) and some other basins. Pore-water chemistry has a variety of important effects and roles in burial diagenesis. For one thing, pore waters transmit, by physical fluid flow or ion diffusion, the solute ions produced by dissolution or consumed by cementation and replacement reactions. Moreover, water chemistry seems to have a strong influence on major burial-diagenetic processes . One of the most important of these, pressure-solution (chemical compaction), seems to operate less rapidly in Mg-rich (e.g. , marine-derived) pore water - a factor that may be partly responsible for the retention of high poros ity in many chalks (Neugebauer, 1973, 1974).
and eventually to hydrocarbons and other organic compounds (see review by Barnes et aI., 1984). In dirty sandstones, the reactions in this complex series are widely thought to release CO2 which can combine in carboxylic and other organic acids capable of dissolving carbonates (e.g. , Schmidt and MacDonald, 1979; Hayes, 1979; BjorIykke, 1984 ; Surdam et et. , 1984) . Material-balance calculations have led some workers to suggest, however, that CO2 produced only by decarboxylation , at least in shales of the Oligocene Frio Formation in the Texas Gulf Coast, cannot account for most of the dissolution porosity found in associated sandstones; other mechanisms and/or longdistance transport of acidic waters must also be responsible (Lundegard et el., 1984; Lundegard, 1985). Dissolution porosity of burial origin is being recognized increasingly in limestones since its discovery in the early 1980s (Moore and Druckman, 1981; Elliott, 1982; see also Druckman and Moore, 1985; Moore, 1985). Although the minerai and fluid systems must be different, mechanisms like those invoked for sandstones have generally been suggested for the limestones. Interestingly, the temperature range in which peak generation of hydrocarbons is believed to occur, 60-150oC (Tissot and Welte, 1980), is also the temperature range in which many other burial diagenetic processes involving inorganic reactions, pressure-solution,
A
Geoscience Canada Reprint Series 4
B TEMPERATURE,oc
PRESSURE, bars 4000
ir-----.;.~--::..:r'---"""""-'-r--__,
0
100
200
300
10
...E
:z: l-
20
'6 ... )(
ll. W
a
c (
c c c c c c c
c c (-
c
c c c c c c c c c c c c c e
100
x1()3 psi
c c c c c
OF
Figure 2 Graphs showing the general ranges of pressure and temperature in the deepburial diagenetic realm . (A) Static pressure ranges. The position of the hydrostatic pressure curve varies depending on the concentration and density of pore waters. Most pore-fluid pressures in the subsurface would plot on curves in the stippled area. Commonly the change from normal to excessive porefluid pressure is a transition marked by more rapid increases in both pressure and (graph B). (B) Temperature ranges assuming different geothermal gradients. Most gradients are 15·35°C, and most are non-linear and commonly steepen (decrease) with increasing depth.
c..
c C l
c.. t
C
c
c c c: c c ( o c c
(
(
o o o o o o o o ( ( (
Limestones - The Burial Diagenet ic Environment
The introduction of liquid hydrocarbons into the pore systems of limestones during burial diagenesis , especially if the pore walls are oil-wet, inhibits pressure-solution and can preclude the formation of stylolites; Dunnington (1967), who first suggested these effects, also noted that the emplacement of oil in some limestone reservoirs of the Middle East might actually have prevented their compac tion and created differential-compaction traps for hydrocarbons. Liquid hydrocarbons in pores can also shut off or prevent cementation. This mechanism has been invoked to explain differences in extent of pressure-solution cementation and porosity reduction in some chalk reservoirs in the North Sea (Feazel et al., 1985; Feazel and Schatzinger, 1985). PROCESSES AND PRODUCTS In the sections that follow, we examine the common sedimentological and geochemical processes and products of burial diagenesis - the records of their passage into the deep subsurface (Table 1). Physical Compaction. As sediments compact under load they dewater, lose porosity, and decrease in thickness while sedimentary part icles and sedimentary structures are modified and/or rearranged, depending on how much porosity they had and whether they were grain-supported or mud-supported initially.
o o o
DEPOSITION BURROWS DESICCATION CRACKS EARLY LITHIFIED LAYERS FOSSILS
~:~r---.,r--.,-r='--I
~~;;~=3 ~ I
@)~ ~
BURROWS
GRAINSTONE LENSES
Processes. Physical compaction in marine sediments buried in marine pore waters can be visualized as taking place in three main stages which probably overlap. Stage 1: Initial particle settling and repacking with resulting dewatering occurs in the first metre or so of burial, where porosities of lime muds may be reduced a small amount, say from 80-75% (Ginsburg , 1957; Brown, 1969; Schlanger and Douglas, 1974; Enos and Sawatsky, 1981). Stage 2: Particles are rearranged and elongate grains re-oriented in grainy sediments, while muddy sediments are compacted with continued dewatering , until a self-supporting "framework" is achieved at porosities of approximately 40%. The relative importance of this stage of compaction is proportional to the original ratio of lime mud to coarse silt size and larger grains. In shallow sub-tidal sediments th is compaction often takes place within a metre or so below the sea floor ; one can stand on the sediment rather than sink into it and one can generally core it (L.S. Land. pers. comm., 1986). Stage 3: Overburden stress is applied at grain contacts, and is accommodated by grain deformation in the form of either ductile squeezing or brittle fracturing and breakage. Lime mud between framework-forming grains , in shelter cavities beneath elongate particles, or within fossils, is generally protected from compaction.
PHYSICAL COMPACTION
•
SQUASHING
!::~ ~ ~
EARLY L1THIFIED LUMPS OR I" (
79
_,_~
:~
I~
~---
.
OVERPACKING SPALLED
0010 CORTEX
FRACTURED MICRITE ENVELOPES .. GRAINS
Figure 3
Sketches illustrating features produced by physical (mechanical) compaction.
A common conclusion of experiments involving compaction of muddy carbonate sediments is that thickness reductions can be substantial and in some cases greater than porosity reductions, especially in stage 2 compaction (Terzaghi, 1940; Hathaway and Robertson, 1961; Fruth et al. , 1966; Robertson , 1967; Ebhardt, 1968; Shinn et aI., 1977; Bhattacharyya and Friedman, 1979). Shinn and Robbin (1983) compacted modern shallow marine and peritidal muddy carbonate sediments with 47-83% initial porosity down to as little as 27% (mostly 30-65%) of their orig inal thickness, with final porosities reduced to 35-45%. Although some chem ical compaction effects were observed in cores subjected to the highest pressures, physicalor mechanical compaction was by far the dominant process. At pressures of 66 to almost 1000 kg' cm·2 (93313,600 psi), compaction as gauged by thickness reductions essentially ended when porosities were reduced to 35-45%. The results of these experiments indicate that carbonate sediments can compact under as little as 100 m of overburden to one-half their original thickness, with accompanying porosity losses of 50-60% of original pore volume. Products. Sedimentological products of physical compaction are almost as varied as the constituents of the limestones (Figure 3). In mudsupported or pelleted-mud sediments (mudstone and wackestone) common features include : thinning of laminae between and draping over early concretions (Figures 4 and 5), early-lithified layers , and other hard objects ; squashed burrows and crumpled fenestrae, gas-escape structures, and desiccation cracks (Figure 6); shells and other elongate grains that have been rotated, packed more closely, or in some instances brought into proximity with other rigid grains against which they have been crushed (Figures 7 and 8); and flattened pellets or other grains (Figure 9). Physical compaction also produces a variety of swirling structures as well as the textural "telescoping" noted by Shinn and Robbin (1983), converting grain-poor lime mud to wackestone or packstone (Figure 8). Physical compaction can close and obliterate desiccation cracks and fenestrae or birdseyes in peritidal sediments (Figure 6).
80
Figure 4 A slab of early burial-diagenetic displacive-fibrous calcite (DFC, arrows) developed on both sides of a mudstone bed that was formed by coalescence of microspar concretions. The bed was thinned a litt/e between concretions by physical compaction prior to formation of the DFC. From Cambro-Ordovician Cow Head Group, western Newfoundland (Coniglio , 1985). Photo courtesy of M. Coniglio, published by permission .
Figure 5 A concretion formed by earlyburial microspar cementation of peloidal wackestone (darker) and grainstone (lighter). The uncemented laminae to right have been compacted to 36% of their thickness as preserved in the concretion . Compaction was probably physical rather than chemical. A dilation crack (arrow), now filled by ferroan calcite spar (black in this picture), opened due to differential compaction of the concretion . Photo is a negative print of a thin-section. Bar scale is 1 cm. From Cambro-Ordovician Cow Head Group, western Newfoundland (Coniglio, 1985). Photo courtesy of M. Coniglio, published by permission.
Figure 6 (A) Uncompacted supratidal sediment from the Holocene of south Florida (plastic-impregnated core). The openings are birds-eye or fenestral pores, vertical shrinkage cracks, and burrows, and horizontal cracks and large cavities probably formed artificially. (8) Artificially compacted sediment from the same core. Only the thin storm layers and desiccation cracks in A are clearly preserved. Virtually all the voids have been squeezed shut, and the overall fabric has been strikingly changed. From Shinn and Robbin (1983). Photos courtesy of E.A. Shinn, published by permission .
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c (
c c c c c c c c c ( c c c c c c c c C E
C C C
C
c:
c c c c
C C l
C
c
c
Limestones - The Burial Diagenetic Environment
81
(
( ,
o o ( c (
·0········ . --:.-
(
-~.' ~
'. ...
~
.
'. ':' ~..@:.~~'. ~ ;;;2..
C' C
.. ~.: .:.f(~-~~ ... ~ ~" . ' .. . . . .' .• f'::\ .
o o o o o o
~
\!!../ .
.
.
(
o
o o c o o o
Figure 7 (A) Uncompacted subtidal bioclastic muddy sediment from the Holocene of south Florida (plastic-impregnated core). The mollusk shells and Halimeda plates (white) lack preferred orientation . (R, horizontal grass rhizomes; G, marine turtle grass. (8) and (C) Artificially compacted sediments from the same core, under different loads. Mollusk and Halimeda fragments have been reoriented and commonly broken, drape-like swirls have appeared, and all of the "open " and uncalcified structures have been squeezed shut. From Shinn and Robbin (1983). Photos courtesy of EA Shinn, published by permission.
Figure 8 Diagrams illustrating the transformation of lime mud with sparse allochems to Iithified, more allochem-rich limestone, in this case wackestone, as a result of physical compaction alone. Proportions of mud particles to pore space are shown schematically by the dots.
l
l C Figure 9 A pisolitic grainstone that has suffered grain flattening by physical compaction, and also pressure-solution interpenetration along microstylolites now marked by dolomite rhombs and voids (white). Bar scale is 1 mm. From the Mission Canyon Formation (Mississippian), Williston Basin, North Dakota (Elliott, 1982). Photo courtesy of IL. Elliott, published by permission.
Figure 10 Ooids which have been squashed by physical compaction ; outer concentric laminae on the ooids to right and left of center are fractured and spalled. Some pressure-solution also has occurred, embaying outer laminae as in the large ooid to right of center. Bar s..cale is 1 mm. Upper Cambrian, Port au Port Peninsula, Western Newfoundland. Photo courtesy N. Chow.
82
Organic matter is commonly squashed and deformed into irregular stringers that drape over rigid grains and look deceptively like stylolites. Relatively brittle laminae and thin mud-cracked beds between layers of plastically compacted mud may fracture and rotate (Figure 8). In grain-supported clean and muddy carbonate sands (grainstone and packstone), plastic deformation of soft pellets, coated particles and skeletal grains is common (Figure 9). Ubiquitous, however,arefractured andcrushed grains of various kinds such as spalled and broken cortical layers of ooids (Figure 10), broken thin rinds of cement around grains that have been squashed, crushed fossils (Figure 11), and fractured micrite envelopes surrounding grain molds. Also typical are planar or curviplanar grain contacts resulting from plastic deformation; care must be taken, however, to be sure these contacts are due to physical compaction and not pressure-solution. Compaction of lime mud between grains, in sheltered patches either beneath broad particles such as phylloid algal plates and brachiopod valves or inside fossils, is generally very slight, so that original textures and particles are preserved (Figure 12). Thin ning of limestone strata by physical compaction itself is becoming more widely recognized. One ofthe more convincing large-scale examples is reported by Beach and Schumacher (1982) in the Mission Canyon Formation (Mississippian) in a part of the Williston Basin in North Dakota. There, pisolitic grainstones interpreted as "island facies" that haveopen, little compacted fabrics are surrounded by compacted marine bioclastic and peloidal wackestone and packstone. A series of isopach maps suggests post-depositional, non-erosional thinning of the marine limestones by as much as 31 %, and crushed fossils and other features (Figures 11 and 12), unaccompaniedby dissolution seams (to be discussed in the next section), imply that physical compaction contributed to the thinning. Differential compaction at hand-sample scale has been estimated from changes in the shapes of Chrondites and other burrow tubes that were cylindrical before they were buried (Byers and Stasko, 1978; Gaillard and Jautee, 1987; Richen, 1987). Compaction in fine-
Diagenesis
grained carbonate sediments is generally related directly to the abundance of clay and organic matter so that the relatively pure CaC031ime muds tend to compact the least. Ricken (1987) has suggested that this inverse relationship
Geoscience Canada Reprint Series 4
noted by many workers (e.g., Hallam, 1964; Byers and Stasko, 1978; Bathurst, 1987) between percent carbonate and amount of compaction is rigorous and unvarying in mud-rich sediments. In such sediments, however, particularly if
c c c c c c C
r
C C C
C C C C C C C (
Figure 11 Part of a crushedrugose coral, the victim ofphysical compaction of the surrounding packstone matrix. Near-shore-marine skeletal limestone from the Mission Canyon Formation (Mississippian), Williston Basin, North Dakota (Beach and Schumacher, 1982). Bar scale is 1 mm. Thin-section loaned by D.K. Beach.
C C C C C C C C C C C C C (
e c c c c c c. c (.
Figure 12 Sheltering effects in compacted shallow-marine foraminiferal-peloidal limestone. A whole, but broken, brachiopod shell sheltered the grain-rich sediment inside it from physical compaction, and this sediment is now normally packed grainstone (lower right), while the same sediment outside the shell was compacted and is overly packed packstone. Bar scale is 1 mm. Mission Canyon Formation, Williston Basin, North Dakota (Beach and Schumacher, 1982). Thinsection loaned by D.K. Beach.
(
(
Limestones - The Burial Diagenetic Environment
83
r C C
o c (
( (
c (
o o o o o o o o o c o o o o o o o o o C
Chemical Compaction (Pressure-solution). Chemical pressure-solution compaction is of great significance in carbonate rocks. It is known to be responsible for reductions of bed and sequence thicknesses in the range of at least 20-35%, thickness reductions that took place after physical compaction. Pressuresolution (or pressure-dissolution) is also becoming more widely recognized, despite the common difficulty in proving cause and effect, as an important
aragon ite and Mg-calcite are abundant , the amount of compaction must surely be dependent as well on the chemistry of the pore waters during early burial. Although the relative importance of physical and chemical compaction is often difficult to assess even where the products of the two can be clearly identified (cf. Figures 3 and 13),what is now clear is that stratigraphic thinning and many smaller-scale changes produced by physical compaction are widespread in the carbonate rock record.
CHEMICAL
COMPACTION
•
, . - --
-
• STYLOLITES
-
',
' .
./
SINGLE
( ,.
»>:~
1"'-/1-/'-"'\,-<,'' /
/"
f~~~~:::L,J~'v.-=",'~~
MULTIPLE
~~~.___
~~~.~: _-- SOLUTION SEAMS NODULES
I :. ... -.
FITTED NODULES
~NOD""
~• •...-:_--~
MICROSTYLOLITES
I- - · -·..;...._~ .;, ;;.. NODULE
~ .~ ."""'. · ·--- '~I PHYSICAL COMPACTION ." MICROSTYLOLrrES ....
r:
~
:;...-
-
-~._~=-=
--
STYLOLITE
Figure 13 Sketches of features produced by chem ical (pressure-solution) compaction.
SOLUTION
•
,
l r
'-r:
'-.-.'
r:
Figure 14 A conceptual sketch of elements on a microscale involved in pressure-solution of two grains pressed together at a sutured contact.
source of carbonate cement in deepburial diagenes is (e.g., Bathurst , 1975; Hudson, 1975; Wong and Oldershaw, 1981; Scholle and Halley, 1985). Furthermore, it has been demonstrated that spaced cleavage of tectonic origin in folded limestones is in fact a pressuresolution phenomenon, and that spacecleavage surfaces are actually stylolites or non-sutured seams (e.g. , Geiser, 1974; Groshong, 1975a,b; Rutter, 1976, 1983; Alvarez et al., 1976; Drexler and Schaer, 1979; Engelder et a/. , 1981 ; Geiser and Sansone, 1981; Marshak and Engelder, 1985). The term "chemical compaction" was proposed by Lloyd (1977) for the loss of porosity and decrease of thickness brought about in limestones by the generation of " autochthonous" carbonate cement internally through pressure-solution. Carried to completion, cementation of pores by this process should be capable of completely occluding whatever porosity remains in the wake of physical compaction and near-surface diagenesis. In Bathurst's (1971) words a limestone undergoing chemical cornpaction serves as both a donor and a receptor of carbonate cement. Pressure-solution and its most obvious products, stylolites , have attracted the attentions of geologists for well over a century, beginning apparently with the inquiries of Thompson (1862) and Sorby (1879, 1908). Quantitative studies of thickness reductions due to stylolites were made more than a halfcentury ago by Stockdale (1922, 1926), and more recently have been carried out by Dunnington (1954, 1967), Mossop (1972), Playford (1980), Hurley (1985), Bathurst (1987) and many others; our experience and discussions with other geologists suggest that major thickness reductions also have been well documented in a large number of unpublished studies. The time and space relationships between pressure-solution , physical compaction and cementation have been examined most recently by Shinn and Robbin (1983), Bathurst (1984), and Scholle and Halley (1985). Processes. Pressure -solution is a process wherein pressure due to load or tectonic stress, transmitted to and concentrated at contact points or surfaces between grains, crystals or larger entities (limestone beds, for example), increases the solubil ity of the stressed mineral(s) due to increasing elastic
84
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c c c (
C
C C C
Figure 16 Bedding-parallel stylolites (sutured seams) in thin beds of lime mudstone separated by thin shaly partings (lighter gray). Etched core-slab surface, bar scale 1 cm. Ste. Genevieve Formation (Mississippian), Illinois Basin, Illinois.
Figure 15 Stylolitic contact between ooid grainstone above and microcrystalline dolomite below. Finely particulate sulphides and pyrobitumen(?) form the dark sutured seam. Core-slab surface, bar scale 1 cm. Ste. Genevieve Formation (Mississippian), Illinois Basin, Illinois.
Figure 17 Swarms of microstylolites and non-sutured wispy seams in a wackestone with lenticular burrow fillings of grainstone. Bar scale is 1 em. Bajocian (Jurassic), Lincolnshire, UK. Photo courtesy of R.G.C. Bathurst, published by permission.
C C C C G
C C C C C C C
C C C C C C Figure 18 (lower left) Outcrop photo of lime wackestones that show "pressuresolution bedding" at fissilepartings made up of.concentrations of microstylolites. Scale is 0.5 m long. (upper right inset) Thin-section photo of partings in an area such as that shown by arrow, illustrating a concentration of dissolution seams and microstylolites. Brachiopod shelf was broken during physical compaction, then thinned somewhat by pressure solution. Bar scale is 0.25 mm long. Visean (Carboniferous), Clwyd, North Wales, UK. Photos courtesy of R.G.C. Bathurst, published by permission.
C
c e ( c l
C (
C
c
(
c c: ( o o (
(
c (
C' (
o ( o o ( o (
o
c o o o o (I
l
Limestones - The Burial Diagenetic Environment
strain , and causes dissolution at the contact. The phenomenon therefore involves a change of state from solid to liquid, creating a microscopic solution film (Weyl, 1959;De Boer, 1977; Robin, 1978)a few water molecules thick at the contact. This film mayor may not be connected with the solution-filled pore system of the rock (Figure 14). For CaC0 3 , because the solubilityproduct constant Ks p = [Ca 2 + l x [CO~- l increases at the locus of maximum stress due to increased strain energy, dissolution releases new ions of solute to the solution film. This pressureinduced solubility increase sets up a chemical potential gradient which declines away from the locus toward areas of lower stress. Ions move down this gradient by solution transfer (Durney, 1976)or diffusion , to be either combined and precipitated as new calcite cement at nearby sites where stress and solubility both are lower, or transported in moving pore waters to more distant sites. Each mole of CaC0 3 dissolved by this mechanism will produce a mole of solute ions and therefore, potentially, a mole of CaC0 3 cement. The analogy is nearly exact with ice which is gradually stressed at a contact point , say by a metal-tipped pole: ice dissolves at the contact and a depression forms , with an intervening film of water between metal and ice; when the pressure is released and the metal point removed, provided the air is below the freezing temperature of water, ice forms again in the depression. The main departure from this analog of course is that in the subsurface the pressure is usually maintained . Products. Pressure-solution features are amazingly diverse (Figure 13). Even stylolites, the most spectacular products. vary so much in amplitude and in form from one occurrence to another that they have stimulated a series of elaborate classifications (e.g., Stockdale, 1922;Park and Schot, 1968; Wanless, 1979). The common denominator of stylolites and other sutured seams (Wanless, 1979) is a jagged surface, generally coated by insoluble clays, organic matter and/or other minerals, and made up of interlocking pillars, sockets, and variously shaped "teeth". Sutured seams include stylolites, many microstylolites, and networks of anastomosing micro-
stylolites (the " fitt ed fabrics" of Logan and Semeniuk (1976) and Buxton and Sibley (1981). Some stylolites develop select ively at contacts (Figure 15) between beds or nodules of highly contrasting solubilities such as wellcemented oolite and microcrystalline dolomite (e.g., Choquette and Steinen, 1980). Often they propagate most extensively in relatively clay-free, lime mudstone and wackestone, though they may have been initiated at clay/ shale partings between " clean" limestone beds (Figure 16). On the other hand, non-sutured seams (Wanless, 1979), have been recognized only since the 1970s (e.g., Mossop, 1972) as features probably caused by pressure-solution (Figure 17). These are "wispy" surfaces that may extend laterally a few centimetres to hundreds of metres. They are marked, like sutured seams, by concentrations of organic matter, clays, pyrite and detrital silicates, and commonly dolomite of residual or syn-stylolite origin. Non-sutured seams develop typically in limestones (and some finely crystalline dolomites) that have significant amounts (more than about 10%) of clay and silt-sized quartz and/or several percent or more of organic matter. Often it is difficult to be sure that these wispy to planar seams are the result of pressure-solution rather than physical compaction, and in fact very similar features have been produced experimentally by physical compaction and termed pseudo-stylolites (Shinn and Robbin, 1983).It can be equally hard to ascertain whether thinning and draping of laminae over and under rigid nodules and other objects, without obvious seams of residues, are due to the "non-seam dissolution " of Wanless (1979) or to purely physical compaction as noted in Shinn and Robbin's (1983) experiments. In some sequences of stratified limestones the seams are so abundant and closely spaced as to impart a distinct pseudo-bedding strikingly like depositional bedding , as illustrated in Figure 18 (Bathurst, 1987; Simpson, 1985). Bathurst (1987)has reckoned that solution thinning along these non-sutured seams may be comparable to the thinning caused by stylolite-forming dissolution . It would seem, therefore, that there is an important, previously unrecognized source of CaC0 3 in addition to obvious stylolites.
85
Recognition that non-sutured seams (also called clay seams - Barrett, 1964; horsetails - Roehl, 1967; Mossop, 1972; wispy laminae - Lucia, 1972; wavy laminae - Reinhardt and Hardie, 1976; pseudo-stylolites - Shinn et al., 1977; and "microstylolites" - Wanless, 1979) are pressure-solution features has led to useful new classifications (Garrison and Kennedy, 1977; Wanless, 1979; Buxton and Sibley, 1981 ; Koepnick, 1984; Scholle and Halley, 1985), as well as a more radical classification and terminology (Logan and Semeniuk, 1976). The organization of pressure-solution features outlined in Figure 19 is modified from a conceptually simple classification by Wanless (1979). From this way of organizing the observations of many workers and from other considerations, some points of interest emerge as one tries to understand the origins and dynamics of pressure-solution features. (1) Both sutured and non-sutured seams can be either bedding-parallel and single or multiple, or reticulate, as implied in Figure 19 by the multiple stylolites and fitted nodules. (2) Reticulate systems of anastomosing microstylolites (fitted fabrics) and non-sutured seams occur as a rule in limestones made up of abundant rigid entities of sand size or coarser (the "idens" of Logan and Semeniuk, 1976). These may be in grain-to-grain contact, like ooids and skeletal debris in poorly cemented grainstones (Buxton and Sibley, 1981; Meyers and Hill, 1983); or, the entities may be surrounded by more soluble sediment, like burrowcontrolled, resistant nodules of chert or ' carbonate-cemented sediment in a matrix of clayey or orqanlc-rich lime wackestone. Pressure-solution presumably starts at grain contacts (often preceded by physical compaction). The microstylolites then propagate irregularly through their host to create a fitted-fabric network. The network style is a result of a high degree of heterogeneity. (3) Bedding-parallel seams typically develop in well-bedded and/or more homogeneous limestones. The homogeneity commonly exists either because the limestones are Iime-mudrich with few large grains, or because they are grain-rich but have been extensively cemented by calcite and thus have less contrasting solubilities than
c
Geoscience Canada Reprint Series 4
Diagenesis
86
ISTYLES OF PRESSURE-SOLUTION I INON-SUTURED SEAMS I BEDDING-PARALLEL SINGLE SWARMS
SUTURED SEAMS * BEDDING- PARALL EL
-------
SINGLE SWARMS
NON -PARALLEL RETICULATE
MULTI-GRAIN
I~~
~
" GRAIN -CONTACT t
. ' .
~ ""'" "', -
LA RGE AMPL SM ALL AMP
NON -PARALLEL RETICULATE
€I>
GRAIN -CONTACT
<
"MULTI-G RAIN t , " NOD ULE-B OUNDING t
".
-c::zm~·
, stylo lites , • microstytontes t may be Med fabrics
Fig ure 19 Styles of pressure solution . Modified from Wanless ( 1979).
2
1
3
~--- - -- -- -
~
A
-
... . -..
- -=:, =-= ~ -..; =---
-"'-=-~-
€) - -. - .:.,: - - •
..;.;:.:.;:-::;:;:- . ........;.;...
-- @JJ)--::~- .:::,:,,::,::::. BURROWED LIM E MUDSTONE w JORGANIC-R ICH NON · SUTURED SEAMS
B
PARALLEL TO RETICUL ATE MICROSTYLOLITES & SEAMS
PARALLE L TO RETICULATE MICROSTY OliTES
LIME M UDSTONE w JSHALE INTERBEDS
PARALLEL S"" ALL STYLO LITES
PARAL LEL STYLOLITES & M ICROSTY LOLITES
PACKSTONE OVE R MICRODOLO MITE
MICROSTYLOL ITES & STYLOLITES CONT ACT
RET:CULA TE NETWORK OF' MICROST YLOLITES IFITT ED F ABRIC,
- '-e;;==:..
~
-
c
Figure 20 Sketches to suggest pathways along which pressure-solution features may develop and evolve. The initial sediments (stage 1) are assumed to have been physically compacted already, so that further compaction is entirely chemical. The microcrystalline dolom ite in Cremains essentially uncompacted.
c c c c c c c
c c c c c c c c c c c c c c c c c c c c c c c c c e c c c c..
c
c c
r c o r r:
o o o
o c o r:
o r:
r:
,-.
Limestones - The Burial Diagenetic Environment
before cementation. Well-bedded lime mudstones , therefore, tend to have stylolites of fairly uniform ampl itudes. Rudstones commonly have stylolites that parallel stratification but vary in their amplitudes because of grain-tograin solubility contrasts. Sequences in which relatively pure limestones are interlayered with organic-rich and/or clay-rich limestone intervals tend to undergo strong compaction and intense development of stylolites in the less pure intervals (e.g., Figure 18). (4) Although recent experimental work suggested to Baker et al. (1982) that clays reduce the solubility of calcium carbonate with which they are admixed, a large body of field, petrographic and other observations by many other workers indicate the opposite, namely that clays (and/or organic matter) seem in fact to be required in order for pressure-solution to occur. Furthermore, as just noted, many solution seams show signs of having been initiated along clay partings or clayey zones. Notwithstanding that many aspects of pressure-solution are poorly understood, still it is useful to try and put together the observations just considered, in a way that may suggest some dynamics for the initiation and propagation of pressure-solution features. As shown diagrammatically in Figure 20, stylolites and microstylolites may begin either as sutures along contacts between rigid grains, nodules and the like, or as non-sutured seams in clayrich zones and clay partings. Once initiated, the sutures propagate at apparent rates and in patterns that appear to vary in space and time. It was once thought that stylolite geometr ies are governed in part by solubility contrasts between grains and other rigid entities (Stockdale , 1922, 1943; Coogan and Manus , 1975;Chilingaretal., 1979). Dunnington (1954) showed, however,that this factor alone cannot be enough because it fails to explain the variations in geometry found among stylolites. A more plausible kinematic control on the development and form of stylolites has been proposed by Guzzetta (1984) . This kinematic model of migrating dissolution domains is based on the hypothesis that dissolution can only take place on one side at a time of a pressure-solution contact surface, and will shift from one side to the other through time. Where
multiple stylolites are initiated, some propagate parallel as well as normal to the direction of maximum stress, and these then accommodate the stress selectively and continue to grow at the expense of the others. Factors that Influence Pressure-Solution. The factors that determine when in a sediment's burial history pressuresolution will become significant, under what conditions it will slow or cease, and its role relative to physical compaction and cementation , all are poorly understood. What is clear is that a large number of factors can be influential. Among these are the amount of overburden (Iithostatic pressure), early cementation and dolomitization, water chemistry, water flow rates, pore-fluid pressure, presence of liquid hydrocarbons, CaC0 3 mineralogy, clay and organic matter, porosity and permeability and possibly temperature because of its effects on carbonate minerai solubilities. Many of these factors also influence carbonate cementation, and most of them have been considered recently in that context by Feazel and Schatzinger (1985). Burial Depth. Chemical compaction probably starts in marine sediments within the first few hundred metres of burial. There and until the sediments are firmly cemented , particle-toparticle pressure-solution and pointcontact cementation (the " s p otwelding" of Mapstone, 1975) seems to be the dominant style of pressuresolution . Non -sutured and sutured seam pressure-solution features may begin to form toward the lower end of this depth range and certainly become obvious after about a kilometre of burial. In carbonates being buried in dilute pore waters, the minimum burial depth needed to initiate chemical compaction is likely to be more variable and may often be less . Pressure-solution in shallow-marine grainstone and packstone of Mississippian age in southeastern New Mexico apparently began after only tens to hundreds of metres of burial (Meyers and Hill , 1983), resulting in fitted fabrics of microstylolites. Stylolites are well developed in Mississippian lime mudstones in the Illinois Basin subsurface, about 150 m below the unconformity separating the Mississippian and Pennsylvanian systems there, and
87
buried now under 550 m of overburden (Choquette and Steinen, 1980). In limestones of the uppermost Madison Group (Mississippian) in parts of western Wyoming, irregular stylolites are intensely developed in zones a few metres to tens of metres beneath the postMadison unconformity and rather sharply diminish in intensity downward. This distribution suggests that they may have been formed in a near-surface meteoric zone. Thus, the suggestion of Dunnington (1967) that pressuresolution becomeseffectiveand stylolites become abundant after burial to around 600-900 m is probably an oversimplification. Pore-Water Composition. Pres sure-solution seems to be accelerated, and physical compaction may be inhibited, by Mg-poor meteoric waters as opposed to marine-derived pore waters. Neugebauer (1973, 1974), drawing on work by Lippmann (1973) and some others, concluded that high Mg2+ activities and high Mg2+ /Ca2+ ratios in the marine-derived pore waters of the North Sea and deep-sea chalks were partly responsible for inhibiting cementation and porosity reduction by inhibiting pressure-solution. The inhibiting effect is apparently due to the high hydration energy of Mg2+, which decreases with rising temperature (Bjorlykke, 1983). The converseof these ideas is that pressure-solutionshould be favoured in burial settings characterized by input of meteoric water. Metastable Mineralogy. Wherever aragonite and high-Mg calcite persist into the burial environment they should favor pressure-solution,giving their host limestones a high "diagenetic potential" for alteration (Schlanger and Douglas, 1974). Deep-sea nannofossil-foraminiferal oozes have low diagenetic potentials because their calcium carbonate is all low-Mg calcite. This is a factor in accounting for their slow rates of pressure-solution, cementation, and porosity reduction. Clay Minerals. On balance, as discussed earlier, clays appear to foster pressure-solution. Marshak and Engelder (1985) concluded from petrographic and X-ray diffraction evidence that in order for spaced, pressure-solution cleavage of tectonic origin to develop in limestones, clays must be present between grains or along crystal boundaries to provide "interconnectivity" of
Diagenesis
88
dissolution sites at grain boundaries with the larger-scale, free-fluid system. Bathurst has concluded (pers. comm. , 1986) that the amount of clay and associated fines has an influence on the style of pressure-solution: Suturedseam stylolites of the sort illustrated in Figures 15 and 16 are favoured where clay/silt content is significant but less than about 10%, whereas non-sutured seams like those in Figures 17 and 18 tend to develop if clay/silt content exceeds about 10% and/or there is a significant amount (more than a few percent) of organic matter. Where there is neither clay nor organic matter, stylolites are unlikely to develop. Stylolites common ly seem to start developing along clay-shale partings. The observations of Marshak and Engelder (1985) are relevant in this connection. They found that pure lime grainstones deform primarily by twinning, with only isolated tectonic stylolites, and rarely develop tectonic (pressure-solution) cleavage. They also observed that sutured, stylolitic seams develop in relatively claypoor limestones and non-sutured seams in the more clay-rich limestones. Early Dolomitization. Early, preburial or very shallow-burial dolomitization can inhibit or preclude stylolites from forming . For example , lime mudstones in the Ste. Genevieve Formation (Miss issippian) of the Illinois Basin have superbly developed beddingparallel stylolites, but closely associated microcrystalline dolomites that formed by local alteration of similar precursor lime muds have no stylelites (Choquette and Steinen, 1980). The dolomites acquired a "grainsupported" crystal framework that resisted stylolitic pressure-solution. Liquid Hydrocarbons. Early introduction of oil into the pore systems of limestones still undergoing chemical compaction tends to inhibit or shut off the process, and can entirely prevent the formation of stylolites (Dunnington, 1967) as well as other pressure-solution effects including "autocementation ". Under what conditions, then, apart from the factors just considered, can pressure-solution no longer take place? If the solution-film model is in fact the mechanism for pressure-solution in limestones, and diffusion due to stress at this film is the rate-controlling step (Brown, 1987), pressure-solution should continue at diminishing rates
until there is no longer an ion-activity gradient away from the stressed-solid contact (along the solution film toward the pore-water system). This can only obtain where the pore-fluid pressure equals or exceeds Iithostatic (or other directed-stress) pressure, in which case the porosity and permeability must effectively be zero. If, as now seems likely, the porosities of sedimentary carbonates approach zero asymptotically with increasing burial depth, pressuresolution may in fact continue, though at rates approaching the infinitesimal, to depths well below 6 km in the lithosphere. All that is required to sustain some pressure-SOlution, given aqueous pore fluids, is enough stress to exceed the pore-fluid pressure. Burial Cements It is now clear that much of the carbonate cement in many well-lithified limestones was emplaced during burial diagenesis. Essentially all cementation in deep-basinal sediments that were never uplifted into the meteoric, chalks included, must be of burial origin (see
Geoscience Canada Reprint Series 4
Schlanger and Douglas (1974) for a thoughtful review of chalk diagenesis in DSDP cores, and Coniglio (1985) for an in-depth study of the petrology of ancient deep-marine limestones). Much of the cement in ancient shallowmarine, peritidal, and platform-margin limestones is of burial-diagenetic origin (Talbot, 1971; Meyers, 1974, 1978; Grover and Read, 1983; Mattes and Mountjoy, 1980; Moore and Druckman, 1981; Elliott , 1982; Frank et al., 1982; James and Klappa , 1983; Burruss et al., 1983; Walkden and Berry, 1984; Choquette and Steinen, 1985; Druckman and Moore, 1985; Mruk, 1985; Brinton, 1986). The importance of burial-diagenetic cementation has also been forcefully argued on grounds of stableisotope compositions by Hudson (1975, 1977)and on this and other evidence by Bathurst (e.g., 1971, 1984) and Scholle and Halley (1985). Shallow Burial "Marine" Cements. Cements of early, shallow burial origin (first few metres to tens of metres) in deeper marine settings include displacive fibrous calcite (also called "beef"
BURIAL CALCITE CEMENTS
BLADED PRISMATIC
NON-LUMI NESCENT BRIGHT
I
c c c c c c c c c c c c c c c
c c c c c c c
c c c c C E C C C C
c e 1.._
COARSE MOSAIC CALC SPAR PLANE LIGHT
DULL
C (
CATHODOLUMI NESCENCE
Figure 21 Sketches of burial-diagenetic cements as they appear in plane light and cathodoluminescence.
(
C (
C
cr: Limestones - The Burial Diagenetic Environment
c o o
o o
o o
o o
o o o
and " cone-in-cone"; see Figure 4), a variety peculiar to highly argillaceous mudstones; and low-Mg calcite cements of micrite, microspar and slightly coarser crystal sizes (ca. 1-200 m) found in fine-grained sediments including chalks. Coniglio (1985) has recently documented in detail the sedimentology, petrography and geochemistry of all these shallow-burial cements in Cambro-Ordovician slope and basinal limestones of western Newfoundland. Deeper-Burial Cements. In chalks the burial cements are microto-nannocrystalline, from around 1to 10 micrometres except in internal cavities of some skeletal grains, and so have been studied best with the scanningelectron microscope (e.g., Matter, 1974; Scholle, 1977;Feazelet al., 1985). Chalk cements have not yet been investigated in terms of cement zonation and geochemistry because of their ultra-small crystal sizes. The principal types of burial cements found in shallow-marine/peritidal limestones include: clear coarse calcite (calcspar or coarse equant spar); clearto-turbid coarse dolomite or dolospar ("saddle" dolomite of Radke and Mathis , 1980; " baroque" dolomite of Folk and Aseretto , 1974); the xenotopic dolomite of Gregg and Sibley (1984); and coarse anhydrite. Various lines of evidence including aqueous fluid inclusions, oxygen-isotope compositions and cement zonation demonstrate that these burial cements can form over a substantial range of conditions. These include temperatures from around
40-50°C to perhaps 200°C or higher, burial depths from a few hundred metres or less (Meyers, 1980) to kilometres; formation water compositions ranging from brackish to highly saline, from acidic to strongly alkaline pH values, and from moderately to strongly negative Eh values; water flow rates that probably range from almost stagnant to moderately rapid for ground water (order of less than 10° to 101 rn-yr-i): and time intervals of anywhere from a few thousand(?) to tens of millions of years. The general characteristics of coarse burial cements (Figure 21)are now well established. The cements are commonly ferroan (more than 500 ppm ferrous iron) and enriched in Mn2+ (more than 100ppm) but impoverished in Sr2+ compared with associated earlier diagenetic cements , and have oxygen-isotope compositions depleted in 180 . Their cathodoluminescence is characteristicallydull and mayor may not show compositional zonation. Aqueous fluid inclusions (generally containing both gas and liquid phases) and hydrocarbon inclusions are common. The significance of iron and manganese in the 2 + valence state is that they require reducing conditions to exist. Although in theory the waters may have positive Eh values (Carpenter in Frank et al., 1982), in actuality those deep formation waters for which analyses have been made (e.g., White, 1975;Collins, 1975; Drever, 1982) have at least moderately negative Eh. Consequently, ferroan and manganoan carbonate cements must precipitate from at least
89
moderately reduc ing pore waters (Oglesby, 1976; Pierson, 1978; Frank et al., 1982). Near-surfacemeteoricground water is normally well oxygenated (Evamy, 1969), but the distal parts of intermediate and deep carbonate aquifers remote from recharge areas can be anoxic (e.g., Grover and Read, 1983). Cathodoluminescence (Cl) is favoured by the presence of Mn2 + but inhibited by Fe2 + . The ratio of these two cations seems to be more important than their absolute concentrations in determining Cl (Pierson, 1978; Frank et al., 1982; Grover and Read, 1983;Fairchild, 1983; ten Have and Heijnen, 1985). Other cations such as Pb2+ and Ce2+ may serve as activator ions and Ni2+ and C02+ as inhibitor ions (Machel, 1985), though they are clearly subordinate to Mn 2+ and Fe2 + and iti s unclearwhether they affect primarily Cl or photoluminescence. Bladed-prismatic calcite. This cement consists of elongate scalenohedral crystals , generally found growing directly on grain surfaces or atop earlier, marine cements. Crystals are a few tens of micrometres across and up to a few hundred micrometres long, and have prismatic terminations . Cl is typically dark with one or more bright lines (Figure 22). Generally where more than one burial-cement type occurs, bladedprismatic calcite is the oldest. In fact, although bladed and prismatic calcite cements have been observed to contain two-phase fluid inclusions which suggest precipitation from hot brines (O'Hearn, 1985),some workers believe that these cements are "pre-burial" in
r: ~ (
Figure 22 Multiple generations of burial calcite cement in a small reef limestone cavity rimmed by Renalcis and early marine cements : (left) plane light, illustrating bladed prismatic (arrows)and coarse mosaic calcspar. (right) cathodoluminescence, illustrating black, non-luminescent prismatic crystals (arrows) with brightly luminescent tips and dUll, but zoned, luminescent coarse mosaic calcspar. Bar scale is 1 mm. Lower Cambrian, Labrador.
c 90
origin and form in the sub-Ienticularmixing zone (e.g., Moore and Druckman, 1981; James and Klappa, 1983; Grover and Read, 1983;Saller, 1985;Druckman and Moore, 1985). Coarse mosaic calcspar. This cement typically consists of plane-sided, equant crystals ranging from tens to hundreds of micrometres in size, which mayor may not be noticeably ferroan and generally have dull, zoned or unzoned CL (Figure 22). Poikilitic calcite. These crystals can be several millimetres in size and large enough to enclose several depositional grains, and otherwise have dull CL like the coarse mosaic calcspar. Coarse dolomite cement. Saddle or baroque dolomite and xenotopic dolomite are other varieties of burial cement which often areferroan. Saddle dolomite (Radke and Mathis, 1980) is typified by curved cleavage traces and crystal faces, a peculiar and distinctive sweeping extinction in cross-polarized light caused by a distorted crystal lattice, and often zonation in terms of Fe2+ and Mn2+ concentrations as well as CL (Figure 23). Xenotopic dolomite (Gregg and Sibley, 1984) is typically of mosaic style and consists of anhedral crystals with irregularorcurved boundaries, fine to coarse crystal sizes (ca. 0.1-1 mm), and as a rule undulatory extinction. From experimental and geochemical considerations it appears to form at temperatures of about 50°C and higher. Late-stage burial dolomite cements have been reported in limestone and dolomite sequences in many regions (Choquette, 1971; Radke and Mathis,
Diagenesis
1980; Mattes and Mountjoy, 1980; Choquette and Steinen, 1985; Moore and Druckman, 1981; Grover and Read, 1983; James and Klappa, 1983; Druckman and Moore, 1985; Prezbindowski, 1985). Commonly the dolomite is closely associated with burial calcspar cement which it either pre- or postdates. Saddle dolomite partly of replacement origin has been discovered recently in grainstones of the Jurassic Smackover Formation in the Louisiana Gulf Coast (D.H. Mruk, pers. cornm., 1986).The precipitation of saddle dolomite in preference to calcspar seems to be favoured by the presence of dolomite-crystal substrates, and at the relatively high temperatures of the burial diagenetic realm can take place in waters with relatively low Mg2+ICa 2 + ratios (Rosenberg and Holland, 1964). Coarse anhydrite cement. Large crystals of anhydrite, in some instances poikilitic forms, can be seen in many buried limestones associated with units of anhydrite, often in carbonateevaporite cycles (Brinton, 1986). Paragenetically, anhydrite cement may predate or postdate the burial carbonate cements (Figure 23). Its presence suggests formation, or conversion from gypsum, at depths of something over 300 m since above that approximate depth gypsum is the more stable form of calcium sulphate (Murray, 1964). Criteria tor Recognition of Burial Cements. Although the characteristics just outlined are generally reliable indicators of deep-burial origin, the surest criteria are those based on determining that
Geoscience Canada Reprint Series 4
cements actually postdate other burialdiagenetic features, such as physical compaction features, stylolites and microstylolites, and hydrocarbons or their alteration products (Figure 24). Generally speaking, cements can be safely dated as "late" if they are involved in one or more of the following kinds of time-space relationships: (1) Cement crystals crosscut stylolites, or microstylolites or pressure-solution seams end at cement crystals. (2) Cements heal fractured grains or spalled ooid cortices. (3) Cement crystals postdate the physicalor chemical compaction of grains (filling "compacted pores"), or enclose or replace compacted grains. (4)Cement crystals enclose or postdate . hydrocarbon alteration products such as asphalt or pyrobitumen. (5) Cements fill tectonic fractures or fill dissolution pore spaces which themselves formed by removal of both grains and early burial cements. Conditions of Cement Precipitation. Most burial cements, as outlined earlier, formed over a wide range of conditions. Temperature. Burial calcite and saddle dolomite cements evidently precipitate over a range of temperature from -40°C to 200°C or somewhat higher but most commonly between - 50°C and 150°C, and at burial depths between a few hundred metres and several kilometres. Precipitation temperatures have been estimated by three indirect methods, from (1) the temperatures at which primary two-phase fluid inclusions homogenize to a single, vapor phase when heated, (2) stable-
c c
c
c c c
c (
C C
C C C C C C ~
C C C C C C C C C C C C
C (
c e
c c c
Figure 23 Burial cements in a phylloid-algal mold in skeletal grainstone. (left) plane light, illustrating mosaic calcspar (e), saddle dolomite (D) with inclusion-rich centres, and anhydrite (A) partly replacing a zoned dolomite rhomb. Dark areas in central part of photo are pore spaces. (right) cathodoluminescence, illustrating unfilled pore space (black) and outer zones in saddle dolomite cross-cut by non-luminescent anhydrite. Bar scale is 1 mm. Paradox Formation (Pennsylvanian), southeastern Utah (Brinton, 1986). Photo courtesy of L. Brinton.
C (
C C
c
c c
Limestones - The Burial Diagenetic Environment
91
c
oxygen isotope compositions, and
t:
(3) inferred geothermal gradients (see EQUANT -MOSAIC CALCSPAR POST-DATES MICROSTYLOLITES
(i (
SYNTAXIAL OVERGROWTH
o (.
o o o o o o o
CEMENT HEALS SPALLED
0010 CEMENT POST-DATES FRACTURE
c
c c
o
o l
o o ( c
CEMENT HEALS MICRITE ENVELOPE
Figure 24
Criteria for interpreting cements as "late" burial features.
4 0
v.
III Q
a.
2
0
...
CO)
00
Z
0
0
III
a: C 0
-12
-10
-8
-6
-4
-2
0
2
OXYGEN, &180pDB'%o
l
C
r '-
Figure 25 A graph showing the hypothetical trend of systematic changes in isotopic composition which accompany the diagenesis of a marine carbonate sediment (A), first in the meteoric realm (B), then in possibly a mixing zone with precipitation ofprismatic cement (e), and finally in three successively deeper domains of the burialrealm with precipitation of coarse calcspar (D.E) andcoarse saddle dolomite spar(F). Thesequence ofcements from B through E was precipitated from progressively hotter waters. This general trend, although hypothetical, has been observed in carbonates - both limestones anddolomites, with both calcite anddolomite cements - from a great many areas and stratigraphic units. It may well be the " normal " trend of cement stratigraphies in the deep-burial domain.
Figure 2)and burial depths at the time of cement precipitation. Salinity. The solutions from which burial calc ite and dolomite cements precipitate, judging from a large number of determinations of freezing and initial melting temperatures of fluid inclusions, are generally brines of varying salinities. Most of these brines seem to be of Na-Ca-CItype, and the common salinity range is about 10 to 100 parts per thousand (seawater = 35 ppt), Some of the high salinities reported are from inclusions in cements from sequences associated with evaporites, such as the Jurassic Smackover Formation in the US Gulf Coast (e.g., Klosterman, 1981; O'Hearn, 1985), where the ambient pore waters probably reflect dissolution of evaporites (Collins , 1975; Carpenter, 1978; Moore, 1985). Furthermore, not all cements judged to have formed in burial settings have two-phase aqueous fluid inclusions. Nor does it follow that brines were responsible for al/ burial cements. Furthermore, Read and his associates (e.g., Grover and Read, 1983; Dorobek, 1987; Mussman et al., 1988) have suggested from cathodoluminescence and isotopic evidence that shallow-burial cements may be precipitated from meteoric waters in regions of especially active recharge and deep circulation . Despite these important exceptions, however, the data acquired thus far point toward brines as probably the main sources of burial cements. Isotopic compositions of cements and waters. Burial cements typically show a progression, from older to younger cement generations and also within generations, toward lighter and lighter stable-isotope compositions: their 013C values decline only slightly, whereas their 018 0 compositions decrease sharply, by as much as 10-15per mil (%0) (Figure 25). This same progression was observed by Dickson and his coworkers (Dickson and Coleman, 1980; see also Moore, 1985) in a series of compositional zones painstakingly sampled within single crystals of burial cement. Moreover, burial cements are virtually without exception isotopically "lighter" (depleted in the heavier isotopes of C and 0) than earlier marine and meteoric cements as well as marine depositional constituents in the same carbonates.
c 92
These relationships have been observed in a large number of studies both for calcite-dominated cement systems (e.g., Tanand Hudson, 1974; Walls etal., 1979;Meyers and Lohmann, 1980; Prezbindowski, 1985; Moore, 1985; Brinton, 1986),and for systems of multiple dolomite cements in early-diagenetic dolomite and limestone (e.g., Choquette, 1971; Fritz and Jackson, 1972; Mattes and Mountjoy, 1980; Choquette and Steinen, 1980; Brinton, 1986). However, Halley has cautioned that some burial cements may have relatively heavy 018 0 values and go unrecognized simply because we expect strongly negative values; for example, reservoir water in the Sunniland Field, south Florida has 018 0 of + 10%0 and at present reservoir temperature would make calcite cement with only -2 to -3%0. Land and Prezbindowski (1981) report values of + 12 to + 29%0 for waters from deeply buried Cretaceous carbonates of south Texas (R.B. Halley, pers. comm., 1986). The progressive decreases in oxygen-isotope composition primarily reflect increasing temperatures of the precipitating waters. These can be estimated by using assumed or determined values for the oxygen-isotopic compositions of the waters. Graphic charts of the sort published by Prezbindowski (1985) for calcite-water systems based on equilibrium relationships worked out by O'Neil et al. (1969), and by Land (1985) for dolomite-water systems can be consulted for this purpose. Using the chart by Prezbindowski (1985) one can get a sense of the temperature ranges involved in individual suites of burial cements. For example, cements with 018 0 values8%0 apart wouldhaveprecipitated at temperatures about 70°C apart, assuming a water of constant isotopic composition. Since progressivelyyounger burial cements commonly have precipitated from progressively more saline and presumably more 180-enriched waters, temperature ranges like the one just cited must be minimum values. Origin of Burial Cements. Passive cementation. As in any other carbonate-water system, CaC0 3 dissolved in one place can be transported over large distances before conditions are right for precipitation. It has been shown (e.g., Pray, 1966; Dunham, 1969; Bathurst, 1971, p. 440-441; Scholle and Halley, 1985)that in order to precipitate one volume unit of calcite
Diagenesis
cement from an imported aqueous solution with reasonable CaC0 3 content, something on the order of 10,000 to 100,000 unit volumes of solution is required. Is water flux likely to be large enough to meet these needs? In relatively shallow aquifers with flow rates of - 5-20metres per year such as the south Florida Tertiary aquifer (Back and Hanshaw, 1971) and the Madison Limestone aquifer of Wyoming, Montana and the Dakotas (Hanshaw et al., 1978),cementation appears to be an active presentday process. But in strata that are buried at depths of several kilometres such as the Smackover in much of the US Gulf Coast, and are unconnected with active aquifers, formation water flow rates are likely to be very small. In the opinion of some workers (e.g., Wood and Hewett, 1984) no formation waters are likely to be completely static even if overpressured, and some circulation with flow rates of at least 1 metre per year must take place by convection alone. Autocementation. An increasingly convincing body of evidence points toward the view that in partly closed systems pressure-solution may be the main agent for cementation in limestones undergoing burial diagenesis (Hudson, 1975, 1977; Scholle and Hailey, 1985). Most studies that have attempted to find cause-effect relationships between stylolites and burial cements, however, have not clearly established such relationships. A few authors have noticed that stylolites commonly occur in intervals of anomalously low porosity (Harms and Choquette, 1965;Dunnington, 1967;Nelson, 1981), but have not established that the cement occluding the porosity was genetically related to the stylolites. The only investigation known to us which demonstrated a direct link unequivocally was by Wong and Oldershaw (1981). Those authors showed not only a spatial relationship between stylolites and intervals of low porosity (Figure 26), but a time relationship as well, in which cement zones were found to be genetically related to stages in stylolite growth (Figure 27). The cements are very coarse and the zones, revealed by staining with KFeCN SOlution, are so spectacular that they can be seen with the unaided eye. Is it possible that similar time-space relationships exist in other stylolitic and once-porous limestones but have not
Geoscience Canada Reprint Series 4
been identified? To discern them in more finely crystalline cements would require very critical observations and some measure of luck. Buxton and Sibley (1981) argued that flow rates in limestone aquifers are too high to favour the transport of ions released by pressuresolution solely by diffusion along solution films to nearby precipitation sites. Wherever solution films are connected to water-filled pore systems in which flow is significant, the films should discharge their load of ions into the flow system. Factors that promote burial cementation. It is generally accepted that supersaturation with respect to both calcite and dolomite prevails in most subsurface pore waters. Exceptions are the shallow "upstream" (upgradient) parts of aquifers that are being actively recharged, like the Mississippian Madison Formation in the northern
c
c c c c c c
c:
c
c
c c c
c c c c c c c c
c c
c
c c
C E C C C (
c
Figure 26 A core-slab surface illustrating the common association of anomalously low porosity 6%) with stylolites. Porosities above and below the thin stylolite-related interval are 15-20%. The arrows indicate positions of pores that were used to define the boundaries of the "tight" horizon. Devonian Kaybob reef, Swan Hills Formation, Alberta (Wong and Oldershaw, 1981). Photo courtesy of P.K. Wong, published by permission.
«
e (
c l
l
C l
C C
c
c
Limestones - The Burial Diagenetic Environment
93
c
c: (
o
c c
c
c (
c o o
o
o o
o o
o (
o o o o o o o o
US Rocky Mountain and high plains region (Hanshaw et al., 1978) and the confined Biscayne aquifer(Cenozoic) of south Florida (Kohout, 1970; Back and Hanshaw, 1971);there, undersaturation and dissolution are common. The general condition, however, is one of supersaturation. It is likely to be achieved primarily by dissolution of CaC0 3 increasing the activities of Ca2 + and CO~ - in ambient pore waters. Supersaturation is likely to be maintained as long as significant precip itation of carbonate cement or significant dilution of system pore waters by meteoric water do not occur. Increasing temperature should also work to enable the precipitation of calcite , which is retrograde (Helgeson, 1969). Although increasing pressure partially counters the effects of temperature , experimental work by Sippel and Glover (1964) and Sharp and Kennedy (1965) indicates that over the temperature range from about 25°C to 200°C the solubility of calcite in water decreases by about two orders of magnitude. As much as half ofthis decrease may be counteracted by increasing pressure, but the change may be significant nevertheless. Another mechanism that may promote the precipitation of carbonate
cements in situations where fracturing and fault ing sharply alter the hydrology and pore-fluid pressure of a sequence or basin is CO2 outgassing. Hanshaw at al. (1978)suggested this mechanism as a means of precipitating carbonates from meteoric ground water subjected to reduced pressure at the Earth's surface. The same process may operate in special situations in the deep subsurface. There, where pore-fluid pressures are likely to be one to two orders of magnitude larger than any in the nearsurface meteoric, PC0 2 also is likely to be much higher. Any fault or fracture system that suddenly, in terms of geologic time, connects the aqueous porefluid column in the deep-burial domain with that in the shallower subsurface will abruptly reduce the pore-fluid pressure of the system connected with the opened fractures, dropping the PC0 2 and causing carbonates to precipitate both in the fractures and in the adjoin ing water-filled pore systems, Rates of outgassing are highly temperature dependent; a warm bottle of carbonated beverage, for example, loses its CO2 much more quickly than a cold one. The mechan ism is operative where travertine deposits form at the surface vents of hot springs in carbonate terranes. That it may indeed operate
' POROUS'
HORIZON
'POROUS' HORIZON
TIGH T
HORIZON
'POROUS' HORIZON
L (, (
A
r '-, r·· 1....
(
C
c
D
Figure 27 Diagrams showing the development of a stylolite andthe concurrentprecipitation of coarse, zoned cements with CaC03 produced by pressure solution. In its later stages, the stylolite grew at the expense of some of the earlier cement. Fe-rich cement is black, Fe-poor cement is white, and pore spar:e is stippled. Devonian Kaybob reef, Swan Hills Formation, Alberta (Wong and Oldershaw, 1981). Diagram courtesy of P. K. Wong, published by permission.
in the subsurface is strongly suggested by the fact that dilation fractures and breccias in carbonate rocks, whetherof tectonic or other origins, commonly are filled by coarse calcite or dolomite cement for which oxygen-isotope compositions indicate precipitation from hot format ion waters; furthermore, these fracture-filling cements often have the same isotopic composition as cements in the "wall rock" around them (e.g., Murata et al., 1969; Choquette, 1971). Baroque dolomite fillings are common and well known in breccias, fractures and "zebra structures" found in Mississippi Valley-type sulphide-ore gangues and in numerous epigenetic dolomite reservoirs of petroleum, Perhaps CO2 outgass ing through fracture systems was instrumental in their origin. Factors that inhibit cementation. The most clearly identifiable of these (Table 3) have been outlined earlier and are discussedby Feazeland Schatzinger (1985). Entryof hydrocarbonsisoneofthe more important inhibiting factorsand has been documentedin manylimestones, as illustrated in Figure 28. Dissolution in Burial Diagenesis The discovery of solution porosity in sandstones, attributed by some to CO2 and/or H2S generated during burial diagenesis (Schmidt and MacDonald,1979; Hayes, 1979), led shortly to a successful search for solution porosity of likely deep-burial origin in limestones. Pore space believed to be of this origin was reported by Moore and Druckman (1981; also Druckman and Moore, 1985; Moore, 1985) in deeply buried (> 2-3 km) oolitic and pisolitic limestones of the Upper Jurassic Smackover Formation in the US Gulf Coast. Around that time, solution porosity of burial-diagenetic origin was reported and carefully documented by Elliott (1982) in limestones of the Mississippian Mission Canyon Formation in the Williston Basin. The solution pores may be fabric selective or not (Choquette and Pray, 1970) and include molds, vugs, and solution-enlarged interparticle voids. Fabric-selective pores may be relatedto previous diagenetic history. A suite of metastable carbonate sediments which Is -altered and lithified in the meteoric environment may be entirely calcite, but not all its components are the same. Calcite cement is a pure precipitate;
94
calcitized aragonite and Mg-calcite allochems are precipitated calcite with minor embedded organic and mineral relics and high trace element content; original calcite fossils are biogenic calcite containing organic templates . Inthe new burial environment, because of differences in crystal size and composition , microporosity, and amount and kinds of original organic matter, the relative solubilities of these components will be subtly different: biogenic calcite > calcitized aragonite or Mgcalcite > calcite cement. Fabricselective dissolution of biogenic calcite allochems and possibly of calcitized allochems may subsequently take place during burial (Donath et al., 1980; James and Klappa, 1983). This is also manifested by silicification of biogenic calcite and Mg-calcite allochems (bryozoans , brachiopods, echinoderms) in the subsurface, while original aragonite allochems (molluscs, scleraetinian corals) are preserved as calcite. Non-fabric-selective voids may begin as interparticle pores or molds and quickly enlarge. Features that point toward a burial-diagenetic origin of these pore spaces are illustrated in Figure 24, and examples of these types of voids are shown in Figure 29 and 30. Pores commonly cross-cut both carbonate grains and cements, in situations where the cements are clearly of post-compaction, burial origin. In other cases, solution pores occur along and locally cut across stylolites, which they therefore postdate (Figure 30). Some micro porosity (Choquette and Pray, 1970), typified by apparent pore ..diameters" that are generally less than 0.03 mm and as a class of multiple origins, almost certainly forms in burialdiagenetic settings. Much chalk that has never "seen" pore waters other than marine has micropore systems (mainly intercrystal and primary interand intraparticle) which probably are the result of incomplete pressure-solution compaction and associated cementation that were arrested sometime early in the burial process (Moshier, 1987). Other microporosity includes both primary types and solution-enhanced micropores which seem to have originated or been enhanced late enough in the burial process to have postdated, entirely or in part , the development of stylolites and fittedfabric microstylolites (Dravis, 1986, 1987; Moshier, 1987).
Diagenesis
The burial origin of microporosity is often difficult to tie down, but a number of criteria suggested for occurrences in the Jurassic Haynesville Formation of east Texas (Dravis, 1987) are more widely applicable. These criteria include: (1) lack of early, meteoric pore space or cements; (2) evidence of pervasive pressure-solution and absence
Geoscience Canada Reprint Series 4
of pre-compaction cements; (3) close association of microporosity with stylolites and microstylolites; (4) earlier or closely associated burial cements, in some cases containing hydrocarbons; and (5) identical microporosity in biotic materials such as oyster shells made up of ordinarily stable low-Mg calcite . Microporosity of deep-burial origin is
c c c
c
c c c c
C
c
C C C C C
C C C G
C C C
c:
C C
C C C E C C
C C
c e
N~w"C;;' ,"~
"
POROSITY: MEASURED 0.3\>, PREDICTED, 22\>
Figure 28 Scanning-electron micrographs of chalk from a well in the Ekofisk area of the North Sea, illustrating the retarding effects of hydrocarbons on burial cementation. Samples A,S are from high in, C,D from the middle of, and E,F from below the hydrocarbon zone. Pore·fluid pressures are believed to be approximately equal in the three intervals. Cementation by calcite was extensive in the lowest, water-saturated zone, but was slight in the two zones above, suggesting that it was impeded by hydrocarbons. From Feazel et al. (1985b). Photo courtesy of C.l Feazel, published by permission.
(
c c c (
C C
c (
Limestones - The Burial Diagenetic Environment
95
('
C' (
c
c
c
c
c o o o o o o o o o o o o o o o o o o o C l
c , '-.,'
,.--'-..
1"....
L
especially interesting in petroleum geology because its presencehas made economic reservoirs of gas out of some carbonates that have low porosities and permeabilities and might otherwise be "tight", such as some extensively compacted and/or cemented oolitic limestones in the US Gulf Coast Jurassic. Solution porosity of burial-diagenetic or ig in has been ascribed by most workers to formation waters with anomalously high concentrations of CO2 that
were supplied by the thermal decarboxylation of organic matter(kerogen). According to Lundegard (1985), however, a more likely source of CO2 to account for dissolution of carbonate cement in sandstones of the Wilcox Formation (Tertiary) of the Texas Gulf Coast is a series of hydrous pyrolysis reactions between organic carbon and oxygen in water to yield CO2 or organic acids . These reactions would have occurred in connate waters expelled
into the Wilcox sandstones from compacting shales downdip. A third mechanism, the production of strongly acidic waters by sulphate reduction (in reaction with methane, for example) to produce CO2 and H2S, seems possible in format ion waters reducing enough to produce cements, and may have found support in the discovery of abundant iron sulphides in some Smackover limestones (L.M. Walter, oral comm., 1985).
Figure 29 Solution porosity interpreted as burial diagenetic in origin (Elliott, 1982). (left) Solution vug that cross-cuts both grains and dolomite-lined microstylolites. Most grains are flattened pisolites. (right) Smallvug formedby dissolution of micrite matrix in apisolitepackstone. Calcite crystals interpreted as warm -waterprecipitatespartlyline the vug. Bar scales are 1 mm. Miss issipp ian Mission Canyon Formation, Williston Basin, North Dakota . Photos courtesy of T.L. Elliott, published by permission.
Table 3
Factors that influence cementation in the deep-burial environment. Based in part on Feazel and Schatzinger (1985). Inhibiting
• • • • •
Stable CaC0 3 mineralogy Dilute pore waters Hydrocarbons displace pore waters Low-flux hydrology Increase in Pco, (higher pore pressure) • Decrease in temperature • Reduced porosity and permeability
Enabling • Metastable CaC0 3 mineralogy • Pressure solution • High-flux hydrology • Highly oversaturated pore waters (if high flux) • Increase in temperature • Decrease in Pco,
Figure 30 Solution pores (P) and calcspar (C) interpreted as burialdiagenetic, formed by dissolution along a stylolite. Bar scale is 1 mm . Mississippian Mission Canyon Formation , Williston Basin, North Dakota (Elliott, 1982). Photo courtesy of T.L. Elliott, published by permission.
c 96
Recently the suggestion has been made that meteoric-water systems, perhaps aided by mixing-corrosion (see James and Choquette, this volume) have more potential on mass-balance considerations for producing significant volumes of dissolution porosity in the deep subsurface (Giles and Marshall, 1986). It is clear that in active aquifers meteoric water can penetrate to great depths (2000 m in the Great Artesian Basin of Australia, Haberheh, 1980) and can travel long distances (up to 120 km off the east coast of Florida, Manheim, 1967). But the saturation states of meteoric water in such distal settings are little known, and the amount of dissolved CO 2 carried by them is likely to be small unless augmented by CO 2 from decarboxylation, sulphate reduction, hydrous pyrolysis reactions, orother reactions. The mechanisms involved in producing large volumes of deep-burial dissolution porosity - and indeed the very existence of large volumes - remain elusive. Whatever the mechanism(s) may be, there is little doubt that (1) solution porosity can be generated in the deep subsurface and may in places be significant, and (2) organic compounds, possibly including liquid hydrocarbons and! or organic acids, are implicated. It is unclear as yet whether solution porosity of this origin is to be found generally in limestones and sandstones situated updip in the flow paths of connate waters being expelled from compacting shales, and the prediction of such late porosity is a difficult challenge.
Diagenesis
published to date, and shows a number of importantfeatures. The DSDP curve , when comb ined with data for North Sea chalks (1 well, Scholle, 1977), has an exponential shape: rates of porosity reduction are highest at shallow depths and decrease exponentially with burial depth. The same is true for South Florida basin limestones, even though they appear to be much more var ied Iithologicallythan the chalks and likely spent part oftheir burial history in the meteoric environment. The exponential aspect was what led Pray (1960) and Schmoker and Halley (1983), respectively, to suggestthe concept of a porosity " half-life" or "half-depth". There is a general porosity versus depth relationship for oceanic chalks in marine pore-water systems. The righthand curve in Figure 32 is based on porosity data from a dozen DSDP sites
down toabout 1.1 km sub-bottom, and on chalks in a North Sea well from about 1.6 to 2.5 km sub-bottom. Building on interpretations by Schlanger and Douglas (1974), the chalk curve has been extrapolated to about 6 km; there it probably converges with a somewhat differently drawn curve for the South Florida limestones. Most Phanerozoic limestones buried to comparable depths lose all but about 5% or less of their porosity (Schlanger and Douglas, 1974; Scholle and Halley, 1985). Limestones with that little porosity generally have so little permeability ( < 0.05 md is typical) that further throughput or expulsion of pore waters is unlikely and compaction and cementation should be arrested. This general curve for chalks in a marine pore-water system also traces, albeit in very generalized form , the progressive lithification of pelagic, nanno-
~
~
o
3
7\:';::;';::''"u"" ,
o
[
11l
10 (:) o o
N '"
SHALLOW - M A RI NE LS. S FLA. ] AVG & DATA ENVELOPE
C
C C
C C C
C
c:
C
C C C E C C
15 POROSITY-DEPTH CURVES CHALKS
6
Figure 31 A plot of percentporosity versus depth for various limestone sequences. All but the deepest curve (to 6 km) are for Cretaceous and Tertiary chalks. (From Scholle, 1977; Lockridge andSCholle, 1978; SchlangerandDouglas, 1974). Chalk curves other than for DSDP carbonates and chalks from the SCotian Shelf and North Sea well L 16/1 probably reflect some diagenesis in the meteoric . Thedeep curve is forshaJlow-marine andperitidal limestones from the south Florida shelf; the shaded band indicates the spread of data for this curve. (After Schmoker and Halley,
1982).
(
C C
--
.: NORTH SEA IL1611l NORMAL CURVE
1,.,-..,.
0W
c c
C C 5
J:
c
C
NIOBRARA L~l U.S. GR. PLAIN AUSTIN .... CHALK, .: U.S . GUll;"
E
c c
C
.'
NORFOLK CHA LK, U.K .
c
(;
POROSITY, % 40 o 80 O,.......---r--.---rr--.~:7r~;rr--r:::;;oo--,
Changes in Porosity with Depth. The sum of all these processes and products should be reflected in changes in porosity with depth. General trends. Most reliable information comes from Cretaceous and Tertiary chalks in oceanic settings as a result of the Deep-Sea Drilling Project (DSDP), synthesized by Garrison (1981).Data for chalks in the North Sea, adjoining areas in Great Britain, and the US High Plains and Gulf Coast has been developed or synthesized by Scholle (1974, 1977), Lockridge and Scholle (1978), Scholle and Halley (1985). All of the porosity-depth information for the South Florida basin comes from studies by Schmoker and Halley (1983) and Schmoker (1985). Figure 31 gives most of the porosity versus depth curves that have been
Geoscience Canada Reprint Series 4
C C
c
e c. c
c (
C C C
('
c
Limestones - The Burial Diagenetic Environment
c c
foss il-foraminiferal ooze (unconsolidated) through chalk (poorly consolidated) into Iith ified limestone . Most deep-sea oo zes, and also sh allowmarine lime muds (Ginsburg , 1957), lose some poros ity in the fi rst metre or so of burial. Afte r this initial reduction from 80% to 75% , physical compaction remains dominant down to around 200250 m, where porosities are reduced to 60% and ooze changes into chalk. From 200-250 m to around 1000 m chemica l compaction becomes more and more dominant as the smaller nann ofossil remains dissolve under load and low-Mg calcite micrite cement precipitates, initially cementing grains at the ir contac ts. From around a kilometre downward, pressure-solution and " autocementation" clearly must continue to dominate, until diminishing pe rmeab ilities end pore-water circulation.
C CJ C'
C r:
('
o o
o
c c
97
If we add average values of initial porosity for modern shallow-marine carbonate sediments (Table 4) at the zerodepth level in Figure 32, we can define an envelope which should allow us to predict in very appro ximate fashion the porosity -depth characte ristics of virtually any CaC0 3 sedimen t buried in a marine pore -water system. This envelope, which assumes convergence at around 5% porosity or less, is in acco rd with data from compaction experiments that show rates of porosity reduction (compacti on) to be lowest for clean carbonate sand (grainstones) and highest for mudstones. It is interesti ng that the south Florida curve falls within thi s envelope. The similarit y in trends can be interpreted in more tha n one way, particular ly since the data for south Florida limestones show a considerable amount of scatte r
POROSITY, %
40
(Figure 31; Halley and Schmoker, 1982; Schmoker and Halley, 1983). From published information about the stratigrap hic age relationships of the oceanic chalks (Schlanger and Douglas, 1974) and the south Florida shallowmarine limestones(Schmoker and Halley, 1983) one can get a crude picture of the rates at which porosity is reduced in th ese Cretaceous to Pleistocene sediments . For chalk (Schlanger and Douglas, 1974), reduct ion from 80 % to 75% poros ity in the top metre or so is accomplished in around 50 ,000 years; reduction from 75% to 60 % by physical compaction to depths of 200-250 m takes on the order of 10 m.y.; and reduction from 60 % to 40% at around 1000 m depth takes as long as 120 m.y. For south Flo rida lim estones the age relat ionships are less well constrained but point to roughly comparable or slower rates of
POROSITY'" 80
OOZE - 1- 2 5 0 m 'PH YS. CO MPAC TI O N 00 "1. ' SOM E PO- CEMENTATION '75 -60 % POR OSITY CH A LK 250 -1000
m
' C HEM. CO MPACTIO N & PO CE M ENTATIO N OO M. , 60 - 40 % PO RO SITY
o o G
E
.x.
J:
~
a... w
o
3
LI M ESTONE - 1- 6+ krn ' CHEM . CO MPA CTI O N & PO C EME NTATION , 4 0 -5 % OR LESS POR O SITY
DISSO LUTION
. NORMA L' BURIA L CURVE
o D l
OOZE-TO-CHALK-TO LIMESTONE CURVE POROSITY vs DEPTH MARINE PORE WATERS
HYPOTHETICAL POROSITY-DEPTH CURVES
r
6 LL.........L._--L-_.........._ : L - _ L . - . . . - - L _..........- - - - ' (
,,' '-....'
'-..
'-..
r
Figure 32 Generalized porosity versus depth curves . The chalk curve Is for DSDP cores of unllthlfied nannofossil-foraminiferal ooze, partly Iithitied ch alk, and indurated limestone and rep resents a general lithificatio n sequence for lime-mud sediments saturated only with marine-derived po re waters. The lithification stages are from Schlanger and Douglas (1974). The south Florida curve is from Schmo ker and Halley (1982). The grainstone curve is hypothetical and is extrapolated from an average value at " zero depth " for modern sediments (Enos and Sawats ky, 198 1).
Figure 33 Hypothetical cur ves of porosity versus depth for five known situa tions: 1. " normal" burial of fine-gralned sediments in marine pore waters; 2. cementation In the meteono (horizontal segments) alternating with burial In marine pore waters; 3. reversalo fnormaltrend by dissolution Inthedeep subsurface fo/lowed by resumption of normal burial; 4. poros ity reduction arrested by build-up of abnorma/ly high pore-fluid press ures.
98
porosity decrease with depth; the first porosity "half-life" (50%-25%) or "halfdepth" would be approximately 50 m.y. (1.5 km) and the second and third (25%6%) would be around 150 m.y. (4-6 km). Rates of porosity reduction appear to be substantially greater in limestone sequences that now reside or in the past resided in the meteoric realm, as pointed out by others (e.g., Bathurst, 1985). Unfortunately, there is very little information about porosity variations over depth intervals of more than a few tens or hundreds of metres through meteoric zones. The porosity-depth curves in Figure 33, which are qualitative, portray the shapes one might expect for different burial-d iagenetic histories: (1) a "normal" curve of the kind shown in Figure 32; (2) a curve for chalk or lime mud that was subjected very early on to meteoric-zone or conceivably marine cementation by imported CaC0 3 ; (3) a curve of the sort expected if solution porosity were created after burial to considerable depth; and (4) a curve caused by early build-up of high pore-fluid pressure. It seems possible that the south Florida data, given its rather wide scatter, could include thin meteoric zone profiles. Volumetric considerations. At this point it may be helpful to summarize the principal ways, and some of their implications, that initially high pore volumes may be reduced. The simple diagrams in Figure 34 illustrate these points. Initial porespacecan bereduced, in principle at least, entirely by physical compaction (case 1), entirely by precipitated cement (case 2), or by combinations of compaction and cementation as suggested in case 3. It is important to distinguish between changes in pore volume and changes in porosity, which isthe percent of bulk volumeoccupied by pore volume. Porosity is the property that is measured whether using wireline logs, standard core-analysis methods, the GRAPE porosity-profiling device (Harmsand Choquette, 1965), orvarious other means. In case 1 the original thickness (to) is reduced to 35% and the original pore volume (PV0) to 16%, but the measured porosity is reduced from 80% to 40%. The distinctions here are important for what they imply about the real pore space now available for filling by cement, in th is example only 16 percent
Diagenesis
Table 4
Geoscience Canada Reprint Series 4
General porosity ranges of common textural types of modern carbonate sediments. Compiled from Enos and Sawatsky (1981), Schlanger and Douglas (1974), and GRAPE porosity profiles by Harms and Choquette (1965) and by Pray and Choquette (unpublished).
Chalks Lime muds (mudstone texture) Muddy lime sands (wackestone and packstone) Clean lime sands (grainstone)
I to
PV o
c c c c c
50% to 38% PV o
80%
j
70-80% 60-70% 50-60% 40-50%
c
35% l c 19% PV o
60% 40%
(
43%
c
57%
~
CASE 1. PHYSICAL COMPACTION ONLY. 75% PV o
c
50% PV o
40% 60% 80%
20%
CASE 2.
20%
20 %
20 %
CEMENTATION ONLY. ALL CEMENT IMPORTED. OPEN SYSTEM.
38% PV"
35% 10 16% PV "
60%
c c c c c c c C E C
50 % 10 80%
c(' c c c c c c ( c
40 %
20%
C C l (:
CASE 3. PHYSICAL AND CHEMICAL COMPACTION. ALL CEMENT INTERNAL, FROM PRESSURE SOLUTION. PARTLY CLOSED SYSTEM, WATERS EXPORTED ONLY.
i: (
c (' Figure 34 Diagrams showing the changes in pore volume , percentporosity, and thickness for three hypothetical cases of compaction and/or cementation. For purposes of these diagrams, all solid constituents are shown by the shaded/stipple pattern, the cement portion of the solid fraction is indicated by the hachured pattern, and pore space is white .
\... ..
(
C (
C
c
c c
or c ( r
c o C) o o o o c o o o o o
o c
o (
o o C)
o o r ,
r,
.....
Limestones - The Burial Diagenetic Environment
of the original pore space. A quite different picture emerges from this way of look ing at volume changes than that implied in the idea of a " porosity half-life" suggested by Pray (1960) and later reworded to "porosity half-depth " by Scholle and Halley (1985), for those concepts are based on measured porosities - the only measures of pore-space volume that are normalized and so can be compared between different data sets. Case 1, based as it is on the work of Shinn and Robbin (1983), should be considered to apply to fine-grained sediments that are subjected to increasing load while saturated with marine pore waters only. It seems likely that if the same sediments had been saturated with meteoric water when subjected to such extreme stress , some cementation would have occurred. The other extreme example is shown in case 2 of Figure 34. Here the same initial pore volume is assumed, but it is reduced entirely by precipitated cement unaccompanied by compac tion, with no significant change in thick ness. As in case 1, the measured porosity at one stage in pore-volume reduction is 40 percent. In this case also, a true "porosity half-life" is represented, because the real reduction in pore volume is exactly the same as the reduction in measured porosity. The important difference is that the real pore volume filled and remaining to be filled by cement is far greater here than in case 1. Furthermore, in case 2 all of the cement required is assumed to have been imported from cement donors somewhere else; none is supplied by the sediment or receptor itself. Case 2 may be a fair representation of those shallow-marine, peritidal (e.g., beachrock) and perhaps eolian carbonate sediments that "set up" and/or become partly cemented at the sediment-water or sediment-air/water interface, then during meteoric diagenesis acquire cement quickly enough to withstand a significant (though not predictable) amount of load stress and compaction . Cementation may be mineral-controlled or water-controlled as discussed earlier in this series (James and Choquette, 1984). The sediment is, however, entirely a receptor of cement. Case 3, in which both physical compaction and cementation produced by
pressure-solution work to reduce initial porosity, is probably closerto reality than the two preceding, extreme situations . The first three stages shown there probably model closely what happens in chalks and other fine-grained oceanbasin sediments that undergo marineonly burial diagenesis (Schlanger and Douglas, 1974; Garrison, 1981), as discussed in the next section. The fourth stage, with 5% porosity, is assumed from extrapolated porosity versus depth curves shown in Figures 31 and 32. Cement, which at the end stage occupies about half of the "rock volume", is internally generated by pressure-solution ; the sediment is both donor and receptor, and in the sense that pore waters are expelled both from the sediment and through it from underlying sediments, the system is partly closed with respect to ambient pore water. Another way of analyzing the effects of early meteoric cementation, pressure-solution, and compaction on the evolution of porosity with depth has been developed in an extensive thinsection point-count analysis of oolitic grainstones in the Smackover Formation of the US Gulf Coast (Grabowski et al., 1987). The total intergranular pore volume (IGV) of these rocks, that is, open plus cement-occluded intergranular volume, appears to become "set" partway through the bur ial process. Grainstones with porosity inversion and freshwater cements, analyzed from well depths of 600 to 2800 rn, have IGV = 40% (equal to the assumed average initial porosity); grainstones with minor marine and freshwater cements from 1800 to 5000 m have IGV = 30% ; and more extensively compacted grainstones without early cements from 1800 to 7300 m have IGV = 20%. During the compaction process , grain interpenetration stopped before 1800 m but pressure-solution continued to the maximum depth of the core samples , 7300 m. In each of the three diagenetic groups of grainstones, reduction of porosity was by cementation, and while the porosity decreased exponentially with depth, the total of occluded and unfilled pore space remains essentially constant, indicating that porosity reduction by compaction alone is in these rocks essentially " frozen" at relatively shallow burial depths.
99
Burial-Diagenetic Models In spite of obvious deficiencies in our data and understanding, it is nevertheless important to try and summarize what we know or can surmise about burial diagenesis. For this purpose it is useful to envision two general endmemberdomains(Figure 1and Table 5): (1) a basinal burial domain involving relatively deep-marine sediments which are being buried in undiluted, marine pore waters ; and (2) a shelf/platform burial domain involving shallow-marine and peritidal/coastal sediments which are being buried in varied pore waters ranging from meteor ic to marine to occasionally evaporitic, in response to episodes of emergence and exposure that interrupt their ongoing subsidence and burial. Burial loading may be more or less uninterrupted , as in the deeper marine parts of many sedimentary basins (1 above), or repeatedly interrupted as appears to be the case in many shallow-marine carbonate shelfl platform provinces bordering basins (2 above); or initially uninterrupted as a sequence is progressively buried, then arrested or reversed by basin-margin or intrabasin uplift. Although the general course of burial diagenesis seems generally similar in many respects, there are important differences between the two domains which we attempt to summarize in the discussion that follows (Table 5). Basinal Burial Domain. General characteristics. Sediments that accumulate in basinal environments, apart from sand-sized and coarser allochthonous debr is flows, are dominantly fine-grained. They carry marine pore waters into the subsurface, which are modified only by reactions between pore fluids and the organic and inorganic constituents of the sediments. Meteoric ground water may impinge on sediments of the outer shelf. or on sediments short distances downslope from exposed shelf/platform and ramp margins, but should not interact with basinal sediments even during low sea-level stands. The diagenet ic system is partly closed hydrochemically, in the sense that pore waters are being expelled only upward and toward the basin margins (Bonham, 1980) during the phases of majo r thickness and porosity reduction. There should be no significant transport in countercurrent
c 100
directions, unless by hydrologic convection (Wood and Hewett, 1984). Sedimentation and burial rates are slow,in the range of 2-15 metres per m.y. Build-up of abnormal pore-fluid pressure may occur nevertheless, for example beneath hardgrounds or zones of intense stylolite development or physical compaction or beneath zones of abundant carbonate concretions and layers. Geothermal gradients should be low, in the range of -15-20°C ' km-1• Nature of sediments entering the burial realm. Most of these will be unlithified, exceptforthin hardgrounds. The boundary between shallow (few metres to a few tens of metres?) and deep is somewhat arbitrary and is perhaps the lower limit at which sulphate reduction takes place, biogenic methane is produced, early-burial carbonate concretions and layers form, and chert nodules originate. Some workers place the boundary in the transition from bacterially mediated reactions to thermally mediated reactions, at approximately 60-80°C (L.S. Land,pers. comm., 1986). Carbonate fractions of pelagic sediments will be composed of low-Mg calcite, without aragonite or Mg-calcite, and thus stable, with little or no "diagenetic potential" (Schlangerand Douglas, 1974). Aragonite and Mg-calcitein allochthonous carbonates should survive well past the shallow-burial domain . Displacive- and replacive-fibrous calcite cements such as cone-in-cone may also have formed (see Coniglio (1985) for detailed information about shallow-burial features in fine-grained limestones). Trends in diagenesis. These will be dominated in the basinal domain by the expulsion of pore waters (dewatering), physical compaction initially and later pressure-solution compaction, the consequent long-term reductionof thicknessand pore volume, the production of carbonate cement by pressure-solution, and the thermochemical conversion of organic matter to hydrocarbons. Burial-dlagenetic features. Porosity generally decreases sharply in the upper 200-250 metres as outlined in Figure 32, mainly due to physical compaction. Through the first 800-1000 metres(?) of burial the sediments will show decreasing Sr2+, Mg2+ and bulk 018 0 as some of the finer-grained constituents dissolve initially under graincontact linear pressure (pressure-solution) and low-Mg, low-Sr finely crystalline calcite cements begin to precipitate.
Diagenesis
Cementation will start somewhere below 200-250 metres and become dominant below 1000 metres, will be directly due to pressure-solution (autocementation), and should be volumetrically proportional to the amount of pressure-solution unless flow rates of expelled pore fluids are high enough to move solute ions quickly away from sites of pressure-solution. Because Mg2+ concentrations and Mg/Ca ratios are relat ively high in marine pore waters, stylolites may begin to form at greater depths than they would in shelf/platform burial domains involving more dilute waters. However, pressuresolution "bedding" (Figure 18) due to intervals of closely spaced stylolites and non-sutured seams should be common, sequences with clay- and/or organic-rich strata. Table 5
Geoscience Canada Reprint Series 4
The cathodoluminescence of cements produced in sub-basin burial diagenesis is speculative. Early-burial displacive fibrous calcite may be dully to rather brightly luminescent (Coniglio, 1985) . Later burial calcite microcements should have dull CL with little or no zonation, reflecting the general monotony of basinal pore-water systems. Some burial dolomite may form, especially if the sediments are argillaceous, and should be ferroan, with dull, possibly zoned CL. Shelf/Platform Burial Domain. General characteristics. The sediments are far more varied and may include, in various combinations, limestones of mudstone to grainstone texture , reef carbonates, peritidal limestone and dolomite, coastal plain evaporites, and coastal plain to shallow
Similarities and differences between basinal (B) and shelf/platform (S/P) burial domains. SIMILARITIES
GENERAL TRENDS with increasing burial include: • TEMPERATURE and PRESSURE INCREASE • POROSITY REDUCTION, to less than 5-6% after 5-6 km burial. (Many thin , exposed shelf/platform carbonates are cemented tight before entering the shelf/platform burial domain.) • EXPULSION OF PORE WATERS • HYDROUS TO LESS HYDROUS OR ANHYDROUS MINERALS • MORE SALINE PORE WATERS • PHYSICAL COMPACTION FOLLOWED BY CHEMICAL COMPACTION CEMENTS: mostly dull CL and terroan in shelf/platform domain ; less well known but finer-crystalline in basinal domain DIFFERENCES BURIAL RATES: slow in basinal (B), high in shelf/platform (SIP) with interludes of exposure STARTING PORE WATERS: B: marine SIP: meteoric to brackish, marine and hypersaline STARTINGSEDIMENTS: B: mostly stable fine-grained CaC03 with low diagenetic potential and high physical compactibility; some metastable CaC0 3 in shelf-derived debris flows ; thin lith ified intervals; often argillaceous/ organic rich SIP: metastable to stable CaC0 3 and dolomite, fine- to coarse-grained, commonly lithified, with variable diagenetic potential; subt idal muds, evaporites and shales compactible ACCESS BY METEORIC WATER: B: nil unless uplifted SIP : usual due to eustatic and/or tecton ic uplift CARBONATE FOR CEMENTS: B: pressure solution dominant source SIP: meteorically derived water at shallow/proximal depths, pressure solution more important with burial; sulphate cement from dissolution of evaporites POROSITY REDUCTION: B: follows exponential curvets) of type in Figure 32 unless arrested or reversed SIP: can follow similar curvets) but also can go more swiftly in high-flux settings; rates more variable; reduction by physical compaction variable with facies and access of meteoric water
c c
c c c C \_-
c:
C C C C C C
C C C
c
C C C (
C C
C C C E
C C C (
C C (
L ( (: (
C C
c c
Limestones - The Burial Diagenetic Environment
101
c
c'
c
('
C
r
( (
c. c () o o
c (
o o () o (
o l
o o G o o c o c C L~
C
c ~
C. (
marine sandstones and shales. Because these shallow marine to coastal sediments are commonly exposed intermittently during accretion , they are periodically filled with meteoric ground water. Consequently, before they exit the meteoric environment they may have seen multiple generations of fresh to brackish to marine or even hypersaline waters related to low and high stands of sea level. Sedimentation and burial rates will vary from low to very high in the case of reef carbonates. The pre-burial diagenetic system will in general be open, since during times of exposure parts of the shelf/platform sequence will be recharge areas and others will be confined or unconfined surface aquifers linked to recharge areas. With burial, diagenetic systems may become partly closed as fluid flow directions become dominantly upward and laterally toward the updip margins of these shelf basins. Build-up of abnormal pore-fluid pressure may occur locally beneath evaporites and compacted shales, early-cemented permeability barriers of various kinds, and zones of intense stylolite development. Geothermal gradients may range from low to quite high. Nature of sediments entering the burial realm. The limestones will vary in CaC0 3 mineralogies from metastable to completely stabilized (see James and Choquette, 1984). Because some carbonates will be Iithified and others not, their resistance to physical compaction and/or chemical compaction will range from low in marine lime muds and uncemented marine lime sands, for example, to very high for such carbonates as calcrete, beachrock, shallow-marine hardgrounds, Iithified reef carbonates, and dolomites of peritidal/supratidal or mixing-zone origin. Porosities also will vary widely. Porewaters will range from fresh to marine, or in evaporitic sequences to hypersaline. Limestones associated with karst terranes will have acquired stable mineralogies and a wide variety of water-controlled alteration features (James and Choquette, 1984) including vugs, caves and solution-enlarged fractures, internal sediment, and a variety of speleothems. Limestones that still retain their metastable mineralogies, or retained them while in the meteoric or shallow-marine setting, may already have undergone mineralcontrolled alteration, perhaps in the
form of low-Mg calcite cements or of selective dissolution of allochems . It is unclear in any detail what the mineralogies and cements in the carbonates may be when they exit the lenticular mixing zone and enter the sublenticular, shallow burial realm. The limestones may have acquired nonluminescent , bladed and prismatic calcite cements with brightly lumines cent sub-zones (Figure 22). Early-diagenetic finely crystalline dolomite will probably have formed in some meteoric (mixed-water?) settings, and both the dolomite and some limestone may have acquired some pore-lining limpid dolomite cement. Trends in diagenesis. The dominant trends will include the evolution of pore waters from a varied suite to mostly saline to hypersaline brines; greatly varying amounts of physical compaction during relatively shallow burial and of chemical compaction during shallow(?) to deeper burial; cementation with both locally sourced CaC0 3 from pressure-solution and more remotely sourced CaC0 3 produced by dissolution beneath exposure surfaces; dehydration of certain minerals such as gypsum and smectite; and long-term reduction of porosity. Burial-diagenetic features. Porosity should vary considerably in the shallow reaches of the burial domain, but generally should decrease overall, at rates that will vary, if controlled by cementation, according to the amount of alteration accomplished earlier in the meteoric realm. Theduration and extent of physical compaction must vary considerably too, but cannot be predicted in any but very general terms . Uncemented marine sediments will suffer substantial compaction. Pressure-solution features are widespread in shallowmarine limestones, and include grainto-grain microstylolites and non-sutured seams, swarms of non-sutured seams in nodular arrays, and bedding parallel stylol ites and non-sutured seams. Pressure-solution features in limestones may begin developing at shallower depths than in sub-basin diagenesis - perhaps after was little as a few tens of metres in some cases because of the more dilute waters with lower Mg/Ca ratios and lower Mg2+ concentrations. Cementation should continue from the lenticular mixing zone downward.
Burial cements precipitated either below the reach of fresh meteoric ground water, or in deep aquifers at long distances down-gradient from recharge areas (see Back and Hanshaw, 1971; Hanshaw et al. , 1978; Grover and Read, 1983), will by and large have dull cathodoluminescence which mayor maynot be zoned. These cements will include both coarse calcite spar and saddle or baroque dolomite, and many of them will be ferroan. Older to younger burial cements will vary systematically in oxygen-isotope composition toward lighter o,ao values, and in trace-element compositions toward lowerconcentrationsof Sr2+ and higher Fe2+ and Mn2+ . Fluid inclusions will generally record progressively higher temperatures of precipitation from aqueous pore fluids that become more saline with depth and are dominantly of Ca-Na-CI type. Cement Sequences and Burial History Many different burial-history scenarios of course apply in sedimentary basins, and often can be discerned in the intricate "cement stratigraphies" (Meyers, 1974) that result. Many authors (e.g. , Meyers, 1974, 1978 ; Walls et al., 1979; Mattes and Mountjoy, 1980; Dickson and Coleman, 1980; Moore and Druckman, 1981; Grover and Read, 1983; James and Klappa, 1983; Moore, 1985; Walls and Burrowes, 1985 ; Burruss et al., 1985; Mruk, 1985; Brinton, 1986) have used various aspects of cement stratigraphy and geochemistry to reconstruct burial histories. Figure 35 is an attempt to portray the sequences of carbonate cement types, with CL characteristics, that might be precipitated in various diagenetic realms starting with the seafloor and meteoric settings and progressing into the basinal and shelf/platform burial domains , during three commonly encountered stages: (1) widespread marine transgression (high sea level), (2) lowered sea level with consequent exposure, and (3) continued subsidence and burial, all without tectonic uplift. Where sedimentary basins, containing a spectrum of facies from shallowmarine/peritidal or continental to shelf basin or basinal, are deformed and their margins exhumed and exposed to recharge by meteoric waters into aquifers, a whole array of major changes
c Geoscience Canada Reprint Series 4
Diagenesis
102
c c
( must then take place: the hydrology of the system commonly becomes dominated by meteoric waters which are modified with transport into the subsurface; geothermal gradients may be altered, becoming lower in the recharge areas and changing laterally to steeper gradients across the subsurface parts of the deformed basin - changes that may influence the course of carbonate cement precipitation; and the recharge portions of limestone aquifers may then become major donors of cement because of large-scale dissolution. All of these modifications to the basin dynamics will inevitably be recorded in the burial-diagenetic record. Investigations ofthe present-day and past hydrology and the geochemistry of many basins and deep aquifers, notably the Western Canada or Alberta basin (e.g., Hitchon, 1963a-d, 1969a,b; Garven and Freeze, 1984a,b), provided the basis for a model sketched in Figure 36. As in many other basins, the general paleohydrology, hydrochemistry, and burialdiagenetic history of this particular basin have also been intimately bound up in the generation and migration of hydrocarbons in the basin and the origin of lead-zinc sulphide ore deposits and their associated epigenetic dolomite host rocks in the Pine Point and other districts.
Conclusion Much of what we see when we look ata limestoneis the resultof its residence inthe burialdiagenetic environment. The most dramatic changes are brought about through compaction. Physical compaction, occurring first, may transform depositional texture from wackestone to packstone, compress unlithified sediment, squash and obscure critical depositional features, and through plastic and brittle deformation rotate, fracture and crush sedimentary particles. Chemical compaction or pressuresolution, taking place deeper, results in stylolites or non-sutured seams, which can create a new style of diagenetic bedding/stratification, pressure-solution pseudo-bedding. Dissolved carbonate is precipitated locally (autocementation) and probably accounts for much burial cementation. The bladed, coarsely crystalline, poikilotopic, commonly ferroan burial cements have distinctive isotopic, cathodoluminescence and fluid-inclusion signatures which
c
MARINE-FLOODED SHELF
c
_1 __L _ _____ J ____ l_l"••
c
SHALLOW AOUIF~
~~
Prism-NL
e Bolry-MOD
SEA WATER,
c
e Mosa ic-DU
EVOLVING WITH DEPTH
eM icrile-DU?
c
•
LOWERED SEA LEVEL
----..:;----\:.---1----:;-----'-.::-------'= -----7 - . "..:.lr------~--~
~
l
e
6~sR~~~~ON
DEEP ACTIVE
~QUIFER
!
L
t
h
"---
\ \ , - - - - e Mosaic-DU (Fe) '--- "-.--Biaded-NLIDU
BRACKISH
•
-:
~
eMlcrlte-DU Micrile-DU
DEWATERING
(
c
-- ---
-~-----1
SLUGGISH CIRCULATION
eBolry-MOD ' \ Prlsm-NL Spar-DU (Fe)
"o:::::::-~~~ e Mosaic-DU Bladed-NLIDU Spar-DU (Fe)
"
eMicrite-D Mlcrile-DU Spar-DU
Figure 35 A diagram showing the cement stratigraphies that might develop in different settings with lowered sea level followed by prolonged subsidence and burial of a carbonate sequence. The basinal burial domain is at right and the shelf/platform domain at left. Abbreviations: DU, dull cathodoluminescence; NL, non-luminescent; MOD, moderately luminescent. Prism, prismatic; Bofry, botryoidal; Spar, coarse calcspar.
c c c c c
c
c c c c c
c c
~---~.....
c c
W IW
::;; LL
o
(,
~
o
o
c..
CEMENTATION LIKELY
o
MAY BE UNDERSATURATED ~ o o
c c c c c c
c
PROLONGED SUBSIDENCE & BURIAL
e Prism-NL CORROSION Spar-DU (Fe)
(
LOW T GRAD ...........- - - - -
HIGH T GRAD 10's-100 's KILOMETERS
•
Figure 36 A schematic cross-section of an asymmetricalforelandbasin, the Alberta Basin, showing hydrologic flow lines and temperature gradients . (After Hitchon, 1969).
c
( ( (
C C
( (
Limestones - The Burial Diagenetic Environment
103
(
'-
(
C1 C
C , (
o o r
o o
(
o
r
(,
r.
e
.....
point to precipitat ion from brines , promoted by gene ral carbonate supersaturation and elevated temperature. The dominant, prevailing trend in bur ial diagenesis, as pointed out many years ago (e.g., Pray and Choquette, 1966; Choquette and Pray, 1970), is toward complete or nearly complete occlusion of porosity. Some porosity may be retained, however, for a variety of reasons: if the sediments were already low-Mg calcite or dolomite prior to burial; or formation-water flux was sluggish over much of the sediment's burial history, and/or residence time in the burial setting and depth of burial were insufficient for pressure-solution cementation to proceed to completion; or perhaps the geothermal gradient was low and the temperature effect on carbonate solubility was suppressed; or again, perhaps most desireable and serendipitous of all, if the occlusion of porosity was shut off by the influx of hydrocarbons. Dissolution may also take place, though perhaps on a limited scale, through reactions between formation waters and maturing organic matter. Although the processes and general products of burial diagenesis give us an idea of the features and the kinds of limestones to expect in deep-burial settings, the actual fabrics and structures that develop in limestones are largely dependent on differences in the original composition and early diagenetic history of the sediments. Two prelim inary diagenetic models representing burial of basinal versus shelf/platform sediments, respectively removed from or associated with early " exposure" to waters of meteoric origin, are proposed to highlight these differences. Only with hindsight are we likely to know whether or not in the 1980s we have been on a plateau or a steep climb in the search for unifying principles in the burial diagenesis of sedimentary carbonates. It does seem clear that our understanding of the dynamics and our . ability to forecast or even adequately chart the different courses of diagenesis in this largely occult realm are still quite primitive. There is much to learn but we have come a long way; the next steps in the search are sure to be excit ing.
Acknowledgements This art icle has benefited from the ideas and suggestions of many colleagues . We are grateful to Robin G.C. Bathurst, Robert B. Halley, and Lynton S. Land, and to our colleagues Mario Coniglio , Dexter H. Craig and Neil F. Hurley for critical reviews of the manuscript. Photographs were graciously lent us by Robin Bathurst, David K. Beach, Lise Brinton, Nancy Chow, Mario Coniglio, Charles T. Feazel, Eugene A. Shinn , and P.K. Wong. We owe much to the thoughtful treatises on aspects of burial diagenesis by three distinguished friends , Robin Bathurst, Robert Halley and Peter A. Scholle . Finally, we thank our wives, Jean and Judith, for assistance in the preparat ion of this review and the two that preceded it (James and Choquette, 1983,1984).N.P.Jo acknowledges the Natural Sciences and Engineering Research Council of Canada for ongoing support. PoW.C. acknowledges the management of Marathon Oil Company, particularly S.G. Andrew and F.G.Knight, for support and permission to publish.
REFERENCES General References and Reviews ADREF- , 1984, Stylolites and associated phenomena - relevance to hydrocarbon reservoirs: - Abu Dhab i National Reservoir Research Foundation, Special Publication, P.O. Box 6838, Abu Dhabi, U.AoE., 304 p. A compendium of recent papers on the burial diagenesis of sediments in general and on stylolites in particular. Barnes , M.A., Barnes, W.C. and Bustin , R.M., 1984, Diagenesis and chemistry and evolution of organic matter: Geoscience Canada, v. 11, p. 103-114. An overview of the alteration of sedimentary organic matter. Bathurst, R.G.C ., 1975, Carbonate sed iments and their diagenesis: Elsev ier Scientific Publishing Co., Amsterdam, Developments in Sedimentology 12, Second Edition, 658 p. The most comprehensive synthesis ever written on diagenes is in carbonates, and like its predecessor first edition (Bath· urst, 1971), the standard reference on the subject. The bibliography, although now in need of updating, is extensive and unusually valuable.
Bathurst, R.G.C., 1980a, Lith ification of carbonate sediments : Blackwell Scientific Publications, Oxford, Science Progress, p.451-471. A review of research on cementation in carbonates. Bathurst, R.G.C., 1980b, Deep crustal diagenes is in limestones : Revista del Instituto de Investigac iones Geologicas Diputacion Provincial , Universidad de Barcelona, v. 34, p. 89-100. Probably the first recent review specifically on burial diagenesis . Bathurst , R.G.C ., 1985, Carbonate diagenesis and reservoir development conservation, destruction and creation of pores : Notes for a short course: Colorado School of Mines , Golden, Colorado, April 8-10, 1985, 63 p. A useful review on carbonate diagenesis and porosity. Soon to be published, with other short-course notes from the same series , in the Colorado School of Mines Quarterly. Bricker, O.P., 1971, ed ., Carbonate Cements : Johns Hopkins Press, Baltimore, MD, 376 p. The last third of this book contains articles on burial cements and cementation . Other parts have examples of marine and meteoric carbonate cementation. Chanda , S.D., Bhatlacharyya , A. and Sarkan , S., 1983, Compact ion in limestones - a reappraisal: Journal of the Geolog ical Society of India, v.24, p.73-92. An extensive review that forcefully argues the case for commonly extensive compaction in fine-grained limestones. Choquette, P.W. and Pray, L.C., 1970, Geologic nomenclature and classification of porosity in sedimentary carbonates: American Association of Petroleum Geologists, Bulletin, v. 54, p. 207-250. An outline of the basic concepts of porosity evolution in carbonate rocks. Contains a comprehensive descriptive classification that has been widely accepted. Enos, P. and Sawatsky, L.H., 1981, Pore networks in Holocene carbonate sediments : Journal of Sedimentary Petrology, v. 51, p. 961-985. The best single reference to the porosity of modern carbonate sediments, with abundant data. Feazel , C.T. and Schatzinger, A.A., 1985, Prevention of carbonate cementat ion in petroleum reservoirs , in Schneidermann N. and Harris , P.M., eds., Carbonate Cements : Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 97-106. A conc ise overview on factors responsible for preserving carbonate poros ity is preserved in the geologic record.
c 104
Diagenesis
Geoscience Canada Reprint Series 4
C
r
Friedman , G .M ., 1975 , The mak ing or unmak ing of limestones or the ups and downs of porosity: Journal of Sedimentary Petrology, v. 45, p. 379-398. A general review on both cementation and porosity. Garrison , R.E , 1981, Diagenesis of oceanic carbonate sed iments - a review of the DSDP perspective, in Warme , J.E ., Douglas , R.G . and Winte rer, E.L., eds ., The Deep Sea Drilling Project -a decade of progress: Society of Economic Paleontologists and Mineralogists, Spec ial Publication 32, p. 181-207. A synthes is of results from the DSDP project concerning diagenesis of chalk and other deep-marine carbonate sediments in the marine burial environment. Halley, A.B., 1984, Post-depositional diagenesis of metastable carbonates - an overview: Palaeontographia Amer icana, no . 54, p. 245-253 . A brief review emphasizing characteristics of starting carbonate skeletalconstituents and sediments. Halley, A.B. , 1985, Burial diagenes is of carbonate rocks : Notes for a short course (unpublished): Colorado School of Mines , Golden, Colorado, April 1985, 44 p. An excellent review covering many of the same topics discussed in this article, but from a different perspective. Soon to be published in the Colorado School ofMines Quarterly. Meyers, w.J ., 1974, Carbonate cement stratigraphy of the Lake Valley Formation (Mississippian), Sacramento Mountains, New Mexico: Journal of Sedimentary Petrology, v. 44. p. 837-861. Thefirst paper to use cement stratigraphy based on cathodoluminescence to unravel the burial diagenetic history of a group of shallow-water carbonates. This paper stimulated a new generation of studies and discoveries in diagenesis. Roehl, P.O. and Choquette, P.W., 1985, eds ., Carbonate Petroleum Reservoirs : Springer-Verlag , Heidelberg, 622 p. Numerous papers in this volume contain excellent illustrations of diagenesis in the burial realm. Schneidermann, N. and Harris, P.M., 1985, eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, 379 p. Many papers in this votume specifically address burial diagenesis and cementation. An important collection of works on the subject. Schlanger, S.O. and Douglas , R.G., 1974, The pelagic ooze-chalk-Iimestone transition and its implications for marine stratigraphy, in Hsii, K.J. and Jenkyns, H.C., eds., Pelag ic Sediments on Land and Under the Sea: International Association of Sedimentologists, Special Publication 1, p. 117-148.
The first paper to outline succinctly the diagenesis ofpelagic carbonates and propose the concept of diagenetic potential. Scholle, PA and Halley, R.B., 1985, Burial diagenesis - out of sight, out of mind!, in Schneidermann, N. and Harris, P.M., eds. , Carbonate Cements: Society of Economic Paleontolog ists and Minera logists , Special Publication 36, p.309-334. An outstanding review of burial diagenesis of carbonate sediments. Argues persuasively the case for extensive cementation of pressure-solution, burial-diagenetic origin. Has an excellent bibliography. Shinn, E.A . and Robbin, D.M., 1983, Mechanical and chemical compaction in fine-grained shallow-water limestones: Journal of Sedimentary Petrology, v. 53, p.595-618. Thefirst detailed experimentalstudy using cores of modern carbonate sediments with their original in situ pore fluids. A wellillustrated work that helps explain many features seen in fine-grained limestones. Tissot, B.P.and Welte, D.H., 1978, Petroleum Formation and Ocurrence - a new approach to oil and gas exploration: Springer-Verlag, Berlin, 538 p. Probably the standard reference on the geochemistry, maturation, and genera tion of organic matterandhydrocarbons in sedimentary rocks , emphasizing shales more than carbonates. Physical Compaction
Baldwin, B. and Butler, C.O., 1985, Compaction curves: American Association of Petroleum Geologists, Bulletin, v. 69, p.622-626. Beach, D.K . and Schumacher, A., 1982, Stanley Field, North Dakota - a new model for a new exploration play,in Christopher, J.E. and Kaldi, J., eds., 4th International W illiston Basin Symposium, Saskatchewan Geological Society, Special Publication 6, p. 235-243. Bhattacharyya, A. and Friedman, G.M., 1979, Experimental compaction of ooids and lime mud and its implication for lithification during burial: Journal of Sedimentary Petrology, v.49, p. 1279-1286. Brown, P.A., 1969, Compaction of finegrained terrigenous and carbonate sediments - a review: Bulletin of Canadian Petroleum Geology, v. 17, p. 486-495 . Byers, C.w. and Stasko , L.E., 1978, Trace fossils and sedimentologic interpretation - McGregor Member of Platteville Formation (Ordovician) of Wisconsin: Journal of Sed imentary Petrology, v. 48, p. 1303-1310.
Coogan , A.H., 1970, Measurements of compaction in oolite grainstone : Journal of Sedimentary Petrology, v. 40, p. 921-929. Ebhardt, G., 1968, Experimental compaction of carbonate sediments, in Muller, G. and Friedman, G.M., eds., Recent Developments in Carbonate Sedimentology in Central Europe : Springer-Verlag, Berlin, p.58-.65. Fruth, L.S., Jr., Orme, G.R. and Donath, FA , 1966, Experimental compaction effects in carbonate sediments: Journal of Sedimentary Petrology, v. 36, p. 747-754. Gaillard, C. and Jautee, E., 1987, The use of burrows to detect compact ion and sliding in fine-grained sediments : an example from the Cretaceous of SE France: Sedimentology, v. 34, p. 585-593. Ginsburg, R.N., 1957, Early diagenesis of lithification of shallow-water carbonate sediments in south Florida, in LeBlanc, R.J. and Breeding , J.G., eds., Regional Aspects of Carbonate Deposition: Society of Economic Paleontologists and Mineralogists, Special Publication 5, p.80-98. Hallam, A., 1964, Origin of the limestoneshale rhythms in the Blue Lias of England: a composite theory: Journal of Geology, v. 72, p. 157-168. Hathaway, J.C. and Robertson , E.C., 1961, Micro-texture of artificially consolidated aragonite mud : United States Geological Survey, Professional Paper 424-G , p.301-304. Meyers, W.J., 1980,Compaction in Mississippian skeletal limestones , southwestern New Mex ico : Journal of Sedimentary Petrology, v. 50, p. 457-474. Meyers, w.J . and Hill, B.E, 1983, Quantitative studies of compaction in Mississippian skeletal limestones, New Mexico: Journal of Sedimentary Petrology, v. 53, p.231-242. Oertel , G. and Curtis, C.D. , 1972, Clay ironstone concretion preserving fabrics due to progressive compaction: Geological Society of America Bulletin, v.83, p. 2597-2606. Pray, L.C., 1960, Compaction in calcilulites (abstr.): Geolog ical Society of America Bulletin, v. 71, p. 1946. Rieken, w., 1987, The carbonate compaction law: a new tool: Sedimentology, v. 34, p.571·584. Robertson, E.C., 1967, Laboratoryconsolidation of carbonate sediment, in Richards, A.F., ed., Marine Geotechnique: lnternational Research Conference on Marine Geotechnique, Urbana, Illinois, University of Illinois Press, p. 118-127. Shinn, EA., Halley, R.B., Hudson , J.H. and Lidz, B.H., 1977, Limestone compaction an enigma : Geology, v. 5, p. 21-14.
(
C C C (
(
C (
C C C C C C C C (
C C
C C C
C C C E C C C C (
i: (
c
c c (
c.
c (
Limestones - The Burial Diagenetic Environment
(1 1....
Q (
(
c
o o o
o ()
r
r r
Shinn, E.A. and Robbin, D.M ., 1983 , Mechanical and chemical compaction in fine-grained shallow-wate r limestones: Journal of Sedimentary Petrology, v. 53, p.595-618. Terzaghi, R.D., 1940, Compact ion of lime mud as a cause of seconda ry structure: Journal of Sedimentary Petrology,v. 49, p.437-462. Weller, J.M., 1959, Compaction of sediments: American Associat ion of Petroleum Geolog ists, Bullet in , v.43, p.273-319. Zankl , H., 1969, Structural and textura l evidence of early lithification in finegra ined carbonate rocks : Sedimentology, v. 12, p. 241-256. Chemical Compaction, Stylolites and Cementation Alvarez , W, Engelder, T. and Lowrie , W , 1976, Formation of spaced cleavage and folds in brittle limestone by dissolution : Geology, v. 4, p. 698-701. Baker, P.A., Gieskes, J.M. and Elderfield, H., 1982, Diagenesis of carbonates in deep-sea sediments - evidence from SrI Ca ratios and interstitial dissolved Sr data: Journal of Sedimentary Petrology, v. 52, p. 71-82. Baker, P.A., Kastner, M., Byerlee, J.D. and Lockner, D.A., 1980, Pressure solution and hydrothermal recrystallization of carbonate sediments - an experimental study: Marine Geology,v. 38, p. 185-203. Barrett, P.J., 1964, Residual seams and cementation in Oligocene shell calcarenites, TeKuiti Group :Journal of Sedimentary Petrology, v. 34, p. 524-531. Bathurst, R.G.C., 1984, The integration of pressure-solution and mechanical compaction and cementation , in Stylolites and Associated Phenomena - relevance to hydrocarbon reservoirs: Abu Dhabi National Reservoir Research Foundation , Special Publication, Abu Dhab i, U.A.E., p. 41-56. Bathurst, R.G.C ., 1987, Diagenetically enhanced bedding in argillaceous platform limestones : stratified cementation and selective compaction : Sedimentology, v. 34, in press. Bochon , P. , 1976, L'importance des joints stylolithiques dans la compaction des carbonates: Centre des Recherches de Pau, Bulletin , v. 10, p. 627-644. Brown, A.A., 1987, Mechanisms of porosity loss in limestones: applications to numerical models of porosity loss : Society of Economic Paleontologists and Mineralog ists, Abstracts v. IV, 1987 Midyear Meeting , p. 10. Buxton , T.M. and Sibley, D.F., 1981, Pressure solution features in a shallow buried limestone: Journal of Sedimentary Petrology, v. 51, p. 19-26.
Choquette, P.W. and Steinen, R.P. , 1980, Mississippian non-supratidal dolom ite, Ste . Genevieve Limestone , Ill inois Basin: evidence for mixed-water dolo mitization, in Zenger, D.H., Dunham , J.B. and Ethington, R.L., eds., Concepts and Models of Dolomit izat ion - a symposium: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 163-196. DeBoer, R.B ., 1977a , On the thermodynamics of pressure solution: interaction between chemical and mechanical forces : Geochimica et Cosmochimica Acta, v. 41, p. 249-256. DeBoer, R.B., 1977b, Pressuresoluton ,theory and experiments: Tectonophysics, v. 39, p. 287-30t Dunnington , H.V., 1954, Stylolite development post-dates rock induration: Journal of Sed imentary Petrology, v. 24, p.27-49. Dunn ington, H.V., 1967, Aspects of diagenesis and shape change in stylol ite limestone reservoirs: 7th World Petroleum Congress, Proceed ings, Mexico City, v. 20, p. 339-352 . Droxler, A. and Schaer, J.P., 1979, Deforma tion cataclastique plastique lors du plissement, sous faible couverture, de strates calcaires: Eclogae Geologicae Helveticae, v. 72(2), p. 551-570. Durney, D.W., 1972, Solution-transfer, an i mpo r t ant geological deformation mechanism : Nature, v. 235, p. 315-317. Durney, D.W., 1976, Pressure-solution and crystallizat ion deformation : Royal Society of London, Philosophical Transactions , v. 283A, p. 229-240. Eder, W, 1982, Diagenetic redistribution of carbonate, a process in forming limestone-marl alterations (Devonian and Carboniferous, Rheinisches Schiefergebirge , West Germany) , in Einsele, G. and Seilacher, A., eds., Cyclic and Event Stratification: Springer-Verlag , New York, p. 98-112. Engelder, T., Geiser, P.A. and Alvarez, W , 1981, Role of pressure solution and dissolution in geology: Geology, v.9, p.44-45. Fletcher, R.C. and Pollard, D.D., 1981, Anti crack model for pressure solution surfaces: Geology, v. 9, p. 419-424. Garrison, R.E. and Kennedy, WJ ., 1977, Origin of solution seams and flaser structure in Upper Cretaceous chalks of southern England: Sedimentary Geology, v. 19, p. 107-137. Geiser, P.A., 1974, Cleavage in some sedimentary rocks of the Valley and Ridge province, Maryland: Geological Society of America Bulletin, v. 85, p. 1399-1412. Geiser, P.A. and Sansone , S., 1981, Joints , microfractu res and the format ion of solution cleavage in limestone: Geology, v. 9, p. 280-285.
105
Groshong, R;H., Jr., 1974, "Slip" cleavage caused by pressure solution in a buckle fold: Geology, v. 3, p. 411-413. Groshong , R.H., Jr., 1975, Strain , fracture and pressure solution in natural singlelayer folds: Geological Society of America Bullet in, v. 86, p. 1363-1376. Guzzetta, G., 1984, Kinematics of stylolite formation and physics of the pressuresolution process : Tectonophysics, v. 101, p. 383-394. Harms, J.C. and Choquette, P.W., 1965, Geologic evaluation of a gamma-ray porosity device: Society of Professional Well Log Analysts, 6th Annual Symposium, Transactions, v. 2, p. C1-C37. Hurley, N.F., 1986, Geology of the Oscar Range Devonian reef complex, Canning Basin, Western Austral ia, unpublished Ph.D. Thesis, University of Michigan, Ann Arbor, Michigan, 269 p. Koepnick, R.B., 1984, Distribution andvertical permeability of stylolites within a Lower Cretaceous carbonate reservoir, Abu Dhabi, U.A.E. , in Stylolites and Associated Phenomena - relevance to hydrocarbon reservoirs : Abu Dhabi National Reservoir Research Foundation , Special Publication, Abu Dhabi , UAE. , p. 261-278. Lloyd, A.M ., 1977, Porosity reduct ion by chemical compaction - stable isotope model (abstr.): American Association of Petroleum Geologists, Bulletin , v. 61, p.809. Logan , B.W. and Semeniuk, V.• 1976, Dynamic metamorph ism : processes and products in Devonian carbonate rocks , Canning Basin, Western Australia : Geological Society of Australia, Special Publication 16, 138 p. Manus, A.W. and Coogan, A.H., 1974, Bulk volume reduction and pressure solution derived cements: Journal of Sedimentary Petrology, v. 44, p. 466-471Marshak, S. and Engelder, T., 1985 , Development of cleavage in limestones of a fold-thrust belt in eastern New York: Journal of Structural Geology, v. 7, p.345-359. Merino, E., Ortoleva, P. and Strickholm, P., 1983. Generation of evenly-spaced pressure-solution seams during (late) diagenesis - a kinetic theory : Contributions to Mineralogy and Petrology, v. 82, p.360-370. Meyers, W.J. and Hill, B.E., 1983, Quantitative studies of compaction in Mississippian skeletal limestones, New Mexico: Journal of Sedimentary Petrology, v. 53. p.231-242.
106
Morse, J.W., 1983, The kinetics of calc ium carbonate dissolution and precipitation, in Reeder, R.J., ed., Carbonates: Mineralogy and Chem istry: Mineralogical Society of America, Reviews in Mineralogy, v. 11, p, 227-264. Mossop, G.D., 1972,Origin olthe peripheral rim, Redwater Reef, Alberta: Bulletin of Canadian Petroleum Geology, v.20, p.238-280. Nelson, R.A., 1981, Significance offracture sets associated with stylolite zones: American Associat ion of Petroleum Geologists, Bulletin , v. 65, p. 2417-2425. Park, W.C. and Schot, E.W., 1968, Stylolitization in carbonate rocks, in Muller, G. and Friedman, G.M., eds., Recent Developments in Carbonate Sedimentology in Central Europe: Springer-Verlag, Berlin , p. 66-74. Playford, P.E., 1980, Devonian " Great Barrier Reef" of Canning Basin, Western Australia: American Association of Petroleum Gecloqists, Bulletin, v. 64, p.814-840. Pratt, B.R., 1982, Limestone response to stress: pressure solution and dolomitization - Discussion and examples of compaction in carbonate sediments: Journal of Sedimentary Petrology, v. 52, p.323-328. Rheinhardt, J . and Hardie, L., 1976, Selected examples of carbonate sedimentation, Lower Paleozoic of Maryland : Maryland Geological Survey, Guidebook 5, 53 p. Robin, P.F., 1978,Pressure solution at grainto-grain contacts: Geochimica et Cosmochimica Acta, v. 42, p. 1383-1389. Roehl, P.O., 1967, Stony Mountain (Ordovi cian) and Interlake (Silurian) facies analogs of Recent low-energy marine and subaerial carbonates, Bahamas: American Assoc iation of Petroleum Geologists, Bulletin, v. 51, p. 1979-2032. Rutter, E.H. , 1976, The kinetics of rock deformation by pressure solution: Royal Society of London, Philosophical Transactions, v. 283A, p. 203-219. Rutter, E.H., 1983, Pressure solution in nature, theory and experiment: Journal of the Geological Society of London, v. 140, p. 725-740. Sorby, H.C., 1879, On the structure and origin of limestones : Quarterly Journal of the Geological Society of London , v. 35, p. 56-95. Sorby, H.C., 1908, On the application of quantitative methods to the study of the structure and history of rocks: Quarterly Journal of the Geological Society of London, v. 64, p. 171-233. Simpson, J., 1985, Stylolite-controlled layering in an homogeneous limestone: pseudo-bedding produced by burial diagenesis: Sedimentology, v. 32, p.495-505.
Diagenesis
Sprunt , E.A. and Nur, A., 1977a, Experimental study of the effects of stress on solution rate : Journal of Geophysical Research, v. 82, p, 3013-3022. Sprunt, E.A. and Nur, A., 1977b, Destruction of poros ity through pressure solution: Geophysics, v. 42, p. 726-741. Stockdale, P.B., 1922, Stylolites - the ir nature and origin: Indiana Univers ity Studies, v. 9, p. 1-97. Stockdale, P.B.,1926,The stratigraphic significanceofsolutions in rocks: Journal of Geology, v. 34, p. 1-97. Stockdale, P.B., 1943, Stylolites: primary or secondary?: Journal of Sed imentary Petrology, v. 13, p. 3-12. Thompson, J., 1862, On crystallization and liquefaction , as influenced by stresses tending to change of form in crystals : Royal Society of London, Proceedings , v. 11, p. 473-480. Trurnit, P., 1968, Analysis of pressure-solution contacts and classification of pressure-solution phenomena, in Muller, G. and Friedman, G.M . , eds ., Recent Developments in Carbonate Sedimentology in Central Europe: Springer-Verlag, Berlin, p. 75-84 . Wanless, H.R., 1979, Limestone response to stress - pressure solution and dolomitizat ion: Journal of Sed imentary Petrology, v. 49, p. 437-462. Wanless, H.R., 1982, Limestone response to stress - pressure solution and dolomitization - reply: Journal of Sedimentary Petrology, v. 52, p. 328-332. Wanless, H.R., 1983, Burial diagenesis in limestones , in Parker, A. and Sellwood, B.W., eds ., Sediment Diagenesis : J. Reidel and Company, Lancaster, p.379-417. Weyl, P.K., 1958, The solution kinetics of calcite: Journal of Geology, v.66 , p.163-176. Weyl, P.K., 1959, Pressure solution and the force of crystallization - a phenomenological theory: Journal of Geophysical Research, v. 64, p. 2001-2025. Wolfe, M.J., 1968, Lithification of a carbonate mud: Senonian Chalk in northern Ireland: Sed imentary Geology, v. 2, p.263-290. Wong, P.K., 1984, Origin of low-permeability zones in the Upper Devonian Kaybob reef, Alberta, Canada , in Stylolites and Associated Phenomena - relevance to hydrocarbon reservo irs: Abu Dhabi National Reservoir Research Foundation , Special Publication, Abu Dhabi, U.A.E., p. 247-260. Wong, P.K. and Oldershaw, A., 1981, Burial cementation in the Kaybob reef complex, Alberta, Canada: Journal of Sedimentary Petrology, v. 51, p. 507-520.
Geoscience Canada Reprint Series 4
Burial Cements - Petrography Burruss , R.C., Cercone, K.R. and Harris, P.M., 1983, Fluid inclusion petrography and tectonic burial history of the AI Ali No. 2 well: evidence for the timing and diagenesis and oil migration, northern Oman foredeep : Geology, v.11 , p.567-570. Cayeux, L., 1935, Carbonate rocks - limestones and dolomites, [1970 translation by A.V. Carozzi [: Habner Publ ishing Company, Connecticut, 506 p. Cercone, K.R. and Lohmann, K.C., 1987, Late burial diagenesisof Niagaran(MiddleSilurian) pinnacle reefs in Michigan Basin: American Associationof Petroleum Geologists, Bulletin, v. 71, p. 156-166. Choquette, P.W. and Steinen, R.P., 1985, Mississ ippian oolite and non-supratidal dolomite reservoirs in the Ste. Genevieve Formation, North Bridgeport Field, Illinois Basin, in Roehl, P.O. and Choquette, P.w., eds., Carbonate Petroleum Reservoirs: Springer-Verlag, New York, p. 207-227. Coniglio, M., 1985,Origin and diagenesis of fine-gra ined slope sed iments, Cow Head Group (Cambro-Ordovician), western Newfoundland, unpublished Ph.D. dissertation, Memorial University of Newfoundland, v. 1 and 2, 684 p. Dorobek, S.L., 1987, Petrography, geochemistry and origin of burial diagenetic facies, Siluro-Devonian Hulderberg Group (carbonate rocks), central Appalachians: American Association of Petroleum Geologists, Bulletin , v.71 , p.492-514. Druckman, Y. and Moore, C.H., 1985, Late subsurface secondary porosity in a Jurassic grainstone reservoir, Smackover Formation , Mt. Vernon field , southern Arkansas, in Roehl, P.O. and Choquette, P.w.,eds. , Carbonate Petroleum Reservo irs: Springer-Verlag, New York, p.369-384. Elliott, T.L., 1982,Carbonate facies, depositional cycles, and the development of secondary porosity duringburial diagenesis, in Christopher,J.E. andKaldi,J., eds., 4th International Willisto n Basin Symposium : Saskatchewan Geological Society, Special Publication 6, p. 131-151. Folk, R.L. and Assereto, R., 1974, Great aragon ite rays and baroque white dolomite in tepee fillings, Triassic of Lombardy, Italy : American Association of Petroleum Geologists, Abstracts with Programs, Annual Meeting, San Antonio, Texas, p, 34-39. Frank, J.R., Carpenter, A.B. and Oglesby, T.W., 1982, Cathodoluminescence and composition of calcite cement in the Taum Sauk Limestone(UpperCambrian), southeast Missouri: Journal of Sedimentary Petrology, v. 52, p. 631-638.
c c
c
c
c c c (
(
C C C (
C C C C
C C
C C C (
C C
C C C E: C C C C (
C (.>--
C
C ( (
(
c
Limestones - The Burial Diagenetic Environment
r:
c
c. (
C r
'-
, r
r .. J
(
o (
o C)
L
c(, (,
r
Franks , P.C., 1969, Nature , orig in and significance of cone-in-cone structures in the Kiowa Forma tion (Ear ly Cretaceous), north-central Kansas: Journal of Sedimentary Petrology, v. 39, p. 1438-1454. Grover, G., Jr. and Read, J.F., 1983, Paleoaquifer and deep-burial-related cements defined by regiona l cathodoluminescent patterns , Middle Ordovician carbonates, Virginia: American Assoc iation of Petroleum Geologists, Bulletin , v. 67, p. 1275-1303. Hallam, A., 1960, A sedimentary and faunal study of the Blue Lias of Dorset and Glamorgan: Royal Society of London , Philosophical Transactions , v. 243B, p. 1-44. Hallam, A., 1964, Origin of the limestoneshale rhythm in the Blue Lias of England • a composite theory: Journal of Geology, v. 72, p. 137-169. Harris , D.C., 1982, Carbonate cement stratigraphy and diagenesis of the Burlington Limestone (Mississippian), southeastern Iowa and western Illinois : M.S. Thesis , State University of New York, Stony Brook, 220 p. Heckel, P.H., 1983, Diagenetic model for carbonate rocks in midcontinent Pennsylvanian eustatic cyclothems: Journal of Sedimentary Petrology, v. 53, p. 733759. Irwin , H., 1980, Early diagenetic carbonate precipitation and pore-fluid migration in the Kimmeridge Clay of Dorset, England: Sedimentology, v. 27, p. 577-591. James, N.P. and Klappa, C.F., 1983, Petrogenesis of early Cambrian reef limestones, Labrador, Canada: Journal of Sedimentary Petrology, v. 53, p. 10511096. Loucks, R.E. and Budd, D.A., 1985, Diagenesis and reservoir potent ial of the upper Jurassic Smackover Formation of south Texas: Gulf Coast Association of Geological Societies, Transactions , v. 31, p.339-346. Marshall , J.D., 1982, Isotopic composition of displacive fibrous calcite veins , reversals in pore-water composition during burial diagenes is: Journal of Sedimentary Petrology, v. 52, p. 615-630. Matter, A., 1974, Burial diagenesis of pelitic and carbonate deep-sea sediments from the Arabian sea: Deep Sea Drilling Project Initial Reports , v. XXIII. Mattes, B.Wand Mountjoy, EW., 1980,Burial dolomitization of the Upper Devonian Miette buildup, Jasper National Park, Alberta, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontolog ists and Mineralogists , Spec ial Publication 28, p.259·297.
Meyers, w.J ., 1974, Carbonate cement stratigraphy of the Lake Valley Formation (Mississipp ian) , Sac ramento Moun tains, New Mexico : Journal of Sedimentary Petrology, v. 44, p. 837-861. Meyers, W.J., 1978, Carbonate cements their regional distr ibution and interpretation in Mississ ippian limestones of southwestern New Mexico : Sed imentology, v. 25, p. 371-400. Moore, C.H., 1985, Upper Jurassic subsurface cements - a case history, in Schneidermann , N. and Harris, P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineral ogists, Special Publication 36, p.291-308. Mruk, D.H., 1985, Cementat ion and dolomitization of the Capitan Limestone, McKittrick Canyon, west Texas: M.S. Thesis, University of Colorado/Boulder, Boulder, CO, 153 p. Oldershaw, A.E. and Scoffin, T.P.,1967, The source of ferroan and non-ferroan calcite cements in the Halkin and Wenlock limestones: Geological Journal , v. 5, part 2, p. 309·320 . Prezbindowski, D.R., 1985, Burial cementation - is it important? A case study, Stuart City trend , south-central Texas, in Schneidermann , N. and Harris, P.M., eds., Carbonate Cements: Society of Econom ic Paleontologists and Mineralog ists, Special Publication 36 , p.241-264. Saller, A.H., 1984, Diagenesis of Cenozoic limestones on Enewetak Atoll : Ph.D. Thesis, Louisiana State University, Baton Rouge, LA, 363 p. Saller, A.H., 1986, Radiaxial calcite in Lower Miocene strata , subsurface Enewetak Atoll: Journal of Sedimentary Petrology, v. 56, p. 743-762. Talbot, M.R., 1971, Calcite cements in the Corallian Beds (Upper Oxfordian) of southern England : Journal of Sedimentary Petrology, v. 41, p. 261-273. Walkden, G.M. and Berry, J.R., 1984, Syntaxial overgrow1hs in muddy crinoidal limestones; cathodoluminescence sheds new light on an old problem : Sedimentology, v. 31, p. 251-268. Walls, R.A. and Burrowes, G., 1985, The role of cementation in the diagenetic history of Devonian reefs, in Schneidermann , N.and Harris , P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 185-220. Woodland, B.G., 1964, The nature and origin of "cone-in-cone" structure: Fieldiana, Geology (Field Museum of Natural History, Chicago), v. 13, p. 185-305. Yurewicz, D.A. and Dravis, J .J., 1984, Improved recogn ition of sedimentary fabrics using fluorescence microscopyimplications for interpreting carbonate
107
facies and diagenetic history : American Association of Petroleum Geologists , Bulletin, v. 68, p. 542. Burial Cements Hydrology
Water Chemistry and
Back, W. and Hanshaw, B.B., 1970, Comparison of chemical hydrogeology of the carbonate peninsulas of Florida and Yucatan peninsulas : Journal of Hydrology, v. 10, p. 330-368. Back, W. and Hanshaw, B.B., 1971, Ratesof physical and chemical processes in a carbonate aquifer : American Chemical Society, Advances in Chemistry, v. 106, p.77-93. Baker, P.A., Gieskes ,J.M.and Elderfeld, H., 1982, Diagenesis of carbonates in deepsea sediments - evidence from Sr/Ca ratios and interst iti al dissolved Sr 2 .. data: Journal of Sedimentary Petrology, v. 52, p. 71-82. Berner, R.A., 1975, Diagenetic models of dissolved species in the interstiti al waters of compacting sediments : American Journal of Science, v. 275, p. 88-96. Bonham, L.C., 1980, Migration of hydrocarbons in compacting basins, in Roberts, W.H., III and Cordell , R.J., eds., Problemsof PetroleumMigration: American Assoc iat ion of Petroleum Geologists, Studies in Geology,v. 10, p. 69-88. Bredehoeft , J.D. and Papadopulous, I.S., 1965, Rates of vertical groundwater movement estimated from the earth 's thermal profile : Water Resources Research, v. 1, p. 325-328. Carpenter, A.B. , 1978, Origin and chemical evolut ion of brines i n sed imentary basins: Oklahoma Geological Survey, Circular 79, p. 60-77. Champ, D.R., Gulens,J . and Jackson, R.E., 1979, Oxidat ion-reduction sequences in ground-water flow systems: Canadian Journal of Earth Sc iences , v.16, p. 12-23. Clayton, R.N. , Friedman, I., Graf, D.L., Mayeda, T.K., Meents, WF. and Shimp, N.F.,1966,Theoriginofsalineformation waters. I. Isotopic composition: Journal of Geophysical Research, v. 71, p. 38693882. Collins , A.G., 1975, The Geochemistry of Oil Field Waters: New York, Elsevier Scientific Publishing Company, 496 p. Cook, P.J., 1974, Geochemistry and diagenesis of interst itial fluids and associated oozes, Deep Sea Drilling Project, Leg27, Site 262, Timor Trough,in Veevers, J.J., Heirtzler, J.R., et a/., Initial Reports of the Deep Sea Drilling Project, v. XXVII, p.463·480 . Drever,J.I., 1982,The Geochemistry of Natural Waters: Prentice-Hall, Englewood Cliffs, J, 388 p.
c 108
Dunham, R.J., 1969, Early vadose silt in Townsend mound (reef), New Mexico, in Friedman , G .M ., ed., Depositional Env ironments in Carbonate Rocks : a symposium: Society of Economic Paleonto logists and Mineralogists, Special Publication 14, p. 139-181. Galloway, W.E. , 1985, Hydrogeologic regimes of sandstone diagenesis, in McDonald, D.A. and Surdam, RC., eds., Clastic Diagenesis: American Assoc iation of Petroleum Geologists, Memoir 37, p. 3-13. Garven, G., 1985, The role of regional fluid flow in the genesis of the Pine Point depos it, Western Canada sedimentary basin : Econom ic Geology, v.80, p.307-324. Garven, G. and Freeze, RA. , 1984a, Theoretical analysis of the role of groundwater flow in the genesis of stratabound ore deposits. 1. Mathematical and numerical models : American Journal of Science, v. 284, p. 1085-1124. Garven , G. and Freeze, RA. , 1984b, Theoretical analysis of the role of groundwater flow in the genesis of stratabound ore deposits. 2. Quantitative results: Amer ican Journal of Science, v. 284 , p. 1125-1174. Graf, D.L. and Anderson, D.E., 1981, Geochem ical inputs for hydrologic models of deep-lying sedimentary units - loss of mineral hydrat ion water: Journal of Hydrology, v. 54, p. 297-314. Hanor, J.S., 1979, Sedimentary genesis of hydrothermal fluids, in Barnes, H.L., ed., Geochemistry of Hydrothermal Ore Deposits , Second Edition: Wiley, New York, p. 137-168. Hanshaw, B.B., Busby, J. and Lee, R., 1978, Geochemical aspects of the Madison aquifer system, in Rehrig, D., ed., The Economic Geology of the Williston Basin : Montana, North Dakota, Saskatchewan, Manitoba: Montana Geological Society, 24th Annual Conference, 1978 Williston Basin Symposium, p.385-389. Helgeson, H.C., 1969, Thermodynamics of hydrothermal systems at elevated temperatures and pressures: American Journal of Science , v. 269, p. 729-804 . Helgeson , H.C.,1979 , Masstransferamong mine rals and hydrothermal solutions, in Barnes, H.L., ed., Geochemistry of Hydrothermal Ore Deposits, Second Edition : Wiley, New York, p.568-610. Hitchon , B., 1969a, Fluid flow in the western Canada sedimentary basin - 1. Effect of topography: Water Resources Research, v. 5, p. 166-185. Hitchon, B., 1969b, Fluid flow in the western Canada sedimentary basin - 2. Effect of geology: Water Resources Research, v. 5, p. 460-469.
Diagenesis
Hitchon, B., Bill ings, G.K. and Klovan, J.E., 1971, Geochemistry and origin of formation waters in the western Canada sedimentary basin . 3. Factors controlling chemical compos ition : Geochimica et Cosmochimica Acta , v. 35, p. 567-598. Hitchon, B. and Friedman , I., 1975, Geochemistry and origin offormation waters in the western Canada sedimentary basin. 1. Stable isotopes of hydrogen and oxygen: Geochim ica et Cosmochimica Acta, v. 33, p. 1321-1349. Holland, H.D. and Malinin, S.D., 1979, The solubility and occurrence of non-ore deposits, in Barnes, H.L. , ed., Geochemistry of Hydrothermal Ore Deposits, Second Edition: Wiley, New York, p.461-508. Kharaka, Y.K., Hull, R.W. and Carothers, W.W., 1985, Water-rock interactions in sedimentary basins, in Gautier, D. L., Kharaka, Y.K. and Surdam, RC., eds., Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralogists, Notes for Short Course No. 17, p.79-176. Klosterman , M.J., 1981, Application of fluid inclusion techniques to burial diagenesis in carbonate rock sequences: M.S. Thesis, Louisiana State University, Baton Rouge, LA , 102 p. Kohout, EA ., 1965, A hypothesis concerning cyclic flow of salt water related to geothermal heating in the Floridian aquifer: New YorkAcademy of Sciences, Transactions, Series II, v. 28, p. 249-271. Kohout, EA ., 1967, Groundwater flow and the geothermal regime of the Floridian plateau : Gulf Coast Association of Geological Societies, Transactions, v. 17, p.339-354. Land, L.S. and Prezbindowski, D.R, 1981, The origin and evolution of saline forma tion water, Lower Cretaceous carbonates, south-central Texas: Journal of Hydrology, v. 54, p. 51-74. O'Hearn, T.C., 1985, A fluid inclusion study of diagenetic mineral phases, Upper Jurassic Smackover Formation, south west Arkansas and northeast Texas: Applied Carbonate Research Program, Lous iana State University, Technical Series, Contribution No. 24, 189 p. Plummer, L.N . and Back, W., 1980, The mass balance approach - application to interpreting the chemical evolution of hydrologic systems : American Journal of Science, v. 280, p. 130-142. Pray,L.C., 1966, Informal comments on calcium carbonate cementation: Society of Economic Paleontologists and Mineralogists, Technical session on lithification and Diagenesis, St. Louis meetings , April 1966 (unpublished).
Geoscience Canada Reprint Series 4
Roehl, P.O., 1981, Dilation brecciation - a proposed mechanism of fracturing, petroleum expulsion and dolomitization in the Monterey Formation, California, in Garrison RE ., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. and Ingle, J.C., eds., The Monterey Formation and Related Siliceous Rocks of California: Society of Econom ic Paleontologists and Mineralogists, Pacific Section, Symposium , p.285-315. Rosenberg, P.E. and Holland, H.D., 1964, Calcite-dolomite-magnesite stabil ity relations in solutions at elevated temperatures: Science, v. 145, p. 700-701. Sayles, E.L. and Manheim, F.G., 1975,Interstitial solutions and diagenesis in buried marine sediments : results from the Deep Sea Drilling Project: Geochim ica et Cosmochimica Acta , v. 39, p. 103-127. Secor, D.T., Jr., 1965, Role of fluid pressure in jointing: American Journal of Science, v. 263, p. 633-646. Sharp , W.E. and Kennedy, G.C., 1965, The system CaO-C0 2-H 20 in the two-phase region calc ite + aqueous solution: Journal of Geology, v. 73, p. 391-403. Sippel , R.F. and Glover, E.D., 1964, The solution alteration of carbonate rocks, the effects of temperature and pressure : Geochimica et Cosmochimica Acta , v. 28, p. 1401-1417. White, D.E., 1975, Saline waters of sedimentary rocks, in Young, A. and Galley, G .E . , eds ., Fluids in Subsurface Environments: American Association of Petroleum Geologists, Memoir 4, p.342-366. Wood, J.R and Hewett, T.A., 1984, Reservoir diagenesis and convective flu id flow, in McDonald, D.A. and Surdam, RC., eds., Clastic Diagenesis : Amer ican Assoc iation of Petroleum Geologists , Memoir 37, p. 99-110. Burial Cements - Trace Elements and Cathodoluminescence Carpenter, A.B. and Oglesby, TW., 1976, A model for the formation of luminescently zoned calcite cements and its implications: Geological Society of America, Abstracts with Programs, v.8, p.469-470. Fairchild, I.J., 1983, Chemical controls of cathodoluminescence of natural dolomites and calcites - new data and review: Sedimentology, v. 30, p. 579-583. Frank, J.R, Carpenter, A.B. and Oglesby, T.W., 1982, Cathodolum inescence and composition of calcite cement in the Taum Sauk limestone (Upper Cambrian), Southeast Missouri: Journal of Sedimentary Petrology, v. 52, p. 631-638. Machel , H.G., 1985, Cathodoluminescence in calcite and dolomite and its chemical interpretation: Geoscience Canada, v. 12, p. 139-148.
c c
c
c
c C
c
( C
C C C
C
r
'-
.[
C
C
c
C C C C C
C C C C C C C
C (
C
-:t
C C
G
..
C C
c
(~
c c ( (
c
(
c ( (~
o
o
o o (
o
o o o
o o o o o )
l l
o C (~
C' (
:'r .
'-.
c f"
\.
Limestones - The Burial Diagenetic Environment
Oglesby, T.W., 1976, A model for the distribution of manganese, iron and magnesium in authigen ic calcite and dolomite cements in the upper Smackover Formation in eastern Mississippi : M.S. Thesis, University of Missouri/Columbia, Columbia, MO, 122 p. Pierson , B.J., 1981, The control of cath odoluminescence in dolomite by iron and manganese : Sedimentology, v. 28, p.601-610 . ten Have, T. and Heijen, w., 1985, Cathodoluminescence activation and zonation in carbonate rocks: an exper imental approach: Geologie en Mijnbouw, v. 64, p.297-310. Burial Cements - Stable Isotopes Brinton, L., 1986, Deposition and diagenesis of phylloid algal banks, Ismay Field and San Juan Aiver Canyon, Paradox Basin, Utah: M.S. Thesis, Colorado School of Mines, Golden , CO, 316 p. Burke , W.H., Denison, A.E., Hetherington, A.B., Koepnick, A.B ., Nelson, H.F. and Otto, J.B., 1982, Variation of seawater 87Sr/ 86Sr throughout Phanerozoic time: Geology, v. 10, p. 516-519. Choquette, P.W., 1971, Late ferroan dolomite cement, Mississippian carbonates, Illinois Basin, U.S.A., in Bricker, O.P., ed., Carbonate Cements: The Johns Hopkins Press, Baltimore, MD, p.339-346. Choquette, P.w. and Steinen, A.P., 1980, Mississippian non-supratidal dolomite, Ste. Genevieve Limestone , Illinois Basin : evidence for mixed-water dolomitization, in Zenger, D.H., Dunham , J.B. and Ethington , A.L., eds., Concepts and Models of Dolomitization - a symposium: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 163-196. Czerniakowski, L.A., Lohmann , K.C. and Wilson, J.L., 1984, Closed-system marine burial diagenesis - isotopic data from the Austin chalk and its components: Sedimentology, v. 31, p. 863-877. Dickson, J.A .D. and Coleman, M.L., 1980, Changes in carbon and oxygen isotope composition during limestone diagenesis: Sedimentology, v. 27, p. 107-118. Fritz, P. and Jackson, S.A., 1972,Geochemical and isotopic characteristics of Middle Devonian dolomites from Pine Point, northern Canada: 24th International Geological Congress, Montreal, Proceedings, Section 6, p. 230-242. Hudson, J.D., 1975, Carbon isotopes and limestone cement: Geology, v.3 , p.19-22. Hudson, J.D., 1977, Stable isotopes and limestone lithification: Journal of the Geological Society of London, v. 133, p.637-660.
Irwin, H., Curtis, C.D . and Coleman , M., 1977, Isotopic evidence for source of diagenetic carbonates formed during burial of organic-rich sed iments: Nature, v. 269, p. 209-213. Land, L.S., 1980, The isotopic and traceelement geochemistry of dolomite - the state of the art, in Zenger, D.H., Dunham, J.B. and Eth ington , A.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication No. 28, p. 87-110 . Land , L.S., 1985, Environments of limestone and dolomite diagenesis - some geochemical considerations: Notes for a short course on carbonate diagenes is, Colorado School of Mines, Golden, CO, April 8-10, 1985. Lohmann, K.C., 1983, Unravelling the diagenetic history of carbonate reservo irs integration of petrographic and geochemical techniques, in Wilson, J.L., et et., Notes for a short course, American Association of Petroleum Geologists Annual Meeting , Dallas, Texas, April 1983. Matter, A ., Douglas, A .G. and PerchNielsen, K., 1975, Fossil preservation, geochemistry, and diagenesis of pelagic carbonates from Shatsky Aise, northwest Pacific, in Larson, A.L., Moberly, A., et aI., Initial Aeports of the Deep Sea Drilling Project, v. XXXII, p. 891-921. McKenzie , J.A., Bernoulli , D. and Garr ison, A.G., 1978, Studies on lithification and diagenesis - lithification of pelagichemipelagic sediments at DSDP Site 372, oxygen-isotope alteration with diagenesis, in HsO, K.J., Montadert, L., et al., Initial Aeports of the Deep Sea Drilling Project, v. XLII, p. 473-478. Meyers, W.J.and Lohmann, K.C., 1985, Isotope geochemistry of regionally extensive calcite cement zones and marine components in Miss issippian limestones, New Mex ico , in Schneidermann , N.andHarris, P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 223-240. Moldovanyi, E.P. and Lohmann, K.C., 1984, Isotopic and petrographic record of phreatic diagenesis - LowerCretaceous Sligo and Cupido Formations: Journal of Sedimentary Petrology, v. 54, p. 972-985. Murata , K.J., Friedman, I. and Madsen , B.M., 1969, Isotopic composition of diagenetic carbonates in marine Miocene formations of California and Oregon : United States Geological Survey, Professional Paper 614-B. O'Neil, J.A., Clayton , A.N. and Mayeda, T.K., 1969,Oxygen isotope fractionation in divalent metal carbonates: Journal of Chemistry and Physics, v. 51, p. 55475558.
109
Prezbindowski , D.A ., 1985, Burialcementation - is it important? A case study, Stuart City trend , south central Texas,in Schneidermann, N. and Harris, P.M., eds., Carbonate Cements : Society of Economic Paleontologists and Mineralogists, Special Publication 36, p.241-264. Sofer,Z., 1978, Isotope compositionof hydration water in gypsum: GeochimicaetCosmochimica Acta, v. 42, p. 1141-1149. Stueber, A .M. and Pushkar, P., 1983, Application of stront ium isotopes to origin of Smackover brines and diagenetic phases, southern Arkansas (abstr.): American Associat ion of Petroleum Geologists, Bulletin , v. 67, p. 553-554. Tan, F.C. and Hudson , J.D., 1974, Isotopic studies of the paleoecology and diagenesis of the Great Estuarine Series (Jurassic) of Scotland : Scottish Journal of Geology, v. 10, p. 91-108. Veizer,J. and Compston, W., 1974, 87Sr/ 86Sr composition of seawater during the Phanerozoic: Geochimica et Cosmochimica Acta, v. 38, p. 1461-1484. Veizer,J. and Hoefs, J., 1976, The nature of '80/'60 and '3C/'2C secular trends in sedimentary carbonate rocks: Geochimica et Cosmochimica Acta, v. 49, p. 1387-1395. Walls, A.A., Mountjoy, E.W. and Fritz, P., 1979, Isotopic composit ion and diagenetic history of carbonate cements in Devonian Golden Spike reef, Alberta, Canada: Geological Society of America Bulletin, v. 90, p. 963-98. Organ ic Geochemistry and OrganicInorganic Interactions Barnes, M.A., Barnes, W.C. and Bustin , A.M., 1984, Diagenes is and chemistry and evolution of organic matter: Geoscience Canada, v. 11, p. 103-114. Berner, A.A., 1975, Diagenetic models of . dissolved species i n the interstitial waters of compacting sediments: American Journal of Science , v. 275, p. 88·96. Bjorlykke , K., 1984, Formation of secondary poros ity: how important is it? , in McDonald, D.A. and Surdam, A.C., eds., Clastic Diagenesis: American Association of Petroleum Geologists, Memoir37, p. 127-151. Curtis, C.D., 1977, Sedimentary geochemistry - environments and processes dom inated by involvement of an aqueous phase : Aoyal Society of London , Philosophical Transactions , v. 286A, p. 353-372. Curtis , C.D., 1978, Possible links between sandstone diagenesis and depthrelated geochemical react ions occurring in enclosing mudstones: Journal of the Geological Society of London, v. 135, p. 107-117.
110
Diagenesis
Geoscience Canada Reprint Series 4
c c c
Deroo , G . , Power, T.G ., Tissot, B., McCrossan , R.G. and Hacquebard, P.A., 1977, The origin and migration of petroleum in the western Canadian Sedimentary basin - a geochemical and thermal maturation study : Geological Survey of Canada, Bulletin 262. Dow, w., 1977, Kerogen studies and geological interpretations: Journal of Geochemical Exploration, v. 7, p. 79-99 . Espitalie, J., Laporte, J .L., Madec , M., Marquis , E , Leplat , P., Paulet , J. and Boutefen, A., 1977, Methode rapide de characterisation des roches-meres, de leur potentiel petrolier et de leur degre d'evolution: Revue de l'lnstitut Francais du Petrote, v. 32, p. 23-42 . Foscolos, A.E. and Powell, T.G., 1980, Mineralogical and geochemical transformations of clays during catagenesis and the ir relation to oil generation, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Society of Petroleum Geologists, Memoir 6, p. 153-172. Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A ., Heath, G.R ., Cullen, D., Dauphin, P., Hammond, D., Hartmann, B. and Maynard, V.,1979, Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis: Geochimica et Cosmochimica Acta, v. 43, p. 1075-1090. Hayes, J.B., 1979, Sandstone diagenesisthe hole truth, in Scholle, P.A. and Schluger, P.R., eds ., Aspects of Diagen esis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 127-140. Hood, A. and Castano, J.R., 1974,Organ ic metamorphism - its relationship to petroleum generation and application to studies of authigenic minerals: United Nations ESCAP, CCOP, Technical Bulletin 8, p. 87-118. Hue, A.Y. and Durand, B.M., 1977, Occurrence and significance of humic acids in ancient sediments: Fuel, v. 56, p. 73-80. Irwin, H., Curtis, C.D. and Coleman, M., 1977, Isotopic evidence for source of diagenetic carbonates formed during burial of organic-r ich sediments : Nature, v. 269, p. 209-213. Kharaka, Y.K., Hull , R.w. and Carothers, W.W., 1985, Water-rock interactions in sedimentary basins, in Gaut ier, D.L., Kharaka, Y.K. and Surdam, R.C., eds., Relationship of Organic Matter and Minerai Diagenesis: Society of Economic Paleontologists and Mineralogists , Notes for Short Course No. 17, p. 79-176. Lundegard, P.A., 1985, Carbon dioxide and organic acids: origin and role in burial diagenesis, Texas Gulf Coast Tertiary: Ph.D. Thesis, University of Texas/Austin , Austin, TX, 145 p.
Mackenzie, A.S., Patience, R.L. and Maxwell , J.R ., 1981, Molecular changes and the maturation of sedimentary organ ic matter, in Atkinson, G. and Zuckerman, J., eds., Origin and Chemistry of Petroleum: Pergamon Press, Oxford , p. 1-31, McBr ide, E.E , 1977, Secondary porosity importance in sandstone reservoirs in Texas: Gulf Coast Association of Geological Societies, Transactions, v. 27, p. 121-122. Mitterer, R.M ., 1968, Amino acid composi tion of organic matrix in calcareous oolites: Science, v. 162, p. 1498-1499. Powell, T.G., Foscolos, A.E ., Cunther, P.R. and Snowdon, L.R., 1978, Diagenesis of organic matter and fine clay minerals - a comparative study: Geochimica et Cosmochim ica Acta, v. 42, p. 1181-1198. Schmidt, V. and McDonald, D.A., 1979, The role of secondary porosity in the course of sandstone diagenesis, in Scholle, P.A. and Schluger, P.R. , eds., AspectsofDiagenesis: Society of Economic Paleontologists and Mineralogists, Spec ial Publication 26, p. 175-207. Snowden , L.R . and Powell , T.G., 1982, Immature oil and condensate - modification of hydrocarbon generation model for terrestrial organic matter: American Association of Petroleum Geologists, Bulletin , v. 66 , p. 775-788. Suess, E., 1970, Interaction of organic compounds with calc ium carbonate. I. Association phenomena and geochemical implications: Geoch imica et Cosmochimica Acta, v. 34, p. 157-168. Suess, E. and FOtterer, D., 1972, Aragonitic ooids - experimental precipitation from seawater in the presence of humic acid : Sedimentology, v. 19, p. 129-139. Surdam , R.C., Boese, S.w. and Crossey, L.J., 1984, The chem istry of secondary porosity, in McDonald, D.A. and Surdam , R.C ., eds ., Clastic Diagenesis: American Association of Petroleum Geologists, Memoir 37, p. 127-149. Tissot , B.P., Bard, J.F. and Espitalie, J., 1980, Principal factors controlling the timing of petroleum generation, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence : Canadian Society of Petroleum Geologists, Memoir 6, p. 143-152.
Burial Diagenesis of Chalks and Other Fine-Grained Oceanic Carbonates Beall , A.O. and Fischer, A.G ., 1969, Sedimentology, in Ewing , M., et a/., eds. , Initial Reports of the Deep Sea Drilling Project, v. I, p. 521-593. Feazel, C.T., Keany, J. and Peterson , R.M., 1985, Cretaceous and Tertiary chalk of the Ekofisk Field area , central North Sea, in Roehl, P.O. and Choquette, PW., eds. , Carbonate Petroleum Reservoirs : Springer-Verlag, New York, p. 495-511.
Hancock, J.M., 1976, The petrology of the Chalk : Geological Soc iety of London, Proceedings, v. 86, p. 499-535. Mapstone , N.B., 1975,Diagenet ic history of a North Sea chalk: Sedimentology, v. 22, p.601·613. Matter, A ., Douglas, R.G . and PerchNielsen, K., 1975, Fossil preservation, geOChemistry, and diagenesis of pelagic carbonates from Shatsky Rise, north west Pacific, in Larson , R.L., Moberly, R., et ei., Initial Reports of the Deep Sea Drilling Project, v. XXXII, p. 891-921, McKenzie , J.A., Bernoulli, D. and Garrison, R.E., 1978, Studies on lithification and diagenesis - lithification of pelagichemipelagic sediments at DSDP Site 372, oxygen- isotope alteration with diagenesis, in HsG, K.J., Montadert, L., et aI., Initial Reports of the Deep Sea Drilling Project, v. XLII, p. 473-478. Mimran, Y., 1977, Chalk deformation and large-scale migration of calcium carbonate: Sedimentology, v. 24, p. 333-360. Neugebauer, J., 1973, The diagenetic problem of chalk - the role of pressure solution and pore fluid: Neues Jahrbuch fUr Geologie und Palaontoloqie, Abhandlungen, v. 143, p. 223-245. Neugebauer, J., 1974, Some aspects of cementation in chalk, in HsG, K.J. and Jenkyns, H.C., eds ., Pelagic Sediments on Land and Under the Sea: International Association of Sedimentologists, Special Publication 1, p. 149-176. Scholle , P.A., 1974, Diagenesis of Upper Cretaceous chalks from England, Northern Ireland and the North Sea, in HsO, K.J. and Jenkyns, H.C., eds ., Pelagic Sediments on Land and Under the Sea: International Assoc iation of Sedimentologists, Special Publication 1, p.177·210. Scholle, P.A., 1977, Chalk diagenesis and its relation to petroleum exploration - oil from chalks, a modern miracle?: American Association of Petroleum Geologists, Bulletin, v. 61, p. 982-1009. Van den Bark, E. and Thomas, O.D., 1980, Ekofisk : first of the giant oil fields in western Europe, in Halbouty, M.T., ed., Giant Oil and Gas Fields of the Decade 1968- 1978: Amer ican Assoc iation of Petroleum Geologists, Memoir 30, p.195-224. Wolfe, M.J., 1968, Lith ificat ion of a carbonate mud - Devonian chalk in Northern Ireland: Sed imentary Geology, v.2, p.263-290.
Porosity in Burial Diagenesis Choquette, P.w. and Pray, L.C., 1970, Geologic nomenclature and classification of porosity in sedimentary carbonates: American Association of Petroleum Geologists, Bulletin , v. 54, p. 207·250.
c c
c-
C r: (~
C C
r
C
C C ·C
C
C
c
C
c c
c c
,C C C C C C C C (
c
c:
c c
c ( c
c c (' c
Limestones - The Burial Diagenetic Environment
(
c(. cr c
o c o o o o ( o o o ('
o c () o (, o
o
l
o
o
c o (
/-
,
r: , '-..' /-
'-
c
Donath, EE ., Carozzi, A.V., Fruth, L.S., Jr. and Rich, DW., 1980, Oomold ic porosity experimentally developed in Mississippian oolitic limestones : Journal of Sedimentary Petrology, v. 50, p. 1249-1260. Dravis, J.J., 1987, Reservo ir poros ity in Upper Jurassic Haynesville limestones in east Texas: evidence for deep-burial generation: Society of Economic Paleontologists and Mineralogists, Abstracts, v. IV, 1987 Midyear Meeting , p. 22. Elliott , T.L., 1982, Carbonate fac ies, depositional cycles and the development of secondary porosity during burial diagenesis, in Christopher, J.E and Kaldi,J., eds. , 4th International Williston Basin Symposium: Saskatchewan Geological Society, Special Publication 6, p. 131-151. Enos, P.and Sawatsky, L.H., 1981, Pore networks in Holocene carbonate sediments: Journal of Sedimentary Petrology, v. 51, p.961-985. Feazel, C.T. and Schatzinger, A.A. , 1985, Prevention of carbonate cementation in petroleum reservoirs , in Schneidermann , N. and Harris , P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 97-106. Giles, M.A. and Marshall, J.D., 1986, Constraints on the development of secondary porosity in the subsurface: re-evaluation of processes : Marine and Petroleum Geology, v. 3, no. 3, p. 243-255. Grabowski, G.J., Jr., Wendte, J.C., Williams, S.C.,Anderson,J.H ., Kick, A.M., Harrison, W.J., McFarlan, E, Jr., Reeckmann, SA and Kaufman,J., 1987, Grain interpenetration, pressure solution and cementation of Upper Smackover lime grainstones, Gulf of Mexico: Society of Economic Paleontologists and Mineralogists, Abstracts, v.I\I, 1987 Midyear Meeting, p. 30-31. Halley, A.B. and Harris, P.M., 1979, Freshwater cementation of a 1000-year-old oolite: Journal of Sedimentary Geology, v. 49, p. 969-988. Halley, A.B . and Schmoker, J.M., 1982, High poros ity Cenozoic carbonate rocks of South Florida : progressive loss of porosity with depth: American Association of Petroleum Geologists, Bulletin , v. 67, p.191-200. Harrison, A.S., 1975,Porosity in Pleistocene grainstones from Barbados: some preliminary observations: Bulletin of Canadian Petroleum Geology, v. 23, p. 383-392. Lockridge, J.P. and Scholle , P.A., 1978, Niobrara gas in eastern Colorado and northwestern Kansas: Rocky Mountain Association of Geologists 1978 Symposium , p, 35-49. Moore, C .H., 1985, Upper Jurassic subsurface cements: a case history, in Schneidermann, N. and Harris, P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 291-308.
Moore, C.H. and Druckman, Y., 1981, Burial diagenesis and poros ity evolution, upper Jurassic Smackover, Arkansas and Louisiana: American Assoc iat ion of Petroleum Geologists, Bulletin , v. 65, p.597-628. Moshier, S.O., 1987, Microporosity in micritic limestones: a review: Society of Econom ic Paleontologists and Mineralogists, Abstracts, v. IV, 1987Midyear Meeting, p. 58. Schmoker, JW., 1984, Empirical relation between carbonate porosity and thermal maturity - an approach to regional porosity predict ion: Amer ican Association of Petroleum Geologists, Bulletin, v. 68, p. 1697-1703. Schmoker, J.w. and Halley, R.B., 1982, Carbonate porosity versus depth - a predictable relation for south Florida: American Association of Petroleum Geologists, Bulletin , v. 66, p. 2561-2570. Scholle , PA and Halley, A.B. , 1985, Burial diagnesis: out of sight , out of mind! , in Schneidermann, N. and Harris, P.M., eds., Carbonate Cements: Society of Economic Paleontologists and Mineralogists, Special Publication 36, p. 309-334 . Yanguas, J.E. and Dravis, J.J., 1985, Blue fluorescent dye technique for recognition of micro porosity in sedimentary rocks: Journal of Sedimentary Petrology, v. 55, p.600-602.
Minerai Transformations, and Other References Beach, D.K., 1982, Depositional and diagenetic history of Pliocene-Ple istocene carbonates of northwestern Great Bahama Bank: evolution of a carbonate platform : Ph.D. Thesis, University of Miami, Miami, FL, v. 1, 447 p. Boles, J.R. and Franks , G., 1979, Clay diagenesis in Wilcox sandstone of southeast Texas: implications of smectite diagenesis on sandstone cementation: Journal of Sedimentary Petrology, v.49, p.55-70. Burst, J.E, 1969, Diagenesis of Gulf Coast clay sediments and its possible relation to petroleum migration: Amer ican Assoctation of Petroleum Geologists, Bulletin, v. 53, p. 73-93. Enos, P. and Perkins, R.D., 1977, Quaternary Sedimentation in South Florida: Geological Society of America, Memoir 147, 198 p. Gregg, J.M. and Sibley, D.E, 1984, Epigenetic dolomit ization and the origin of xenotopic dolomite texture : Journal of Sedimentary Petrology, v. 54, p. 908-931. Hower, J., Eslinger, EV., Hower, M.E. and Perry, EA. , 1976, Mechan ism of burial metamorphism of argillaceous sed iments - 1. Mineralogical and chemical evidence: Geological Society of America Bulletin, v. 87, p. 725-737.
111
Isaacs, C.M., 1981, Porosity distribution during diagenesis of the Monterey Formation , Santa Barbara coastal area, California, in Garr ison, R.E., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. and Ingle, J.C., The Monterey Formation and Related Siliceous Rocks of California: Society of Economic Paleontologists and - Mineralogists, Pacific Section, Symposium, p. 257-272. James, N.P. and Choquette, P.W., 1983, Limestones - The Sea-floor Diagenetic Environment: Geoscience Canada, v. 10, p.162-179. James, N.P. and Choquette, PW., 1984, Limestones - The Meteoric Diagenetic Environment: Geoscience Canada, v. 11, p. 161-194. Kendall, A.C., 1984, Evaporites, in Walker, A.G., ed., Facies Models, Second Edition: Geological Association of Canada, Geoscience Canada Reprint Series 1, p.259-298. Lippman, E, 1973, Sedimentary Carbonate Minerals: Springer-Verlag, Berlin, 228 p. Murray, A.C., 1964, Origin and diagenesis of gypsum and anhydrite : Journal of Sedimentary Petrology, v. 34, p. 512-523. Perry, E.A. and Hower, J., 1970, Burial diagenesis in Gulf Coast pelitic sediments: Clays and Clay Minerals, v. 18, p. 165-177. Radke, R.M. and Mathis, R.L., 1980, On the formation of saddle dolomite: Journal of Sedimentary Petrology, v. 50, p. 1149-1168. Sando, W.J., 1974, Ancient solution phenomena in the Madison Limestone (Mississippian) of north-central Wyoming : United States Geological Survey, Journal of Research, v. 2, no. 2, p. 133-141. Sando, w.J ., 1985, Revised Mississippian time scale, western interior region, conterminous USA: United States Geolo!]ical Survey, Bulletin 160SA, 18 p. Schlee, J.S ., 1974, ed., Interregional Unconformities and Hydrocarbon Accumulation: American Association of Petroleum Geologists, Memo ir 36, 184 p. Steinen, R.P., 1974, Phreatic and vadose diagenetic modification of Pleistocene limestone: petrographic observations from subsurface of Barbados, West Indies: American Association of Petroleum Geo logists , Bulletin, v.58, p. 1008·1024. Steinen, R.P. and Matthews, R.K., 1973, Phreatic versus vadose diagenesis : stratigraphy and mineralogy of a cored borehole on Barbados, w.1.: Journal of Sed imentary Petrology, v.43 , p. 1012-1020.
Originally published in Geoscience Canada v. 14 Number 1 (March 1987) Revised 1987
c
112
Diagenesis
Geoscience Canada Reprint Series 4
c
C C
C
c-
C
C ('
C C {
C C C C
C
C ~
c
c c c c c C c C Photographs of diagenetic dolomite of the Manetoe Facies developed in the Middle Devonian Nahanni Formation in the Kotaneelee Gas Field of Southeastern Yukon Territory. The box outline on the core photograph on the left indicates the area covered by the photomicrograph on the right (image reversed). Coarsely crystalline "saddle" dolomite with curved crystal faces infills pore spaces (P) in a dolomitized pelletal packstone laminite. Stylolites (S) postdate dolomitization. Photograph courtesy of D. W. Morrow (Geological Survey of Canada).
C (.
C C
C
c
e
c c
c
c.
c
c c
C l
Dolomites - Chemistry and Precipitation
113
~
(, (
G (
(
c (
C Cl (
o o ( (
(
c
o o o C)
o C C
r:
'-....
c
Dolomite - Part 1: The Chemistry of Dolomitization and Dolomite Precipitation D.W. Morrow Institute of Sedimentary and Petroleum Geology 3303-33rd Street N. W. Calgary, Alberta T2L 2A7 INTRODUCTION Dolostones form a large part of the Proterozoic and Phanerozoic successions in many areas of Canada, but are particularly abundant in lower Paleozoic strata (Figure 1). These strata are predominantly carbonate unlike the overlying Mesozoic and underly ing Proterozoic strata that are dominated by terr igenous sediments. This reflects the fact that during early Paleozoic time Canada was part of the Laurentian landmass that was situated in the tropical equatorial belt (Ziegler et el., 1979). The economic importance of dolostones parallels their abundance in the stratigraphic record (Figure 1). Dolostones form important reserveif:&.fGr...oil (e.g., the Leduc Formation or 03) and gas (e.g., Manetoe Facies, Figure 2) and are the main host rock in Canada foriea(j': ZinC mineral deposits {e.g. , the Presqu'i1e Facies of the Pine Point deposit). The importance of dolostones in Canada is a true reflection ofthe world-wide importance of dolostones as reservoirs for hydrocarbons and as hosts for leadzinc deposits (e.g., see Davies, 1979 and Macqueen , 1979). Given the major, albeit indirect, role that the mineral dolomite has played in contributing to the well being and growth of the Canadian and world economies, one might assume that the process of dolomitization and the precipitation of dolomite are well understood. Unfortunately, although there is a consid-
erable amount of experimental and theoretical data concerning the chemistry of the dolom itization process and the precipitation of the mineral dolomite from solution, no consensus exists regarding the general chemical conditions required for dolomitization. The absence of a widely. acc.ep.te_d~th~.or.y concernin g the chemistrY.Ql(loJQJIlit.im:. tion is due Rr imatil¥-toJhe..dif.ficuLt.y..i~ precipitatin9 dolomite from aePJ:QB.!iate • solutions at tem w~tu~e.s..less-than 1OO° C.• This has prevented direct interpretation of the physico-chemical conditions that prevailed during the precipitation of most dolostones in nature with the exception of those that were formed at high temperature. lJJg.in.;, abilityto precip.!t?t~99Jqmite undetcojitrolled laboratoryconditions constitutes the essence of the g!Qtllem 01the Qr)gio of dolomite. The interpretation of the origin of a particular dolostone is therefore dependant upon inferences that are derived mainly from comparisons with a variety of dolomitization models, such as the Evaporative Pumping Model, the Reflu x Model , the Cooron g ~el, the Solution-Cannibalization Model, the Mixed-Water or Dorag Model, the Compaction Model and other, less commonly encountered ones. These models are discussed in a following paper on dolomite entitled "Dolomitization Models and Ancient Dolostones". Taken together, they present a bewildering complexity of choice and it is commonly not apparent or universally agreed upon as to what are the essential features of these models that renders them agents of dolomitization (e.g., high Mg/Ca solution ratios, low salinities, high salinities, high [C0 3 2-1, large organiccontent,low sulphate content) or even why such features promote dolomitization . For example, does a lowering of salinity favour dolomitization from a kinetic standpoint (Folk and Land, 1975) or largely from a thermodynamic stand oint Badiozamani, 1 3? It is not sufficient merely to state that a drop in salinity promotes dolomitization, we must know why and under what circumstances this is true if we are to make full use of this insight in our interpretations. In other words, the formidable array of current models for dolomitization is not clothed by any certain or complete understanding of the precipitational conditions of the mineral
dolomite and, like the emperor who had no clothes, it stands before us embarrassingly naked. In this review, I will assemble and stitch together the experimental, field and theoretical evidence concerning the chemistry of dolomitization into what I hope will pass as an example of sartorial elegance. DOLOMITE CRYSTAL STRUCTURE AND COMPOSITIONAL VARIABILITY The mineral dolomite has an ideal chemical composition of CaMg(C03)2 and a sxmmeJ,!:y..§tructure3 in the trigonal subsystem of the hexagonal GLY.s!& ~ste!I! . The unit cell is a rhombohedron elongate parallel to the "c" crystallographic axis (Figure 3) similar to that of calcite but more highly ordered, and has a lower degree of symmetry because of the substitution of magn~· sium atoms for half of the calcium atoms. eiKe calcite, the dolomite crystal l'SbliiitU poflayersofcations(Ca2+ ,Mg2+) alternating with anions (CO~-) perpendicular to the c-axis direction with the distinction that half of the cation layers are magnesium layers. It is this regular alternation of calci~Rd-ma9ReSium layers with intervening carbonate layers mat uniquely character rzesthe r nlnerar dolomite. Dolomite crystals in nature and those synthesized experimentally are better represented by the formula: Ca(1 + x) Mg(1-x)(C03)2 and commonly depart considerably from an ideal composition and structure. Three distinct factors contribute to this departure. The degree of order of a composi- 5"0 : . tionally ideal (i.e., stoichiometric) dolomite crystal is affected by the degree to which calcium and magnesium are segregated into their respective layers . Firstly, l! calcium atoms from calciLUIl_ layers are i nter£han~~ an equa). .D-Q number of m ~gnesium atomsfrom mag~ neslum layers the degree of order of the crystal is lowerea artnough its comlLQ.S.L tion is unaffected. Most natural dolomites display some degree of mixing of calcium and magnesium between cation layers (Goldsmith and Graf, 1958; Carpenter, 1980). Secondly, dolomites commonl y depart from a stoichiometric-zfC com position because of an excess Of.~ calcium which is accommodated in the magnesil ,mlayers {Goldsmith and Graf, 1958; Lumsden and Chimahusky, 1980). Finally, a multitude of cations, primarily Fe, Sr, Na and Mn, substitute for cal-
-.
.
.-
c Diagenesis
114
Geoscience Canada Reprint Series 4
(
C C
C TRIASSIC-JURASSIC PERMIAN 1-:-"':::';:'=:='-=-=---"" TURNER VALLEY _ _..., MI SS.-PENN, ""::"~=~~:=-'-PRESQU'ILE MANETOE KEG RIVER WINNIPEGOSIS DEVONIAN 31 LEDUC NISKU GROSMONT (SWAN HILLS) (PINE POINT) SILURI AN
C
SOUTHEASTERN CANADA AND OFFSHORE
WESTERN REGION
30-
1
ARCTl C CANADA
(LODGEPOLE)
20-
0- T-J P M-P D S
2(DUNDEE AND LUCAS)
3
17 GUELPH
2 ALLEN BAY (BLUE FIORD)
9 (TRENTON) (ST. GEORGE)
2 (THUMB MOUNTAIN)
6 THER£SA
8
1
5 SOCIETY CLIFFS
0 € P
16
CAMBRIAN
PROTEROZOIC
9 \
2
11
3+unnamed parts of 4 groups
Figure 1 (top) A chart showing the numerical distr ibution of Canadian dotostones that have been formally named. Units that are named in the chart have economic importance both as reservoirs for oil and gas or as hosts for Pb-Zn deposits. Names in brackets are partly dolomitized limestone units in which the dolomite phase has economic importance. Inset graph shows distribution of Canadian dolostones according to geolog ic age.
7+
(
c
10-
CANADIAN SHIELD ORDOVICIAN
c c
unnamed
parts of 5 groups
c c c c c
c
c
C c
c c c c c c C c c c (
Figure 2 (right) An SEM photomicrograph of the gas-bearing dolomite of the Manetoe Facies in the Yukonand Northwe st Territories. Well-formed (Le., idiotopic) crystals of white dolomite (D) project into a vug that is lined in part by reniform bitumen (B). The surrounding dolomite has very low porosity and permeability. Dolomitization preceded the introduction of hydrocarbons.
c c c
c
c (
L
c (
C \...
C
Dolomites -
(
CO
r
r:
o
Chemistry and Precipitation
cium in many dolomites (e.g., Veizer et al., 1978). Iron in particular may be present in concentrations of up to 4 or 5 mole% in ferroan dolomites. These departures from the ideal composition also diminish the ordering of the crystals (Carpenter, 1980). Intracrystal variations in composition and degree of order are also common, particularly in larger crystals (e.g., Fritz and Katz, 1972). There is even some evidence that compositional var iations have influenced the shape of the coarse crystals with curved faces that are common in some late diagenetic dolostones (Radke and Mathis, 1980). The degree of disorder and of departure from stoichiometry also affects the unit cell dimensions and the solubility of dolo..!!!!ill (Land, 1980). Many authors have applied the name protodolomite or even pseudodolomite (e.g ., Gaines, 1977, 1978; Deelman ,
115
1978; Gidman, 1978) to dolomites that are not ideal in structure or composition. Such names tend to obscure the similarity of these non-ideal dolomites to ideal dolomite and Land (1980) suggested retaining the name dolomite for all such phases that have a composition near CaMg(C0 3 b and display evidence, such as superstructure reflections in x-ray patterns, of the degree of order characteristic of dolomite. THE STABILITY OF DOLOMITE IN NATURAL ENVIRONMENTS Minerals that have survived unchanged for long periods of time in natural environments do so for one of two reasons. The mineral may be stable thermodynamically, whiph isJ.~L~Y. i! wfllriOt § affected by small chan ges in the environment,..QIjrnuninera.li&J.lo.stable ~rmod¥oamica1Ly_buUs..itle.r.t Inert minerals are in the process of change to
DOLOMITE
o
I~
o
/
CaMg(C03 ) 2
SUBSTITUTION OF Fe,Sr,No
<,
SUBSTITUTION OF Co
o r
lAYERED SUPERSTRUCTURE
A
o
5 .4
()
3 2
o
o
1
..... ...
Co
.. C03 r:
'M9
r
Figure 3 Diagram of four rhombohedral unit cells of dolomite showing its layered superstructure perpendicular to the "c " axis direction composed of regularly alternating sheets of calcium, magnesium, andcarbonate. Commonly Ca and Mg are incompletely segregated into their respective layers and there is an excess of calcium. Also many foreign ions substitute for calcium. These defects affect the thermodynamic properties of dolomite.
r
other species that have a lower free energy of formation, but the process may be extremely slow, as, for example in the conversion of diamond to graphite on the earth's surface. Such compounds are commonly called metastable to indicate the fact that they exist for variably long periods of time within the stability field of other compounds of similar composition. The appropriateness of the terms "metastable" and "inert" are obviously dependent on the period of time over which a reaction is being considered , which in this case, is geolog ic time. The factors (e.g., catalysts , temperature and concentration) that influence the speed of a chemical reaction are known collectively as reaction kinetic effects . The dolomite-calcite relationship is one of the more spectacular natural examples of the influence of kinetic factors . Under a wide range of surface and near-surface conditions dolomite is favoured thermodynamically ove r co-existing calcite . The dolomitization reaction, as it is cornmonly expressed (e.g., Bathurst, 1975) 2CaC0 3 + Mg2+ ..... CaMg(C03 h + Ca 2+ (1) is characterized by a free energy change of ~Go r = -1.83 kilocallmole for the transformation of aragonite to dolomite of ideal composition and structure (Table 1), which indicates that this reaction has a thermodynamic drive toward dolomitization under standard conditions and unit molar concentrations. We may use the relationship ~Gr = ~Go r + RTlnQ (where, Q = activity ratio of Ca 2+ IMg2+, R = the gas constant, andT = absolute temperature), to derive a value of 21.96 for the activity ratio Ca2+IMg2+ for reaction (1) at equilibrium. The accuracy of this value depends upon the accuracy of estimates for the free energies of formation of the products and reactants in equation (1) (see table 2 in Carpenter, 1980). Solutions with smaller values of Q will favour the replacement of aragon ite by ordered dolomite. Seawater is one such solution with a Q value of only 0.16 (Garrels and Thompson, 1962; Wigley and Plummer, 1976) indicating a strong thermodynamic drive toward the replacement of aragonite and calcite by dolomite in the world 's oceans. What perversity of nature has left the vast present-day submarine blankets of aragonite undolomitized?
c Diagenesis
116
The Mg2+ in equation (1) is necessarily supplied in solution to the reaction site from an external source. Lippman (1973) has emphasized that dolomite precipitates from ions in solution and that the CaC0 3 must dissolve before dolomite can form if the Ca2+ and CO~ ions are derived from the CaC0 3 alone . It is also possible that part of the CO~- in dolomite may have been supplied by the solutionfrom anexternalsourceand Lippman (1973) has proposed the reaction CaC0 3 + Mg2+ + CO§- ...... CaMg(C0 3)2 (2) in which the CO~' anions as well as Mg2+ are supplied by solution to the reaction site as an alternative to the traditional equation of reaction (1). Forthe dolomitization of aragonite, the free energy for reaction (2) is ~Go r = -13.24, favouring the formation of ordered dolomite even more strongly than reaction (1). Reaction (2) also has the advantage that the calcium liberated by dissolution of CaC03is removed entirely from solution by dolomite whereas dolomitization by reaction (1) poses the problem of Ca 2+ removal from the reaction site. If this evolved calcium is not removed its presence will raise the Ca/Mg ratio of solu-
tion and lower the thermodynamic drive toward dolomitization of reaction (2). It may be assumed that, like other diagenetic reactions involving carbonates, dolomitization is entirely a wet chemical process (see Lippman, 1973; Bathurst, 1975; Land, 1980). Solid state reactions between carbonate minerals in the absence of an aqueous phase proceed prohibitively slowly at temperatures lessthan several hundred degrees Celsius (e.g. 1 Fyfe and Bischoff, 1965). The existence of calcium-rich, nonstoichoimetric dolomite with com~
Geoscience Canada Reprint Series 4
tions that fall in the region of the immisci bility gap between calCite and dOlomite (Figure 4) is evidence of slow reac•tion rates. However, recent work has shown that there may be some stable phases of intermediate composition between the low temperature calcite and dolomite stable end members (Reeder and Wenk, 1979). Water of various compositions occupies even sub-micron-sized lamellar spaces between adjacent crystals in carbonate rocks and participates in their diagenesis (Pingitore, 1976; and
c c ( c c ( c c c C "::.
Table 1 Free energies of dolomitization reactions (1), (2) and (3) under standard thermodynamic conditions. ~GO r calcite (kilocal/mole) - 1.38-13.01 - 4.29
Reaction (1) Reaction (2) Reaction (3)
~GO r
aragonite (kilocal/mole) - 1.83 -13.24 - 3.08
- Calculation of free energies from data of Helgeson et a/. (1978) using fully ordered dolomite. These numbers vary depending on the degree of order ofthe dolomite, but ordered dolomite is the mostfavoured phase thermodynamically. One calorie is equivalent to 4.184 joules in SI units.
M
1000
c c
C C C (
(
C C C C C C (
860
0 0 00
•
C C C E
)(
~ ;r;
z
500
0 .8
Q
oU I-AREA OF w
!;r
UNMIXING
~ ~ ~
C C C C
a:
:::>
Ca.MAGNESITES
~ 0 .4
II:
iw ~
SAMPlE 2 ' SEIWATER
25
so
CaMglCO,12
75
100 MgC°3
- 0.8 -0.4
Figure 4 Phase diagram of (water-free) CaC03> CaMg(C0:J2(do/omite). andMgC03 (Magnesite) at temperatures between 0° and 1200°C. Only pure dolomite and calcite are thermodynamically stable at earth surface temperatures . (After Deelman, 1978; Goldsmith and Graf, 1958; and Goldsmith and Heard, 1961). Recent work by Reeder and Wenk (1979) may cause some modification of this simple scenario.
-0.2 0 0.2 0.4 CAlOTE SATURATION INDEX
0.6
Figure 5 Calcite and dolom ite saturation index changes for mixing of sample 1 (Peo. = 10-2 atm; broken fine) and sample 2 (Peo 2 = 10-3 atm; solid fine) with seawater. Shaded region is where Badiozamani (1973) and others have argued that dolomitization is favoured (diagram after Wigley and Plummer, 1976).
Figure 6 A dolomite structure, showing watermolecules preferentiaffy bondedto the cations of the crystal surface. Calcium ions (1) are less strongly hydrated than magnesium ions (2) so calcium ions tend to be incorporated into magnesium layer positions causing the growth of a disordered Carich carbonate. Carbonate ions are unhydrated but must have sufficient energy to displace the water molecules adjacent to the cation layer(e.g ., 3 and 4). Forsimplicity, an NaCI type of crystal structure is shown (after Lippman, 1973,fig. 32).
(
C (
C ( "--
(. (
C C C
c c· c (;
(J
c c' c
c' (
c
( .
Q
o o
Dolomites - Chemistry and Precipitation
(3) The low activity or concentration of Wardlaw, 1976). Most of these natural CO~- relative to Ca2+ or Mg2+ in most solutions are mixtures of waters of natural solutions inhibits the precipitameteoric and marine origin modified by tion of magnesium-bearing carbonate bur ial processes and commonly are minerals in general (Lippman, 1973). supersaturated with respect to dolomite Some have argued that the growth of even when undersaturated with calcite, dolomite crystals is predominantly by except for dilute solutions undersatuthe sequential addition of successive rated in both phases (Figure 5). In warm cation-anion layers (e.g . , Deelman , subsurface environments , the relative stability of dolomite with respect to cal1975, but see also Towe, 1975) . ~ ever, in modern environments of dolocite is even more pronounced than in mite formation dolomite isJr.l1J.al:iabl¥surface environments because the dolomite-calcite equilibrium is shifted present in the s~.Q1rneotas...c.b.ombs.tba.l towards higher Ca/Mg ratios at higher range in size from less tban_1JL to. temperatures (Rosenberg and Holland, about 10JLm. No crystals have been caught in t e act of adding layers as 1964). In other words, the increase of planar sheets perpendicular to the temperature attendant on progressive c-axis. Instead, dolomite crystals apburial in sedimentary basins will inpear to grow as rhombs at more or less crease the saturation level of dolomite equal rates in all directions in spite of with respecttothat of calcite in solutions their layered construction (Figure 3). with constant Ca/Mg ratios . .::J, . In summary, we can state that dolo- y The speed of crystallization inf u~mite is stable in most natural solutions '(" ences the degree of order because ions that the existence of aragonite ~nd that adhere to incorrect lattice positions on the surfaces of rapidly growing cryscalcite in these solutions is a metastable tals are more likely to be entombed in condition. these positions than if the crystals grow THE ROLE OF KINETIC EFFECTS slowly. Slower growth permits a longer period of exposure to e so u Ion 0 an ON THE CRYSTALLIZATION OF Incorrectl sited ion. This increases t e DOLOMITE Several kinetic factors that impede the likelihood that it will be displaced by an precipitation of dolomite from superion of proper type for this lattice site. saturated solution have been identiCalcium and magnesium ions are easily fied and studied in the past several interchanged in dolomite's crystal lattice because of the small differences of decades. These factors involve the molecular mechanics of precipitation lattice energy that are involved (Folk and difficulties associated with the forand Land, 1975). The strength of the electrostatic ion of mation of the regularly alternating sheets of calcium and magnesium magnesium to water molecular bond is atoms in growing dolomite crystals. about 20% greater than that of its calcium counterpart and much greater Many of these kinetic effects interlock than that ofthe carbonate ion (Lippman, and it is difficult to separate cause and 1973, table 20). The relatively strong effect. However, there are three major attachment of the magnesium ion to underlying kinetic barriers to the preits water of hydration shell (Figure 6), cipitation of dolomite: (1) Rapid crystallization from superis even more evident in the data of Samoilov (1965, see Horne, 1969 , saturated, concentrated (i.e. , saline) p. 106) in which magnesium is cited as solutions impedes the segregation of calcium and magnesium ions into their being about 40 times as likely to carry its water of hydration sheath with it during respective layers so that magnesium migration through the liquid. The multilayers are contaminated by calcium tude of metastable, naturally occurring atoms and vice versa and disordered Ca-Mg carbonates are formed (Folk, hydrous magnesium carbonates , such 1979; Folk and Land, 1975; Lippman, as hydromagnesite or nesquehonite, 1973, p. 179); that crystallize in the thermodynamic stability field of anhydrous magnesite, is (2) Pronounced differences in the hydration behavior in aqueous solution also evidence of the relative strength of of the constituent ions of dolomite prothe hydration sheath of the magnesium ion. Magnesite, like dolomite, is highly mote precipitation of calcium-rich phases (Lippman, 1973); and super-saturated in seawater and many
and
-
( r
o
o o [I
o r
8
'-' ,. '-...-.
r.
117
environments of modern dolomite formation, such as the Persian Gulf sabkhas or ephemeral lakes associated with the Coorong Lagoon in Australia (Alderman , 1965; Patterson, 1972), are accompanied by magnesite precipitation (plus assorted hydrous magnesium carbonates). Calcium ions, which are less strongly hydrated than magnesium ions, are more readily incorporated into a growing Ca-Mg carbonate and will tend to occupy sites within magnesium layers as well (Figure 3). Again, rapid rates crystal growth reduces the time available for magnesium ions in solution to displaceerrant calcium ions beforethese defects are buried by continued crystal growth.Thus,morerapidcrystalgrowthor crystallization rates favour the development of more calcium-rich phases. Another important kinetic effect related to the strength of the magnesium ion hydration sheath, is the inability of the relatively_unh drated carbonate ion to batter its wa through this ~ dration barrier ~:tG.omejn contac with the cr stal surfac!i(Figure 6). Only a part of the total CO~- in solution has sufficient energy of motion (translational or vibrational) to break through the hydration barrier surrounding magnesium ions on the surface of a growing dolomite crystal. This low activity of the CO~· ion compared to the activities of the Ca2+ and Mg2+ ions in many neutral and acid natural solutions, such as in seawater, is therefore a kinetic hindrance to the formation of dolomite although the thermodynamic solubility product of dolomite is exceeded by a wide margin(Lippman, 1973, p. 158·161). Consequently, three parameters that will affect the kinetics of the growth of dolomite al~_tbe Mg2+ICa 2 + r atio, the ~y, and the CO~·ICa2+ ratio. The block diagram shown in Figure 7 is a plot of these solution parameters and the dolomite-calcite field boundary is shown as a plane separating a dolomite field on the near side and a calcite field on the far side. This plane represents a generalized kinetic boundary that lies entirely within the region of thermodynamic stability of dolomite. The lower surface is the bivariate plot of Folk and Land (1975).A solution in the calcitefield can move into the dolomite field by increasing its Mg/Ca ratio, decreasingits salinity or by increasing its CO~· content.
¥
c 118
Diagenesis
Geoscience Canada Reprint Series 4
(
( An increase in the solution's Mg/Ca ratio favours the uptake of magnesium into a precipitating Mg-Ca carbonate. This is because an increase in the amount of magnesium in solution probably increases the number of magnesium ions relative to calcium ions that are energetic enough to shed their water of hydration and be incorporated into a growing Ca-Mg carbonate (Figure 8). The laboratory syntheses by Glover and Sippel (1967) and Erenburg (1961) of magnesium calcites containing progressively greater mole% MgC03 contents with increasing Mg/Ca ratios in solution, constitute agood illustration of the effect of independently varying the Mg/Ca ratio of the solution. The sabkha environment of the Persian Gulf is a natural setting where the influence of the solution MG/Ca ratio on dolomit ization of aragonitic sediments is particularly well documented (e.g., Kinsman, 1965; and Patterson, 1972). At high temperature (100° C) the rateordolomite preci pitation is increasea by higher Mg/Ca ratios up to an optimum value (Gaines, 1980). This may apply aiSO"at lower temperatures. The disordering effect caused by rapid precipitation may be overcome by a reduction in the rate of precipitation by dilution of these solutions. This is difficult to verify by laboratory experiments, because of the slowness of crystallization from solutions more dilute than seawater, but has been inferred from natural examples such as modern dolomitization from brackish waters in the Floridan Aquifer (Hanshaw et al., 1971; Fanning et al., 1981), in the sediments of Lake Balaton, Hungary (Muller et al., 1972), and from Pleistocene examples wherethe geologic history is known with a high degree of certainty (e.g., Land, 1973a,b). Calculations of the thermodynamic stability of various carbonate minerals in mixtures of seawater and fresh goundwater show that there is a large region in the mixing continuum in which ". calcite is undersaturated but dolomite remains supersaturated (Badiozamani, 1973; Wigley and Plummer, 1976; and Figure 5). Badiozamani (1973) speculated that the replacement of c~ dolomite would be considerably easier within this mixing region bgcause of the thermodynamic drive causin calcite to ...99 into SoJU1iOUlD~_
It is certainly true that mixing a wide variety of solution compositions can induce limestone dissolution. However, it is less clear why this fact in itself should induce dolomitization. The dissolution of aragonite and ~Ritgtiorf of dolomite takes place in So" .tions that are saturated or su ersaturated with ·D.QS-liIJ. as the ot phases i sabkhaenvironment (Kinsman, 1964) or evaporite lagoons or ponds (Morrow, 1978; Friedman, 1980) because of an appropriate rise in the solution Mg/Ca ratio through the precipitation of gypsum. It seems likely therefore, that the undersaturation of calcite that results from the mixing of saline with dilute solutions is not an important factor for dolomitization. The dilution of saline solutions favours dolomite precipitation simplyoec.!use oOdnetic. eJie.cLoUb.L attendant decrease in th e rate of precipitation. An increase in the CO~- ion concentration of the SOlution, particularly in excess of the Ca2 + concentration, may
also be an important variable for promoting dolomitization in that a greater number of CO~- ions with sufficient energy to shoulder aside the hydration shield of magnesium ions would be present. A variety of laboratory experiments by Lippman (1968a) and Davies et al. (1977) have indicated the importance of the CO~- concentration . Liebermann (1967) also synthesized dolomite from artificial solutions moresaline than seawater and with a Mg/Ca ration approximately that of seawater but with a relatively high pH and CO~' concentration. Sources for naturally occurring alkaline solutions containing high CO~- concentrations include continental groundwaters that have been involved in the weathering of siliceous rocks or that have dissolved continental alkalic carbonate minerals, or marine and continental waters that have been involved in anaerobic bacterial sulphate reduction (Lippman, 1973). The Coorong Lagoon is the bestdocumented modern
....
~----......,:::or-------~
g,
Co Mg(CO,),
C C (
C
C C ~
C (
C
C (
<: C
-(
' u
c
1:5
c
M
CoCO,
(
c ( c
10:1
01
\
C C
1'1
0
C (
~
C (
C C
c
Mg/Ca
Figure 7 Block diagram showing the effect of variation in the three parameters, the MgICa solution ratio , the salinity, and the CO:/Ca solution ratio . The plane represents the kinetic boundary between dolomite and calcite or aragonite and it includes the hidden corner of the Coorong Lagoon waters as a point on the plane. The basalplane is afterFolk and Land(1975). Note that the vertical projection of Coorong Lagoon waters falls largely on the calcite-aragonite side of the stabifity boundary on the basal plane. A decrease in salinity, an increase in the MgI Ca rat io or an increase in the CO:/Ca ratio favours the precipitation of dolomite,
e (
c C C C C C
(
( (
119
Dolomites - Chemistry and Precipitat ion
('
c c c (
c (
c.
o
o«: o o c ( (
C C C (
occurrence of dolomitization under the influence of continental groundwater (Vonder Borch et al., 1975; Vonder Borch, 1976; Von der Borch and Lock, 1979). The composition of groundwater in the Coorong Lagoon Area is shown in Figure 7. Note that the projection of this solution composition onto the Folk and Land (1975) bivariate diagram falls largely outside the region of dolomite kinetic stability.This maybe an indication that the high CO~- concentration in the Coorong ground waters also promotes the precipitationof dolomite and Figure 7 has been constructed on this basis. Generation of the CO~- by anaerobic sulphate reduction may have occurred in many burial environments from depths less than a metre to several hundred metres below the sedimentwater interface. This process has been suggested as a factor in the dolomitization of some Recent burrow fillings (Brown and Farrow, 1978). A supply of SO~- ions in excess of the Ca2 + concentration is probably necessary because the initial rise in alkalinity will precipitate CaC0 3 before proceeding to dolomitization (Lippman, 1973).
::l<'The relati~ase-of-dissolution_of.thEL ~P.Qlyrno~pl:1s.0f.CaC03-1s.also.an.
important rate; deter:minmg.step-in.-t he dOIomitimi9J:l_pr.ocess.(Gaines, 1974). Tile dolomitization of aragonitic sediments proceeds considerably more rapidly than for calcitic sediments (e.g., Schmidt, 1966).The rapid conversion of aragonite to low-magnesian calcite in humid climates has been cited as a factor in the resistance to dolomitization of Pleistocene shelf limestones exposed during low stands of sea level (Sibley, 1980). The presence of certain cations, such as lithium and possibly iron, haveacatalytic effect on the precipitation of dolomite (Carpenter, 1980; Gaines, 1980). Carpenter (1976) has also suggested that because the proportion of strongly hydrated magnesium ions decreases markedly with increasing salinity, dolomitization should be favoured by high salinities and it does appear that dolomites that have formed under co.o.d .L tions of higher salinit y and higher..sAlu.:.... tion Mg/C a ra..Jip.s_t.l:!D.cLto_b.e_Le..§.§... calcium-rich and heDce mace stoichjo~(Fuehtbauer and Goldschmidt, ION DEHYDRATION ENERGY
MEAN ENERGY Mg2+ ION-
( --80 KCAL/MOLE-
(
ENERGY OF MOTION OF Mg "AND Ca "IONS (RELATIVE TO DEHYDRATION ENERGY OF Mg2" AND Ca 2"IONS)
.
Figure 8 Diagram showing energy (of motion) distribution of Mg2+ and Ca2 + ions in solution relative to the ion dehydration energy. It is possible to have an equal number of Mg2+ ions with energies exceeding the dehydration barrier by increasing the amount of Mg2+ in solution relative to the amount of Ca2 + in solution. A similar argument also applies to the C~- ion in solution . A normal distribution of ion kinetic energies is shown here for simplicity. The true shape of the distr ibuiton of the kinetic energies of molecules in a liquid is not known and is probably unique for each ion in solution but probably lies somewhere between a normal distribution and the asymmetric Soltzman distribution for the energies of molecules in a gas.
1966; Glover and Sippel , 1967). But the degree of disorder introduced by the faster rate of precipitation from more saline solutions overshadowsthe effect of the diminution of the hydration barrier of the magnes ium ion with increasing salinity so that, although the Ca/Mg carbonates that are formed approach a dolomite composition, they tend to be disordered (Glover and Sippel , 1967;Folk and Land, 1975; Morrow, 1978). More experimental data are needed to separate the effects of precipitation rate versus the Mg/Ca ratio of the solution . The origins of many dolomite occurrences are consistent with the model outlined in Figure 7 part icularly for those that have formed in near surface settings. However, in subsurface environments, the increase of temperature with depth is probably of dominant importance. The increase of temperature that occurs with burial will reduce the kinetic inhibitions to dolomitization (see Gaines, 1974; Mattes and Moun joy, 1980) and cause dolomite to become more saturated relative to calcite (Rosenberg and Holland, 1964). The amount of dolomiteJ.o..r.me.d will depend on the rate that magnesium is supp lied and oJ1..the_time _available (Hanshaw et al., 1971). RECENT INITIATIVES IN DOLOMITIZATION CHEMISTRY Baker and Kastner (1981) introduced a new dimension to the study of dolomitization with their discovery that sulphate ions in solution exert a strong inhibition on the dolomitization of calcite and aragonite. Their laboratory' experiments , conducted at 200°C, indicated that dolomitization of calcite is prevented by the presence of sulphate in solution at concentrations of only 5% of that of seawater sulphate concentration (0.028 M). Dolomitization of ara . e was somewhat less influenced b sui hate in solution . Dolomitization in many diverse geologic settings has been suggested to be primarily a response to reductions of sulphate concentrations in these settings, such as during the dilution of seawater by fresh groundwater in coastal aquifers (Kastner, 1984) or by the microbial reduction of sulphate within organic-rich sea floor sediments (Baker and Burns, 1985).
120
Diagenesis
Geoscience Canada Reprint Series 4
c c (
However, as discussed by Hardie and CO~- but is not required to remove slightly acid diagenetic environment of (1987), it is difficult to rationalize the any reaction products . the sabkha. Patterson suggested that occurrence of Recent dolomites in setBoth of these equations involve conthe average value of x in reaction (4) is tings characterized by high solution sulsiderable volume changes. A loss of close to 0.1 based on the organic content phate concentrations, such as in the volume of 6% to 13% (aragonite or of sabkha sediments. This value of x calcite replacement) accompanies much-studied sabkhas of the Persian also satisfies the conservation of volGulf(e.g., Patterson, 1972),in terms of a dolomitization according to reaction (1), ume criter ion. sulphate inhibition of dolomitization. A a circumstance commonly cited by Reaction (3) implies that the-e.quiUb... point that has not been discussed overtworkers in the oil industry. On the other rium constant for dolomitization deIy is that a kinetic control on a reaction hand, dolomitization according to reacpends on the CO~- concentration aswell rate, such as the sulphate inhibition tion (2), if applied literally, involves a asfne Mg2+ aod-~c:~mcentrations . effect, which has been determined for a tremendous volume gain of about 75% However, Stoessell (1987, see also to 88% (Lippman, 1973, table 28). part icular temperature (e.g., 200°C), Machel and Mountjoy, 1987) has demshould be more pronounced at lower However, most dolomitized rocks do - onstrated that equilibrium between caltemperature because of the temperanot appear to have undergone volume cite (or aragonite) and dolomite in reacture dependence of reaction rates. We changes during dolomitization indicattion (3) is dependent only on the Ca/Mg may infer that it is not appropriate to ing a volume for volume replacement activity ratio and independant of the apply the same concentration levels of ratherthan a mole for mole replacement activity of CO~- . But, because of kinetic sulphate found to inhibit dolomit ization (Friedman and Sanders, 1967).Deposifactors , the concentrations of Mg2+and tionalfabricssuchasthosedisplayedb at 200°C (Baker and Kastner, 1981) to CO~- in the solution must be higher than algal laminites (Gebelein and Hoffman, lower temperature (e.g., 20-40°C) setthat necessary merely to satisfy the tings for dolomitization (Kastner, 1984). 1973) or by oolitic sediments (Powers, thermodynamic requirements of doloInstead, much lower sulphate concen1962) are not disrupted by dolomitizamitization and the dolomitizing solution trations should suffice to inhibit dolomition. The increase in permeability that must play an active role as a transporter tization at lower temperatures. This facaccompanies total dolomitization may of ions to and from the reaction site.This tor lends weight to the observations of be related to the idiomorphic shape and has important implications for carbon Hardie (1987) concerning the occuruniformity of size of the dolomite crysand oxygen stable isotope studies of renee of dolomite in Recent sulphatetals (Wardlaw, 1979). dolostones. bearing near-surface env ironments. ;it.... The conservation of volume during The release of Ca2+ to solution during Undoubtedly, further laboratory studies -{ 'the dolomitization reaction may be an dolomitization can lead to the local preconcerning sulphate inbibjtion..w.i~ important constraint. For a volume-forcipitation of gypsum as has been docuvolume reaction the net dolomit ization conducted and ~ ps some research mented in Persian Gulf sabkhas (Patterwill be done on the sul ghate content of reaction may be written son, 1972). This may be an alternative ancieii'i dotcmltes., (2-x)CaC0 3 + Mg2+ + xCO~- +-+ explanation also for the minute gypsum Less attention has been focussed CaMg(C0 3 )2 + (1-x)Ca2+ (3) or anhydrite crystal inclusions in the recently on the effect of organic material where, x = 0.11 and 0.25 for the dolocoarsely crystalline white dolostones of on dolomitization. The precipitation of mitization of aragonite and calcite, resthe Mississippi Valley and Pine Point dolomite-similar phases by the action pectively. Any departure from a volume Districts which Beales and Hardy (1977, of transition-metals and organic chelatfor volume replacement will result in 1980) suggested were remnants of former evaporites. ing agents (Mirsal and Zankle, 1985) changes in these numbers which repremay cause more research in this direcsent the net exchange of ions between tion. They found that metal chelates the solution and the solid phases. Reac"THE DOLOMITIZATION PROCESS CONCLUSIONS mediated the transfer of magnesium cartion (3)is characterized by a free energy bonate to crystal surfaces causing the charge of dG r = -3.085 for dolomitiza(1) Dolomite is thermodynamically staincomplete dolomitization of calcite. tion of aragonite (Table1) which is interble in most natural solut ions at earth mediate between the free energies of surface conditions and a thermodyDOLOMITIZATION AND THE namic drive exists for the conversion of reaction of (1) and (2). Reaction (3) is a VOLUME PROBLEM calcite and aragonite to dolomite. composite of reactions (1) and (2)which The chemical reaction for dolomitizamay be regarded as end members of a (2) The precipitation of dolomite is hintion is most commonly given in the form series of reactions dependant on the dered by kinetic factors that are a conseshown in reaction (1) lD wbklll a !D..Q@_of degree to which there is a gain or loss of quence of the high degree of order of dolomite re laces two moles of calcite. the dolomite crystal lattice and of the rock volume during dolomitization. The alternative dolomitization reaction differences in hydration behavior of Patterson (1972, p. 325) has progiven in reaction (2) was proposed by the Ca2+, Mg2+ and CO~- ions. posed the reaction (2-x)CaC0 3 + Mg2+ + xHC03 +-+ Lippman (1973, p. 180) largely because (3) The extent to which these factors . CaMg(C0 3)2 + (1-x)Ca2+ + xW (4) inhibit dolomitization appear to be govof the advantages of a high CO~' concentration in solution. In reaction (1) the based on chemical mass balance conerned largely by the interplay of three dolomitizating solution supplies the siderations for dolomitization of sabkha critical parameters: (a) the rate of crysMg2+ required and removes the Ca 2 + sediments which is similar to reaction (3) tallization governs the degree of order released to SOlution, whereas in reacexcept that carbonate is supplied in the of precipitating Ca-Mg carbonates. Dilution (2)the solution supplies both Mg2+ form of bicarbonate because of the tion of supersaturated natural solutions
c c
(
c
c
c c c
c (;.
c C!::
C C C C C C C C C C C C C C C C C C C C (
C C C C C C C
G C C'
Dolomites - Chemistry and Precipitation
r: ,
'\ J
c
r:
r: V
(
is one method of reducing the crystallization rate; (b) the Mg/Ca ratio of the solution determines the degree to which Mg2+ is incorporated into precipitating Ca-Mg carbonates. This ratio commonly rises during the precipitation of gypsum and/or anhydrite to a level favourable for dolomitization; and (c) a C03/Ca ratio greater than one will aid in the precipitation of dolomite. This ratio may rise in ground water involved in weathering of igneous rocks or where anaerobic bacterial reduction of sulphate occurs. The presenceof sulphate in solution may also inhibit dolomitization. (4) A conservation of volume constraint implies a reaction of the form: (2-x)CaC0 3 + Mg2+ + xCO~- ...... CaMg(C0 3)2 + (1-x)Ca 2+ where, x equals 0.11 or 0.25 for dolomitization of aragonite or calcite, respectively. All three of the constituent ions of dolomite are exchanged between the solution and the solid phases emphasizingthe roleoffluid movements and compositions during dolomitization. It is hoped that this review has succeeded in providing a coherent synthesis of evidence and theories specifically concerning the chemistry of the dolomitization process. The main kinetic factors controlling dolomite precipitation may have been identified, but a more determined effort should be madeto precipitate dolomite in the laboratory in quantities large enough to verify the influence of these kinetic effects and perhaps even to gain a degree of quantitative understanding of them. Only then will we be in a position to fully understand the significance of the isotopic composition and of the trace element contents of dolostones because of their dependence on the chemistry of the dolomitization process. Knowledge ofthe chemical environment of dolomitization also provides a clearer insight into the process of dolomitization in specific models for dolomitization.
ACKNOWLEDGEMENTS r:
Douglas Kent of the University of Saskatchewan and Alan Kendall of Amoco Canada Ltd. critically read the manuscript. I also wish to thank many individuals, such as H.G. Machel (University of Alberta), E. Sass (Hebrew Universityof Jerusalem) and R.K. Stoessell (University of New Orleans), who contributed to this paper through discussions on dolomitization.
121
REFERENCES Dolomite Crystal Structure and Compositional Variability
Carpenter, A.B., 1980, The chemistry of dolomite formation 1: the stability of dolomite, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 111-121. Gaines, A.M., 1977, Protodolomite redefined: Journal of Sedimentary Petrology, v. 47, p. 543-546. Goldsmith, J.R. and Graf, D.G., 1958,Structural and compositional variations in some natural dolomites: Journal ofGeology, v. 66, p. 678-693. Land, L.S., 1980, The isotopic and trace element geochemistry of dolomite: the state of the art, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 87-110. Lippman, F., 1973, Sedimentary Carbonate Minerals: Springer-Verlag, New York, 228 p. The Stability of Dolomite in Natural Environments
Carpenter, A.B., 1980, The chemistry of dolomite formation 1: the stability of dolomite, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 111-121. Gaines, A.M., 1977, Protodolomite redefined: Journal of Sedimentary Petrology, v. 47, p. 543-546. Garrels, G.M. and Thompson, M.E., 1962,A chemical model for sea water at 25°C and one atmosphere total pressure: American Journal of Science, v. 258, p.402-418. Goldsmith, J.R. and Heard, H.C., 1961,.Subsolidus phase relations in the system CaC0 3 : Journal of Geology, v.69, p.45-74. Helgeson, H.C., Delany, J.M., Nesbitt, H.W. and Bird, D.K., 1978, Summary and critique of the thermodynamic properties of rock-forming minerals: American Journal of Science, v. 278-A, 229 p. Land, L.S., 1980, The isotopic and trace element geochemistry of dolomite: the state of the art, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds.,
Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 87-110. Lippman, F., 1973, Sedimentary Carbonate Minerals: Springer-Verlag, New York, 228 p. Rosenberg, P.E. and Holland, H.D., 1964, Calcite-dolomite-magnesite stability relations in solutions at elevated temperatures: Science, v. 145, p. 700-701. The Role of Kinetic Effects on the Crystallization of Dolomite
Badiozamani, K., 1973, The Dorag dolomitization model-application to the Middle Ordovician of Wisconsian: Journal of Sedimentary Petrology, v. 43, p.965-984. Carpenter, A.B., 1976, Discussion: Dorag dolomitization model: by K. Badiozamani: Jour. Sed. Petrology, v.43, p. 965-984: Journal of Sedimentary Petrology, v. 46, p. 258-261. Davies, P.J., Bubela, B. and Ferguson, J., 1977, Simulation of carbonate diagenetic processes: formation of dolomite, huntite and monohydro-calcite by the reactions between nesquehonite and brine: Chemical Geology, v. 19, p. 187-214. Folk, R.G., 1974, The natural history of crystalline calcium carbonate: effect of magnesium content and salinity: Journal of Sedimentary Petrology, v. 44, p. 40-53. Folk, R.L. and Land, L.S., 1975, Mg/Ca ratio and salinity: two controls over crystallization of dolomite: American Association of Petroleum Geologists, Bulletin, v. 59, p. 60-68. Friedman, G.C., 1980, Dolomite is an evaporite mineral: evidence from therock record and from sea-marginal ponds of the Red Sea, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 69-80. FOchtbauer, H. and Goldschmidt, H., 1965, Beziehugen zwishen calcium-gehalt und bild ungsbed ingungen der dolom ite: Geologische Rundschau, v. 55, p. 29-40. Gaines, A.M., 1980, Dolomitization kinetics; recent experimental studies, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 81-86.
122
Diagenesis
Geoscience Canada Reprint Series 4
c c (
Glover, E.D. and Sippel , R.F., 1967, Synthesis of magnesium calcite: Geochimica et Cosmochimica Acta, v. 33, p.603-613. Hanshaw, B.C., Back, W. and Deike, R.G., 1971, A geochemical hypothesis for dolomitization by groundwater: Economic Geology, v. 66, p. 710-724. Land, L.S ., 1973a, Contemporaneous dolomitization of Middle Pleistocene reefs by meteroic water, North Jamaica: Bulletin of Marine Science, v. 23, p. 64-92 . Land, L.S., 1973b, Holocene meteoric dolomitization of Pleistocene limestones, North Jamaica: Sedimentology, v. 20, p.411-424. Liebermann, 0., 1967, Synthesis of dolomite: Nature, v. 213, p. 241-245. Lippman, F., 1968, Synthesis of BaMg (C0 3 b (Norsethite) at 20°C and the formation of dolomite in sediments, in Muller, G. and Friedman, G.M., eds., Recent Developments in Carbonate Sedimentology in Central Europe : Springer-Verlag , New York, p. 33-37. Lippman, F., 1973, Sedimentary Carbonate Minerals: Springer Verlag, New York, 228 p. Morrow, D.W., 1978, The influence of the Mg/Ca ratio and salinity on dolomitization in evaporite basins : Canadian Petroleum Geology Bulletin, v.26, p.389-392. Muller, G., Irion, G. and Forstner, V., 1972, Formation and diagenesis of inorganic Ca-Mg carbonates in the lacustrine environment: Naturwissenschaften, v. 59, p. 158-164. Patterson , R.J., 1972, Hydrology and carbonate diagenesis of a coastal sabkha in the Pers ian Gulf, Ph.D. Dissertation, Princeton Univ., Princeton , New Jersey, 498 p. Sibley, D.F., 1980, Climatic control of dolomitization, Seroe Domi Formation (Pliocene), Bonaire, N.A ., in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of DolomitizatIon: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. p. 247-258. Von der Borch, C.C., 1976, Stratigraphy and formation of Holocene dolomitic carbonate deposits of the Coorong area, South Australia:Journal of Sedimentary Petrology, v. 46, p. 952-966. Recent Initiatives In Dolomlrlzation Chemistry Baker, P.A. and Burns, S.J., 1985, Occurrence and formation of dolomite in organic-rich continental margin sediments : American Association of Petroleum Geologists, Bulletin, v.69 , p. 1917-1930.
Baker, P.A. and Kastner, M., 1981, Constraints on the formation of sedimentary dolomite: Science , v. 213, p. 214-216. Hardie, L.A., 1987, Dolomitization: a critical view of some current views: Journal of Sedimentary Petrology, v.57, p. 166-183. Kastner, M., 1984, Control of dolomite formation: Nature, v. 311, p. 410-411. Mirsal, I.A. and Zankle , H., 1985, Phenomenological aspects of carbonate geochemistry. The control effect of transition metals: Geologische Aundschau, v. 74, p. 367-377. Dolomitization and the Volume Problem Friedman , G.M. and Sanders, J.E., 1967, Origin and occurrence of dolostones, in Chlingar, G.V., Bissell, H.J. and Fairbridge, R.W., eds. , Carbonate Rocks: Elsevier, Amsterdam, Developments in Sedimentology, No. 9A, p. 267-348. Lippman, F., 1973,Sedimentary Carbonate Minerals: Springer-Verlag , New York, 228 p. Powers, R.W., 1962, Arabian Upper Jurassic carbonate reservoir rocks, in Ham, W.E., ed., Classification of Carbonate Rocks -A Symposium: American Association of Petroleum Geologists, Memoir 1, p. 122-192. Wardlaw, N.C., 1979, Pore systems in carbonate rocks and their influence on hydrocarbon recovery efficiency, in Geology of carbonate porosity: American Association of Petroleum Geologists, Continuing Education Course Note Series 11, p. E1-E24. Other References Cited in the Text Alderman, A.R., 1965, Dolomitic sediments and their environment in the south-east of South Australia: Geochimica et Cosmochimica Acta, v. 29, p. 1353-1365. Bathurst, R.G.C., 1975, Carbonate Sediments and Their Diagenesis: Elsevier, Amsterdam , Developments in Sedimentology No. 12, 658 p. Beales, F.W.and Hardy, J.L., 1977, The problem of recognition of occult evaporites with special reference to southeast Missouri: Econom ic Geology, v. 72, p.487-490. Beales, F.W. and Hardy, J.L., 1980, Criteria for the recognition of diverse dolomite types with an emphasis on studies on host rocks for Mississippi Valley type ore deposits, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization : Society of Economic Paleontologists and Mineralogists, Special Publication 28 , p.197-213.
Brown, R.J. and Farrow, G.E., 1978,Recent dolomitic concretions of crustacean burrow orig in from Goch Sunart, West Coast of Scotland : Journal of Sedimentary Petrology, v. 48, p. 825-834 . Davies, G.R., 1979, Dolom ite rese rvoir rocks : processes, controls, porosity developments , in Geology of Carbonate Porosity: American Association of Petroleum Geologists, Continuing Education Course Note Series 11, p. C1-C17. Deelman , J.C., 1975, Dolomite synthesis and crystal growth : Geology, v.3, p.471-472 . Deelman , J.C., 1978, Discussion : protodolomite redefined : Journal of Sedimentary Petrology, v. 48, p. 1004-1067. Erenberg , B.G ., 1961, Artificial mixed carbonates in the CaC0 3-MgC0 3 series : Zhurnal Skrukturny Khimiia , v. 2, p. 178-182. Fanning, K.A., Byrne , R.H., Breland, JA, Betzer, R.R., Moore, W.S., Eisinger, R.J. and Pyle, T.E., 1981, Geothermal springs of the west of Florida Continental Shelf: evidence for dolomitization and radionuclide enrichment: Earth and Planetary Science Letters , v. 52, p. 345-354. Fritz, P. and Katz, A., 1972, The sodium distribution of dolomite crystals : Chemical Geology, v. 10, p. 237-244. Fyfe, W.S. and Bischoff, J.L., 1965, The calcite-aragonite problem , in Pray, L.C. and Murray, R.C., eds., Dolomitization and Limestone Diagenesis - A Symposium: Society of Economic Paleontolgists and Mineralogists, Special Publication No. 13, p. 3-13. Gaines, A.M., 1978, Reply - protodolomite redef ined : Journal of Sedimentary Petrology, v. 48, p. 1009-1011. Gebelein, C.D. and Hoffman, P., 1973, Algal origin of dolomite laminations in stromatolitic limestone: Journal of Sedimentary Petrology, v. 43, p. 603-613. Gidman , J., 1978, Discussion - protodolomite redefined :Journal of Sedimentary Petrology, v. 48, p. 1007-1008. Horne, A.A ., 1969 , Marine Chemistry: Wiley-Interscience, New York, 568 p. Kinsman, D.D .J., 1965, Gypsum and anhydrite of recent age, Trucial Coast, Persian Gulf, in Aau, J.L., ed., Second Sympos ium on Salt, v. 1: Northern Ohio Geological Society, Cleveland, Ohio, p.302-236. Lumsden , D.N. and Chimahusky, J.S., 1980, Relationship between dolomite nonstoichiometry and carbonate facies parameters, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontolog ists and Mineralogists, Special Publication 28 , p. 123-137.
c c c
c c c
c c c c c c c c c c c c c c c c c
c C E
C C C C
c e c c c c C l
C
r C
o c o o (
o (
o o o o o c o o o o
o o o o o (
o o o () (
e r:
r
Dolomites -
Chemistry and Precipitation
Machel , H.G. and Mountjoy, E.W., 1987, Chemistry and environments of dolomitization - a reappraisal (Reply): Earth Science Reviews, v. 24, p. 213-215. Macqueen, R.W., 1979, Base metal deposits in sedimentary rocks : Some approaches: Geosc ience Canada, v. 6, p.3-9. Mattes, BW. and Mountjoy, EW., 1980, Burial dolomit ization of the Upper Devon ian Miette Buildup , Jasper National Park, Alberta, in Zenger, D.H., Dunham, J.B. and Ethington , R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Special Publicat ion 28, p. 259-297. Pingitore, N.E., Jr. , 1976, Vadose and phreatic d iagenesis: Processes , products and their recognit ion in corals: Journal of Sedimentary Petrology, v. 46, p. 60-67. Radke , B.M. and Mathis, M.L., 1980, On the formation and occurrence of saddle dolomite: Journal of Sedimentary Petrology, v. 50, p. 1149-1168. Reeder, R.J. and Wenk, H.R., 1979, Microstructures in low temperature dolomites: Geophysical Research Letters, v. 6, p. 77-80. Samoilov, O.Y., 1965, Structure of aqueous electrolyte solutions and hydration of ions: Consultants Bureau, New York.
123
Schmidt , v., 1965, Facies, diagenesis , and related reservoir properties in the Gigas Beds (Upper Jurassic), northwestern Germany, in Pray, L.C . and Murray, R.C., eds ., Dolomitization and Limestone Diagenesis: Society of Economic Paleontolog ists and Mineralog ists, Special Publication No. 13, p. 124-168. Stoessell, R.K., 1987, Chemistry and environments of dolomitization - a reappraisal (Discussion): Earth Science Reviews, v. 24, p. 211-212. Towe, K.M., 1975, Dolomite synthesis and crystal growth: Comment: Geology, v. 3, p. 612. Veizer, J., Lemieux, J., Jones, B., Gibling, J.R. and Savelle, J., 1978, Paleosalinity and dolomitization of a Lower Paleozoic carbonate sequence, Somerset and Prince of Wales Island, Arct ic Canada: Canadian Journal of Earth Sciences, v. 15, p. 1448-1461. Von der Borch, C.C., Lock, D.E . and Schwebel, D., 1975, Groundwater formation of dolomite in the Coorong region of South Australia: Geology, v.3, p.283-285. Wardlaw, N.C. , 1976, Pore geometry of carbonate rocks as revealed by pore casts and capillary pressure : American Association of Petroleum Geologists, Bulletin, v. 60, p. 245-257.
Wigley, T.M.L. and Plummer, L.N., 1976, Mix ing of carbonate wate rs : Geochimica et Cosmochimica Acta, v. 40, p.989-995 . Ziegler, A.M ., Scotese , C.R., McKerrow, M.S., Johnson, M.E. and Bambach , R.K., 1979, Paleozoic paleogeography : Annual Reviews of Earth and Planetary Sciences, v. 7, p. 473-502.
Originally published in Geoscience Canada v. 9 Number 1 (March 1982) Revised 1988
124
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c c c c c c
c c c c c c c
c
c
c c c c c c c ( SEM photomicrograph of a synthetic dolomite formed in a reaction vessel at 220 0 C. Rhombic dolomite crystals have formed from the reaction of calcite with a CaCI2 - MgCl2solution. Loss of CO2 from the reaction vessel causedprogressive dissolution of this dolom ite. The etched surface of these euhedral dolomite crystals is an early manifestation of CO2 depletion in the reaction solution. Photograph courtesy of D. W. Morrow (Geological Survey of Canada).
c
c
c
c
e (
c c C L
C
c' (
o c' o o c o o ( o o o c o o o o o o o o o o o o
Dolomites -
Models and Ancient Dolostones
Dolomite - Part 2: Dolomitization Models and Ancient Dolostones D.W. Morrow Institute of Sedimentary and Petroleum Geology 3303-33rd Street N. W Calgary, Alberta T2L 2A7 INTRODUCTION The first paper of this review series (Morrow, 1982, this volume) was a synthesis of what is presently known concerning the chemistry of dolomitization. In that paper, parameters of major importance to the dolomitization process were identified: the Mg/Ca solution ratio ; the ionic strength or salinity of the solution; the C03/Ca solution ratio; and the sulphate concentration. Other factors, such as temperature, that influence the dolomitization process also were discussed. This paper focusses the reader's attention on a different aspect of dolomitization - the great variety of models for dolomitization that have been proposed to explain the origin of the wide spectrum of dolostone types. In approximate historical order of appearance , these are the Hypersaline Lagoon and Reflux Model, the Burial Compaction Model, the Solution Can nibalization Model, the Coorong Model, the Sabkha Model, and the Mixed Water (Dorag) or Dilution Model, the Organogenic Sea-floor Model and the Hydrothermal Convection Model. The Sabkha and Coorong Models are specific to well defined environmental settings. The other models are more generalized and emphasize particular aspects of the dolomitizing process that may be operative in a variety of environmental settings. For example, in the Mixed Water Model, the emphasis is on the dilution of sea-
125
water by fresh groundwater which may occur in a multitude of situations both penecontemporaneous with, or long after sedimentation. A number of cond itions must be satisfied for these models to act as agents for dolomit ization (Figure 1). These conditions, in their simplest form are as follows : (1) The amount of Mg2+ available for dolomitization must be adequate to form a given mass of dolostone (i.e., source of Mg2+). (2) A mechanism is required to deliver the available Mg2+ to the dolomitization site. This mechanism may also be required to deliver a small amount of CO~' ions to the dolomitization site and carry away Ca2+ ions . (3) The composition of a proposed dolomitizing solution must be conducive to dolomitization. The first paper of this review series provided some criteria to test this proposition. Each model will be examined in the light of these conditions to place constraints on their application to the stratigraphic record and on their viability in general. A variety of specific factors may contribute to these conditions in particular instances. For example, the delivery mechanism (condition 2) may depend on porosity and permeability relationships, the hydraulic head of the solution and the time over which the mechanism is operative. The application of these models to individual dolostone masses depends
on inferences concerning the attributes of these dolostones. These include considerations of scale and spatial relationships, major stratigraphic and paleogeographic relationships, internal facies relationships, petrographic relationships, and of geochemical data. It is at this stage, where statements are made concerning the application of specific dolomitization models based on the physical and chemical attributes of dolostone bodies, that the confusion concerning the dolomitization process becomes readily apparent. An attempt is made here to determine the relative degree of certainty that can be placed on the diverse criteria used to substantiate the choices of dolomitization models that have been applied to many ancient dolostones in Canada (Figure 2) and elsewhere.
Dolomitization Models The Hypersaline Lagoon and Reflux Model. One of the first models proposed for dolomitization on a large scale was the Seepage Refluxion or more simply, the Reflux Model (King, 1947 ; Adams and Rhodes, 1960) . In this model (Figure 3), open marine water evaporates as it passes landward across a hypersaline shelf lagoon. The resulting increase in density of these evaporated waters causes them to infiltrate the underlying sediment and to move seaward by seepage (i.e., reflux) through the seaward-dipping beds. The cycle of fresh
DOLOMITIZATION MODELSESSENTIAL COMPONENTS
2
1
~ g
MgMg Mg CO:! Mg
1. SUPPLY OF Mg AND C03
~c=7 ''0
d
2. DELIVERY MECHANISM
3. DOLOMITE CONSTRUCTION SITE
Figure 1 A construction site analogue of the process of dolomitization illustrating the three essential elements of any dolomitization model. The external supply of magnes ium, and, to a lesser extent, the supply of carbonate must be sufficient to form the observed amounts of dolom ite. There must be a delivery mechanism, commonly in the form of fluid flow, adequate to deliver these ions to the dolomitization site and lastly there must exist chemical conditions favourable for the precipitation of dolomite at the "dolomite construction site ", Failure to meet any of these conditions results in an inadequate dolomitization model.
Geoscience Canada Reprint Series 4
Diagenesis
126
DOLOMITIZATION MODELS· CANADIAN EXAMPLES
Figure 2 Canadian examples of the dolomitization models discussed in this paper. Some units, such as the Presqu 'ite Formation have been assigned to more than one model.
Model
Example (Formation and author)
1. Hypersaline Basin and Seepage Reflux
Keg River· Pine Point - Presqu 'ile (Skall, 1975) Muskeg - Keg River (Davies and Ludlum, 1973; Bebout and Maiklem, 1973) Prair ie Evapor ite - Winnipegosis (Wardlaw and Reinson, 1971) Lodgepole (Young and Greggs, 1975) Leduc (Wilson , 1975)
(
( ( (
ch¥fie Lake (Roy, 1972) SI. George (Co llins and Smith, 1975) Turner Valley (Murray and Luc ia, 1967) Muskeg-Winnipegosis (Shearman and Fuller, 1969)
2. Sabkha
( ( (
Guelph (Sears and Lucia, 1980) Nisku (Exploration Staff, Chevron Standard Lim ited , 1979) Abenak i (Eluik. 1978) Allen Bay (Land et a/., 1975: Veizer et al., 1978) Thumb Mountain (Morrow and Ker r, 1978) SI. George? (Coll ins and Smith , 1975) Stone (Morrow, 1975: Macqueen and Thompson. 1978) Manetoe (Morrow, 1975)
3. Mixed-Water
C (
C (
4. Coorong
Belcher Group (Ricketts, in press)
5. Burial-Compaction
Presqu'lle (Griffin , 1965; Beales and Jackson, 1969) Leduc (liling. 1959) Miette (Mattes and Mountjoy, 1980) Guelph (Jodry, 1969)
6. Solution Cannibalization
Yeoman (Kendall , 1977)
7. Tectonic or Hydrothermal
Black River - Trenton (Sanford, 1962) Keg River - Rainbow (Lang ton and Chin , 1968) Presqu'ile (Jones , 1980)
C (
C ( (
C C C
DOLOMITIZATION MODELS 2• . SABKHA. MODEL (REFLU X VARIANT)
w
Figure 3 Major dolomitization models summarized in generalized representative cross-sections . The patterned areas represent zones of active dolomitization during the operation of these models. The horizontal dimensions of these cross sections is proportional to their vertical dtmenston» so that systems tens of metres thick may extend several kilometres whereas systems thousands of metres deep may extend hundreds of kilometres laterally. Arrows correspond to directions of fluid movement. Inset table shows estimates of the values of chemical parameters that characterize each of these models.
o « IL.
.tORM TIDEI
1.
a:
1'-""
SABKHA. MODE L ( C A PIL L ARY RECHARGE VARIANT)
a:
I
-e
rn zw
z
0
0
~
N
:I:
3.
COORONG MODEL EPHlMr:R.u
;:) CJ)
w
~ :)
<
REFLUX MODEL
...
~
LAKES
~:::S;;::::::;;;;;;::;;:~;;;;;;= T tO-1l
...,caFo-'C [ao.] • .·n •.•, 0 . "
~
5.
ORGANOGENIC SEAFLOOR MODEL
Ja..h
. -11
,
0 ,.•-
'UUVEl •
0
•
..J ..J
0 .0'
O. I ~t
O'~~ii
&AU!IITY
. .....00 100-4. .
'o~-
2-70
2 .... 0.000...
1"'.0
.-u
1 0~U . ; : - ....
O=-I
o .~~
-
" ,0
"'-2.1
-e
7
~~-. 1.'
O.OO~-
&0-'"
•
I .'
0.004-
1O-2H
CJ)
":::"-",
I
. ...
0.021
as
:I:
I-
6.
a..
c,
0
0
w w W c IW ~
«
a: w c 0
MIXED WATER-DEEP CONFINED AQUIFER WODEl
1
"'£I:""..,OE
\.. Coo"O'O
~ ----'-....
r..... ·· _ t •••
LA
0 •
•
•/
- ; ~
/
BURIAL COMPACTION MODEL
8.
( (
E
l""U.'
EVjP0j"J'ON
, lI . , II GCI _ to ••
MIXED WATER-UNCONFINED COASTAL AOUIFER MODEL
( ( ( ( ( ( (
~~~~~THER"'Al CONVECTION
C (
C (
C (: (
( (
~
(
C ( (
(
c
Dolomites - Models and Ancient Dolostones
127
C' (',
o o C' o o c o o o o o o o o
o
c
o
o o o o o o o o G
marine replenishment of lagoon waters and seaward seepage of brines is the mechanism by which Mg2+ is continually transported to the sediment. The source of Mg2+ is the evaporated seawater itself. The dolomitization reaction probably occurs after a great deal of the gypsum (or anhydrite) has precipitated, sulphate has been removed, and the MglCa ratio approaches 9.0 possibly overshadowing the inhibiting effect of salinity (Morrow, 1978). The increase of the C0 3/Ca solution ratio that occurs during gypsum or anhydrite precipitation would also aid dolomitization. Modern examples of the seepage reflux ion hypothes is are small in scale and operate at or near the shoreline . Muller and Teitz (1971) and Kocurko (1979) have documented shoreline dolomitization in the Canary Islands and in San Andres Island by evaporated sea water that collects in ephemeral pools in the "spray zone", a few metres above high tide. The shallow, but permanent, evaporating pool at Bonaire (Deffeyes et al., 1965; Murray, 1969) is perhaps the best modern analogue of the large-scale reflux model envisaged by Adams and Rhodes (1960). The Mg2+ content of refluxing solutions in the reflux model is extremely high, but it is not clear what the rate of subsurface flow of these dense solutions is likely to be. Deffeyeset al. (1965) calculated a rate of seaward reflux flow of 2 x 10-6 cm3/s/cm 2 at Bonaire, but subsequent work (Murray, 1969) showed that the refluxing brines flow seaward for only a few weeks each year and most of this flow is confined to more permeable conduits bypassing most of the limestone. Model studies or computer simulations might shed light on the feasibility of large-scale reflux scenarios such as that envisaged by Adams and Rhodes (1960). At present, the dynamics of water movement beneath and adjacent to evaporitic shelf lagoons is unknown and a potentially fertile field of investigation. There are many examples of Canadian dolostones explained by variants of the reflux hypothesis (Figure 2). The Coorong and Sabkha Models. The Coorong Lagoon was known to be an area of dolomite formation since the later 1920s (Mawson, 1929), but was ignored until Alderman and Skinner (1957)inaugurated a series of investiga-
tions. Recentwork (Vonder Borch, 1976; Von der Borch and Lock, 1979; Von der Borch et al., 1975 and Muir et al., 1980) has established the Coorong Lagoon as a model for early near-surface dolomitization in many ancient sequences of aphanitic dolomite that are not associated with evaporites (Figure 3). The lagoon and a series of ephemeral lakes extend 200 km parallel to the south coast of Australia and are confined between Pleistocene fossil dunes that extend parallel to the coast. Surface drainage follows the existing topography, but the subsurface groundwater flows seaward perpendicular to the surface drainage (Figure 3). The ephemeral lakes are filled during the humid winter months by groundwater seepage and are evaporated to partial or complete dryness during the summer months. Extremely fine-grained calcian dolomite (yoghurt) mud and magnesite forms in the landward ephemeral lakes under the influence of continental groundwater or in the zone of mixing between seawater and continental water. This dolomite is in the form of 0.5-1 flm , amorphous , spherular aggregates and may have precipitated directly as gel (Von der Borch and Jones, 1976).Aragonite and calcite predominates in the Coorong Lagoon itself which is filled with seawater. Dolomite (and magnesite) precipitation in the ephemeral lakes is aided by groundwater dilution, by a decrease in sulphate concentration and by the high CO~- concentration of the alkaline groundwater. The source of Mg2+ ions has not been identified with certainty; dolomites forming close to the coast may have derived their Mg2+ directly from seawater, but dolomites forming farther inland probably utilize Mg2+ in the groundwater, some of which may have been derived from the weathering of basic volcanic rocks (Vonder Borch et al., 1975). In either case the delivery mechanism is the continual subsurface groundwater flow. Successive regressions of a Coorong-type system of lagoons and ephemeral lakes on a subsiding continental margin can build thick sequences of evaporite-free peritidal dolomite (Muir et al., 1980). Because of the predominant role of fresh groundwater, the Coorong Lagoon Model may be regarded as a specific type of the
more general ized mixed-water or dilution model. Only one documented Canadian example of this model exists (Figure 2), but more undoubtedly will be found. The other well-documented setting for penecontemporaneous dolomite, the sabkha environment , is restricted to those dolomites with accompanying evaporite minerals. Illing et al. (1965) in collaboration with geologists of the Shell Oil Company provided the first comprehensive report on the Holocene intertidal and supratidal dolomite that had replaced aragonitic sediments exposed on sabkha surfaces bordering the Qatar Peninsulaon the PersianGulf. This was followed by the studiesof Kinsman (1966), Butler (1969), de Groot (1973),Bush (1973), Hsu and Schneider (1973), Patterson (1972) and Patterson and Kinsman (1981, 1982) of the Abu Dhabi Sabkhas east of the Qatar Peninsula. Dolomite in the sabkhas of the Persian Gulf is most abundant as 1 to 5 flm euhedral crystals in the high intertidal zone close to the strand line and is strongly controlled by the network of flood channels across the sabkhas . Storm driven flood tides reach furthest inland along these channel courses (Patterson, 1972). The frequency of flooding decreases landward across the sabkha but the MglCa ratio of the floodwaters rises uniformly landwardby means of gypsum precipitation so that the zone of optimum dolomite formation is less than a kilometre wide straddling the boundary between high intertidal and supratidal areas (Patterson, 1972; Bush, 1973; and Figure 3). The dense Mg2+-bearing hypersaline floodwater brines sink downward and flow seaward through the sediment by seepage refluxion . Dolomitization of the underlying intertidal and subtidal sediments occurs to a depth of 2 to 3 m beneath the sabkha surface in regions landward of the continuous algal mat (Patterson, 1972; Bush, 1973; de Groot, 1973). Hsu and Siege nth alar (1969) , Hsu and Schneider (1973) and McKenzie et al. (1980) have developed an alternative model for sabkha dolomitization based on a process which they term evaporative pumping. In this process, a continual flow of seawater moveslandward through the sabkha sediments to replace groundwater lost by evaporation at, or near the sabkha surface. This flow
c 128
Diagenesis
Geoscience Canada Reprint Series 4
c (
c
pattern is the reverse of seepage refluxion (Figure 3), butthe field evidence for evaporative pumping in the sabkhas of the Persian Gulf is contradictory (see Hsu and Schneider, 1973; McKenzie et al. , 1980; Patterson, 1972; Patterson and Kinsman, 1981). It is unlikely that evaporative pumping has contributed greatly to sabkha dolomitization because evaporation of sabkha groundwater is slowed considerably by a number of factors such as absorption of water at night and high relative humidity ofthe overlying air mass. Consequently, this mechanism does not provide the volume of flow of ion-bearing groundwater necessary for dolomitization. The Sabkha Model may be regarded as a specific example of the Reflux Model modified slightly by the process of evaporative pumping. Many Canadian examples of dolomitization by the Sabkha Model have been identified (Figure 2). The Mixed·Water Aquifer Models. These models arose in the period following 1970. Before that time, there were a few reports of dolomite that formed in low salinity environments such as in lakes (e.g., Klahn, 1928) or in caves (e.g., Hoiland et al., 1964), but Hanshaw et al. (1971) documented the first large-scale system of dolomitization under the influence of fresh groundwater in the deep confined Tertiary carbonate aquifer of Florida (Figure 3). They emphasized the role of active groundwater circulation in supplying marine-derived Mg2+ for dolomitization. Land (1973) argued convincinglyfrom field relationships and petrographic and chemical evidence, that the Middle Pleistocene reef rocks of Jamaica were dolomitized penecontemporaneously by the interaction of marine and fresh groundwater in a shallow coastal aquifer. Ward and Halley (1984) provide similar examples of mixing zone dolomitization in the Upper Pleistocene of the northeast coastal limestones of Yucatan. Land (1973) suggested that many ancient platform dolomites which lack evaporites and are depleted in both trace elements (e.g., Na+, Sr 2+) and heavy isotopes ('80, 13C) may have formed in a similar manner. This suggestion has been followed recently by many authors (e.g., Badiozamani, 1973; Land et al., 1975; Veizer et al., 1978; Randazzo and Hickey, 1977; Dunham and Olson, 1980; Choquette and
Steinen, 1980). In most of these examples, thick platform dolomite sequences consist of many individual regressive carbonate sediment wedges that have been penecontemporaneously dolomitized by the shallow coastal freshwater aquifers that accompanied their seaward progradation (Figure 3). The vast differences in the chemical and isotopic composition between dolomites that have formed under the influence of freshwater and dolomites that have formed from hypersaline brines has led to an emphasis on these types of data as criteria in choosing between the Mixed-Water Aquifer Models and other models. However, uncertainties in the interpretation of chemical and isotopic data often preclude their use in model discrimination in the absence of other data (Land, 1980). In the Mixed-Water Aquifer Models dolomitization occurs within the zone of mixing of fresh groundwater with phreatic seawater. The Mg2+ ions for dolomitization are derived primarily from seawater and the delivery of mechanism is the continual circulation of seawater induced by the flow of fresh groundwater (Land, 1973). Dilution of saline solutions causes slow precipitation and a lowering of the sulphate concentration which favour dolomite precipitation (Folk and Land, 1975 ; Kastner, 1984) and the high CO~- concentration in many dilute continental groundwaters may also promote the precipitation of dolomite (Lippman, 1973; Morrow, 1982). These models have been used to explain the origin of many Canadian dolostones (Figure 2). The Burial Compaction Model. Compaction of fine-grained sediments during burial involves the progressive expulsion of pore water. Commonly, large shale masses contain carbonate reef bodies and are underlain by or are bordered by thick shelf carbonate sequences. A part of the Mg2+-bearing compaction water may pass through adjacent limestones and cause dolomitization (Figure 3). The Mg2+ required for dolomitization is derived primarily from the pore water. IIling (1959), Jodry (1969) and Griffin (1965) were early proponents of this model (Figure 2). The popularity of the Burial Compaction Model waned for a long period because of the upsurge in studies of Holocene penecontemporaneous dolo-
mites . But recent developments in the understanding of diagenetic changes in clay minerals and associated amorphous material, and the parallel maturation of organic material to hydrocarbons during burial has led to a renewed interest in this model (Davies, 1979; Mattes and Mountjoy, 1980). Clay minerals undergo a sequence of depth-related mineralogical changes that provides an additional source of pore water and of a variety of ions, including Mg2+, released from clay minerals to the pore fluids. These transformations have been cited as additional causes for burial-compaction dolomitization. There is little doubt that the kinetic inhibitions that bedevil the formation of dolomite at near-surface conditions are greatly reduced in deep burial environments largely because ofthe increase of temperature with depth. Also, the time available for dolomitization is much greater in deep subsurface environments than in the near-surface environments of early dolomitization (Mattes and Mountjoy, 1980). However, the problem of the supply of magnesium for dolomitization is an important constraint in deep burial environments. About 5.5 x 10-3 moles of Mg2+ are required to dolomitize 1 cm3 of calcite at 10% porosity. If this Mg2+ is supplied by compaction of adjacent shales and if the pore solution has the composition of seawater, then about 32 cm 3 of shale, compacted from 70% down to 2% porosity, are required to provide this amount of Mg2+. Th is assumes that all the water of compaction passes through the carbonates. In reality more than half of shale compaction water is lost vertically upward into the body of water from which the shale was deposited and most of the remainder passes downward (Magara, 1976, p. 92-95) . On a basin-wide scale , therefore, a more reasonable estimate would be that on the order of hundreds of cubic centimetres of compacted shale are required for the dolomitization of 1 cm 3 of laterally adjacent limestone. Therefore, shale compaction water is insufficient to form large masses of dolomite such as the Paleozoic platform dolomite sequences of Western Canada and elsewhere. Open space cementation imposes an even more extreme limitation on the volume of solution required to form a given amount of dolomite. This is apparent
c c c ( c c c c c c c c c c c c c c c c c c C
E C C
C C C
c:
C
C C C C C C. C
r Dolomites -
Models and Ancient Dolostones
129
r:
o r:
o r-t
o o u
o o
o o
o o .-... U
o r-. ....,
,.
o
when one considers that most solutions saturated with dolomite contain the equivalent of only about 10-5 moles of dolomite per litre of solution. In environments of burial greater than 2000 m, the transformation of montmorillonitic clays to illite is accompantedbythereleaseot'Sr-.Fes" ,Ca 2 + , and Mg2+ ions to the pore solution so that authigenic ankerite, ferroan dolomite and chlorite are precipitated as cements in some sand-shale sequences (Boles, 1981). In many instances, however,there is an insuff icient volume of fluid at these depths to move these ions laterally and to participate in the solution of large amounts of limestone and contemporaneous precipitation of dolomite. Shale porosity is less than 20% at burial depths exceeding 2000 m and processes such as aquathermal pressuring or the release of structured water from clays donot add significantly to this volume of interstitial pore water (Magara, 1976). The existing evidence strongly indicates that the late diagenetic release of ions from clay minerals is accompanied by local cementation rather than the long distance transportation of ions to shelf carbonates facing shale basins. Another factor limiting the dolomitizing potential of these deep burial diagenetic solutions is the fact that Ca2+ ions are released to solution in amounts nearly equal to, or greater than Mg2+ (Boles, 1981; Foscolos and Kodama, 1974).There is little, if any,excess Mg2+ available for dolomitization . In the most favourable situation, one might expect the void spaces in porous carbonates facing shales (e.g., Wong and Oldershaw, 1981) to be partly occluded with ferroan dolomite, but little actual dolomitization of pre-existing limestone. Although more work needs to be done to quantify the many variables involved in the Burial-Compaction Model, as a first approximation it does not appear that regional bodies of dolomite can form by this mechanism. The observation that many deeper water slope and basinal carbonates commonly are not dolomitized even where they are overlain by thick shale masses tends to confirm this impression (Wilson, 1975). Small amounts of dolomite may be commonly associated with this model, but in general the greater ease of dolomitization at the higher temperatures of deeper burial is countered by the pro-
gressive loss of the raw material for dolomitization (i.e., the pore fluids and their ions) as burial compaction proceeds. Nevertheless, some workers have suggested that large volumes of marine-derived basinal fluids in large basins can be focussed through more porous limestone layers by post-depositional regional ground-water systems dominated by gravity flow (Gregg , 1985). This scenario might be regarded as an extension of the simple burial compact ion model in that connate burial compaction fluids are mobilized by regional ground-water systems after uplift of the region above sea level. The Hydrothermal-Convection Model. This model has emerged from the recent application of the study of fluid inclusions to the origin of regional dolomite bodies in Canada. The essential features of this model were outlined by Aulstead and Spencer (1985) where they suggested that dolomite cements in the Keg River Formation of Alberta precipitated from evaporitic brines of the Elk Point Basin which had been recirculated by thermal convection from great depths. Crustal involvement ofthe . evaporitic brines that precipitated the regionally extensive Manetoe Dolomite ofthe NorthwestTerritories was inferred by Morrow et a/. (1986) on the basis of strontium isotope data which showed that the brines probably had interacted with crustal rocks. In this model, dense hypersaline brines that have migrated to great depths in the earth's crust are recirculated to shallow depths by thermal convection where they can dolomitize porous limestones. Dolomite is supersaturated in such solutions, but the exact composition of these solutions can only be inferred from fluid inclusion data or by comparison with evaporitic brines of other settings (Figure 3). Variants of this model have the potential to explain the occurrence of many regional late-stage dolomites, such as the late-stagedolomitesin some limestone petroleum reservoirsof Saudi Arabia (Broomhall and Allan, 1985). Other Settings for Dolomitization. The Organogenic Sea-floor Model for penecontemporaneous to early dolomitization (Figure 3), which has emerged as a distinct model in only the past several years, is unusual in several respects. In this model, dolomite nodules and beds form by precipitation from
marine-derived fluids only a few metres beneath the sediment-water interface in offshore deeper marine continental margin settings (Baker and Burns, 1985). Unlike other models for early or penecontemporaneous dolomitization, the magnesium required for dolomitization is supplied primarily by diffusion from the overlying water massin a deep marine setting and dolomitization takes place in fluids with a nearly normal marine Mg/Ca ratio and salinity. The intrasediment diagenetic environment is strongly reducing leading to sulphate depletion by sulphate reduction and the oxidation of organic material. The Organogenic Sea-floor Model derives its name from the fact that carbonate ions generated from the organic material are incorporated into the precipitated dolomites. Both the reduction in sulphate concentration and the generation of carbonate alkalinity are thought to enhance dolomitization (Kastner, 1984 ; Baker and Burns, 1985). This model for dolomitization is less significant in a volumetric sense but, as Baker and Burns (1985) pointed out, it is important as an indicator of organic-rich source rocks for hydrocarbon generation. This model awaits application to many ancient deeper water dolostones. Solution Cannibalization (Goodell and Garman, 1969) refers to the derivation of Mg2+ for dolomitization from the dissolution of magnesian calcite and reprecipitation of low-magnesian calcite. Some examples of partial dolomitization, such as dolomitized burrow fillings (Kendall, 1977) have been explained by solution cannibalization, but for more complete dolomitization other sources of Mg2+ are required (Hsu , 1966). Recently, the phenomenon of pressure solution along solution seams and stylolites has been cited as a major cause of dolomitization (Logan and Semeniuk, 1976; Wanless, 1979). The source of Mg2+ionsin pressuresolution dolomitization is by solution cannibalization of Mg2+ from pre-existing magnesian calcite during pressure solution (Wanless, 1979). Complete dolomitization is thought to result from the concentration of dolomite along solution seams as a relatively insoluble residue. Intriguing as this dolomitization model is, it suffers from the difficulty in reconciling the deep burial setting of most stylolites (e.g., Mossop, 1972) with the
c
130
common pre-burial conversion of magnesian calcite to low-magnesiancalcite. For large amounts of dolomite other sources of Mg2+ are required. Many coarsely crystalline, white sparry dolostones that have cross-cutting relationships with the enclosing strata have been designated as "tectonic" or "hydrothermal" dolomites in older studies, particularly in instances where these dolostones occur near known faults (e.g., Sandford, 1962). However,the common absence of other hydrothermal minerals implies that many of these dolostones have other origins. The relationship of some dolostone masses to faults may merely reflect the fact that ground-water movement in aquifers tends to be much more rapid along subsurface fault zones (i.e., ground-water cascade zones), enhancing diagenesis. ANCIENT DOLOSTONES CRITERIA FOR RECOGNITION OF DOLOMITIZATION MODELS Criteria that may guide the researcher in his choice of dolomitization models to apply to ancient dolostones can be divided into three broad categories stratigraphic, petrographic and geochemical criteria. Few, if any, of these criteria are unequivocal in themselves and it is only after the accumulation of muchcomplementarydatathat a truly definitive choice can sometimes be made. Stratigraphic and Facies Criteria Size and shape. Small intraformational bodies of dolomite may have local sources of Mg2+, derived from the sediment itself as in the Solution-Cannibalization Model (e.g., Kendall, 1977; and Gebelein and Hoffman, 1973), but large masses, such as entire formations or major parts of a formation, invariably require external sources of Mg2+. Dolostone masses that conform to rock stratification may indicate that dolomitization occurred before burial, but nonconformable dolostone bodies that intersect primary stratification and facies at high angles strongly indicate that dolomitization is post-lithification and post-burial, particularly where these dolostones are associated with solution-collapse breccias (e.g., Collins and Smith, 1977; Morrow, 1975). The form and location of some dolostone masses conform to known faults indicating a late post-burial model of origin (e.g., Skall, 1975; Jones, 1980).
Diagenesis
Associated evaporites and shales. Laterally adjacent or overlying evaporites provide sources of Mg2+-bearing hypersaline solutions capable of dolomitizing by seepage-reflux ion particularly where shelf carbonates separate evaporite bodies from more open marine sediments (e.g., Skall, 1975; Wilson, 1975, p. 317). Dolomitized reefal masses encased in evaporites is a
Geoscience Canada Reprint Series 4
common situation where early diagenetic brine reflux may be suspected (e.g., Sears and Lucia, 1980). Late diagenetic movements of Mg2+bearing connate brines derived by expulsion during burial compaction from evaporite masses into laterally adjacent carbonates (Jodry, 1969; Broomhall and Allan, 1985), or overlying carbonates (Jones, 1980) may also
-
O.5mm
Figure 4 A hand specimen of a bright yellow, supratidal flat dolomicrite that has undergone synsedimentary folding after partial lithification, from the Rowatt Formation, a Precambrian unit in the Belcher Islands, Hudson Bay. Thicker laminae are about one centimetre thick. The photomicrograph on the left shows the aphanocrystalline character of this type of early replacement dolomite with preservation of microscopic detail. (Photographs suppliedby B.D. Ricketts).
c ( c c c c ( c c c c c c c c c c c c c c c c c c (
C
E C C C C C
c:
C l Figure 5 Medium to coarsely crystalline "late" diagenetic sucrosic dolomite in the Ordovician Allen Bay Formation of the Arctic Islands. Macroscopically identifiable Halysitid-type colonial corals populate this outcrop from the south coast ofDevon Island. Thispreservation of macroscopic textures indicates that no macroscopic voids were created before or during dolomitization. Microscopic details of these corals have been destroyedby dolomitization that formed the hypidiotopic fabric of subhedral crystals shown in the photomicrograph in the upper left with a field of view of 1.2 mm. (Photographs supplied by U. Mayr).
C C C
C (
C
c c o C' r
o (J (
o o o
c
o o
Dolomites - Models and Ancient Dolostones
cause dolomitization. Similarly shales adjacent to dolomitized carbonates are also suspect as sources of dolomitizing solutions during their compaction. Unconformities. Dolostone bodies are commonly developed beneath regional unconformities (Sonnenfeld, 1964). Many of these dolostones are best explained by the Mixed-Water freshwater aquifer (e.g., Collins and Smith, 1975 and Figure 3). In other instances, such unconformities are considered merely to represent a period of karst development that enhances the porosity of the underlying carbonates so that dolomitizing solutions, whatever their origin , subsequently were localized in the strata beneath the unconformity (e.g., Skall, 1975). Facies development. Models for penecontemporaneous dolomitization such as the Sabkha and Coorong Models are verified most readily by the identification of associated facies charac-
131
teristic of these environments. The sabkha depositional model with its peri tidal and suprat idal vertical sequence of dolomite that culminates in an interval of displacive anhydrite, forms a very distinctive cyclic sequence of lithologies and sedimentary structures (e.g., Wood and Wolfe, 1969).Similarly, environmental conditions of the Coorong Model generates an evaporite-free, cyclic sequence of very finely crystalline, penecontemporaneous dolomite with different sedimentary structures (Muir et al., 1980). In many instances, reef buildups or carbonate mud mounds are facies that coincided with paleotopographic "h ighs" and thus were susceptible to dolomitization by the Reflux Model or by the Mixed-Water Model after exposure by relative changes in sea level (e.g., Eliuk, 1978; Sears and Lucia, 1980;Exploration Staff, Chevron Exploration Staff Ltd., 1979). Paleogeography. There is a strong
G
o
c o o u
B
o o o (
A ,.-
Figure 6 Megacrystalline white sparry dolomite of the Manetoe Facies (Formation) in the Nahanni Formation of the Yukon Territory. Near centimetre-sized crystals of white sparry dolomite have partly filled spaces created by solution-collapse in the core slab (A). Photograph B, at the same scale , is the exposedinner surface of one ofthese vugs displaying crystals with markedlycurved crystalfaces. This type ofdolomite is totally fabric destructive. This sample is a core specimen taken 11,707.0 feet below sea level in the gas-producing Columbia et al. Kootanee/ee H-38 well .
correlation between paleogeography, as expressed by rock facies, and dolomitization reflecting a common tectonic control (Wilson, 1975; Dunham and Olson, 1980). Shelf areas tend to be exposedduring regression and become the sites for freshwater recharge and mixed-water dolomitization. Deeper water carbonates which are never exposed, are not subject to this type of dolomitization. Thick sequences of shelf dolomite may be built up as a consequence of multiple regressions. The seaward progradation of each regressive cycle is accompanied by seaward expansion of the coastal aquifer so that a dolomitization front also moves seaward as progradation cont inues (Figure 3). Petrographic and Textural Criteria Crystal size. Previous rev iews (e.g., Folk, 1973)have emphasized that petrographically, dolostones are of two basic types separated by a crystal size boundary of 10 to 20 ILm. 0010stones more finely crystalline than this were characterized as primary or very "early", pre -burial replacements in which sedimentary and organic fabrics are well preserved (Figure 4), whereas more coarsely crystalline dolostones that occur mainly as stratform replacements of limestone with poorly preserved fossil remains were considered to be "secondary", "sucrosic" or "late replacement" dolomites (Figure 5). A third major petrographic variety of dolomite, which has received considerable attention in recent years, is the white, very coarsely crystalline dolomite or megacrystalline (i.e., ranging from millimetres to centimetres) that, in many places, is associated with lead-zinc mineralization (Figure 6). Much of this type of dolomite is void-filling and displays curved crystal faces (e.g., Radke and Mathis, 1980). Dolomicrlte. The very finely crystalline fabric preservation type of dolomite in many ancient dolostones is directly analogous to the fine-grained dolomitic sediments that form as early replacements of aragonite in the intertidal and supratidal areas of modern carbonate tidal flats such as in the Caribbean (dolomitic crusts), in the sabkhas of the Persian Gulf and in the Coorong region of Australia . This petrographic type of dolomite usually is inferred to havereplaced aragonite mud before burial (e.g., Schmidt, 1966 and Figure 4).
c ( Diagenesis
132
Medium crystalline sucrosic dolomite. The origin of the finely to medium crystalline, suerosie dolomites (generally hypidiotopic - an aggregation ofsubhedra) ismuch moreproblematic because of the apparent absence of modern analogues. However,this type of dolomite does occur as replacements of Pliocene-Pleistocene limestones on some Caribbean Islands (Land, 1973; Supko, 1977; Sibley, 1982).Unlike similar dolostones in more ancient rocks, the geological history of these Caribbean dolostones is known with a high degree of certainty and this history in conjunction with other data, suggests that most, if not all, of the Caribbean dolomite originated by the Mixed-Water Model (e.g., Land, 1973). Crystals of this type of dolomite commonly contain cloudy centres surrounded by clear rims and the Caribbean dolostones are no exception. Sibley (1980)suggested thatthis phenomenon, in the Caribbean examples at least, is due to the evolution ofthe diagenetic pore fluids from a state of near calcite saturation to a state of calcite undersaturation, as grain growth continues under the influence of mixedwater diagenesis. it is likely that the interlocking "hypidiotopic" texture of subhedra and anhedra that characterizes ancient sucrosic dolostones is the result of grain growth from an initial state where very finely crystalline dolomite euhedra are scattered throughout. These crystals grow and eventually join along irregular compromise boundaries (e.g., Morrow, 1978). The euhedral, micrometre-sized dolomite rhombs that are scattered in the recent sediments forming Andros Island (Gebelein at a/., 1980) may record the beginning of a progression toward atotally dolomitized fabric of sucrosic dolostone (Figure 7). Although many ancient, stratiform, platform dolostone sequences of sucrosic, dolostone have been interpreted to be the consequence of mixedwater diagenesis (e.g., Randazzo and Hickey, 1978 ; Dunham and Olson, 1980), the Reflux Model has been invoked in situations where sucrosic dolostones are adjacent to evaporites (e.g., Sears and Lucia, 1980) or adjacent to shales (Mattes and Mountjoy, 1980). Megacrystalline white dolomite. This type of dolomite is less abundant than the more finely crystalline varieties, but is economically important as a
host rock for lead-zinc deposits (e.g., Presqu'ile Dolomite) and hydrocarbon accumulations (e.g., Manetoe Dolomite). The origin of megacrystalline, white dolomite is the least certain of the three major petrographic types because it has not been observed in recent sediments or in Quaternary rocks. However, the universal association of white dolomite, both as a void-filling and as a replacement mineral , with solutioncollapse breccias and solution cavities indicates that it formed after lithification and probably after at least some burial. Several workers have emphasized the common association of lead-zinc mineralization and white dolomite (e.g., Beales and Hardy, 1980; Davis, 1979; Radke and Mathis, 1980) suggesting that both the sulphides and the dolomite precipitated together by means of a single, overall-sulphate reduction reaction attemperatures between 600 and 150°C (Barton, 1967; Radke and Mathis, 1980). But any theory linking the origin of leadzinc mineralization with the associated white sparry dolomite in specific areas must also account for the widespread distribution of dolomite beyond the mineralized area (e.g., Skall, 1975; Gorody, 1980). It must also account for the fact that mineralization in most examples post-datedthe emplacement of much of the white dolomite, as, for example in the Pine Point deposit (Skall, 1975) and in the Viburnum Trend of the Mississippi Valleydistrict (Gorody, 1980;Hagni and Trancynger, 1977). The lead-zinc deposits in the St. George Formation in Newfoundland are a notable exception to this generalization (Collins and
Geoscience Canada Reprint Series 4
Smith, 1975). Clearly, the processes responsible for the emplacement of widespread white dolomite masses beyond the mineralized areas are not dependent on the processes of mineralization, although in some places the converse may be true (e.g., Collins and Smith, 1975;Davis, 1977). Probably only the relatively small amount of white sparry dolomite that is actually intergrown with the ore mineralswascoprecipitated with these minerals. Variations of Mixed Water Aquifer Models have been cited asthe modes of origin of white sparry dolomites in many studies (e.g., Lyle, 1977;Gorody, 1980; Morrow, 1975) but the Reflux Model (Davis, 1977) and the Burial Compaction Model (Griffin, 1965; Jackson and Beales, 1967) have also been cited, in addition to the model based on the late diagenetic sulphate reduction reaction (Radke and Mathis, 1980). Recently, also, the Hydrothermal Convection Model has been adopted to explain the high temperatures of fluid inclusion homogenization in some megacrystalline white dolomites (Aulstead and Spencer, 1985; Morrow et a/., 1986). Unfortunately, it is evident that little overall agreement exists between workers concerning the origin of megacrystalline white dolomite either in general, or with regard to specific occurrences. Paragenesis, texture and fluid inclusions. Detailed petrographic relationships and textural data in some instances can be usedto inferthe timing of dolomitization with respect to early events such as the recrystallization of
THE NATURAL HISTORY OF DOLOMITE CRYSTALS
c
¢
o
¢¢¢¢¢¢ ¢ ¢¢¢
I
1 PARTIAL EARLY DOLOMITIZATION - BY SCATTEREO MICROMETRE-SIZEO DOLOMITE EUHEORA THAT ARE INFLUENCED BY THE CHEMISTRY OF MARINE FLUIDS (NA, SR: 500-1500 PPM, 61RO: -1 to +4)
~
~
C C
C C (
C
C C C C
C C C C
C C
C C C C C C C
C C
E C C C C (
BURIAL
. ..... .. . .
C
2 COMPLETE LATE DOLOMITIZATION PROGRESSIVE GROWTH OF TRACE ELEMENT DEPLETED, 16 0 ENRICHED SYNTAXIAL RIMS AROUND EARLY DOLOMITE EUHEDRA RESULTS IN TOTAL OOLOMITIZATION UNDER THE INFLUENCE OF DILUTE SUBSURFACE SOLUTIONS. POLYHEDRAL CRYSTALS TEND TO HAVE CLOUDY CENTRES BUT CLEAR RIMS.
Figure 7 Many dolomites conform to this type of diagenetic history. The dilute subsurface solutions are probably mixed-water or burial-compaction in origin. It is not known whether trace element depletion in crystal rims could also be the result of slow precipitation from more saline solutions.
e c c
c c
c ( c c
c (
C C'
n o (
o o ( o o o o o o o o o o 0, o o o o o o o o o
Dolomites - Models and Ancient Dolostones
aragonite and high-magnesian calcite (e.g.•Schmidt. 1965; Kendall , 1977;Sibley, 1980) or to later events such as the development of stylolites (e.g., Macqueen and Thompson , 1978; Wong and Oldershaw, 1981). Iron-sensitive stains (Figure 8) and cathodolominescent techniques (Choquette and Steinen , 1980) can reveal additional intracrystalline textural details. Early dolomitization is often controlled by original permeability and the commonly noted preferential dolomitization of finer grained sediments (Murray and Lucia, 1967) is probably due in large part to their great abundance of nucleation sites. Increasingly, multiple episodes of dolomitization are being recognized based mainly on the recognition of a wide variety of petrographic types of dolomite within the three-fold petrographic classification outlined above. Many of these varieties have been
related to the operation of specific dolomitization models (e.g. , Sears and Lucia, 1980; Mattes and Mountjoy, 1980), so that a given dolostone may have been influenced by the operation of a succession of such models . However, it is not possible at present to relate any of these petrograph ic types to any particular model in the absence of other data, butthepetrographic character of a given dolomite may help to limit the choice of models. Analysis offluid inclusions in coarsely crystalline dolomites (Figure 8) has recently been shown to be effective in determining precipitational temperatures, salinities and some compositional aspects of dolomitizing solutions (Aulstead and Spencer, 1985; Broomhall and Allan, 1985). This method is limited to those coarsely crystalline dolomites with suitable inclusions. Some recent studies have ind icated
o l
-,
r. Figure 8 Staining textures and fluid inclusions in megacrystalline dolomites. (A) Successions of dolomite cement zones (numbered 9 to 13) in dolomite that fills a
.>:
"
.-
.
gastropod pore in the Swan Hills Formation. Zones are revealed by staining with potassium ferricyanide. (location: 10-3-67 at 8807 feet; photograph supplied by Christian Viau). (8) Two phase aqueous liquid-vapour inclusions (arrows) in dolomite cement of the Manetoe Facies. These elongate inclusions are commonly less than 10 ?m long (location: 62.506°N let.; 122.991 oW long. at 710.3 m; photograph supplied by Kathy Aulstead).
133
that there is a correlat ion between the degree of dolomite crystallin ity and the temperature at which the dolomite precipitated. Gregg and Sibley(1984)found that xenotopic crystal textures resulted from limestone dolomitization at temperatures greater than 200°C. Geochemical Criteria Geochemical parameters. Geochemical parameters include compositional variations of the major elements Ca and Mg and the degree of concomitant crystal disorder, variations in the content of minor and trace elements such as Fe, Sr, Na, Mn and other less commonly studied elements (Ba, Si, S, Zn, Pb),and the isotopic composition of carbon and oxygen . A considerable body of literature has arisen in the past two decades dealing with these parameters. The most effective of these studies have combined stratigraphic and petrographic data with geochem ical data. Stratigraphic and petrographic data provide a foundation of more readily accessible facts upon which a framework of interpretations involving geochemical data may be built. Compositional variations of major elements and crystal disorder. Compositional variation of major elements in dolomites is commonly expressed as a mole percent of CaC0 3 (Figure 9). Calcium is almost invariably present in excess of Mg because of the relative ease with which Ca is incorporated into growing dolomite crystals. It is recognized that the degree of enrichment of Ca in dolostones can reflect the Mg/Ca ratio of the precipitational solution (e.g., Fuchtbauer and Goldsmith, 1965; Morrow, 1978; Lumsden and Chimahusky, 1980). Typically, dolomicrites or finely crystalline dolostones that are assoc iated with evaporites have Ca contents that are lower than for finely crystalline dolostones not associated with evaporites (i.e., Groups 2 and 3 in Figure 9). Dolomites in both of these groups are early diagenetic in that dolomitization occurred within the influence of the depositional environment (e.g., Patterson, 1972; Morrow, 1978; and Lumsden and Chimahusky, 1980). The relatively high Mg/Ca ratios of solutions in evaporitic sett ings (e.g., Reflux and Sabkha Models) induces the formation of nearly stoichiometric dolomites, because the abundance of Mg ions compensates for their relative difficulty in precipitating from solution to
c
Diagenesis
134
Geoscience Canada Reprint Series 4
c c C
form magnesium-bearing compounds (see Morrow, 1982). On the other hand, the lower MglCa ratios of solutions in more humid, non-evaporitic settings (e.g., Mixed-Water and Coorong Models) results in the precipitation of more calcium-rich dolomites. Medium to coarsely crystalline ancient dolomites tend to be more stoichiometric (Group 1 in Figure 9). One explanation is that scattered, early diagenetic dolomite crystals in incompletely dolomitized limestones continue to grow very slowly in late diagenetic subsurface environments, under the influence of very dilute solutions and possibly elevated temperatures . Slow precipitation may overshadow the influence of low solution MglCa ratios and cause the growth of the trace element depleted, more stoichiometric, coarsely crystalline rims that enclose Ca and trace element-enriched cores of dolomite crystals (e.g., Land et al., 1975, Morrow, 1978). The net effect of continued crystal growth during later diagenesis is the formation of dolostones whose bulk composition approaches that of stoichiometric dolomite (e.g., Fritz and Jackson, 1971, and Figure 7). Certainly this interpretation explains the depletion in most trace elements that commonly occurs toward the periphery of dolomite crystals and implies that they are the products of variants of the Mixed-Water Aquifer or the Burial-Compaction Models. However,it is now known from fluid inclusion evidence that some coarsely crystalline late stage stoichiometric dolomites were precipitated from hypersaline rather than dilute solutions (e.g., Morrow et al., 1986). The reason for their stoichiometric composition is not appar. ent, but if these precipitating brines had elevated MglCa ratios like the evaporitic brines on sabkhas then they might precipitate stoichiometric dolomites in a similar manner. Some data concerning the departure from ideal cation ordering within dolomite crystals has been gathered, particularly for dolomites forming in modern environments (e.g., Patterson, 1972). Unfortunately few ancient dolostones have been analyzed for this parameter. Comparisons of the degree of cation ordering as determined by the analysis of superstructure reflections on x-ray patterns with the degree of Ca enrichment might impart some insight
into how closely these parameters are correlated. For example, crystal disorder not associated with Ca enrichment would reinforce the conclusion that finely crystalline, stoichiometric dolomite was rapidly precipitated from saline solutions with a high MglCa ratio. Trace element variations. The concentration of trace elements in dolomite ideally provides a means for estimating the composition of the precipitational medium because the fractionation of an element , such as Sr, from a liquid phase into a solid phase is proportional to its concentration in the liquid phase (amongst a variety of other factors). This implies that dolostones with greater concentrations of trace elements, such as strontium or sodium, precipitated from more saline solutions; but quantitative interpretation of solution compositions based on dolomite trace element data are not feasible at present. Estimates of low temperature partition coefficients based on high temperature experiments (Katz and Matthews, 1977; Jacobson and Usdowski, 1976) or derived from element concentrations in Holocene dolomites (Patterson, 1972; Land and Hoops, 1973; Behrens and Land, 1972) have not been satisfactory (Land,
1980; Veizer et a/. , 1978) and direct determinations from precipitation experiments are of course not possible at reasonable temperatures. The extreme difference in the contents of trace elements in ancient versus modern dolomites is the salient fact to emerge from trace element studies.The two most commonly studied elements, Sr and Na, are present in concentrations approaching, orgreaterthan, 1000 ppm in modern dolomite but in ancient dolomites, they commonly are less than a few hundred ppm (Land et a/., 1975; Mattes and Mountjoy, 1980). Other elements such as S, K, AI, Ba, Zn, Mn and Fe have been analyzed only sporadically in ancient dolostones and, little, if any data exist for modern dolomites. The phenomenon of intracrystal variations or zonations with respect to trace elements within ancient dolomite crystals is commonly observed, even in crystals smaller than 10 /Lm (e.g., Choquette and Steinen, 1980). Studies of element zonations in dolomite crystals have also concentrated on the elements Na and Sr, and in addition, there is some data for Fe and Mn. A general pattern exists of high Na and Sr contents in the cores surrounded by rims with much lower contents. (Fritz and
TRENDS IN DOLOMITE STOICHIOMETRY PRECIPITATIONAL CONDITIONS Mg :Ca
MOLE 010 CaC03
50
52
54
56
58
60
62
t
3:1
1
t
2·5 -5:1
i
<35
1
1
6-22:1
»35 I
t I
t
r
C C (
(
C C C C
C C C C C C
C C C C C C C C C E C C C C <-
e c l
Figure 9 A summary diagram incorporating data compiled by Morrow (1978), Mattes and Mountjoy (1980), and Lumsden and Chimahusky (1980) concerning the stoichiometry of dolomite both modern and ancient. Group 1 is composed of ancient sucrosic and sparry dolomites, Group 2 is composed of finely crystalline modern and ancient dolomites not associated with evaporites and Group 3 are finely crystalline modern and ancient dolomites associated with evaporites. Group 3 also may include some late stage sparry dolomites that contain hypersaline fluid inclusions .
C (..
c c
(
C
c
Dolomites -
Models and Ancient Dolostones
135
(\
o o o
c (
o
o
o o
o
o o o
o
o c
o o o o (,
o o o o 8 C
Katz, 1972; and Figure 7). On the other hand, Fe, and possibly also Mn, tend to exhibit an opposite zonation with higher concentrations toward the periphery of crystals. Bulk sample analyses, of course, are weighted averages of these zoned crystals. Some stud ies have shown slight statistical differences in the bulk trace element contents of a variety of dolostone types (Veizer et al., 1978). Any explanation of dolomitization must account for both the element zonation of ancient dolomite crystals, and their reduced trace element contents , particularly for the aphanocrystalline and fine- to medium-crystalline, sucrosic replacement dolomites that commonly have cloudy centres and clear rims. It is difficult to avoid the conclusion that as dolomite crystal growth continues, it does so under the influence of progressively more dilute solutions that dissolve pre-existing calcite and precip itate dolomite . It seems likely that the cores of ancient dolomite crystals, in largely or totally dolomitized rocks, represent individual crystals that formed in shallow early diagenetic environments under the influence of seawater or modified seawater solutions. Subsequent burial of these rocks into deeper subsurface environments may have been accompanied by the continued precipitation from more dilute solutions of dolomite overgrowths that are relatively inclusion-free, more stoichiometric and lower in Sr and Na (Figure 7). Such dilute and subsurface solutions are consistent with the operation of either the Mixed-Water Model or the Burial-Compaction Model. Ion filtration during burial compaction of shales can yield dilute solutions that may participate in the dolomitization of nearby carbonates, although there are limitations on the potential amount of dolomite that can be formed. The commonly observed high Fe concentration of up to several mole per cent in the rims of large crystals, particularly in those that have precipitated as open space fillings with complex zonation (as revealed by staining and cathodoluminescence, Figure 8) have been related qualitatively to diagenesis in the reducing subsurface phreatic zone, with fluctuations in the supply of Fe2+ (Wong and Oldershaw, 1981; Gorody, 1980). These observations are also consistent with either a Mixed-Water or BurialCompaction Model.
Coarsely crystalline , ferroan and non-ferroan, white, sparry dolomites directly associated with Pb-Zn mineralization or with hydrocarbons have also been interpreted to have been precipitated from saline host fluids , based primarily on the occurrence of very saline flu id inclusions in associated minerals (Radke and Mathis, 1980 ; Gregg, 1985) and in the dolomite crystals themselves (Aulstead and Spencer, 1985). Certainly subsurface prec ipitation of dolomite at the inferred temperature of greater 100 D C is less subject to kinetic inhibitions. However, it is puzzling that the few occurrences of white, sparry dolomite associated with Pb-Zn mineralization which have been analyzed systematically for their trace element contents (e.g., Fritz and Jackson, 1971) are characterized by low trace element contents in conformance with the general rule regarding trace elements in dolostones. Saline solut ions should precipitate dolomites markedly enriched in trace elements. This major contradiction concerning the origin of white megacrystalline, sparry dolomite cannot fail to stimulate further research and the correlation of trace element and fluid inclusion studies will undoubtedly playa key role in resolving this paradox. Distribution coefficients for some elements , such as strontium (Jacobson and Usdowski, 1976) have been determined for the high temperature (> 1DO D C) precipitation of dolomite which might enable quanititative modelling of trace element distributions. Trends of trace element concentrations within dolostone bodies have also been used as indicators of the direction of fluid migration during diagenesis (Land, 1980). This type of data may allow discrimination between models for dolomitization that require different directions offluid migration during dolomitization, such as the Burial-Compaction Model versus the Mixed-Water Model. Correlations between ratios of major and trace elements with stratigraphic position have been shown to be more useful in the discrimination between dolomitization models than the absolute concentrations of these elements themselves in some cases (Sass and Katz, 1982) . Oxygen and carbon isotopes. Land (1980) has recently summarized the status of stable isotope studies (0180 , 013C) of dolomite and has
pointed out the considerable difficulties in their interpretation. Qualitative interpretation of oxygen and carbon isotope data involves an appraisal of the influence of the variables of temperature and solution compositions . Higher temperatures and dilution by meteoric water favour the incorporation of the light 160 isotope into the solid phase so that dolomites that are greatly enriched in 160 may have been influenced by high subsurface temperatures (Mattes and Mountjoy, 1980) or by dilute 160 _ rich ground water (Land et al., 1975). The interplay of these factors is complicated by the probable existence of secular changes in the isotopic composition of ancient seawater (Veizer and Hoefs, 1976). One certainty with regard to oxygen isotopes in ancient dolostones is that their bulk 0180 is much lower than that of modern dolomite crystals (Land, 1980). Probably dolomite crystals are zoned with respect to their 160 content in a manner analogous to that described for trace elements. Although intracrystal variations in 160 have not been described, there does appear to be a correlation between increasing crystal size and decreasing alSOvalues for dolomite crystals (Fritz and Jackson, 1971; Land et a/. , 1975; Mattes and Mountjoy, 1980) which is consistent with the peripheral enrichment of 160 . This may corroborate the interpretation that the growth of dolomite crystals continues under the influence of progressively more dilute solutions or it may record progressive temperature increases during dolomite precipitation. Regional trends of 0180 within dolostones may also corroborate trends in their trace element distributions (Land et al., 1975). In contrast to oxygen isotopes, the ratio of stable carbon isotopes (013C) in most ancient dolostones is virtually the same as in Holocene dolomites with perhaps only a slight tendency for ancient dolostones to be enriched in 12C (Land, 1980). This may simply reflect the fact that the amount of carbon in dissolved species in natural solutions is very small compared to the amount of carbon in the carbonate rocks that are being replaced. There are a few dolostones enriched in 12C that may have been involved in bacterial reduction reactions.
c ( 136
SUMMARY AND CONCLUSIONS The description and classif ication of dolostones previous to about 1960wasa comparatively simple matter as dolostones were commonly classified as "primary" or " secondary" depend ing primarily on whether the sedimentary fabric of the rock was preserved or destroyed by dolomitization . The additional category of"hydrothermal" dolomite was used is still used (Jones, 1980) occasionally to characterize masses of coarsely crystalline, space-filling white dolomite. Little was understood concerning processes of dolomitization and consequently most dolostones were subjected to comparatively casual investigations in spite of their economic importance. The decades following 1960 were marked by the discovery of many examples of modern dolomitization in a variety of environmental settings (e.g., sabkha, ephemeral lake, deep carbonate aquifer) and the characterization of the chemistry of these environments has generated a greatly increased understanding of the chemistry of dolomitization. At the same time, new analytical techniques (e.g., x-ray fluorescence, electron microprobe, etc.) have permitted a better overall characterization ofthe mineral dolomite itself in both ancient rocks and modern occurrences. The rapid increase in knowledge of other aspects of geoscience such as fluid dynamics and chemical changes during the burial compaction of sediments and the hydrology and chemistry of coastal aquifers have had a profound influence on the development of theories concerning dolomitization . This multidirectional increase in knowledge has resulted in the development of a variety of dolomitization models and an increase in the number of criteria available to discriminate between models. The objectives of this review have been twofold. First, to outline clearly the models for dolomitization that have arisen and to review the criteria commonly employed in choosing between these models. The multitude of models for dolomitization that are now recognized are moderately to well defined. It seems evident to this reviewer that variants of the Mixed-Water Aquifer Models are the most widely applicable - the first among equals - although the Hydrothermal Convection Model may prove to have a similar broad applica-
Diagenesis
bility. The other models are certainly viable but perhaps not to the same degree and in less regional settings. The importance of a model such as the Burial-Compaction Model lies not in its potential for forming vast amounts of dolomite but rather in its capacity for modifying pre-existing porosity. Unfortunately, the criteria for choosing between models are not asdefinitive as could be desired and many of these criteria have been interpreted in more than one way. Indeed it could be argued that the same criteria in some dolostones such as the Presqu'ile Formation (or facies) have been used to support a succession of different models as these models successively rose in popular esteem. Hopefully, the time of fads is past and the choice of a model for dolomitization by present day workers is based on an equitable consideration of all available criteria and by practicing the spirit embodied in the phrase "multiple working hypothesis".
ACKNOWLEDGEMENTS Douglas Kent of the University of Saskatchewan and Alan Kendall of Amoco Canada Ltd. critically read the manuscript and offered many useful suggestions. I would also like to thank A.E. Foscolos (Technical University of Crete) and R.J. Spencer (University of Calgary) for discussions that contributed to this review.
REFERENCES Reflux Model Adams, J.E. and Rhodes, M.G., 1960,Dolomitizat ion by seepage reflux ion: American Association of Petroleum Geologists , Bulletin, v. 44, p. 1912-1920. Deffeyes, K.S., Lucia, F.J. and Weyl. P.K., 1965, Dolomit ization of recent and PlioPleistocene sediments by marine evaporite waters on Bonaire, Nethe rlands Antilles, in Pray, A.C. and Murray, R.C., eds., Dolomitization and Limestone Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 13, p. 71-88. King, R.H., 1947, Sedimentation in Permian Castile Sea: American Association of Petroleum Geologists, Bulletin, v. 31, p.470·477. Morrow, D.W., 1978, The influence of the Mg/Ca Ratio and salinity on dolomitization in evaporite bas ins: Canadian Society Petroleum Geologists, Bulletin , v. 26, p. 389-392 .
Geoscience Canada Reprint Series 4
Murray, R.C., 1969, Hydrology of south Bonaire , N.A. - a rock selective dolomitization model: Journal of Sedimentary Petrology,.v. 39, p. 1007-1013. BurIal-Compaction and Hydrothermal Conl/ection Models Aulstead, K.G. and Spencer, R.J., 1985, Diagenes is of the Keg River Formation, northwestern Alberta: fluid inclusion evidence : Bulletin of Canadian Petroleum Geology, v. 33, p. 167-183. Broomhall, R.W. and Allan , J.R., 1985, Regional caprock-destroying dolomite on the Middle Jurassic to Early Cretaceous Arabian shelf: Society of Petroleum Engineers , SPE 13697, p. 157-163. Gregg, J.M ., 1985, Regional epigenetic dolomitization in the Bonneterre Dolomite (Cambrian) , southeastern Missouri: Geology, v. 13, p. 503-506. Griffin, D.G., 1965, The Devonian Slave Point, Beaverhill Lake, and Muskwa Formations of northeastern British Columbia and adjacent areas: British Columbia Department of Mines and Petroleum Resources, Bulletin, v. 50, p. 90. IIling, L.V., 1959,Deposition and diagenesis of some upper Paleozoic carbonate sediments in western Canada: 5th World Petroleum Congress, New York, Proceedings, Section 1, p. 23-52. Jodry, R.L., 1969, Growth and dolomitization of Silurian reefs, S1. Clair County, Michigan: American Association of Petroleum Geologists, Bulletin, v. 52, p.957-981. Mattes, BW. and Mountjoy, E.W., 1980, Burial dolom itization of the Upper Devonian Miette Buildup, Jasper National Park, Alberta, in Zenger, D.H., Dunham , J.B. and Ethington, A.L., eds., Concepts and Models of Dolomitization: Society of Econom ic Paleontologists and Mineralogists, Spec ial Publication 28, p.259-297. Morrow, D.W., Cumming, G.A. and Koepnick, A.B ., 1986, Manetoe Facies - a gas-bearing, megacrystalline, Devonian dolomite, Yukonand Northwest Territories, Canada: American Association of Petroleum Geologists, Bulletin, v. 70, p.702-720. Wilson , J.G., 1975, Carbonate Facies in Geologic History: Springer-Verlag, New York, p. 471 . Wong, P.K. and Oldershaw, A., 1981, Burial cementation in the Devonian, Kaybob Reef Complex, Alberta, Canada : Journal of Sedimentary Petrology, v. 51, p.507-520.
C C C
C C C C C C
C C
C C C
C C
c:
C C C
C C C C
C C E
C C
C C
c e
c c c
c c c c
c' (
c
Dolomites -
Models and Ancient Dolostones
137
(
r
Coorong and Sabkha Models
Mixed Water Model
(
Bush, P. , 1973, Some aspects of the diagenetic history of the sabkha in Abu Dhabi, Persian Gulf , in Purser, B.H., ed., The Persian Gulf-Holocene Carbonate Sedimentation and Diagenesis in a Shallow Epicontinental Sea : Springer-Verlag , New York, p. 471. Hsu, K.J. and Siegenthaler, C., 1969, Preliminary experiments on hydrodynamic movement induced by evaporation and their bearing on the dolomite problem: Sedimentology, v. 12, p. 11-25. IlIing , G.U., Wells, A.J. and Taylor, C.M., 1965, Penecontemporary dolomite in the Persian Gulf, in Pray, A.C. and Murray,R.C., eds., Dolomitization and Limestone Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 13, p. 89-111. Kinsman , D.J.J., 1966, Gypsum and anhydrite of recent age, Trucial Coast, Persian Gulf, in Rau, J.L., ed., Second Symposium on Salt , v. 1: Northern Ohio Geological Society, p. 302-326. Muir, M., Lock, D. and Von der Borch, C., 1980, The Coorong Model for penecontemporaneous dolomite format ion in the Middle Proterozoic McArthur Group, Northern Territory, Australia, in Zenger, D.H., Dunham,J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization : Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 51-68. Patterson, R.J., 1972, Hydrology and carbonate diagenesis of a coastal sabkha in the Persian Gulf, Ph.D. Thesis, Princeton University, Princeton , New Jersey, 498 p. Patterson , R.J. and Kinsman, D.J.J. , 1981, Hydrologic framework of a sabkha along Arabian Gulf: American Association of Petroleum Geolog ists , Bulletin, v. 65, p. 1457·1475. Patterson, R.J. and Kinsman, D.J.J., 1982, Formation of diagenetic dolomite in coastal sabkha along Arabian (Persian) Gulf: American Association of Petroleum Geologists, Bulletin, v.66, p.28-43. Von der Borch, C.C., 1976, Stratigraphy and formation of Holocene dolomitic carbonate deposits of the Coorong area, South Australia : Journal of Sedimentary Petrology, v. 46, p. 952-966. Von der Borch, C.C. and Jones, J.B., 1976, Spherular Modern dolomite from the Coorong area, South Australia: Sedimentology, v. 23, p. 587-591. Von der Borch, C.C. and Lock, D.E., 1979, Geological significance of Coorong dolomites: Sedimentology, v.26, p.813-824.
Badiozamani, K., 1973, The Dorag dolomitization model-application to the Middle Ordovician of Wisconsin: Journal of Sedimentary Petrology, v. 43, p.965-984. Choquette, P.w. and Steinen , R.P., 1980, Mississippian non-suprat idal dolomite, Ste. Genevieve Limestone , Illinois Basin: evidence for mixed-water dolomitization, in Zenger, D.H., Dunham , J.B . and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists, Spec ial Publication 28 , p. 163-196. Folk, R.L. and Land, L.S. , 1975,Mg/Ca ratio and salinity: two controls over crystallization of dolomite: American Association of Petroleum Geolog ists, Bulletin , v. 59, p. 60-68. Hanshaw, B.C., Back, W. and Deike, R.G., 1971, A geochem ical hypothesis for dolomitization by groundwater: Economic Geology, v. 66, p. 710-724. Land, L.S., 1973, Contemporaneous dolomitizat ion of Middle Pleistocene reefs by meteoric water, North Jamaica: Bulletin of Marine Sciences , v. 23, p. 64-92. Land, L.S., Salem, M.R.!. and Morrow, DW., 1975, Paleohydrology of ancient dolomites - Geochemical evidence : American Association of Petroleum Geologists, Bulletin, v. 59, p. 1602-1625. Lippman, F., 1973, Sedimentary Carbonate Minerals: Springer-Verlag, New York, p.228. Ward,W.C.and Halley, R.B., 1984,Dolomitization in a mixing zone of near-seawater composition , Late Pleistocene, Northeastern Yucatan Peninsula: Journal of Sedimentary Petrology, v. 55, p. 407420.
r (
C
o ( o ( o
o o
o (
l
Solution-Cannibalization and PressureSolution Goodell, H.G. and Garman, R.K., 1969,Carbonate geochemistry of Superior deep test well, Andros Island, Bahamas: American Association of Petroleum Geologists, Bulletin, v. 53, p. 513-536. Hsu, K.J., 1966, Origin of dolomite in sedimentary sequences: a critical analysis: Mineralium Deposita, v. 2, p. 133-138. Logan, B.W. and Semenu ik, V., 1976, Dynamic metamorphism ; processes and products in Devonian carbonate rocks Cann ing Basin, Western Aus tralia: Geological Society of Australia , Special Publication 6, 138 p. Wanless, H.R., 1979, Limestone response to stress : Pressure solut ion and dolomitization : Journal of Sedimentary Petrology, v. 49, p. 437-462.
Ancient Dolostones - Stratigraphic and Petrographic Criteria Beales, F.W. and Hardy, J.L., 1980,Criteria for the recognition of diverse dolomite types with an emphasis on studies on host rocks for Mississippi Valley - type ore deposits , in Zenger, D.H., Dunham, J.B. and Ethington , R.L., eds., Concepts and Models of Dolomitizat ion: Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 197-213. Collins, J.A. and Smith, L., 1977, Zinc deposits related to diagenesis and intrakarstic sedimentation in the Lower Ordovician St. George Formation, western Newfoundland: Canadian Society of Petroleum Geologists, Bulletin , v. 23, p.393-427. Davis, J.H., 1977, Genesis of the southeast Missouri lead deposits: Economic Geology, v. 72, p. 443-450. Eliuk, L.S., 1978, The Abenaki Formation, Nova Scotia Shelf , Canada - A depositional and diagenetic model for a Mesozoic carbonate platform : Canadian Society of Petroleum Geologists, Bulletin, v. 26, p. 424-514. Exploration Staff, Chevron Standard limited, 1979, The geology geophysics and significance of the Nisku Reef discoveries , West Pembina area, Alberta, Canada: Canadian Society of Petroleum Geologists, Bulletin, v. 27, p. 326-359. Folk, R.L., 1973, Carbonate petrography in the Post-Sorbian age, in Ginsburg, R.N., ed., Evolving Concepts in Sedimentology: The John Hopkins University, Studies in Geology, no. 21, p. 118-158. Gebelein, C.D., Steinen, R.P., Garrett, P., Hoffmann , E.J., Queen, J.M. and Plummer, L.N., 1980, Subsurface dolomitization beneath the tidal flats of central west Andros Island, Bahamas, in Zenger, D.H., Dunham , J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontolog ists and Mineralogists, Special Publicat ion 28, p. 31-49. Gregg,J.M. and Sibley,D.F.,1984, Epigenetic dolomitization and the origin of xenotopic dolomite texture: Journal of Sedimentary Petrology, v. 54, p. 908-931. Jodry, R.L., 1969, Growth and dolomitization of Silurian reefs, St. Clair County, Mich igan: American Association of Petroleum Geologists, Bulletin , v. 52, p.957-981 . Jones, R.M.P., 1980,Basinal isostaticadjustment faults and their petroleum significance: Canadian Society of Petroleum Geologists, Bulletin, v. 28, p. 211-251. Lyle, J.R., 1977, Petrography and carbonate diagenesis of the Bonneterre Formation in the Viburnum Trend Area, Southeast Missouri: Economic Geology, v.72, p.420-434.
( ( 138
Macqueen, R.w. and Thompson, R.I., 1978, Carbonate-hosted lead-zinc occurrences in northeastern Brit ish Columbia with emphas is on the Robb Lake deposit: Canadian Journal of Earth Sciences, v. 15, p. 1737-1762. Radke, B.M. and Mathis R.L., 1980, On the formation and occurrence of saddle dolomite: Journal of Sedimentary Petrology, v. 50, p. 1149-1168. Ricketts, B.D., 1983,The evolution of a Middle Pre-Cambrian dolostone sequence a spectrum of dolomitization regimes : Journal of Sedimentary Petrology, v. 53, p.565-586. Schmidt, V., 1965, Facies, diagenesis and related reservoir properties in the Gigas Beds (Upper Jurassic), northwestern Germany, in Pray, L.C. and Murray, R.C., eds., Dolomitization and Limestone Diagenesis : Society of Economic Paleontologists and Mineralogists, Special Publication 13, p. 124-168. Sears, S.O. and Lucia, EJ., 1980, Dolomiti zation of northern Michigan Niagara reefs by brine refluxion and freshwater/ seawater mixing, in Zenger, D.H., Dunham, J.B. and Eth ington , R.L. , eds., Concepts and Models of Dolomitization : Society of Econom ic Paleontologists and Mineralogists, Special Publication 28, p. 215-235. Sibley, D.E , 1980, Climat ic control of dolomitization, Seroe Domi Formation (Pliocene), Bonaire, N.A., in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Econom ic Paleontologists and Mineralogists, Special Publication 28, p. 247-258. Sibley, D.F., 1982, The origin of common dolomite textures: Clues from the Pliocene: Journal of Sedimentary Petrology, v. 52, p. 1087-1100. Skall , H., 1975, The paleoenvironment of the Pine Point lead-zinc district: Economic Geology, v. 70, p. 22-47. Supko, P.R., 1977, Subsurface dolomites, San Salvador, Bahamas:Journal of Sedimentary Petrology, v. 47, p. 1063-1077. Wong, P.K. and Oldershaw, A., 1981, Burial cementation in the Devonian, Kaybob Reef Complex, Alberta, Canada: Journal of Sedimentary Petrology, v. 51, p.507-520. Wood, G.V. and Wolfe, M.J., 1969, Sabkha cycles in the Arab-Darb Formation off the Trucial Coast of Arabia : Sedimentology, v. 12, p. 165-191. Ancient Dolostone - Geochemical Criteria
Behrens, E.W. and Land, L.S., 1972, Subtidal Holocene dolomite, Baffin Bay, Texas: Journal of Sedimentary Petrology, v. 42, p. 155-161.
Diagenesis
Fritz, P. and Jackson, SA, 1971, Geochemical and isotopic characteristics of Middle Devonian dolom ites from Pine Point, northern Canada: 24th International Geological Congress , Montreal, Sec . 6, p.230-243. Fritz, P. and Katz, A., 1972, The sodium distribution of dolomite crystals: Chemical Geology, v. 10, p. 237-244. FOchtbauer, H. and Goldschmidt, H., 1965, BeZiehungen Zwishchen calciumgehalt und bildungsbed ing-ungen der dolomite : Geologische Rundschau , v. 55, p.29-40. Land, L.S., 1980, The isotopic and trace element geochemistry of dolomite: the state of the art, in Zenger, D.H., Dunham, J.B. and Eth ington, R.L. , eds. , Concepts and Models of Dolom itization : Society of Economic Paleontologists and Mineralogists, Special Publication 28, p. 87-110. Land, L.S. and Hoops, G.K., 1973, Sodium in carbonate sediments and rocks: a possible index to the salinity of diagenetic solutions: Journal of Sedimentary Petrology, v. 43, p. 614-617. Land, L.S., Salem, M.R.1. and Morrow, D.W., 1975, Paleohydrology of ancient dolomites: geochemical evidence: American Association of Petroleum Geologists, Bulletin, v. 59, p. 1602-1625. Lumsden, D.N . and Chimahusky, J.S., 1980, Relationship between dolomite nonstoichiometry and carbonate facies parameters, in Zenger, D.H., Dunham, J.B. and Ethington , R.L., eds. , Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists , Special Publication 28 , p. 123-137. Morrow, D.W., 1978, The influence of the Mg/Ca ratio and salinity on dolomitization in evaporite basins: Canadian Society of Petroleum Geologists, Bulletin, v. 26, p. 389-392. Sass, E. and Katz, A., 1982, The origin of platform dolomites: new evidence : American Journal of Science , v. 282, p. 1184-1213. Veizer, J. and Hoefs, J., 1976, The nature of 180 / 160 and 13C/12C secular trends in sedimentary carbonate rocks: Geochemica et Cosmochimica Acta, v. 40, p. 1387-1395. Veizer,J., Lemieux, J., Jones, B., Gibling, M. and Savelle, J., 1978, Paleosalinity and dolomitization of a Lower Paleozoic carbonate sequence, Somerset and Prince of Wales Islands, Arctic Canada: Canadian Journal of Earth Sciences, v. 15, p. 1448-1461. Other References Cited in Text
Alderman, A.R. and Skinner, H.C.W., 1957, Dolomite sedimentation in the southeast of South Australia: American Journal of Science , v. 255, p. 561-567.
Geoscience Canada Reprint Series 4
Barton, P.B., Jr., 1967, Possible role of organic matter in the precipitation of the Mississippi Valley areas, in Brown, J.S., ed., Genesis of Stratiform Lead-ZincBarite-Fluorite Deposits : Economic Geology, Monograph 3, p. 371-378. Beales, F.W. and Jackson, SA, 1969,Pine Point - a stratigraphic approach : Canadian Institute of Mining and Metallurgy, Transactions, v. 61, p. 1-12. Bebout , D.G. and Maiklem, W.R., 1973, Ancient anhydrite facies and environments, Middle Devonian Elk PointBasin, Alberta : Canadian Society of Petroleum Geologists , Bulletin, v. 21, p. 287-343. Boles, J.R., 1981, Clay diagenesis and effects on sandstone cementation (case histor ies from the Gulf Coast Tertiary), in Longstaffe , EJ ., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p. 148-168. Butler, G.P., 1969, Modern evaporite deposition and geochemistry of co-existing brines, the sabkha, Trucial Coast, Arabian Gulf: Journal of Sedimentary Petrology, v. 39, p. 70-89. Davies, G.R ., 1979, Dolomite reservoi r rocks processes , controls , porosity development, in Geology of Carbonate Porosity: American Association of Petroleum Geologists, Short Course Note Series no. 11, Houston, p. C1-Cl7. Davies, G.R. and Ludlam, S.D., 1973,Origin of laminated and graded sediments, Middle Devonian of Western Canada: Geological Society of America , Bulletin , v. 84, p. 3527-3546. De Groot, K., 1973, Geochemistry of tidal flat brines at Umm Said, Qatar, S.E., Persian Gulf, in Purser, B.H., ed., The Persian Gulf - Holocene Carbonate Sedimentation in a Shallow Epicontinental Sea: Springer-Verlag, New York, p.377-394. Dunham, J.B . and Olson, E.R., 1980, Shallow subsurface dolomitization of subtidally deposited carbonate sediments in the Hanson Creek Formation (Ordovician-Silurian) of Central Nevada, in Zenger, D.H., Dunham, J.B. and Ethington, R.L., eds., Concepts and Models of Dolomitization: Society of Economic Paleontologists and Mineralogists , Special Publication 28, p. 139-161. Foscolos, A.E. and Kodama, J., 1974, Diagenesis of clay minerals from LowerCretaceous shales of northeastern British Columbia: Clays and Clay Minerals, v. 22, p. 319-335. Gebelein, C.D. and Hoffman, P.,1973,Algal origin of dolomite laminations in stromatolitic limestones : Journal of Sedimentary Petrology, v. 43, p. 603·613.
C C C C C
C (
C
C (
C C
C ( (
C C C C C C C C C C C E C C C C
c e ( \.
C (
C l
C C
(
Dolomites -
(
o
r
c (
c O· (
o o
o o
o
o o
c (
o o o
o o o o
c r,
".
'-...
.
Models and Ancient Dolostones
Gorody, A.W., 1980, Dolomitization and paleo-hydraul ic history of the Lower Ordovician Mascot Formation , Upper Knox Group , in north central Tennessee : Geological Society of America , Abstracts with Program , v. 12, p. 435. Hagn i, R.D. and Trancynger, T.C., 1977, Sequence of deposition ofthe ore minerals at the Magmont Mine, Viburnum Trend, southeast Missouri: Economic Geology, v. 72, p. 451-464. Holland, H.D., Kirsipu, T.V., Heubner, J.S. and Oxburgh, U.M. , 1964, On some aspects of the chemical evolution of cave waters : Journal of Geology, v. 72, p.36-67. HsO,K.J. and Schneider, J., 1973, Progress report on dolomitization hydrology of Abu Dhabi Sabkhas, Arab ian Gulf, in Purser, B.H., ed., The Persian GulfHolocene Carbonate Sedimentation in a Shallow Epicontinental Sea : SpringerVerlag, New York, p. 409-422 . Jackson , SA and Beales, F.w., 1967, An aspect of sedimentary bas in evolution: The concentration of MississippiNalleytype ores during late stages of diagenesis: Canadian Society of Petroleum Geologists, Bulletin, v. 15, p. 383-433. Jacobson, R.G. and Usdowski , H.E., 1976, Partit ioning of strontium between calcite dolomite and liquids : Contributions to Mineralogy and Petrology, v.59 , p. 171·185. Katz, A. and Mathews, A., 1977, The dolomitizat ion of CaC03 : an experimental study at 252°C - 295°C: Geochemica et Cosmochimica Acta, v. 41, p. 297-308.
Kendall, A.C., 1977, Origin of dolomite mottling in Ordov ician limestones from Saskatchewan and Manitoba: Canadian Society of Petroleum Geologists, Bulletin, v. 25, p. 480-504. Klahn , H ., 1928 , SOsswasserkalkmagnes ia Gesteine und KalkmagnesiasOsswasser: Chemie der Erde, v. 3, p.453-587. Kocurko, J.M ., 1979, Dolomit ization by spray-zone brine seepage San Andres , Columbia : Journal of Sedimentary Petrology, v. 49, p. 209-214. Langton, J.R . and Chin, G.E., 1968, Rainbow Member facies and related reservoir properties, Rainbow Lake, Alberta: Canadian Soc iety of Petroleum Geologists, Bulletin, v. 16, p. 104-143. Magara , K., 1976, Compaction and flu id migration: practical petroleum geology: Elsevier, Amsterdam , Developments in Earth Science , v. 9, 319 p. Mawson, 0 ., 1929, South Australian algal limestones in the process of formation: Quarterly Journal of the Geological Society of London, v. 85, p. 613-621. McKenzie, J.A., HsO , K.J. and Schneider, J.F., 1980, Movement of subsurface waters under the sabkha , Abu Dhabi; UAE, and its relation to evaportive dolomite genesis , in Zenger, D.H., Dunham , J.B. and Ethington, R.L., eds., Concepts and Models of Dolomit ization: Society of Economic Paleontologists and Mineralogists , Special Publication 28, p. 11-30. Morrow, D., 1975, The Florida Aquifer : a possible model for a Devonian paleoaquifer in northeastern British Columbia: Geological Survey of Canada, Paper 75-1 B, p. 261-266.
139
Morrow, D. and Kerr, J.W., 1978, Strat igraphy and sedimentology of Lower Paleozoic Formations near Prince Alfred Bay, Devon Island: Geological Survey of Canada, Bulletin 254, 122 p. Mossop, G.D., 1972,Originofthe peripheral rim, Redwater Reef, Alberta: Canadian Society of Petroleum Geologists, Bulletin, v. 20, p. 238-280. MOiler, S. and Teitz, G., 1971, Dolomite replacing " cement A" in biocalcarentiles from Fuerteventura, Cana ry Islands, Spain, in Bricker, D.P., ed., Carbonate Cements: John Hopkins Press, Baltimore, 376 p. Murray, R.C. and Lucia, F.J., 1967, Cause and control of dolomite distribution by rock selectivity : Geological Society of America , Bulletin, v. 78, p. 21-36. Randazzo, A.F. and Hickey, E.w., 1978, Dolomitization in the Floridan Aquifer: American Journal of Science , v. 278, p. 1177-1184. Roy, K.J., 1972, The Boundary Member; a buried erosional remnant of Triassic age in northeastern British Columbia : Canadian Society of Petroleum Geologists, Bulletin, v. 20, p. 27-56. Sanford , B.V., 1962, Sources and occurrences of oil and gas in the sedimentary basins of Ontario : Geological Association of Canada, Proceedings , v. 14, p.59-89.
Originally published in Geoscience Canada v. 9 Number 2 (June 1980) Revised 1988
c
140
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c c c c c c c c c ( (
C
c
c
c ( c c c c SEM photomicrograph of an assemblage of delicate and fibrous authigenic Illite in the Lower Silurian Whirlpool Sandstone Formation of Southeastern Ontario. This illite has partly filled pore spaces between planar crystals of detrital smectite and may have formed by direct precipitation from saline pore fluids after leaching microcline feldspars. Photograph courtesy ofK.J. O'Shea andS.K. Frape (University of Waterloo).
C E C C C
C
c e c (
c
c c c c c
r: Evaporites
141
\
(
,...J
) r:
,
ever, extent of this contribution may be controversial (compare the interpretations of Hitchon et al. (1971) with those of Spencer (1987) forthe origin offormation waters in the WesternCanada Sedimentary Basin). Residual evaporite brines have also beencalled upon asasourceof Mg for the dolomitization of limestones in shallow settings (the solution reflux model of Adams and Rhodes, 1960)or in deeper circulating systems (Aulstead Evaporites and Spencer, 1985; Morrow et al., 1986; Aulstead et al., 1988; inthe Western Canada Sedimentary Basin). Similarities in Ronald J. Spencer r: I the sulphur isotopic composition of Department of Geology and anhydrites from Devonian evaporites Geophysics and anhydrite cements in carbonate The University of Calgary ( units of western Canada (Nahnybida et Calgary, Alberta, Canada T2N 1N4 aI., 1982;Uedaetal., 1987) also suggest a T.K. Lowenstein connection between these evaporites Department of Geological Sciences and carbonate diagenesis. • and Environmental Studies .- It has long been reco gnized that State University of New York at ~evaporite minerals are easily altered on Binghamton burial by even modest increases in .temperature. As a consequence, much Binghamton, New York, U.S.A. 13901 attention in the past was given to such burial metamorphism (see Stewart, • Much of the material presented here 1963, p. 42-43; Borchert and Muir, was originally published in the Sixth 1964, p.93-147, 158-159, 237-254; International Symposium on Salt; we Braitsch, 1971, p. 108-130, 173-195). thank L.A. Hardie and the Salt Institute Unfortunately, little agreement eme~ for permission to use the material here. ged as to which features were due t burial alteration and which were deposiINTRODUCTION tionalfeatures. However,in recent years Evaporites are rocks formed by prethere has been considerable progress cipltatlon of salts from aqueous soru- <.t in the recognition of primary and early Q9ns..MJNateris romo'/oEi ano:tGAlG'§p8S diagenet~features in ' .e va p o r i t e~ cies become more concentrated. Apart throu gh studies of modern envi ronfrom their economic value, evaporites ments. In contrast, little is currently ~_ are significant indicators of l3aleeeliknown about the detailed diagenetic UF" mate and of rift-vallgy tectonic settings... processes involved in the alteration of In terms of the global chemical budget evaporites during burial. If we are to they are massive subsurface "wareproperly exploit the information conhouses" tOr storage of the soluble bytained in evaporites, particularly those products of chemical weathering of the involving paleochemistry and paleoencrust. Most significantly, of all deposits vironments, it is absohItely essential to in the sedimentary record, evaporites be able to separate those minerals, tex~ are the premierrecorders ofthe chemjs~es and structures that are of primary tr~ of ancient s..e..E!...YtaleI.5, lake w~ epositional origin from th ose that are. and other surface waters. IYl tJ-'I ~ ue to diagenetic alteration. The ma~.n m.addltton.evaporttes may be major urpose 0 IS paper is to present a se , { " sources of solutes for deep-circulaf working criteria for distinguishing ting hydrothermal brines and sedibetween syndepositional and later bur mentary basin formation waters, both . I features in evaporites. Early diageof which may be involved in the dianetic features are emphasized beca'iSe genetic alteration of other basinal sedithey are best documented. The criteria ments. 1JJ-e-e0ntribution of solutes from fCiri5i:mat alteration are more uncertain evaporites in sedimentary basin form.a:.. because they must be inferred without l ion waters associated with ore d~osits • the aid of active modern systems. These and oil fields is discussed by Rittencriteria are updated from those .prehouse(1967)an~.f!!]~enter (197a)...l:foW:: sented by Hardie et al. (1985).
7'\
V
j'J
b
.0/1 ': .it
;JtJfi:-
Much of the controversy over the origin of evaporites ang the influence of diagenesis or burial metainorphism~ fwtres around potash deposits. I Potash deposits are economica lly irlrportant as a raw material used in the production offertilizers. The largest potash reserves in the world are in the Prairie , FO rmat ion of Saskatchewan. Mayr-C 0 ~J hofer (1985) gives estimates of 522 billion tons in situ and 37 billion tons of mineable K2 0 worldwide. Of this, 507 billion tons in situ and 34 billion tons of mineable K2 0 are from the Prairie Formation. We address the origin of potash deposits, in terms of depositional and diagenetic processes, using textural~ criteria and fluid inclusion data. A brief discussion of the use of fluid inclusions in the stud y of evapo rites is al so inciudeQ,..
PRIMARY VERSUS SECONDARY FEATURES: DEF.INITIONS There is no consensus on the meanings Ef the terms primary and secondar y as applied to evaporites. At the 1962InternatIOnal Conference on Saline Deposits primary minerals were defined as - those :'lJre.£!.f,1itated directly,from tne so lutiol;l":': and se£2,.ndarY-'l!inerals as !!:!Qgl.;19LITIgQ, later than the j:lrimary. ~(Ingerson, 1968, p. 671). Stewart held similar views in that he dealt with .....~71 both "penecontemporaneous changes" f) ll' ..... and J2,Q§Lconsolidation changes" / ...... under secondary changes (Stewart, 1963,p. 42). Braitsch (1971, p. 92), on the other hand, included early diagenetic ~ ~ J alteration in his definition of "primary precipitation". These differences are in one sense merely semantic, since all acknowledge that initial precieitates may be altered soon after deposition, soltr7o/-J1 it becomes a matter of ereference as to whetheL!Q iocl'lde early aUeratioFlo .!:!.Dder erimary (as does Braitsch) or under secondary (as does Stewart). On the other hand, the distinction between early diagenetic changes and later burial changes is a truly. pivotal decision and we need terms that explicitly acknowledge the timing of such changes. In terms of timin g, a mineral, mineral assemblage. texture, fabric, 1!.uid inclusion or ~.t[lJe.tllre COl lid (1) depositional, i.e., formed at the time of deposition of a sedimentation unit or deposited in its existing form; (2) post-depositional but pre-burial, i.e., formed diagenetically soon after
be
142
Diagenesis
deposition by processes controlled by the existing depositional environment; or (3) post-burial, i.e., formed by late diagenetic or metamorphic-metasomatic processes controlled by the subsurface burial environment. ~ The first two classes, one primary - ~" (depositional) and the other secondary (diagenetic), both depend on processes operating in the depositional environment. They both house valuable information about primary environmental parameters and for this reason we have grouped them together under syndepositional features, as distinguished from post-burial features. In a depositional context, then, we could regard all syndeggsjtioAaUeatufes....inch .ding e!!'ly diagenetic (penecontemporane:. ous) alteration ~Rtimar~ Post-burial features would be " secondary". However, we prefer to emphasize the timing aspect of the problem and so in comPili ng criteria we have followed the syndepositional versus post-burial classification. But there remains a class of features that are equivocal in that there are no decisive criteria that point to their time of origin. We have grouped these under ambiguous features. (]) "[Q. complicate matters. it is common to find extensive secondary changes iQ. mineralo.,gy_wiJb-no..cbange..io..tbe..pJ:·~-lll.xtur.e_(e. g. , pseudomorphous replacement by halite of vertically oriented gypsum euhedra, Lowenstein, 1988, figure 6) or alteration of texture that does not entirely obliterate tfie ~Jnal depOSItional structure (e.g., Dellwig, 1955, figure 2; Hardie and Eugster, 1971, figure 11 A). !?espite the alteration overprints we regard S'lch recognizable primary textures and structyres as syndepositional features ~ca~ carry legibJejnf.ounatioo.about.depgsitional prgcass.es..aod C;:llViLQnmflDl~. Finally, the terms primary and secondary are not easily applied to evaporite deposits as a whole because many deposits carry both syndepositional and secondary alteration features (and commonly in the same bed or crystal , as just discussed above). Should the need lllise to use such terms for the overm) deposits, then we suggest "primar\! _evaporite" might he ••sed to desC[i~ those deQositsnot sufficiently altered to oDscureJl:ul_ptima c.y....sYRdeposiiion I 'res. Or perhaps "modified primary evaporite" might be better where altera-
t
tion is extensive but where it has still ( been possible to "read through" the alteration. Secondary evaporites would be those deposits so thoroughly altered on burial that few if any syndepositional features can be unambiguousl identified. CRITERIA FOR SYNDEPOSITIONAL FEATURES Criteria for syndepositional features comprise (a) mechanical sedimentary structures with detrital textures and fabrics formed by deposition from bedload or suspended load; (b) crystal textures and fabrics produced as chemically precipitated minerals grew in situ on and within bottom sediment; and (c) features indicative of syndepositional disSOlution, reprecipitation and cementation of evaporite minerals. The first two types of information (a and b above) are summarized in brief. The reader is referred to Hardie et at. (1985) for further details. The latter (c) can be considered evidence for early diagenetic changes, and is discussed more completely below.
Geoscience Canada Reprint Series 4
a
r (
Sedimentary Structures
It is well documented that mechanical sedimentary structyres typical of clastic deposition are common in evaporite deposits, particularly those involving gypsum-anhydrite and, to a lesser extent, halite (Dellwig, 1968; Dellwig and Evans, 1969; Hardie and Eugster, 1971; Parea and Ricci-Lucchi, 1972; Ricci-Lucchi, 1973; Schreiber et a/., 1976; Schlager and Bolz, 1977; SchreiQ ber and Hsu, 1980). These are essentially stratification and bed form structures, but other primary structures such ~ m p foldin g _(Schlager and Bolz, 1977, p. 602), sole markings (Parea and Ricci-LucChi, 19(' , fi gure 24), mudcracks. and flat pebble con lomerates (Hardie and Eugster, 1971, figure 20) have been reported. The preservation of all these structures in evaporite sedimentation units not only testifies against pervasive destructive recrY.§- ~ tallization, replacement and deformation oft e eposit, bUtitaIlQWS.\laIUabl1 interpr~tations .to .be made about-depe ~ sitional processes and -the -envleen- 11 ~s in Wtftc.5Jhey operated.
f
( ( (
,'i
( (
:f (
r
(
( (
C C (
(
c
c c (
c l
~
,mati
( (
(
( ( Figure 1 Thin-section photomicrograph of part of a wave ripple composed of prismatic
l
gypsum pseudomorphs, viewed perpendicular to the strike of the ripple crest. Clear crystal fragments with gypsum morphologies butnow composed of halite, are aligned along foresets. Dark surrounding sediment is microcrystal/ine anhydrite and polyhalite. Vertically oriented prismatic gypsum pseudomorphs (bottom), composed of halite and some sylvite, are the same size and shape as the gypsum pseudomorphs comprising the wave ripples. Sample from Salado Formation laminated anhydritelpolyhalite rock. Scale bar is 5 mm long.
l (
(
Evaporites
-
Detrital Framework Textures. As with other sedimentary rocks, the classic criterion of a framework of grains with point contacts should firml¥-estabIish a primary detrital texture in evaporites For example, in gypsum grain: stones the detrital framework may be made up of abraded fragments of single selenite crystals (e.g., Hardie and Eugster, 1971, figures 15and 16)or of stacks of gypsum needle euhedra or iented parallel to foreset laminae (Figure 1; Lowenstein, 1988, figure 9). Halite detrital textures are subtle features easily overlooked. Such textures may be of
(
(
.----------...
--:==-::- ~ ._=:::,-:~' (od ~ two types: (a) mechanical accumulations of cubes that precipitated in the water column and settled to the bottom such as those preserved in modern saline pans (Lowenstein and Hardie, 1985,figure 11)and in certain beds in the Permian Salado Formation of New Mexico (see Lowenstein , 1988, figure 14); and (b) frameworks of abraded crystals sorted by bottom currents, as for example, those reported byWeilereta/. (1974) along the shores of the South Basin of the Dead Sea (see their figures 2 through 6).
(
o c o r
,-
I
,-,' r\
".-
'--. \.....
Figure 2 Photomicrograph of vertically oriented halite chevrons (upward-directed edges and corners) and cornets (upward-directed cube faces) , Within single crystals, dark bands rich in fluid inclusions alternate with lighter bands containing few fluid inclusions, White areas are clear inclusion-poor halite, usuallypresent as rims on chevrons and cornets. Sample from Salado Formation mud-free halite. Scale bar is 3 mm long.
--- -------:\ / ->
~.I/// ./'
<,
143
<:>
Crystalline Framework Fabrics . Fabrics diagnostic of in situ! open-space crystal growth on the bottom of brinepools are commonly preserved in mOd~ ern and ancieQt evaporites particularly for gyp sum and halite (Schaller and Henderson, 1932; Stewart, 1949, 1951; Hardie and Eugster, 1971; Schreiber and Kinsman, 1975; Caldwell , 1976; Schreiber and Schre iber, 1977; Arakel , 1980; Warren, 1982; Lowenstein, 1982, for gypsum, and Gottesmann, 1963; Jones, 1965; Wardlaw and Schwerdtner, 1966; Shearman, 1970; Arthurton , 1973 ; Handford, 1981; Lowenstein, 1982, 1988; Hovorka, 1987; Brodylo and Spencer, 1987 for halite). Such fabrics for other salts have been reported (see Eugster, 1970, for trona, and Wardlaw, 1972, for carnallite), but, unfortunately, there are no complete descriptions. The dominant fabric of open-space prec ipitation is a layered, syntax ially grown crystalline framework consisting of vertically oriented and vertically elongated crystals. In essence, this fabric represents competitive crystal growth upward into a free-standing, saturated brine, a process identical in style to void-filling cement crystallization . There-t fore, the cement criteria carefully documented by carbonate petrographers (see Bathurst, 1975, p. 416-425~ as guides to the recognition of open space bottom rowtb_oLa..crystaliine ramework fabric in evaporites. When specifically applied to the evaporite minerals gypsum and halite as viewed in thin section, the basic fabric crltesia for syndepositional bottom growth are as follows : (1) for gypsum, verticall y oriented prisms with Auhedral terminations (commonly " swallow-tail" twins) , and for halite , vertically directed "chevrons" (syntaxia'i overgrowths on upward-fae;.. ing cube coigOS) and " cornets" (syntaxial overgrowths on upward-facing cube faces) with euhedral terminations and fluid inclusionbanding parallel to cry~ tal growth faces (Figure 2); (2) I,Jpward-coarsenin and widenin of crys a s Figure 3; Hardie and Eugster, 1971, figure 21 ; Arakel, 1980, figure 9) where , during competitive growth, crystal bottoms abutthe substrate and sides of crystals abut one another along compromise boundaries (Shearman, 1970; Bathurst, 1975, p. 421-425 and figure 303). Crystal growth is therefore most rapid on upward-directed faces . For
*
c 144
Diagenesis
Geoscience Canada Reprint Series 4
(
C example, for halite, the upward-facing cube faces and coigns of foundered hoppers that settled randomly on the bottom become the dominant growth directions, producing "cornets" and "chevrons", respectively. Additional textural evidence that supports a syndepositional origin of vertically oriented crystal fabrics is (3) growth originating from a common, laterally continuous substrate, particularly where the basal zone of the growth layer consists of detrital seedcrystals; (4) mud drapes or crystal cumulates immediately above vertically oriented crystal layers thicken in depressions and thin over tops of the framework crystals (Figure 4). This demonstrates that vertically directed growth must nave occurred before deposition of overlying layers (e.g. , Hardie and Eugster, 1971, figure 21). (5) dissolution or erosion surfaces that truncate vertical crystals and act as substratesforgrowthofa newcrystallinelayer (see Lowenstein and Hardie, 1985). Crystalline framework fabrics produced by syndepositional open-space growth on the bottom of a brine body preserve a wealth of information on primary environmental conditions. We believe that syntaxially grown crystalline frameworks are not only primary features but strong evidence for shallow unstratified evaporite environments. The presence of dissolution - reprecipitation features provide additional confirmation of this view because such features require either ephemeral conditions or shallow enough brines to allow significant brine dilution by undersaturated flood water influxes. The abundance of crystalline frameworks showing dissolution features in modern salt pans (Lowenstein and Hardie, 1985) amply support this contention. Finally, the fluid inclusions that outline growth layering in halite crystalline frameworks, for example, can add significant information about the syndepositional physicochemical conditions (see later). Dissolution - Reprecip/tation Features. Syndepositional dissolution - reprecipitation features in gypsum-anhydrite (e.g., Hardie and Eugster, 1971, figure 21;Schreiber and Kinsman, 1975,figure 7; Warren, 1982, figures 12, 15B) and halite deposits (e.g., Dellwig, 1955, figure 11; Wardlawand Schwerdtner, 1966,
plate 3, figure 4; Shearman, 1970,figure 6; Arthurton, 1973, figure 9; Hovorka, 1987;Brodylo and Spencer, 1987, figure 3; Lowenstein, 1988,figure 13)indicate dramatic fluctuations in solution composition during crystal growth and layer accumulation . Evidence of syndepositional dissolution is discussed by Lowenstein and Hardie (1985) and is basically of two types: (1) sharp but smooth truncations parallel to bedding of vertically oriented crystal frameworks overlain in turn by layers of detrital mud, crystal cumulates or vertically oriented crystals (e.g., Figure 5; see also Wardlaw and Schwerdtner,
1966,plate 3, figure 4; Arthurton, 1973, figure 9); (2) rounding ofeuhedral terminations of single crystals followed by syntaxial rehealing (Figure 6). For example,large selenite crystals have been found with rounded terminations draped by thin mud laminae and rehealed by renewed syntaxial growth that preserves both the rounded faces and the mud laminae within the vertically grown crystals (Hardie and Eugster, 1971, figure 21; Schreiber and Kinsman, 1975,figure 7). More complex dissolution and re~ cipitation features accoJD.Qan~ abOVe Qllil.n..o..r.rnma...i.o..moJ:l..er.n.epbemeral salt pans where halite cr ystalline
----
C C
C C C
C C C
C C C
C C C
C ~
C C C C C C
C C C E C C C C
Figure 3 Photomicrograph (crossed nicols) of upward-coarsening and widening halite crystals (black) outlined by gray microcrystalline polyhalite. Halite cornets (centre), with upward-facing cube faces, are flat-topped (arrow) and overlain by smaller vertically oriented crystals. Sample from Salado Formation mud-free halite rock. Scale bar is 2 mm long.
e c c c c
c
c c
0
( .
Evaporites
145
C· (
0
0 ( ( (.
C ('
C C (
Figure 4 Photomicrograph of halite and sylvite from the lower potash zone (Couche inferieure), Mulhouse Basin, Tertiary Rhine Graben. Halite layer (H) is composed of centimetre-size, vertically oriented chevrons that contain fluid inclusion growth-banding (dark gray to black). Sylvite layer (5) contains millimetre-size crystals of sylvite with equigranular mosaic texture. Millimetre-size cumulate halite crystals occur in sylvite layers. Mineralogical sequence of dolomitic mudstone (black) -> halite + minor anhydrite -> sylvite + minor halite. Note that sylvite crystals form a drape above projecting crystals of chevron halite; the top of the sylvite layer is nearly flat.
0 0
0 0 0
0 0 (
C
0 0 0 0 0 0 0 0
0
(\
8 C ('
,~
'-...
r'-L.-
Figure 5 Photomicrograph of halite crust (depth < 0.5 m) from Salina Omotepec, Baja California, Mexico (collected 1972). Dark areas are rich in fluid inclusions. Dissolution surface (between arrows) truncates tops of vertically oriented chevrons. Subsequent syntaxial overgrowth rebuilt the chevrons above the dissolution surface. Verticallyoriented voids (V) are partly filled with clear halite cement. Scale bar is 2 mm long.
Diagenesis
146
frameworks are forming . Ephemeral salt pans go through a repeating threestage cycle of events: (1) an initial flooding stage when undersaturated flood waters inundate the pan after a storm, (2) a saline lake stage when evaporative concentration leads to growth of a crystalline framework layer, (3) a desiccation stage when the pan surface dries up and only subsurface, intracrustal brines remain (Lowenstein and Hardie, 1985). The initial flood stage, in addition to producing dissolutional truncation of the framework and rounding of individual halite crystals, causes extensive tubular networks of vertical and hori,fA." ~c \,-a ontal dissolution cavities that follow ): (ht.\~ grain boundaries between verticallyorill)"'P ented crystals and along bedding Planes (Figure 5). Subsequent evaporative concentratlofi'Bt t e troodwaters results in halite saturated brines and renewed crystal growth. DissoluiiCi'i'1' ~ C"aVitiesbeccimeTIned with inward growing clear halite crystal cements and rounded and dissolved euhedra are rehealed by syntaxial growth. At the desiccation stage, the brine sinks below the pan surface and the voids in the upper halite crust become partly filled with a cement of clear halite (Lowenstein and Hardie, 1985). All these features are syndepositional , formed concurrently with deposition of the halite framework layers, but they may not be easy to distinguish from some burial dissolution-reprecipitation features in ancient evaporites. Burial dissolutipn, cementation and re ~r~aUization could give rise to rounded voids filled with clear halite cements as well as to pateheSOf clear halite mosaics. Care must be taken to look for other evidence of the timing of dissolution and cementation, such as internal sediment overlx!!Jg cement, truncation of void fills by primarl.!ayering, etc. -Cements. Cements have gone essentially unnoticed in evaporites despite the fact that undeformed and uncompacted buried salts typically have negligible porosities, and yet their analogous unburied primary deposits may contain more than 50% void space (e.g., Lowenstein and Hardie, 1985). This anomaly suggests that cementation must be an important process in evaporite diagenesis. Cements are post-depositional fea-~ tures, but they could form either syndepositionally (as is common in salt
lU U-
pans, Shearman, 1970; Lowenstein and Hardie, 1985)oron burial. Insome cases, it may be possible to decide on the timing of cementation from petrographic evidence. For example, meniscus cements surely call for a vadose setting and this in turn implies a syndepositional origin . Also, if erosion or dissolution surfaces cut across cements, or if internal sediment overlies a cement, then there can be little doubt about the syndepositional nature of the cements. We have recognized two basic of cementation in modern and ancient evaporites which we interpret J!§ sYO=. de os' . These are cases where
stYle~
Geoscience Canada Reprint Series 4
primary framework is enclosed by (1) intergranular cements and (2) vug and cavity cements. Intergranular cements usually consist of a mineral lining or filling the pore spaces ofa primary crystalline or granular framework. For example, in Figure 7 coarse sylvite cement fills the voids between cubes of halite in the Permian Salado Formation . These voids are identical in size and morphology to primary unfilled voids so typical of porous crusts of halite in modern salt pans (Figure 8; Lowenstein and Hardie, 1985). Other common features of this type of cementation are isopachous colour
I
(
C C
C C C C C C C C C
C C C
--
C C C
C C
C C C C C C
C E: C C C C
.fjJt::-
~
c
c
e
c c
c
Figure 6 Photomicrograph of dark gray fluid inclus ion-banded vertically oriented halite. Chevron halite crystal contains internally rounded cube coigns that have syntaxially rehealed (arrow). Clear, inclusion-poor halite located between chevrons and cornets. Sample from Salado Formation mud-free halite rock. Scale bar is 3 mm long.
c
c c c c
Evaporites
(
c c c o c (
(J
C
o
o (?
o
o o o c o o c o c. o o c o
Figure 7 Photomicrograph of muddy halite rock. Clear halite crystals contain incorporated mud and few fluid inclusions. Dark areas between halite cubes are composed of zoned crystals of sylvite cement and minor mudstone. Scale bar is 3 mm long.
C
l\ b C
L
c: ~
..
L
Figure 8 Photomicrographs of halite in a muddy halite layer from just beneath the surface of the modern salt pan of Saline Valley, California. The halite occurs as randomly oriented interlocking cubes with incorporated mud (black) and cloudy patches dense with fluid inclusions (arrows). Primary voids (V) surround halite crystals. These voids are equivalent to the space occupied by sylvite and carnallite cements in the Salado and Prairie Formations (compare with Figures 7 and 22). Aggregates of interlocking small rhombs of glauberite and mud (black) are also present. Scale bar is 5 mm long.
147
bands within cement crystals thatfollow original void boundaries and coarse poikilotopic crystals that may enclose many framework grains. Cavities and vugs lined or filled with cement are common in modern evaporite deposits. In halite salt pans, vertical and horizontal networks of dissolution cavities and shelter vugs are lined with euhedral cement crystals made of clear halite (Lowenstein and Hardie, 1985, figure 16). In modern gypsum pans, shelter vugs beneath rumpled surface crusts may be lined with masses of gypsum "swallow-tail" twins (ct. Warren, 1982, p. 628). Such cavity cementation has not been commonly reported from ancient evaporites (see however Hovorka, 1987; Casas and Lowenstein, in press).It isalso likely that the clear rims of halite on cloudy chevrons and cornets so common in ancient halite rocks (Wardlawand Schwerdtner, 1966, plate 2, figure 4; Lowenstein, 1982,figure 6C) are cavity cements (ct. Shearman, 1970). In modern salt pans, clear halite rims form syndepositionally as overgrowths on porous chevron and cornet frameworks during the saline lake and desiccation stages (Lowenstein and Hardie, 1985). However, similar textures can just as easily be produced on burial, assuming the open framework of halite can survive burial. A strong case for syndepositional halite and potash cements may be made based on porosity profiles from modern salt pans (Casas and Lowenstein, 1987, in press).The high porosities of surface salt crusts from modern salt pans decrease sharply in the upper few metres from 50% to less than 10%; porosity is negligible below 30 m. It appears that the open framework of salt pan halite is cemented very early in the burial history. Therefore, based on comparison with modern deposits, we conclude that cements which line or fill a primary granular or crystalline framework in ancient halite deposits probably formed early, and are included as syndepositional features. CRITERIA FOR BURIAL ALTERATION FEATURES
Our concern here is to identify features that clearly reflect alteration,poth diagenetic and metamorphic, due to the Increased pressyres and temperatures Of burial. as well as changes due to m etasomatism by invading subsurface
148
Diagenesis
Geoscience Canada Reprint Series 4
c c C
waters . We are on somewhat uncerlimestones. More insight into the signifianhedral mosaic consists of clear cance of sutured mosaic textures in grains that meet at triple junctions that tain ground here because roaoy...o.Ub.e Cflteria must be inferred . raJb.~~ r~ evaporites must await future work. approach 1200 angles in typical annealobtained direG-tlv. frOl!LG-Qmp'ar:~tW ~ Polygonal Mosaic Textures. ling recrystallization polygonal mosaic dies of active modern systems. NonePolygonal mosaic textures have long fashion (Stanton and Gorman, 1968,figthe ess, we aveputtogetheratentative been documented in experimentally ures 2,3 and 7). The crystals have few working set of criteria, some of which we treated metals and metallic ore minerals inclusions (large rectangular fluid-filled cavities and blebs of mud), as most of have listed more as a basis for discus(Buckley, 1951; Burgers , 1963; Stanton sion than as accomplished facts . In the and Gorman, 1968). Similar textures the "foreign " matter appears to have main the criteria involve textures and are typical of metamorphic rocks (VolI, been purged to the halite crystal boundstructures that record disruption, defor1960). There is a consensus that these aries during recrystallization (cf. Kuhn, mation and destruction of primary polygonal mosaic textures are a product 1968, plate 2, figure 2): These are all features. of annealling recrystallization whereb y features typical of "foam structure" Disruption and Destruction of Sedi,gr~ins optiMize:' their §Ize, snaReand produced by experimental rscrystalllzV mentary Structures. orientation to minimize the energy in the ation of metals and ores and attests to a .nannerof bubbles in foam (Stanton and destructive burial recrystallization proThe absence of primary beddin~ or the /\. ,~ ~esence of only scattered rem ants of Gorman, 1968). cess that leads to complete loss of any , v V IC/'primary beddin g or other sedimentary Evaporites, because of the ease with primary signature. However, such restructures in massivecrystalline mosaic which they are altered at the temperacrystallization may be very patchy, as tures and pressures of even mode~ we have seen in the Permian Salado salts points to destryctive seGGFlgefy burial, should be very susceptible to. evaporite of New Mexico and Devonian crystal growQ!...];he early stages of this SiiCli'anneaIljOQ-f.EtCLY..stallizatlao.. We Prairie evaporite of Canada where pridestructive growth will be seen as irregular isolated crystalline mosaic patches have found halite in the Permian Samary, unaltered "chevron" halite co(see next section) that cross-cut prilado formation of New Mexico and the exists at the same or greater burial mary sedimentary structures (e.g., Devonian Prairie evaporite of Canada depths as "foam" structured halite. Schreiber et al., 1976, figure 18). This with equigranular mosaic texture (Fig- ~ Deformation Features. criterion is most persuasive of burial ure 9; see also Lowenstein and Hardie, Apart from the obvious macroscopic alteration when found in monomin1985).This halite notably lacks vertical deformation structures such as folds, erallic (or essentially monominerallic) orientation, cumulate textures, banded fractures, veins, flow banding, pressure fluid inclusions or any other feature diasolution sllrfaces, etc. (e.g., Balk, 1949; rocks because then the crystalline mosaics must surely be the result of gnostic of primary growth. Instead, the Goldman, 1952;Mulleretal., 1981)there destructive recrystallization. Such crystalline mosaics will have suturecror polygonal grain boundaries see below) quite distinct from porphyroblasts and from-tl;e--isolated disp.lacive, incorporative or poikilotopic growthofeuhedra and nodules typical of syndepositional intra~ sediment crystallization (see below). '® Sutured Mosaic Textures. Crystalline mosaic textures in which )It.~~ 3i n boundaries ar sutured in the fw 'lj manner of neomor hlc s r of c a e roc s compare Hardie and Eugster, "jf~ 1971, figure 11A, with Bathurst , 1975, figures 336 and 338) clearl y indicate a secondary grain growth origin. However, the conditions under which this secondary mosaic formed are Rot ~SuChan interpenetration of grains may represent a low temperature , non-equilibrium texture (VolI, 1960); it certainly stands in marked contrast to the equilibrium polygonal mosaic textures formed experimentally by annealling recrystallization (next section). Supportive of this contention Figure 9 Photomicrograph ofhalite with equigranularmosaic texture ("foam" texture). Note are the contrasts between the polygonal the number of curved crystal boundaries that meet at triple junctions with angles approaching texture of marbles and the tortuously 120°. Mud (black) is included within halite or located at halite crystal boundaries. Small gray sutured boundaries of most diagenetbubbles scattered throughout were (artificially) trapped during thin-section preparation. Sample from Salado Formation muddy halite rock. Scale bar is 5 mm long. ically altered, but unmetamorphosed,
0'
7
r
C C (
C C C C C C C C C C
C C
t
c C C C C C C C G C C C C
c e c c..
c c c C \.
C
( ( (
Evaporites
149
c:
o
c (
(
c o (
o
o o t o o o o o o o o (
o
o c C (1
O· l
fJ
are more subtle microscopic deformation features to look for in thin section or Tn etched, polished slabs of evaporites ~ (Schlichta , 1968). Stanton and Gorman 1I~ ( 1 9 6 8) have discusssed the main types of deformation textures recognized in metallurgical studies and applicable to rocks (see also VolI, 1960, 1976). These deformation textures are (1) deformation twins, (2) slip lines and slip bands, ....-r:t. _ (3) lattice distortion features such as 0 ' -.!:!Ddu!ose extinct ion, bent cleavages, etc., (4) SUb-graindomains and (5) flattening of grains leading to formation of foliation. We leave the reader to seek help from the above and other works (e.g., Prucha, 1968; Mossop, 1979) in recognizing deformation features in evaporites.
AMBIGUOUS FEATURES There are a number of features that could have formed either syndepositionally or after bllW3J..Lo.tbe"~sence of -otFiercriteria to aid intheir interpretation the time of origin of such featllCes !llY§t ~main uncertain. To this class of ambiguous features belong (1) mineral growth within the sediment as euhedra or nodules,(2) pseudomorphous replaceFigure 10 Core slab photograph of halite cubes which have displaced the surrounding ments, (3) coarse crystalline salt, and carbonate-rich mud. Mud is also incorporated within some of the growth bands in the halite (4) certain cavity cements. crystals. Sample from the Prairie Formation. (0 Intrasediment Growth of Euhedra. Evaporite minerals that have grown within sediment commonly occur as randomly oriented , isolated or interlocking crystals with a euhedral or subhedral habit. Such crystals may incorporate sediment during growth or they may form displacively by pushing aside the host sediment during crystallization . The Devonian Prairie Formation . contains examples of gypsum (anhydrite) and halite (Figure 10)with displacive or incorporative growth habits. The Permian Salado evaporite of New Mexico (Lowenstein, 1982, 1988) carries many good examples of this style of intrasediment growth by the minerals gypsum (Figure 11), halite, glauberite (Figure 12) and langbeinite (Figure 13). Incorporative growth provides unequivocal evidence of post-depositional crystallization, but the difficulty is~.n distinguish ing between early-syriaepositional growth and later metamorph ~ ' ~rp hyr~astisg rowth. Intrasediment growth of minerals such as gypsum, halite and mirabilite as euheFigure 11 Photomicrograph of mud-incorporative gypsum crystals (pseudomorphously dral crystals is a common near-surface replaced by microcrystalline anhydrite). Surrounding mud contains magnesite and anhydrite. feature in many modern playas and Sample from Salado Formation laminated anhydrite rock. Scale bar is 2 mm long.
'-i:'I'
r'-,'.
"-..
r: '-..
c
150
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c (-
c c c
Figure 12 Photomicrograph (crossed nicols) of randomly oriented, zoned, incorporative glauberite euheare, now pseudomorphously replaced by halite (dark gray to black) and anhydrite (light er gray). Surrounding sediment is microcrysta lline anhydrite and magnesite-rich mud. Sample from Salado Formation laminated anhydrite rock. Scale bar is 5 mm long.
c c c c c c c
c c
c
c c c c c c c C Figure 13 Photomicrograph of isolated, randomly oriented tetrahedral langbeinite crystals with incorporated mud in crystal cores. The larger crystal contains langbeinite in its core (light gray). Therim and the entire smaller crystalhavebeenpseudomorphously replaced by microcrystalline leonite plus kainite. Surrounding area is mud(black) and halite (light gray). Sample from Salado Formation muddy halite rock. Scale bar is 5mmlong.
E C
C C
C
c e c c c c
c c
c c
c (
c.
C', C" ( ( (
c' (
c
(.
c.
c o c
C-
o c.
c (
c
o c' (,
(
c c o
L r
L
(
Evaporites
151
sabkhas (Hardie, 1968; Butler, 1970a; Thompson, 1968; Logan et al., 1970). We could argue in favor of a syndepositional origin for similar intrasediment growth textures in ancient evaporites by analogy. However, porphyroblastic minerai growth might occur on burial, producing similar textures. On the other hand, intrasediment growth of euhedra of such elevated temperature minerals as langbeinite (Figure 13) and loewite probably indicates burial crystallization, although as we have pointed out below, temperatures high enough to reach the langbeinite stability field have been reported from some modern evaporite surface sediments. G;) /ntrasediment Growth of Nodules. It was the discovery of nodules of anhydrite in the Holocene sediments of the Persian Gulf (Curtis et al., 1963; Shearman, 1966; etc.) that essentially rekindled interest in evaporites among sedimentologists. Since then there have been an astonishing number of papers published that have called on nodular anhydrite, or its presumed pseudomorphs, as a criterion for synoepositional growth of evaporites in a sabhka environment. But we must be very wary here because anhydrite nodules can form under a variety of conditions, such as, for example, from the
Red Sea hot brines in the Atlantis II placement rests on the identification of Deep under 200000 of water (see the crystal morphology of the parent Degens and Ross, 1969, colour plate mineral, and such replacements have between p. 366 and p. 367). One major been widely observed in ancient evapomechanism for tREl ~eRel alib" of rorrites for gypsum (Schaller and Henderhydrite nodules is t ~~ti.o!J...of. son, 1932; Stewart, 1949, 1951, 1953; 9YRsum cr stals to make orous lastic Borchert and Baier, 1953; Kerr and masses of tiny anh r.9.d~..JaJb.§..t.ba1.ar.e Thompson, 1963; Jones, 1965; Nurmi easilydeformedintonQ.!!.y larslug§.:. This and Friedman, 1977; Handford, 1981; conversion may just as easily occur on Presley and McGillis, 1982; Crawford deep burial as during primary surface and Dunham, 1982; Lowenstein, 1982, deposition, The same reasoning holds 1988; etc.) as well as a number of other for all other evaporite minerals that are salts (Schaller and Henderson, 1932; preserved as nodules composed of Stewart, 1963, p. 41; erc.). Pseudomasses of small crystals embedded in morphs commonly preserve a crystala host sediment, such as, for example, line framework and other primary feathe displacive nodular kieserite shown tures of the parent mineral and thus in Figure 14. allow interpretation of the original minWithout independent evidence, then, eralogy, primary processes and deposinodules may not give unambiguous tional environment, despite the minerainformation about their time, mode or logical alteration. For example, the environment of formation. fI.l..i'N w-fl
, "f" ognized in evaporite deposits. These are (1) "dehydration eair" replacement (e.g., gypsum - anhydrite), (2) "reaction pair" replacement involving a common component (e.g., gypsum - polyhalite), and (3) "non-reaction pair" replacement wherein the original and replacing minerals have no common components (e.g.,gypsum - halite). Borchert andMuir (1964, p. 126-144) ~onsider all types of pseudomorphous replacement to be metamorRhic but it has been une uivocally established that both de_h dration ~l!.cJLQ1l.P~ I§cgment§ G.ml
C 8
~
~~
~~
Figure 14 Photomicrograph of displacive kieserite nodules (gray) encircled by mud (black). Kieserite in nodules is a fine-grained granular mosaic. The uncertainties involved in interpreting this texture involve the timing ofkieserite growth andwhetherkieserite is a dehydration product of a more hydrated magnesium sulphate mineral (l.e., epsomite); and if so, what was the timing of dehydration? Sample from Salado Formation muddy halite rock. Scale bar is 5 mm long.
c
form by synde ositional inte !WJion oetween ao..eac1y foc m e d mioecal aod the evaporati her-Ii uors in the normal course of evolution of evaporite brines. For example, both anhydrite and polyhalite can be synthesized at low temperatures (Hardie, 1967, for anhydrite and Autenreith, 1958,for polyhalite), both are predicted from thermodynamic considerations to form from gypsum with progressive evaporation of seawater (Harvie et al., 1980) and
r 152
Diagenesis
Geoscience Canada Reprint Series 4
(
C C C C (
C C C C C
Figure 15 Slab photograph of anhydrite rock. Vertically oriented, elongate gypsum pseudomorphs, now composed of halite (dark gray) originate from a common surface. Laminated anhydrite at bottom of photograph. Scale bar is 5 mm long.
C
C C C (
C C
C C C C C C
C C C C C
C C C
c
e
Figure 16 Core slab photograph of carnallite-halite rock. Small halite cubes and irregular blebs (arrows, smooth surfaces) float in coarser carnallite crystals (gray textured areas). Sample from the Prairie Formation.
Figure 17 Core slab photograph of halite rock. Bedded chevron halite (light-dark bands define bedding) is abruptly truncated perpendicular to bedding by coarse, clear halite. Sample from the Prairie Formation.
c ( c c c c (
c c
Evaporites
153
(
c
c
c r
r. c (
l
o C ( (
c
both have been found as replacements of gypsum in modern sabkhas (see Butler, 1970a, for anhydrite and Holser, 1966, for polyhalite). Nonetheless, we still have no direct and unambiguous way to determine in an ancient evaporite whether such replacements were early or late. The third pseudomorph type, the "non-reaction pair" replacement, involves minerals totally unrelated chemically to each other. For example, a most common non-reaction pair pseudomorph in ancient evaporites is the replacement of euhedral gypsum prisms by halite (Figures 1 and 15) (Stewart, 1963, p. 41 ; Lowenstein, 1982, figures 4 and 5; Brodylo and Spencer, 1987, figure 10). We have no reported examples of this pseudomorph type in modern evaporites so the conditions necessary for such replacement must be inferred. All we can really say about
this replacement process is that it cannot occur during simple equilibrium evaporative concentration of a single solution, as is predicted for chemically related reaction pairs. Thus, for the pseudomorphous replacement of gypsum by halite we must call on the introduct ion of a new solution capable of dissolving gypsum without chemically reacting with it (i.e., without forming a reaction pair). Replacement by halite may either be simultaneous with gypsum dissolution (i.e., a halite saturated but gypsum undersaturated solution) or may occur later. Again the question of timing, whethersyndepositional or after burial, for this replacement sequence observed in ancient evaporites is ambiguous. However, no matter what the timing, a new solution must come into contact with a previously crystalIized mineral. Thus, fluctuating solution compositions and introduction of
c (
o
o
o l r:
c
( .
b
........ r
Figure 18 Photomicrograph of vertically oriented gypsum "swallow-tail" twin pseudomorphs with void-filling cement texture. From the pseudomorph rim to the core are: (1) a mud band, (2) anhydrite stUbby prism lining, (3) microcrystalline polyhalite lining, (4) inward projecting polyhalite prisms, and a clear halite core. This cement texture is interpreted as having formed after a gypsum crystal dissolution event. Sample from the Salado Formation. Scale bar is 3 mm long .
"alien" brines are prerequisites for this type of pseudomorphous replacement. Coarse Crystalline Salt. Some evaporite deposits contain coarse clear crystalline salts. An example of this is the carnallite from the Prairie Formation shown in Figure 16,which poikilitically encloses clear euhedral cubes of halite. The distinction between the fabric of this halite-carnallite rock and the halite-sylvite rock in Figure 7 is important in allowing the timing of the carnallite or sylvite growth to be determined. The halite in Figure 7 forms a primary crystalline framework for the sylvite cement. The halite in Figure 16 occurs as isolated cubes and lacks a primary crystalline framework. The timing of this halite growth, and of the carnallite growth around the halite is unclear. The timing of large-scale cavity or cavern fillings such as the coarse clear halite in Figure 17 is also unclear. The coarse halite crystals clearly formed after deposition of the bedded halite as well as after the intergranular cements, as the boundary of the cavity truncates these features. However, as we have already discussed. cementation of primary bedded halites probably occurs very early. Dissolution pockets in bedded halite may form syndepositionally (salt karst) or upon later burial (the cemented bedded halite framework could withstand significant burial). The coarse clear halite may grow near the surface (as in the case of the coarse euhedral halites which fill cavities in modern salt pans), or at depth. cavity Cements. A type of cavity cement which differs ' from cements discussed as syndepositional features has been commonly observed in ancient evaporites. This is where pseudomorphous replacement takes the form of a void lining and void filling cement. For example, in the Permian Salado Formation, gypsum molds that exquisitely preserve the "swallowtail" twin morphology are lined with an isopachous, euhedrally terminated fringe of anhydrite and polyhalite, which, in turn, are covered by a voidfilling halite cement (Figure 18;see also Schaller and Henderson, 1932,Plate 9). These spectacular features demonstrate that gypsym was first dissolved without destro~n; the external form of the gypsym crystals and th§ mold was~ then filled successivelykJ:lrecipitation
( Diagenesis
154
(
Geoscience Canada Reprint Series 4
( of anhydrite, then polyhalite and then halite. This sequence provides important information on the nature and composition of cement-depositing brines (Lowenstein, 1983), butthetim ing ofthe cementation is unresolved. Temperature Dependent Salts. -Phe use of salts that have restricted /(' temperature stability ranges is an attractive option in deciding a ~ depositional or burial origin. However, a difficulty arises here because there is a large overlap in temperatures experienced by primary depositional environments and secondary burial environments.AlthoughmaD~
.J1L
~
It'
evaporite settings remain at temperatures below 40°C the y-eararound (e.g., Persian Gulf sabkha, Butler, 1970a, p. 125), temperatures near 50°C due to solar heating have been measured in shallow stratified brine ponds (HUdec and Sonnenfeld, 1974). Stewart (1963) reports temperatures in the 60-70°C range for some shallow brine bodies, which is consistent with some of our own field measurements, and values of over 100°C are reached in hot spring brine pools (Holwerda and Hutchinson, 1968). Temperatures of surface sediments on salt flats commonly go above 50°C in the summertime (e.g., Butler, 1970b).J!L the normal geothermal gradient tem peratures of 50°C are reached at about 900 00 dep-th and 100°C at about 2400 m d.ep..tDv This overlap inenvironment temperatures means that "high temperature" salts such as langbeinite, loewite, vanthoffite, etc., stable only at temperatures greater than 40·50°C (Stewart, 1963,figure 10),could be of either syndepositional or bur ial origin. Nonetheless. despite th is ambiguity, the presence of such salts in a eposit tell us that temperatures were a ove at least 40·50 0C_(lhey are " minimu m temperature" geothermometers). These "high temperature" salts have been found in many of the world 's potash deposits as (1) unbedded fine- to coarse-grained mosaic patches cornposed of anhedral crystals, (2) isolated nodules consisting of single crystals or crystalline mosaics, and (3) isolated euhedral crystals. Most important of the magnesium sulphate "high temperature" minerals is langbeinite which is found in mineable amounts in several evaporites (Rempe, 1981). Spectacular examples of large (em-scale) isolated langbeinite tetrahedral crystals that
record a partly displacive, partly incorporative growth style occur in the Miocene of Russia (Lobanova, 1953), Permian of Germany (Borchert and Muir, 1964, figure 20) and Permian of New Mexico (Figure 13, see also Rempe, 1981). FLUID INCLUSIONS Fluid inclusions are extremely abundant in many evaporites,.egrtiC' !Iady i.A halites with a cr ystalline framework fabFluid inclusions may be used to determine the temperature of forwa1ioD of salts antLtbe-compositiQr.l -of-fluids from whicb..tbe.salt&.grAW.~eRoedder 1984a,b, for an overview of fluid inclusion studies). Fluid inclusions not only provide valuable information on t he Physical and chemical environments of salt formation , but also may aid in the clarification of the QrigilLoJ_s.alts..w.ith. ambiguous pet[Qgrap,bicJ eatur.a:i...
JE..
C C C
-¥:
'*
The temperature of formation of salts may be estimated from observation of multi-phase homogenization with in fluid inclusions using a heatin g-freezing sta ge on a getLogr:apbic.Jllicroscope,... The most common observation is the homogenization of liquid plus vapour with in the fluid inclusion to a single liquid phase . Manyfluid inclusions in halite crystals conta in , or nucleate on cooling , a solid daughter crystal. The final dissolut ion temperature of the solid daughter crystal gives a minimum temperature of formation of the fluid inclusion . Studies of fluid inclusions in halite, which on the basis of petrographic evidence appear to be syndepositional (e.g., " raft s" and "chevrons") from the Permian Salado Formation, New Mexico (Roedder and Belkin, 1979) show liqu id-vapour homogenization temperatures with in the range of modern salt pans. Roedder and Belkin (1979)report
(
C C C C C
C C C C C C C
C .)j )vll' ')
C C
~
1>v...Jv..lIV"-"
-/-t II J
l·ll /..J./I/'
.S~J~l/\.\N
f'l ~l tI-~
-
}tv
C C
C
fO{ C C C C C C
C
c c -e c l l Figure 19 Photomicrograph of fluid inclusions outlining primary growth bands in a halite cornet (P) and fluid inclusions within relatively clear halite cement (5). Themelting behavior of each set of inclusions is different, indicating two chemicallv distinct brines. .csr "'-
C C <..
C C
(
( (
Evaporites
c r (
( (
c (
c o o o o c r:
c r
l r: r: r: r:
D
c
,~
.
more than 300 inclusions , the majority of which homogenized between 20 and 50°C. Fluid inclusion liquid-vapour homogenization temperatures in syndepositional halites (crystalline framework fabrics) from the Devonian telegraph salts of Alberta range from 15to 40°C (Brodylo and Spencer, 1987). Although the homogenizat ion temperatures are within the range of modern evaporite basins and salt pans , there is some question as to the reliability of the results. Roedder (1984a,b) discusses the tendency of fluid inclusions in halite to "stretch" during heating, and for inclusions heated to temperatures greater than the homogenization temperature to homogenize at higher temperatures during subsequent runs. Solid-liquid homogenization temperatures are not affected by volume changes due to stretching . The majority of sylvite daughter crystals in halites with syndepositional growth fabrics from the Devonian Prairie Formation of Saskatchewan, Canada, the Permian Salado Formation of New Mex ico, USA, and the Oligocene Rhine Graben of Alsace, France, studied by Lowenstein and Spencer (in press) dissolve completely by 50 to 80°C . These temperatures are within the range of brine temperatures reached in modern evaporite settings as a result of solar heating. We believe that they are an accurate reflection of minimum surface brine temperatures during the formation of these halites. The results from the Prairie Formation are of particular interest when compared to an earlier study of liquidvapour homogenizationtemperaturesby Hartzell (1963). Hartzell (1963) measured temperatures between 51 and 104°C in more than 400 fluid inclusions in halite, but generally discarded the data as unreliable because the temperatures were too high. These halites contain well-preserved chevron and cornet fabrics typical of syndepositional halites. It is possible that heatin on burial has caused the primary inclusions to stretch see oe er an B elkin, 1979) without obliterating the original syndeposit ional fabrics .The filling temperatures would then reflect burial temperatures, but the host would still retain its syndepositional texture. However, the majority of Hartzell's determinations overlap with our determinations from sylvite daughter crys-
155
tals. They may therefore be an accurate reflect ion of depositional temperatures. We must be cautious in using Iiquidvapour homogenization data from fluid inclusions in minerals such as halite which stretch easily, but we also do not want to discard possibly valuable data on the physical and chemical environments of salt formation . Brine compositions may be determined by extraction of fluids from a large number of inclusions, a technique used by Knauth and Beeunas (1986) to determine stable isotopic compositions of fluid inclusions in halite, or by extracting fluids from relatively large (250 JLm diameter) single inclusions for major element analyses (Holser, 1963; Lazar and Holland, 1988; Stein and Krumhansl, 1988).A major drawback to these techniques is that the vast majority of fluid inclusions which are demonstrably primary on the basis of petrographic information are very small (most are less than 20 JLm on a side) and in many instances are intimately associated with larger fluid inclusions in halite cements or with secondary inclusions (Figure 19).These small primaryfluid inclusions are beyond the reach of the techniques of Lazar and Holland (1988). Furthermore, extraction of multiple inclusions leads to problems of mixing fluids from inclusions with different origins and compositions. The origin of some salts with ambiguous petrographic textures may be determined by comparing homogenization temperatures and the melting behaviour of fluids in inclusions with those from demonstrably primary or secondary salts . Observation of the melting behaviour of inclus ions allows data to be obtained from even small single fluid inclusions. As frozen fluid inclusions are warmed a number of different solid phases melt. Comparison of the melting points of these phases gives information on the chemical composition of the inclusion brines. Fluid inclusion data may be used in conjunction with petrographic information to determine the syndepositional or post depositional origin of salts. A summary of the characteristics expected of fluid inclus ions in syndepositonal salt crystals are as follows: 1."Normal shrinkage bubbles" or vapour bubbles that do not change on crushing 2. Low homogenization temperature «100°C)
3. Systematic changes in the melting behavior of inclusions from crystal core to rim (also indicates shallow setting) 4. For crystals with an ambiguous origin, similar behaviour (1, 2 and 3) to demonstrably primary syndepositional fluid inclusions from the same area. We stress that syndepositional salts may contain both primary and secondary fluid inclusions. The criteria for identifying fluid inclusions in evaporites that were trapped under burial conditions arebased on the nature of secondary evaporite brines. Secondary evaporite brines are isolated from the Earth's surface. Temperature in secondary brines may fall in the same range as primary brines, or may be higher. Temperature and pressure conditions may allow miscibility offluids that are immiscible under laboratory conditions. Fluids may be multicomponent or simple salt systems (e.g., material obtained by dissolution). Temperature and pressure changes are the primary drive for crystallization; therefore, fluid composition remains nearly constant. Fabricius (1985) studied fluid inclusions in halite and quartz crystals within salt from the Zechstein 1 and 2 of the Norwegian-Danish Basin. Quartz grew between 75 and 180°C from concentrated (>30 wt.%) Ca-Mg-CI brines. Liqu id -vapour homogenization temperatures and halite daughter crystal homogenization temperatures from fluid inclusions within the quartz crystals also allowed calculation of the pressure ( - 90 MPa) during quartz growth. The fluid inclusion data le..d...t.P.iolEm~re~ tions of the depth and timin g of saltpillow for mation in these salt domes. Fluid inclusions in evaporites altered ~ by burial may havethe following characteristics that are different from inclusions of syndepositional deposits: 1. Vapour under pressure 2. High homogenization temperature 3. Uniform melting behaviour of inclusions. POTASH DEPOSITS
A long recognised problem in evaporite geochemistry is the failure of ancient evaporites that contain the very soluble " potash" or "bittern " salts to match the mineralogical sequences predicted by the evaporation of modern seawater. ){/l' Yet, ~tically every ancient potash ~Qor iten as been [:nteTpYeteCas )/I J
I.
c 156
Diagenesis
Geoscience Canada Reprint Series 4
( (
marine and there is a general consensus that the mai~me[lt.ebem.is try of seawater.bas Dot changed significantly duriOQ..tb.eYhMerozoic (Rubey, 1951; Chave, 1960; Kramer, 1965; HoIland, 1972; Holland et al. , 1986). The discrepancy between the predicted versus observed potash-bearing minerai sequences has been explained by: (1) syndepositional modification of seawater composition by biologic processes, such as sulphate reduction (Borchert and Muir, 1964; Braitsch, 1971; Wardlaw, 1972; Sonnenfeld, 1984); (2) syndepositional modification of seawater by non-marine inflow, either meteoric (D'Ans, 1933; Stewart, 1963; Wardlaw, 1972;Valyashko, 1972),diagenetic, volcanogenic, or hydrothermal waters (Hardie, 1984); (3) non-marine parent waters (Hardie, 1984); (4) metastable equilibrium or non-equilibrium conditions that prevented precipitation of the normal seawater sequence (Valyashko, 1972); (5) syndepositional or post-depositional processes (thermal metamorphism, metasomatic alteration , complex minerai replacements, recrystallization, replacement by high-temperature salts) that have modified the primary marine sequence (Stewart, 1963; Borchert and Muir, 1964;Wardlaw,1968; Evans, 1970; Braitsch, 1971; Holser, 1979; Hardie et al., 1985). In fact, Stewart (1963, p. 4) concludes that "most large evaporite bodies have suffered widespread metasomatic changes since their original formation". We have examined the timing of precipitation of potash evaporites based on petrographic textures , fluid inclusions in associated halites , sedimentary structures and comparison with modern and Quaternary evaporites (Lowenstein and Spencer, in press). A syndepositional origin has been established for some potash salts in the Permian Salado Formation of New Mexico , Oligocene Rhine Graben of Alsace, France and Devonian Prairie Formation of Saskatchewan, Canada. Furthermore, a mechanism of forming syndepositional diagenetic potash cements by cooling of surface brines (Salado and Prairie) is presented in addition to the formation of bedded subaqueous potash deposits (Rhine Graben).
Halite crystals associated with the potash salts in these deposits contain fluid inclusion bands which outline syndepositional growth fabrics. Based on the distinct banding, parallel to growth faces, and similarity with modern salt pan and laboratory grown halites , these inclusions are primary features .The fluids contained in these inclusions are syndepositional brines. Fluid inclusions in halite associated with potash salts in the Salado, Prairie and Rhine Graben evaporites contain daughter crystals of sylvite. Oligocene Rhine Graben. Halite of the lower potash zone (Couche Inferieu re)shows remarkable preservation of depositional textures ,fabrics and primary fluid inclusion banding. Halites commonly exhibit vertically oriented chevron and cornet textures with cloudy fluid inclusion bands outlining crystal growth faces. Other halite layers are made of well-sorted mm-sized equigranular cumulate crystals or larger randomly oriented hoppers, with fluid inclusion banding, that were originally skeletal but are now filled with clear halite cement. The sylvite of the Couche Inferieure occurs in well-defined layers which may also contain halite and anhvdrite. In
many cases, sylvite layers drape the irregular surfaces of underlying chevron halite. Sylvite crystals commonly form mm-sized anhedral mosaics that lack fluid inclusion banding. In some cases, sylvite crystals are flattened and bent around large chevron halite crystals (Figure 20). Sylvite crystals in a given layer are commonly equigranular (well sorted). It is important to note that halite crystals that occur in sylvite layers are about equal in size to the sylvite crystals. The halite crystals in these layers display cumulate textures and may contain well-preserved fluid inclusion banding .The homogenization temperatures of sylvite daughter crystals in fluid inclusions from halite range from 20 to 80°C and average 63°C from a total of 92 measurements. Petrographic features of the Rhine Graben evaporites and fluid inclusion data from associatedhalitesconfirm earlier interpretations that the potash minerai sylvite is of syndepositional origin (Braitsch, 1966, 1971; Kuhn and Roth, 1979). Sylvite layers with fine-grained mosaic textures are interpreted as primary cumulates. The sylvite crystals probably formed as small cubes during cooling of the warm surface brines as orooosed bv Sturmfels (1943) and
C C
C C (
C C
C C
C C
C C
C C C C
C C C
C C C C
C C C ([
C C (
e ( ( Figure 20 Photomicrograph of halite and sylv ite from the lower potash zone (Couche inferieure), Mulhouse Basin, Tertiary Rhine Graben . Halite layer(H) is composedof centimetre size, vertically orientedchevrons that contain fluid inclusion growth bands (dark areas). Sylvite layer (S) contains millimetre-size crystals of sylvite with equigranular mosaic texture. Millimetre-size cumulate halite crystals occur in the sylvite layer. Crystals of sylvite are flattened and bent around the underlying chevrons. Scale bar is 10 mm long.
C C
C C C C
Evaporites
o o D
o
o
Braitsch (1971). The mosaic textures of the sylvite crystals were probably produced by post-depositional modification. Sylvite crystals which are flattened and bent around adjacent halite crystals attest to the ease with which sylvite may be deformed. Permian Salado Formation. Sylvite and carnallite of the Salado Formation are most common and volumetrically abundant in muddy halite rock of the McNutt Zone (Lowenstein, 1988). In these rocks, halite occurs as isolated euhedral and subhedral cubes and subhedral crystalline aggregates. These halite crystals contain incorporated mud and thin bands of fluid inclusions aligned parallel to crystal faces. Sylvite, carnallite, siliciclastic mud and small amounts of polyhalite occur among the halite crystals. Ore zones in the McNutt Potash Zone with up to 50% sylvite and carnallite display poikilitic textures where several halite crystals may be
157
encased in a single sylvite or carnallite crystal (Figure 7). In ore zones containing sylvite, primary fluid inclusions in the associated halites contain sylvite daughter crystals. Data on carnallite are limited to one ore zone in which clear halite crystals contain large isolated fluid inclusions with carnallite daughter crystals. In contrast, sylvite and carnallite daughter crystals are not found in halites away from areas of potash mineralization. Homogenization temperatures of sylvite daughter crystals from fluid inclusions which outline syndepositional growth fabrics in halites range from 28 to 150°C and average 71°C for 119 determinations. The sylvite and carnallite in the Salado Formation occupy areas equivalent to the void spaces in modern salt pan halites. Thus they are interpreted as void-filling cements surrounding a primary crystalline framework (compare
o
,.
o
Figure 21 Halite-sylvite rock from the Devonian Prairie Formation of Saskatchewan displays coarse, equigranular polygonal mosaic texture. Individual crystals are outlined by hematite concentrated along grain boundaries.
Figures 7 and 8). I he monomineralic sylvite and carnallite cements, to§ettier With the fluid inclusion data in asssoclated syndel?ositional halites allow for aeta:iiediriterpretation of the mechanisms of potash minerai formaM n. ihe sylVite (laughter crystalliomogenization temperatures from the fluid inclusions in primary halites establish that the surface brines from which the halite crystallized were saturated with respect to sylvite at moderate temperatures . Following desiccation of the surface brine lake, sylvitecementsformed asthe interstitial brines cooled. Based on petrographic textures, a similar syndepositional origin is suggested for carnallite. We, however, have no independent evidence from fluid inclusions for syndepositional carnallite saturated brines. Devonian Prairie Formation. Halite from the Prairie Formation generally contains well preserved syndepositional textures and fabrics. Crystalline growth fabrics, mainly vertically elongated chevrons and cornets, are present in 50 to 90% of the halite from many intervals (Wardlaw and Schwerdtner, 1966 ; Wardlaw and Watson, 1966; Brodylo and Spencer, 1987). The halite conta ins many features, particularly syndepositional dissolution textures , that are typical of modern salt pans described by Lowenstein and Hardie (1985). Convincing evidence that sylvite of the Prairie Formation formed at the expense of carnallite is the lateral change in stratigraphic thickness between sylvite and carnallite-bearing halites (Wardlaw, 1968). However, Wardlaw (1968) also presents petrographic evidence that some of the carnallite has formed at the expense of sylvite. Halite associated with potash salts of the Prairie Formation is atypical , and generally lacks well-defined syndepositional textures and fabrics. Sylvitehalite rocks are coarsely crystalline and commonly display equigranular polygonal mosaic textures (Figure 21). Sylvite in some cases occupies areas equivalent to void spaces in modern salt pan halites, and is similar to the sylvite in the Salado Formation that cements a primary crystalline framework (compare Figures 7 and 8 with Figure 22). Gradations between the two "end member" sylvite textures shown in Figures 21 and 22 are also found in the Prairie Formation. In many cases, sylvite grain boundaries are highlighted by concentrat ions
( 158
Diagenesis
Geoscience Canada Reprint Series 4
( (
of blood-red hematite . Hematite is less abundant, or absent, in grain interiors which are also clear of fluid inclusions. However, some halite crystals in the polygonal mosaics and primary crystalline frameworks contain cloudy fluid inclusion-rich cores. The fluid inclusions outline remnant syndepositional crystal growth (Figure 23). Carnallite-halite rocks of the Prairie Formation display several textures. Many samples contain isolated centimetre-scale cubes of halite, which are enclosed by large poikilitic carnallite crystals (Figure 16). Most halite cubes are clear, but some contain cloudy, fluid inclusion banded cores, similarto those in Figure 23, which contain patches of syndepositional crystal growth textures . The euhedral, clear halite which overgrows these cloudy cores contains large isolated fluid inclusions with carnallite daughter crystals, and is interpreted as a cement. A second type of carnallite-halite rock consists of alternating centimetre-thick beds of halite and carnallite (Figure 24). The halite layers contain some remnant syndepositional growth fabrics . However, most crystals are clear. Minor amounts of carnallite are present as a void filling cement within the halite beds (Figure 25). The carnallite beds contain centimetre-scale isolated cubes of halite . This type of bedding is similar to halite-sylvite rocksofthe Rhine Graben. The Prairie Formation also contains intergrowths of large (several centimetre) clear halite and carnallite crystals (Figure 26). Neither the halite or carnallite contain any remnant syndepositional growth fabrics. Fluid inclusions in halite from the Prairie Formation contain an aqueous phase and may also contain daughter crystals of sylvite, carnallite, or chlorocalcite (onesample; Figure 27). All three daughter minerals are found in isolated fluid inclusions from clear halite . Sylvite daughtercrystals are found in mostfluid inclusions which outline syndepositional growth banding in the chevron and cornet halite associated with sylvite or carnallite. Carnallite daughter crystals have been observed in fluid inclusions outlining syndepositional growth fabrics associated with interbedded halite and carnallite such as those shown in Figures 24 and 25. 'laughter erystals are Ret present at room tern erature in an fluid inclusions
C
C C C C ( (
C C C C C
C (
C C Figure 22 Photomicrograph of halite-sylvite rock from the Prairie Formation. Halite is present as clearcubes (H), poikilitically enclosedby syivtte (5, gray in photograph). Scale baris 3mmlong.
C C
C C C C
C C C C C
C C C (
C (
Figure 23 Photomicrograph of halite-sylvite rock displaying coarse, equigranular polygonal mosaic texture. Halite crystals (H)contain fluid inclusion bands (dark areas within halite) outlining primary growth structures. Sample from the Prairie Formation. Scale bar is 2 mm long.
c c c l
c C
Evaporites
r:
159
\
(' r:
G C C (" (
(
Figure 24 Coreslab photograph of bedded halite-carnallite rock. A thin carbonate-sulphate layer is present at the base of each halite layer. Halite (h) beds are composedof clearinterlocking crystals with some remnants of chevron growth banding. Carnallite beds (c) contain isolated halite crystals poikilitically enclosed in coarser carnallite crystals. Sample from the Prairie Formation.
Figure 26 Core slab photograph of coarse crystalline halite-carnallite rock. Large halite crystals (h) are clear and do not contain any remnants of primary growth fabric. Carnallite crystals (c) contain inclusions of hematite, but do not display primary growth fabrics . Sample from the Prairie Formation.
Figure 25 Photomicrograph of halite bed from bedded halite-carnallite rock. Fluid inclusion bands outline primary cornet growth fabric halite crys tal at lower left. Carnallite (gray) fills void among clear halite crystals. Sample from Prairie Formation. Scale bar is 3 mm long.
160
awayfrom the potash zones, nor do any of the above daughter minerals nucleate on cooling. Sylvite daughter crystals from fluid inclusions in the cloudy cores of halite crystals associated with sylvite and with the poikilitic carnallite dissolve between 55 and 61°C. Note that sylvite daughter crystals and not carnallite daughter crystals are present in syndepositional fluid inclusions in halite-sylvite and most halite-carnallite rocks. Fluid inclusions in clear halite cements associated with carnallite do contain carnallite daughter crystals. Most of the carnallite daughter crystals dissolve between 50 and80°C, approximately the same temperature range as the sylvite daughter crystals. Sylvite-halite rocks in which the sylvite occurs as a void filling within a syndepositional crystalline framework of halite (Figure 22) probably formed as warm surface brines cooled during desiccation, as proposed for similar rocks from the Salado Formation. Coarse mosaic textures in halite-sylvite rocks appear to have been produced by recrystallization. The present s~lvite..: ~t~tul:ElS-may..r;ap~sJm! .!lqlj!~.-r.ock,..cemented..b.y. e.al;l¥_s~~,
and recrystallized upon later burial. Fluid inclusions which outline primary growth fabrics in halite record a progressive increase in the potassium content of the syndepositional surface brines. The increase is seen on a centimetre-scale within individual depositional sequences and on a larger (several metre) scale immediately below potash salt ore zones. As in the case of the Salado Formation, the surface brines were saturated with respect to sylvite at moderate temperatures. Interbedded carnallite-halite rocks contain halites with fluid inclusions which outline syndeposional growth bands. These fluid inclusions contain daughter crystals of carnallite. Interbedded carnallite-halite rocks were probably formed as bedded subaqueous deposits. Carnallite may have formed as cumulate crystals, similar to the sylvite from the Rhine Graben, and as interstitial cements within the halite framework. Cooling of warm surface brines is a likely mechanism for the carnallite crystallization. The original textures were modified post-clepositionally. The other types of carnallite-halite rocks from the Prairie Formation also
Diagenesis
appear to have been modified by or formed by post-depositional processes. These rocks may have originated as syndepositional bedded subaqueous deposits or as void filling cements around a framework of crystalline halite which has since recrystallized. However, most of the halite-carnallite rocks examined from the Prairie Formation contain no remnant syndepositional fabrics indicative of bedded subaqueous deposition. Another problem is that much of the potash in the Prairie Formation iscarnallite, butthedaughter crystals in fluid inclusions from most remnant syndepositional growth bands in associated halites are sylvite. Carnallite daughter crystals are found in clear halite associated with the carnallite. However,the inclusions and the clear halite in which they are found may have formed syndepositionally or during burial. Therefore, the origin and timing of much of the carnallite from the Prairie formation is unclear. SUMMARY In this paper, we have presented a set of petrographic criteria aimed at dist.i!lQ.Y.i.S.b.D.Q..Synde.p.asuionaJ..featur.es from burial features found in evaporites. The criteria are- -based pnman y on observations from modern near-surface evaporite environments. Nearsurface processes which influence eva-
Geoscience Canada Reprint Series 4
porites and the results of these processes have been more extensively studied than those related to the burial environment. Some bias toward interpretation of petrographic features as '"' being the result of syndepositional, rather than burial, processes may be present, especially for petrographicfeatures which might form either syndepositionally or as a result of burial. There is clearly a need for further study of the burial environment. Syndepositional sedimentary features which might be found in evaporites include sedimentary structures* and crystalline framework fabrics. Syndepositional alteration or early diagenetic modification of evaporites include dissolution-reprecipitation features and cement formation. Extensive cementation of modern evaporites occurs within a few metres of the surface. Burial alteration features that might be present in evaporite deposits include disruption and destruction of sedimentary structures, sutured or polygonal mosaic structures and deformation features. In addition to these features, there are_se~etal_reatur.e'S ound in evaporites whose timing and/or origin is ambiguous. These include intrasedi-ment grow!.., of euhedral crystals or nodules, coarse crystalline salt pseu~m-orpbQus rep acem~n~s a~d_ some ~s _
( ( ( ( ( (
C ( r: '-r
( ( (
C C (
c c ( c c (
( ( ( (
l
C ( r:
( Figure 27 Photomicrographs of daughter crystals in fluid inclusions in halite from the Devonian Prairie Formation. The daughter crystals indicate highly concentrated brines with diverse chemical compositions were present.
( (
l
C
(
00 0
c c c
C G
C
o (
CO (
Evaporites
~
Data obtained from fluid inclus,i0nsin
If' evaporite minerals can De used In con-
jUfiCtlo n with the p.At rog r.~pl1i..9J!Jfc>'"rriia.:-~ tion to hel p--deterJD ine..the ~timir:1g~ar:1d possible ..!!Jg~J:\an isms_oLm io.~@Lfllit. cipitation. Fluid inclusion studies from potas n aeposits have helped to determine the timing of potash mineral growth in some deposits. The analysis of brine compositions from fluid inclusions may also yield information on chemical environments evaporites have encountered during burial. Brine composition and hydrologic factorsare important f~ta1UJ:esjD~tt!.MJ.jrla.Ldi~g~ esis of evapoLiteswbjcb.reguire fyr.tb~r stud~
c
c o o o o o
c o o c c (
o o o , C b ·
'-"'
REFERENCES Adams, J.E. and Rhodes, M.L., 1960, Dolomitization by seepage refluxion: American Association of Petroleum Geologists , Bulletin, v. 44, p. 1912-1921. Arakel , A.V., 1980, Genesis and diagenesis of Holocene evaporitic sediments in Hutt and Leeman Lagoons, Western Australia: Journal of Sedimentary Petrology, v. 50, p. 1305-1326. Arthurton, R.S., 1973, Experimentally produced halite compared to Triassic layered halite-rocksfromCheshire,England:Sedimentology, v. 20, p. 145-160. Aulstead, K.L. and Spencer, R.J., 1985, Diagenesis of the Keg River Formation , northwestern Alberta: fluid inclusion evidence: Bulletin of Canadian Petroleum Geology, v. 33, p. 167-183. Aulstead, K.L., Spencer, R.J. and Krouse, H.A., 1988, Fluid inclusion and isotopic evidence on dolom itization, Devonian of Western Canada : Geochimica et CosmochimicaActa, v. 52, p. 1027-1035. Autenr ieth, H., 1958, Untersuchungen am Sechs-Komponenten-System KO, Nao, Mg· , Ca·, S04'" (CI'), H 20 mit Schlussfolgerungen fUr die Verarbeitung der Kalisalze: Kali und Steinsalz, v. 2, p. 181-200. Balk, R., 1949, Structure of Grand Saline salt dome, Van Zandt County, Texas: American Association of Petroleum Geologists, Bulletin , v. 33, p. 1791-1829. Bathurst, R.G.C., 1975, Carbonate sediments and their diagenesis, Second Edition : Elsevier, Amsterdam, 658 p. Borchert, H. and Baier, E., 1953, Zur metamorphose ozeaner Gipsablagerungen: Neues Jahrbuch fiir Mineralogie Abhandlungen, v. 86, p. 103-154. Borchert, H. and Muir, A.O ., 1964, Salt Deposits: The Origin, Metamorphism and Deformation of Evaporites: D. Van Nostrand, Princeton, NJ, 300 p.
yt,.\)
n
( I
IVI) y\,..R. _
cr;A./\
~ ( U
I '"
v
( 11/1.'
\
Braitsch, 0 ., 1966,Brom ine and rubidium as indicators of environment during sylv ite and carnallite deposit ion of the Upper Rhine Valley evaporites, in Rau, J.L., ed., Second Symposium on Salt: Northern Ohio Geological Society, Cleveland, Ohio, p. 293-301. Braitsch, 0 ., 1971, Salt Deposits . Their Or igin and Composition: Springer-Verlag, New York, 297 p, Brodylo , L.A. and Spence r, R.J., 1987, Depositional environment of the Telegraph salt: Bulletin of Canadian Petroleum Geology, v. 35, p. 186-196. Buckley, H.E., 1951, Crystal Growth : John Wiley and Sons, New York, 571 p. Burgers , W.G., 1963, Principles of recrystallization, in Gilman, J.J., ed., The Art and Science of Growing Crystals: John Wiley and Sons , New York, p.416-450. Butler, G.P., 1970a, Holocene gypsum and anhydri te of the Abu Dhabi sabkha, Trucial Coast: an alternative explanation of origin: Third Sympos ium on Salt, Northern Ohio Geological Society, Cleveland , Ohio, p. 120-152. Butler, G.P., 1970b, Secondary anyhydrite from a sabkha, northwest Gulfof California, Mexico: Third Symposium on Salt , Northern Ohio Geological Society, Cleveland , Ohio, p. 153-155. Caldwell , A.H., 1976,Holocene gypsum deposits of the Bullara Sunkland, Carnarvon Basin, Western Australia, Ph.D . thesis, University of Western Australia, Nedlands, Western Australia, 123 p. Carpenter A.B., 1978, Origin and chemical evolution brines in sedimentary basins : Oklahoma Geolog ical Survey Circular, v. 79, p. 60-77. Casas, E., 1988, Diagenesis of salt pan halite , unpublished Masters thesis , State University of New York at Binghamton , 82 p. Casas, E. and Lowenste in, T.K., 1987, Shallow burial diagenesis of Quaternary halite rocks, Basin and Range Province, U.S. (abstr.): Society of Economic Paleontologists and Mineralogists, Annual Midyear Meeting , v. 4, 14 p. Chave, K.E., 1960, Evidence on history of sea water from chemistry of deeper subsurface waters of ancient basins: American Association of Petroleum Geologists , Bulletin, v. 44, p. 357-370. Crawford , G.A. and Dunham , J.B. , 1982, Evaporite sedimentation in the Permian Yates Formation, Central Basin Platform, Andrews County, West Texas, in Handford, C .R ., Loucks , R.G. and Davies, G.R. , eds ., Depositional and Diagenetic Spectra of Evaporites- A Core Workshop: Society of Economic Paleontologists and Mineralogists, Core Workshop No.3, Calgary, Canada, 1982, p.238-275.
---.. \
161
Curtis , R., Evans, G., Kinsman, D.J.J. and Shearman, D.J., 1963, Association of dolomite and anydrite in the Recentsediments of the Persian Gulf: Nature, v. 143 (3607), p. 679-680. D'Ans, J., 1933, Die losungsgleichgewichte der systeme der salze ozeanischer salzblagerungen: Kali-Forschungsanstalt, Verlagsgesellschaft fiir Ackerbau , Berlin, 254 p. Degens, E.T. and Ross, D.A., 1969, 1969, Hot Brines and Recent Heavy MetalDeposits in the Red Sea: Springer-Verlag, New York, 600 p. Dellwig, L.F., 1955,OriginoftheSalinaSaltof Michigan : Journal of Sedimentary Petrology, v. 25, p. 83·110. Dellwig, L.F., 1968, Significant features of deposition in the Hutchinson Salt, Kansas, and their interpretation, in Mattox, A.B., ed.,SalineDeposits:GeologicalSociety of America, Special Paper 88 , p.421-426. Dellwig, L.F. and Evans, R., 1969, Depositional processes in Salina salt of Michigan, Ohio and New York: AmericanAssociation of PetroleumGeologists, Bulletin, v. 53, p. 949-956. Eugster, H.P., 1970, Chemistry and origin of brines of Lake Magadi, Kenya:American Mineralogical Society,SpecialPaperv. 3, p.213-235. Evans, R., 1970, Genesis of sylvite and carnallite-bearing rocks from Wallace, Nova Scotia, in Rau,J.L. and Dellwig, L.F.,eds., Third Symposium on Salt: Northern Ohio Geological Society, Cleveland , Ohio p.239-245. Fabricius, J., 1985,Studies of fluid inclusions in halite and euhedral quartz crystals from salt domesin the Norwegian-Danish Basin, in Schreiber,B.C., ed., Sixth International Symposium on Salt: Salt Institute , Alexandria, Virginia, p. 247-255. Goldman , M.I., 1952, Deformation, metamorphism , and mineralizat ion in gypsumanhydrite cap rock, Sulphur Salt Dome, Louisiana: Geological Society of America, Memoir 50, 163 p. Gottesmann , W. von., 1963, Eine Haufig auftretende struktur des Halits in Kalifloz Stassfurt: Geologie, v. 12, p.576-58t Handford, C.R., 1981, Coastal sabkha and salt pan deposition of the Lower Clear Fork Formation (Permian), Texas: Journal of Sedimentary Petrology, v. 51, p.761-778 . Hardie, L.A., 1967, The gypsum-anhydrite equilibirum at one atmosphere pressure: American Mineralogist, v. 52, p. 171-200. Hardie, L.A., 1968, The origin of the Recent non-marine evaporite deposit of Saline Valley, Inyo County, California: Geochimica et Cosmochimica Acta, v. 32, p. 1279-130t
Diagenesis
162
Hardie, L.A., 1984, Evaporites: marine or non-marine? American Journal of Science, v. 284, p. 193-240 Hardie, L.A. and Eugster, H.P., 1971, The depositional environment of marine evaporites : a case for shallow, clastic accumulation: Sedimentology, v. 16, p.187-220. Hardie, L.A., Lowenstein, TK and Spencer, R.J., 1985, The problem of distinguishing between primary and secondary features in evaporites , in Schreiber, B.C., ed., Sixth International Symposium on Salt: Salt Institute, Alexandria, Virginia . p. 11-39. Hartzell, W.G., 1963, Liquid inclusions in halite, B.Sc. thesis, University of Saskatchewan, Canada. Harvie, C.E., Wear, J.H., Hardie, L.A. and Eugster, H.P., 1980, Evaporation of sea water : calculated mineral sequences : Science, v. 208, p. 498-500. Hitchon, B., Billings, G.K. and Klovan, J.E., 1971, Geochemistry and origin of formation waters in the western Canada sedimentary basin - III. Factors controlling chemical compos ition : Geochimica et Cosmochimica Acta, v. 35, p. 567·598. Holland, H.D., 1972, The geologic history of sea water - an attempt to solve the problem: Geochimica et Cosmochimica Acta, v. 36, p. 637-651. Holland, H.D., Lazar, B. and McCaffrey, M., 1986, Evolution of the atmosphere and oceans: Nature, v. 320, p. 27-33. Holser, 1963, Chemistry of brine inclusions in Permian salt from Hutchinson, Kansas, in Symposium on Salt: Northern Ohio Geological Survey,Cleveland, Ohio, p.86-95. Holser, W.T. , 1966, Diagenetic polyhalite in recent salt from Baja California: American Mineralogist, v. 5, p. 99-109. Holser, W.T., 1979, Trace elements and isotopes in evaporites, in Burns, R.G., ed., Marine Minerals: Mineralogical Society of America, Short Course Notes Volume 6, chapter 9, p, 295-346. Holwerda, J.G. and Hutchinson, RW., 1968, Potash·bearing evaporites in the Danakil Area, Ethiopia: Economic Geology, v. 63, p.124-150. Hovorka, S.D., 1987, Depositional environments of marine-dornlnated bedded halite, Permian San Andres Formation, Texas: Sedimentology, v. 34, p. 10291054. Hudec, P.P. and Sonnenfeld, P., 1974, Hot brines on Los Roques , Venezuela: Science, v. 185, p. 440-442. Ingerson, E., 1968, Deposition and geochemistry work sessions (International Conference on Saline Deposits), Mattox, R.B., ed., Saline Deposits : Geological Society of America , Special Paper 88, p. 671-681.
m .,
Jones , C.L., 1965, Petrograp hy of evaporites from the Wellington Formation near Hutchinson , Kansas : United Sta tes Geological Survey, Bullet in, v. 1201-A, 70 p. Kerr, S.D. and Thompson, A., 1963, Origin of nodular and bedded anhydrite in Permian Shelf sediments, Texas and New Mexico: American Association of Petroleum Geolog ists , Bulletin , v.47, p. 1726-1732. Keys, D.A. and Wright, J.Y., 1966, Geology of the I.M.C. potash depos it, Esterhazy, Saskatchewan, in Rau, J.L., ed., Second Symposium on Salt: Northern Ohio Geological Soc iety, Cleve land, Oh io , p.95-101. Kramer, J.R., 1965, History of sea water. Constant temperature-pressure equilibrium models compared to liquid inclusion analyses: Geoch imica et Cosmochimica Acta, v. 29, p. 921·945. Kuhn, R., 1968, Geochemistry of the German potash deposits, in Mattox , R.B., ed. , Saline Deposits: Geological Society of Amer ica , Spec ial Paper 88 , p.427-504. Kuhn, R. and Roth, H., 1979, Beitrage zur kenntnis der salzlagerstatte am Oberrhein: Z. Geol. Wiss., v. 7, p. 953-966. Knauth, L.P. and Beeunas, M.A., 1986, Isotope geochemistry of flu id inclusions in Perm ian halite with implications for the isotopic history of ocean water and the origin of saline formation waters: Geochimica et Cosmochimica Acta , v. 50, p.419·433. Lazar, B. and Holland, H.D., 1988, The analysis of fluid inclusions in halite: Geochimica et Cosmochimica Acta, v. 52, p.485-490. Lobanova, V.V. , 1953, On the genesis of the langbeinite in the Sub-Carpathian potash deposits: Proceedings Academy of Science , U.S .S .R ., v.88, no . 1, p. 145-148. [ in Russian] Logan, B.W., Davies, G.R., Read, J .R. and Cebulski, D.F., 1970, Carbonate sedimentation and environments, Shark Bay, Western Australia: American Association of Petroleum Geolog ists, Memoir 13,223 p. Lowenstein, T.K., 1982, Primary features in a potash evaporite deposit, the Permian HanSalado Formation of West Texas and New Mexico, in Handford, C.R. , Loucks, R.G. and Davies, G.R., eds., Depositional and Diagenetic Spectra of Evaporites· A Core Workshop: Society of Econom ic Paleontologists and Mineralogists, Core Workshop No.3, Calgary, Canada, 1982, p. 276-304 .
Geoscience Canada Reprint Series 4
Lowenstein , T.K., 1983, Depos it ion and alteratio n of an ancient potash evaporite: the Permian Salado Formation of New Mexico and West Texas, Ph.D. dissertation, The Johns Hopkins University, Baltimore , MD, 411 p, Lowenstein , T.K., 1988, Origin of depositional cycles in a Permian " saline giant" : The Salado (McNu tt Zone) evaporites of New Mexico and Texas : Geological Society of Ame rica , Bulleti n, v. 100, p.592-608. Lowenstein , T.K. and Hardie, L.A., 1985, Criteria for the recognition of salt-pan evaporites : Sed imentolog y, v.32, p.627-644. Mayrhofer, H., 1985, World reserves of mineable potash salts based on structural analysis , in Schreiber, B.C., ed., Sixth Internat ional Symposium on Salt: Salt Ins titute , Alexandria, Virg in ia, p. 141-160. Meijer Drees, N.C., 1986, Evaporitic deposits of western Canada: Geological Survey of Canada , Paper 85-20, 118 p. Mossop, G.D., 1979, The evaporites of the Ordovician Baumann Fjord Formation, Ellesmere Island, Arctic Canada: Geological Survey of Canada, Bulletin No. 298,52 p. Morrow, D.W., Cumming, G.L. and Koepnick, R.B., 1986, Manetoe Facies - a gasbearing , megacrystall ine , Devon ian dolomite , Yukon and Northwest Territories , Canada: American Association of Petroleum Geologists, Bullet in, v. 70, p.702-720. Muller, W.H., Schmid, S.M. and Briegel, U., 1981, Deformat ion experiments on anhydrite rocks of different grain sizes: rheology and microfabrics : Tectonophysics, v. 78, p. 527-544. Nahnybida , C., Hutcheon, I. and Kirker, J., 1982, Diagenesis olthe Nisku Formation and the orig in of late stage cements: Canadian Mineralogy, v. 20, p. 129-140. Nurm i, R.D. and Friedman, G.M., 1977,Sedimentology and depositional environments of basin-center evaporites, Lower Salina Group (Upper Silurian), Michigan Basin, in Fisher, J.H., ed., Reefs and Evaporites: Concepts and Depositional Models: American Association of Petroleum Geolog ists, Studies in Geology, v. 5, p. 23-52. Parea, G.C. and Ricchi-Lucch i, F., 1972, Resedimented evaporites in the Periadriatic trough : Israel Journa l of Earth Sciences , v. 21, p. 125-141. Presley, MW. and McGillis, K.A., 1982, Coastal evaporite and tidal-flat sediments of the Upper Clear Fork and Glorieta Formations, Texas Panhandle: University of Texas at Austin, Bureau of Economic Geology, Report of Investigations No. 115, 50 p.
c c c c c c c c c c c c c c c c c c c c c c c c c
c c c c c c c c c
e c c c c c
c C'
Evaporites
163
C
c. C. G C ( (
c'
(
o o ( o o o o o o
r. (;
o
o o o o
G 5
r.
,.-
Prucha, J.J., 1968, Salt deformation and decollement in the Firtree Point anticline of central New York: Tectonophysics, v. 6, p. 273-299 . Rempe , N.T., 1981, Langbeinite in potash deposits: 17th Annual Forum on the Geology of Industrial Minerals, Albu querque, New Mexico, May 1981. Ricci-Lucchi, F., 1973, Resedimented evaporites : Indicators of slope instability and deep-basin conditions in Periadriatic Messinian (Apennines fore-deep, Italy), in Drooger, C.w., ed., Messinian Events in the Mediterranean: North-Holland, Amsterdam , p. 142-149. Rittenhouse, G., 1967, Brom ine in oilfie ld waters and its use in determining possibilities of origin ofthese waters: American Association of Petroleum Geologists , Bulletin, v. 51, p. 2430-2440. Roedder, E., 1984a, Fluid Inclusions: Mineralog ical Society of America, Reviews in Mineralogy, v. 12, 644 p. Roedder, E., 1984b, The fluids in salt: American Mineralogist, v. 69, p. 413-439. Roedder, E. and Belkin, H.E., 1979, Application of studies of fluid inclusions in Permian Salado salt, New Mexico, to problems of sit ing the Waste Isolation Pilot Plant, in McCarthy, G.J., ed. , Scientific Basis for Nuclear Waste Management: Plenum Press , New York, v. 1, p.313-321. Rubey, W.W., 1951, Geologic history of sea water - an attempt to state the problem : Geological Soc iety of America, Bulletin, v. 62, p. 1111-1147. Schaller, W.T. and Henderson , E.P., 1932, Mineralogy of drill cores from the potash field of New Mexico and Texas: United States Geological Survey, Bulletin , v. 833, 124 p. Schlager, W. and Bolz, H., 1977, Clastic accumulation of sulfate evaporites in deep water: Journal of Sedimentary Petrology, v. 47, p. 600-609. Schlichta, P.J., 1968, Grow1h, deformation and defect-structure of salt crystals, in Mattox, A.B., ed., Saline Deposits: Geological Society of America, Special Paper88,p.59~617.
Schreiber, B.C., Catalano, R. and Schreiber, E., 1977, An evaporit ic lithofacies continuum: Latest Miocene (Messinian) deposits of Salemi Basin (Sicily) and a modern analog, in Fisher, J.M., ed., Reefs and Evaporites - Concepts and Depositional Models: American Association of Petroleum Geologists, Studies in Geology, v. 5, p. 169-180. Schrieber, B.C., Friedman, G.M ., Decima, A. and Schreiber, R., 1976, Depositional environments of Upper Miocene (Messinian) evaporite deposits of the Sicilian Basin : Sedimentology, v.23, p.729-760.
Schreiber, B.C. and Hsu, K.J., 1980, Evaporites, in Hobson , G.D., ed., Developments in Petroleum Geology - 2: Appl ied Science Publications, Barking, Essex, England , p. 87-138. Schreiber, B.C. and Kinsman, K.J.J ., 1975, New observations on the Pleistocene evaporites of Montallegro, Sicily, and a modern analogue: Journal of Sedimentary Petrology, v. 45, p. 469-479. Schreiber, B.C., Roth, M.S. and Helman, M .L. , 1982 , Recognition of primary facies characteristics of evaporites and the differentiation of these forms from diagenetic overprints, in Handford, C.R., Loucks, R.G. and Davies , G.R., eds., Depositional and Diagenetic Spectra of Evapor ites - A Core Workshop: Society of Economic Paleontologists and Mineralogists, Core Workshop No. 3, Calgary, Canada 1982, p. 1-32. Schreiber, B.C . and Schreiber, E., 1977, The salt that was : Geology, v. 5, p. 527-528. Schwerdtner, W.M., 1964, Genesis of potash rocks in Middle Devonian Prairie evaporite Formation of Saskatchewan: American Association of Petroleum Geologists, Bulletin, v. 48 , p. 1108·1115. Shearman , D.J., 1966,Origin ofmarineevaporites by diagenesis : Institute of Mining and Metallurgy (Newcastle-upon-Tyne), Transactions , v. 75, p. 208-215. Shearman, D.J., 1970, Recent halite rock, Baja, California, Mexico: Institute of Mining and Metallurgy, Transactions, v. B79, p.155-162. Sonnenfeld, P., 1984, Brines and Evaporites : Academic Press lnc., Orlando, FL, 613 p. Spencer, R.J., 1987, The or igin of Ca-CI brines in Devonian formations, western Canada sedimentary basin: Applied Geochemistry, v. 2, p. 373-384. Stanton, R.L. and Gorman, H., 1968, A phenomenological study of grain boundary migration in some common sulfides: Economic Geology, v. 63, p. 907-923. Stein , C.L. and Krumhansl, J.L. , 1988, A model for the evolution of brines in salt from the lower Salado Formation, southeastern New Mexico : Geochimica et Cosmochimica Acta, v. 52, p. 1037-1046. Stewart, F.H., 1949, The petrology of the evaporites of the Eskdale No.2 bor ing, east Yorkshire. Part I. The middle evaporite bed: Mineralogical Magazine. v. 29, p.445-475. Stewart, F.H., 1951, The petrology of the Eskdate No. 2 boring east Yorkshire, Part II, The middle evaporite bed: Mineralogical Magazine, v. 29, p. 445-475 . Stewart, F.H., 1953, Early gypsum in the Permian evaporites of north-eastern England: Proceedings of Geological Assoc iation of London , v. 64, p. 33-39 .
Stewart, F.H., 1963, Marine evaporites: United States Geological Survey, Professional Paper 440-'(, 53 p. Sturmfels, E., 1943, Das kalisalzlager von Buggingen (Sud baden): Neues Jahrbuch fur Mineralogie, Geologie und Palaontoloqle, v. 78, p. 131-216. Thompson, R.W., 1968, Tidal flat sedimentation on the Colorado River delta, northwestern Gulf of California: Geological Society of America , Memoir 107, 133 p. Ueda, A., Campbell, FA , Krouse , H.R. and Spencer, R.J., 1987,34S/32S variations in trace sulf ide and sulfate in carbonate rocks of a Devonian reef, Alberta, Canada and the Precambrian Siyeh Formation, Montana, U.S.A.: Chemical Geology, Isotope Geoscience Section, v. 65, p.383-390. Valyashko , M.G., 1972, Playa lakes - a necessary stage in the development of a salt-bearing basin, in Richter-Bernburg, G., ed., Geology of Saline Deposits: UNESCO , Earth Sciences Series, Paris, v. 7, p. 41-51, VolI, G., 1960, New work on petrofabrics: Liverpool and Manchester Geological Journal, v. 2(3), p. 503-56? VolI, G., 1976, Recrystallization of quartzbiotite and feldspars from Eastfeld to the Leventina Nappe, Swiss Alps and its geological significance : Schweizerische Mineralogische und Petrographische Mitteilungen, Bulletin, v.56, p.641-647. Wardlaw, N.C., 1968,Carnallite-sylvite relationships in the Middle Devonian Prairie Evaporite Formation Saskatchewan: Geological Society of America, Bulletin, v. 79, p. 1273-1294. Wardlaw, N.C., 1972,Unusual marine evaporites with salts of calcium and magnesium chloride in Cretaceous basins of Sergipe, Brazil: Economic Geology, v. 67, p. 156-168. Wardlaw, N.C . and Schwerdtner, W.M. , 1966, Halite -anhydrite seasonal layers in the Middle Devonian Prairie Evapor ite Formation, Saskatchewan, Canada: Geological Society of America, Bulletin, v. 77, p. 331-342. Wardlaw, N.C. and Watson , D.W., 1966, Middle Devonian salt formations and their brom ide content , Elk Point area, Alberta: Canadian Journal of Earth Sciences, v. 3, p. 263-275. Warren, J.K., 1982, The hydrolog ical setting, occurrence and significance of gypsum in late Quaternary salt lakes in South Australia: Sedimentology, v. 29, p.609-639 . Weiler, Y., Sass, E. and Zak, I., 1974, Halite oolites and ripples in the Dead Sea, Israel: Sedimentology, v. 21,p. 623-632 . Accepted 1988
c 164
Diagenesis
Geoscience Canada Reprint Series 4
c
C C C C (
C
C (
C C C C C C C C C
C C C
C C C C
Pore cast of a sandstone. All mineral matter has been dissolved and pore spaces stand in relief. Pores are connected by throats which are sheet-like and represent spaces at grain contacts. Scanning electron photomicrograph. Photograph courtesy of N. Wardlaw (University of Calgary).
C C C C C C C
e
c
C l
C (
C C
(
c c
Coarse-grained Siliciclastic Rocks
( r:
(; (
o
c ( (
o o o o o o (
o o o o o o o o o o (
Aspects of the Diagenesis of Coarse-Grained Siliciclastic Rocks Ian Hutcheon Geology and Geophysics The University of Calgary Calgary, Alberta T2N 1N4 SUMMARY The process of diagenesis, by which sediments become rock, has many physical and chemical aspects. Even a cursory examination of the scientific literature shows that mineral reactions, original depositional environment, diagenetic environment, the flow of fluid, the evolution of organic matter, the compaction of sediment and the precipitation , replacement and destruction of cements all play a role in sandstone diagenesis. The importance of sandstone diagenesis is evidenced by the rapid increase of pure and applied research in this area. The oil and gas industry has greatly accelerated research into sandstone diagenesis, primarily because the nature and distribution of porosity and permeability ultimately governs the extent of oil and gas recovery, and diagenes is is a controlling factor in the formation and destruction of porosity. In addition , as our demand for oil and gas grows, and the relatively easily extracted oil is exploited, there is an increasing emphasis on secondary and tertiary recovery of conventional oil and recovery of heavy oil and tar sands. f Enhanced recovery agents, whether physical (injected water or steam) or chemical (surfactants, acids , polymers, caustic and clay stabilizing agents), are in direct contact with the pore space of the rock. It is important to understand the present state of diagenesis of any rock, and how that state will be affected by an agent used to stimulate recovery.
165
While it istemptingtodivide the subject of diagenesisinto those processeswhich are physical and those which are chemical, the interdependence of physical and chemical processes makes this impossible. The reader also should be warned that even sandstone diagenesis itself cannot be separated from the interplay of differing sedimentary environments, lithologic variation within basins, and setting and evolution of sedimentary basins. All of these factors are ultimately governed by the tectonic setting. INTRODUCTION Sandstone diagenesis is a subject which has been investigated by many different methods, and this reflects the nature of the processes governing diagenesis.ln this review paper, the chemical aspects of the diagenesis of coarsegrained siliciclastic rocks are emphasized. In the broadest sense, .Q.iagenesis includes all the changes which affect a sediment , from the time it is physically aeposited, until all the constituent grains come to chemical e uilibrium dirT m-etamor Ism. Diagenesis lies between sedimentation and metamorphism, both in the sequence of events in the rock cycle and in the methods used to study diagenesis. Physical changes, such as compaction, fluid migration and pressure SOlution, are accompanied by chemical changes caused by mineral reactions, mineral replacement, cementation and dissolution. As rocks are buried, they evolve water, hydrocarbons and various gases. As these fluids move through a sedimentary sequence, they will generally follow fractures or the most permeable units. In siliciclastic-dominated sequences , the most permeable units generally are the sandstones, which become the conduit for the evolving fluids. ~ndstone diagenesis is controlled b a maze of interdependent processes, including burial rate, sediment com osition and texture sedimentaryandtectonicenvironment,chemicatreactlon rates, degree of litholtlglc vana ion, hy ro na t erma gra ents and a host of factors . We will enter this maze by start 109 wnFi the observed effects of sandstone diagenesis, and continue by examining pressure solution, mineral authigenesis, cementation, and development of secondary porosity and fluid movement. By this route, we hope to
arrive at some understanding of how all these processes contribute to turning sand into sandstone. Finally, we will briefly examine the practical application of diagenesis to the exploitation of hydrocarbon reservoirs. METHODS OF STUDY The petrologist stUdying diagenesis relies on transmitted light microscopy, much like metamorphic or igneous petrologists. However, because many of the materials he is attempting to observe are extremely fine-grained (1-20 ILm), he relies heavily on the scanning electron microscope (SEM) and special X-ray diffraction techniques to assist in mineral identification and petrography. In many diagenetic studies, extensive use is made of SEM petrography. With an energy dispersive spectrometer, the SEM can provide qualitative mineral analyses suitable, when combined with X-ray diffraction data and morphology, to assist in the identification of clay minerals and other authigenic phases. The scanning transmission electron microscope (STEM) offers the possibility of determining the exact nature of very fine intergrowths of authigenic minerals and the potential for accurate chemical analyses of very small particles. Many other methods may be applied during the study of clastic diagenesis. Resin pore casts and mercury injection tests can provide detailed information about the size and distribution of pores and are especiallly useful in studies relating to the recovery of hydrocarbons. Fluid inclusion studies of clastic rocks are not common (see Currie and Nwachukwu, 1974), but the success of fluid inclusion studies of carbonates implies that theywill become more common. Stable isotopes and methods of bulk chemical analysis (X-ray fluorescence) have been applied to the study of diagenesis, but separating authigenic minerals from allogenic phases is always difficult and may complicate interpretation of the results of such studies. Some very successful studies comb ine organic geochemistry with detailed study of the diagenet ic events, either by characterizing the organic material (Foscolos and Powell, 1980), or by employing properties such as the vitrinite reflectance of coal (Hutcheon et al., 1980), which is related to time and temperature of burial.
166
DIAGENESIS AND POROSITY The most obvious physical change in siliciclastic sediments during burial is the loss of porosity, which was the subject of an early study by Taylor (1950). Figure 1 incorporates data from the Gulf Coast of Texas(Stuart, 1970), northeast Pacific (Galloway, 1974) and the Mackenzie River Delta (Schmidt and McDonald, 1979a). It is obvious that porosit ~ tends to decrease with increasing dee!th but less obvious that tM,temay be zones in which RQLll§i.!,y -d.ecrease is not as rqgjd, or iIJ.w.l;ticb.p.o.tOS;. ity actually increases. The porosity loss, in general, results from compaction of V0'12:::- Q.rains, pressure..soMiOf.\.of.grains..alld ~rowth of authigenic minerals(cements). It can be seen in Figure 1 that there is n single relationshipbetween porosity and depth. Variations occur between basins and between units within basins, reflecting many interacting processes which contribute to diagenesis. Maxwell (1964 attempted to quantify by experime t the relationships between porosity an other factors, and concluded that increasing temRerature....!!ge. pressure . amount of flu id flow and amount of ./
/-
2
...fo
/
/
/
/ !
I
I I
-
/
,,
I
I
I
,
I
I
,, ,,
I
I
-
3
,
i
" '---" ,
s
"
\
\ I
I I
/ 10
20 POROSITY
%
30
Figure 1 Porosity versus depth curves for sandstones can be generalized for a number of sedimentary basins . Data for the Northeast Pacific are from Galloway (1974), and are shown by the long-dashed (geothermal gradient = 39°C·km-')and dot-dash (geothermal gradient = 29°C ·km-') lines. Relationships also are shown for the Gulf Coast (short-dash curve; Stuart, 1970) and the Mackenzie Delta (solid curve ; Schmidt and McDonald, 1979a) . The "P" and "S " on the data from the Mackenzie Delta indicate primary and secondary poros ity, respectively.
Diagenesis
Geoscience Canada Reprint Series 4
matrix all tend to lead to more rapidly decreasing porosity. These conclusions are reinforced by Selly's (1978) study of porosity evolution in the North Sea ~ number of studies indicate that porosIty evolution and the general path of diagen-esis-ate greatly affected by the ere:senceof hydrocarbons, earlyhydrocarbon migration tending to reserve orosit and immature lagenetic assemblages (Lowry, 1956; Wilson, 1977). Galloway (1974) notes that, within the Queen Charlotte Basin, sandstone cementation is accelerated in areas higher geothermal gradient (Figure 1), which implies that the rate of diagenetic alteration is temperature dependent. Sandstones from the northeast Pacific seem to lose porosity more rapidly than their counterparts in more stable tectonic regimes. such as the Gulf Coast of the United States and the Mackenzie Delta. The sands in the northeast Pacific basins, in addition to being in an area of higher heat flow, have higher contents of volcanic rock fragments, feldspars and other immature or chemically less stable constituents than sands in the Gulf Coast or Mackenzie Delta. It would seem that chemical instability (or compositional "immaturity") of sandstones also pla)/s a rola during dia enesis. he data from the Mackenzie Delta are interpreted by Schmidt and McDonald (1979a)to show two different trends of porosity loss with depth. Porosity which these authors refer to as "primary", or porosity represented by the original interstices in the sediment, is lost more ra idly than " secondar " porosit , which results from fractures shrinkage and dissolutio in cements or rains. On Figure 1.the porosity-depth curve for the Gulf Coast sands shows a rapid increase in porosity with depth below 3500 m. The Gulf Coast basin contaiii'SSi9nificant '(olumes of "overpressured" or " geopressured" rocks. This means that rather than the normal depth-pressure curve expected for an open column of water, pore fluids in the Gulf Coast sands are isolated from communication with the surface and can develop and sustain abnormally high water pressures. These "overpressures" affect the evorution of diagenesisjo1be.GuJiCGast. In summary, we see that porosity is occluded with depth , but the rate of porosity destruction (and presumably
attendant diagenetic processes) is strongly affected by geothermal gradients . initial sediment composition. tectonic setting. presence of secondary porosity and presence of overpressure. We shall see that other factors also are important during diagenesis. Many of our ideas about diagenesis are based on the Gulf Coast because of the large amount of data from that area. It would be incorrect to assume that all basins will show the same diagenetic trends as the Gulf Coast.
W
PRESSURE SOLUTION AND QUARTZ OVERGROWTHS Pressure solution is a subject worthy of an entire article, but since this process contributes to sandstone cementation a very brief review is presented here. Two types of petrographic phenomena are attributed to pressure solution, stylolites and dissolution of grains at grainto-grain contacts. Dissolytion at grainto-grain contacts is more commonly observed in sandstones than are stylolites, which are more common in car-
Figure 2 An idealized interface between two spherical quartz grains is shown with the pressure distribution as postulated by Robin (1978). Thepressures p1 to p3, at the grain contact, are not equal, nor are they equal to the fluid pressure, pw. An overgrowth is shown on the grains (shadedarea). The difference in pressures supplies the chemical potential gradient, which leads to the diffusion of silica from thegrain contact to the pore space.
c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c (
C C C (
(
C
r:
c
Coarse-grained Siliciclastic Rocks
167
r:
(
( (
C C
C
o C o o o o o o o
c
o o c o
o
o o o o
c 8 '-...
bonate rocks. In 1863, H.C. Sorby described convincing petrographic evidence to support the concept of pressure solution, and since that time there has been an abundance of petrographic work to bolster Sorby 's original observations . A satisfactory theoretical explanation for pressure solution has been farmoreelusive, beginningwith Reicke's
(1895) work on solids under homogeneous stress in a fluid and continuing to the present-day experimental work of Sprunt and Nur (1977) , DeBoer et al. (1977) and Robin (1978). Robin (1978) gives a brief, but excellent, summary of the development of theories regarding pressure solution . Figure 2 summarizes a theory proposed by Robin. Pressure
along the interface between grains is not constant, nor is it equal to the pressure exerted by the fluid in the pore space . The difference in thermodynamic chemical potential produced by this pressure difference is sufficient to put quartz into solution, allowing it to migrate along the grain boundary and to be precipitated in the pore space. Robin's analysis verifies the conclusion that force across a grain contact is a function of burial depth, grain size,grain shape and grain packing, and all these factors must playa role in the pressure solution process. Figure 3 ~1!9..~S various petrographic features often attributed to pressure solution . In sandstones, interpenetrating grain-to-grain contacts , s~ contacts, concavo-convexcontacts and quartz overgrowths, in many cases, are 'taken as eVidence of pressure solution. Spang et al. (1979) have demonstrated that clay seams, stylolitic in part, may cause soiution .ot.adlacerrt .rock . The subsequent collapse will produce a texture very much like a pressure solution stylolite. In the strict sense, this texture actually results from chemical reactions among minerals, which may proceed in the complete absence of directed stress, and thus is not truly " pressure" solution, Also, the work of Foscolos and Powell (1980) shows the presence of considerable amounts of inorganic amorphous silica at depths up to 2000 m in the Sverdrup and Beaufort -Mackenzie basins. The solubility of amorphous silica, often in the form of animal skeletons, is considerably higher than that of quartz, and dissolution of amorphous silica and reprecipitation as quartz can lead to the formation 9f quartz overgrowths, a feature often ?scribed to pressure solution but which, in this case, is not related to directed pressure. ~rtz overgrowths are commonl~072 '~ost significant porosity-reducin feature in the diagenesis of quartz-rich ..,.--.... sandston~ Reactions -betweenoclay mirJ..erals and reRlacement of silic a: bear ing minerals by other minerals , s ucn as carlJOlfgteCl:Jffcpnurirle-_ sources of silica . The variability in the development of quartz overgrowths Figure 3 These photomicrographs show textures commonly associated with pressure is governed by the nature of the host solution . grain, with monocrystalline quartz (a) quartz grains show " dust rings" where the original grain boundary (gb) has been grains producing different forms of overgrown by euhedral quartz (og) . Theconcavo-convex contactbetween the two grains cuts overgrowths when compared with polythe dust rings and the overgrowths, indicating solution of material at the interface. (60 x) crystallinequartz(Waugh, 1970). The"dust (b) a highlysutured contact has developedbetween two grains in a quartz sandstone. (400 x )
/1
c
168
lines " associated with quartz- overgrowths (Figure 3a) mark the interface between the overgrowth and the host grain and are often voids (Pittman, 1972). The voids seem to form because nucleation is initiated at a number of sites on the surface of the host grain and, as competitive crystal growth proceeds, a small number of the original nuclei will grow to form the overgrowth. The voids represent the interstices between these original nuclei. /
AUTHIGENIC CEMENTATION Wilson and Pittman (1977) emphasize
the necessity and textural criteria for distinguishing between allogenic (introduced during deposition) and aut hi~rowth in situ during diagenesis) clay minerals. Figure 4 shows typical morphologies of kaolinite, chlorite, smec-
Diagenesis
tite and illite when found as authigenic minerals in sandstone. Most workers acceptthe authigenicnatureofcarbonate or other types of pore-filling cements (Figure5). In manyinstances,there is not a sufficient local source within the sediment to produce all the observed cementing material, suggesting that material for ceroa~~~ scavenged from the pore water The material in the pore water may be derived from connate saline water (seawater), dissolution of soluble rocks (i.e., evaporites) by percolating groundwateror expulsion fromshgilles and Otbet rocks undergoing mineral or or anic reactions.CilFfiS-r1978, 0 es 1979) and Bo es and Franks (1979) have recognized the probability that material for cementation may be transferred from shales, via pore waters, to sandstones.
Geoscience Canada Reprint Series 4
The paths of cementation in sandstones are extremely complex. For example, Galloway (1974) recognizes relatively early calcite pore-fill formed near the surface, and a phase of calcite cementation displaying replacment textures which occurs only at depths greater than~ Numerous authors have documented paragenetic sequences for authigenic minerals in sandstones in which appearances and disappearances of minerals overlap (see, for example, Galloway, 1974; Surdam and Boles, 1979; Mankiewicz and Steidtmann, 1979; Tillman and Almon, 1979). This im lies that, rather than the appearance aod..dis.appearance eta. single mineral, there are reactions taking place, via the ore water, amon .~ minerals in a sandstone.
(
t: C C C C C
C
c
c c
c c c c c c c c c c c c
c c C
E C C C C
c e l (
C Figure 4 Scanning electron photomicrographs of (A) smectite honeycombs, (B) euhedral kaolinite booklets. (C) illite fibres (marked by black arrows) growing on altered plagioclase feldspar and (D) rosettes of iron-rich chlorite. The bar scale on each photograph is in um. All photomicrographs were taken by Alan Oldershaw.
c
c c c
r: Coarse-grained Siliciclastic Rocks
169
r
Boles (1982) has recognized active albitization in Frio Sandstones from the Gulf Coast. Using scanning electron microscopy, the reaction
c (
r
2Si0 2 + 'I2H 20 + H+ + Na- + CaAlzSizOe(1) quartz plagioclase = NaAISi30e + 'I2AI2Si205(OH)4
plagioclase
+ Ca 2 +
kaolinite
can be documented. By plotting the activity ratio of Ca2 + INa + versus pH, contoured tor 1000,125° and 150°C, and also plotting the compositions of waters
(in these same variables), Boles convincingly demonstrates that albitization is occurring, probably at the present time, in the subsurface of the Gulf Coast. The reaction shown provides a source of calcium for the production of carbonate cements. Many diagenetic mineral reactions are interrelated in a similar way, and to obtain a complete picture of diagenetic evolution, a complex sequence of textures, indicative of mineral reactions, must be interpreted. The studies of Merino (1975) and Nesbitt (1980)
(
demonstrate that if we consider the minerals with in the pore space as a chemical system (Figure 6), we can treat many of the chemical problems in sandstone diagenesis using equilibrium thermodynamics. We will see that chemical kinetics, mass transfer and fluid movement also may have a major impact on sandstone diagenesis. In examining the published work on chemical aspects of sandstone diagenesis, it would be easy to conclude that minerals an d aqueous solutions are the onl substa c e e 'ui" in diagenesis. From fluid inclusion studies of sedrments, Burruss (1981) cites examples of hydrocarbon-, methane- or carbon dioxide-bearing fluid inclusions associated with diagenetic events. The importance of carbon dioxide in controlling carbonate dissolution and recipi ation has Ion been reco niz d Many authors have noted the production of methane and carbon dioxide during diagenesis of organic matter, While this may be the case, howmuchofthevapour is carbon dioxide and how much is present as other species is a moot point. In 1942,German n and Ayers noted the production of carbon dioxide from limestones with trace amounts of silica, aluminum, iron and magnesium at temperatures as low as 98°C. More recently, Muffler and White(1969), in a study ofthe Salton Sea geothermal field, postulated a carbon dioxide-producin mineral reaction between clay minerals and carbonates. Hutcheon eta/. (1980) havedocumented the mineral reaction 5CaMg(C0 3 ), + AI2Si20s(OH). + Si02 + 2HzO(2) dolom ite kaolinite quartz = Mg sAI2S i30,o(OH)& + 5CaC0 3 + 5CO z chlorite calcite
which is observed in the Kootenay Formation, at approximately 5000 m (strati-
Figure 5 Authigenic, pore-filling cements are shown in both photomicrographs. (a) calcite is shown enclosing angular quartz grains in a sandstone. The calcite is well crystallized (cc) and shows some euhedral crystal faces. (160 x) (b) a pore in a sandstone with well-rounded quartz grains is filled by authigenic kaolinite. (160 x)
Figure 6 The minerals in the pore space, which are in contact with the pore fluids, often can be considered as a system which may come to thermodynamic equilibrium.
( 170
Diagenesis
Geoscience Canada Reprint Series 4
C C
graphic plus structural) burial. Figure 7 shows the topology of reaction (2) as a function of temperature at constant pressure in a mixture of carbon dioxide and water. As temperature increases, reaction (2) proceeds to the carbon dioxide-bearing side, and this carbon dioxide is added to the vapour phase , increasing the mole fraction of carbon dioxide in the vapour and making higher temperatures necessary for the reaction to proceed. The mineral reaction will follow along the curve in Figure 7 until the miscibility surface between carbon dioxide and water is reached and carbon dioxide will no longer dissolve in an aqueous fluid. At that point, carbon dioxide is produced as a vapour. Most fluids observed in sedimentary rocks are aqueous liquids; it is reasonable to expect that the intersection of the mineral reaction and the carbon dioxide - water miscibility surface represents the maximum equilibrium temperature for the mineral reaction in liquid water. Hutcheon et al. (1980) describe this process in more detail, and show calculated phase diagrams which indicate the maximum temperature for reaction (2) in the Kootenay Formation is about 180°C, in agreement with coal maturation data for the Kootenay Formation. A number of interesting observations can be made as a result of chemical studies of diagenesis. Boles and his coworkers have shown that many diagenetic processes appear to be active JlL other w.Q.!:ds.,..we can obseeve.trorn drill core that ankerite cementation and albitization are taking place as th estidiment is buried. By using mineral chemistryand equilibrium thermodynamics , it has been possible to show that material has moved between shales and sandstones during diagenesis. This result is directly comparable to the study by Hower et al. (1976) in which the smectite-illite transformation in mudrocks is interpreted as an active process. The research of many authors demonstrates that, rather than single minerals appearing and disappearing, there are complex mineral reactions with pore fluid in the subsurface. JQ-.pro per.ly ~ d e [~an d diagenesis , it is necesary.to .QPcumenLthe.se reactions.wlth.careful. ~gmp-hY,J:_om b i n i ng X-ray.J.hin.s.ec:tion and electron microbeam_t.e.ch!}iQu.~
The results of Muffler and White (1969) and Hutcheon et al. (1980) show that mineral reactions may produce carbon dioxide. This is an alternate, additional source to carbon dioxide produced by organic reactlons.Oarbon dioxide is an interest ing constituent of diagenelre-pTo'c~'Sse-g-in-thaCsol'Ti'e
authors have attri outed carbonate cementation to campn qj.o~ide. p-rod uc~ (Curtis, 1978), while other authors attribute dissolution of carbonate (and the formation of "secondary porosity") to the effect of carbon dioxide (Schmidt and McDonald, 1979a). Although these authors are referring to processes which take place at different burial depths, the effects of carbon dioxide production on carbonate stability require clarification .
previously, Schmidt and McDonald ascribe dissolution of carbonates to carbon dioxide produced by organic diagenesis, whereas Curtis (1978) blames the identical process for the carbonate precipitation. Which interpretation is correct? As is usual in geological problems, both are, maybe! Using calcium carbonate as an example, dissolution of carbonates is controlled by a reaction similar to (3): CaC0 3 + H + = HCO; + Ca2+ calc ite
(3)
assuming, as is often the case, that the pH range issuch that HCO; is the dominant aqueous carbonate species in solution. If the pH (activity of hydrogen ion) is controlled by mineral reactions with an aqueous phase, or by organic materials, such as acetic acid, the addiSECONDARY POROSITY tion of carbon dioxide will increase the Many authors have examined the imamount of HCO; by dissociat ion of portance of secondary porosity in the H2C03 • This will cause an increase in development of sandstone reservoirs the solubility product for reaction (3) (Hayes, 1979; Loucks et al., 1979). A and lead to carbonate precipitation . comprehensive account of the role and However, if the pH is not fixed by some recognition of secondary porosity is organic matter or minerals in the rock, given by Schmidt and McDonald something different will happen. The (1979a, b). They recognize porosity proaddition of carbon dioxide will produce duced by fracturing, shrinkage , dissoluH2C03 which, being a weak acid, will tion of sedimentary material and disdissociate, producing hydrogen ions solution of auth igenic material, b~tk- and lowering the pH. This will decrease consider the dissalutiOD-Of carbon ~1\ the solubility product for reaction (3), to be a m,.mor contributor:..to.secondar.y. leading to carbonate dissolution. Preporosity. in sJ~ndstones . The data of sent knowledge of water-rock systems -SChmidt and McDonald show a marked does not allow unequivocal determinareductionintherateofporositylosswith tion of pH from mineral assemblage burial depth (Figure 1,line labelled "S") data; therefore recourse lies in obtainover depth ranges in which secondary ing reliable, accurate water chemistry porosity is being destroyed. As stated from subsurface waters.
1
XCOz -
Figure 7 Mineral reactions similar to the one shown will attemptto control the carbon dioxide content of a vapour phase during burial. As the carbon dioxide content in the vapour increases, the reaction is stable at higher temperatures. Once the miscibility surface (solid curve) between carbon dioxide and water is reached, carbon dioxiderich vapour is produced.
GEOLOGICAL CONTROLS OF DIAGENESIS Siever (1979)has summarized plate tectonic controls on diagenesis. He notes source area by rial rate heat flow and residence time as generally important .tartars.While it is obvious that burial environment and original sediment composition and texture, which are determined by depositional environment, are two important geological controls, there is a paucity of studies which relate these two important variables. A study of Foscolos et al. (1982) marks the beginning of a project devised to relate diagenetic alteration to burial and sedimentary environment. Concentration of effort in this area is badly needed.
C C C
c
c
c
c c c c
c c
c c c
c c c c c c c c
C E C.
C C C
c e c c
C l
C C C C
c (
Coarse-grained Siliciclastic Rocks
~
o
r
C ( .
() 0,
o o
o c
o o
o
n
o o o r:
FLUID FLOW AND DIAGENESIS It is clear that as sediments are buried they lose water and other volatile constituents. Many of our concepts of sandstone diagenesis are founded on studies of the Gulf Coast , an area of low topograph ic relief and active burial. Movement of fluid in the Gulf Coast is upward and laterally outward along permeable units (Magara, 1976; Neglia, 1979). In addition, the fluids are largely connate marine water and water from the dissolution of evaporites, modified by diagenetic mineral reactions (Carpenter,1978; Hanor, 1979). Most ancient sedimentary basins are not found in this particular setting. For example, the Alberta Basin is in close proximity to a mountain range which provides sufficient hydraulic potential that meteoric water circulates through permeable units to depths on the order of 3000 m below ground surface (Toth, 1980). This origin for waters has strong influence on water composition and therefore influences the nature of mineral react ions taking place between pore fluids and the enclosing rocks. Concentrations of dissolved solids in formation waters in Alberta are commonly 50,000 rnq- L-\ but can range up to 200,000 mq- L-1 in Devonian rocks (Hitchon et a/., 1971 ; Hitchon , 1980). Since some of these waters can circulate to a great range of depths, fluid movement is a process which can transport large amounts of material, even in the time scale of the Holocene (Toth, 1980). The presence of overpressured zones ~n many basins implies that porous and permeable units may be isolated from communication with other permeable rocks by impermeable bodies. These permeable sands still show evidence of diagenetic features associated with other diagenetic settings, and it appears that fluid movement must have occurred. Also spatial patterns of minerai distribution may be developed, as, for example, in the alteration of volcanic rocks (Surdam and Boles, 1979). Wood and Surdam (1979), Cassan eta/. (1981) and Wood and Hewett (1982) all consider convective fluid movement to be an important process. Cassan et a/. (1981) surmised that convective fluid movement controls the dissolution of feldspar and cementation by quartz and calcite in the Gabon Basin by imposing temperature changes as a result of convection cells.
.J'< 1'1
c r:
Although convective flow is ~ booklets may become detached from hypothesis which requires further vertheir substrate and lodge in the pore ification by careful study of diagenetic throats, shutting off flow in that part of mineral assemblages, there is little the rock. The smectite group of cla¥ doubt that fluid flow, whether by upward minerals tends to swell when in contact Wiffi water due W~Sl1'ltltltswetlii'ig1rnCl moving compaction water, circulating groundwater cells in hydrodynamically 1tle adsorptIon of water molecufes In tne complex basins or convective flow, is a interlayer eosltlon ~ This swelling can oe up to two orders of magnitude of the controlling process of mass transport in solid volume occupied by smectite with porous sandstones. Fluid flow probably dominates over diffusion controlled a corresponding decrease in porosity mass transfer for mass movement on and permeability: Further, all clay minerals have high surface area to volume the large scale . Nevertheless, diffusion ratio~ and tend to have electrically is an important process in pressure ~Iution (Robin, 1978) and may playa char ged surfaces, making them efreC-' greater role in sandstone diagenesis 't1Veat binding water and causing bigh irreducible water saturations. The dif:. than is presently recognized . Studies f ering cation exchange capacities for such as those by Aagaard and Helgeson (1982) mark the beginning of the use of groups of clay minerals indicate difkinetics as a means of understanding ferent surface charge distributions. sandstone diagenesis. Clays with higher cation exchange capacities, such as smectite, tend to be DIAGENESIS AND HYDROCARBON more effective at binding water than RESERVOIRS suc h clays as kaolinite, which has a low c ation exchange capacity. Van Elsberg Intergranular, dissolution, fracture and micro-porosity are important in sand'(1978) has noted that many amorphous hydroxides have similar properties to stone reservoirs. The aythigenic c1a1P! content of sandstone reservoirs can clay minerals , and are very difficult to create mIcro-porosity, which is largely detect. In addition, the role of arnorineffec tive, and low permeabiljties by phous material in sandstone diagenesis decreasing the size of pore apertures- is poorly understood, with very few The high surface area to volume ratio of workers (Foscolos and Powell (1980) most clay minerals and the tendency fOrJ.4-- being notable exceptions) recogn izing expansion of the smectjtes or migratiod'1 the importance of amorphous material. ofkaolinitewhenexpQsedtofresbwa~er Almon and Davies (1981) and Thomas are limiting factors in oil recover y. Many (1981) outline possible problems created by the presence of clay minerals, of these factors are affected by diagenesis (Pittman, 1979). Consequently, and offer general methods for minimizmuch effort is being expended to uning their effect on recovery. Almon derstand the role of diagenesis in hydro(1981) also outlines the necessity for carbon recovery, encouraging an overusing diagenetic trends as a method for all increase of interest in sandstone porosity prediction in oil exploration . . diagenesis. Table 1 summarizes some of the more The recovery efficiency of sandgeneral ways in which clay minerals can stones has strong positive correlations affect recovery and common methods ~ith high porosity, small pore-to-throat of treating these problems. size ratiOS, small mean particle size and HEAVY OIL AND TAR SANDS low percentage of carbonate (Wardlaw and Cassan, 1979).Ofthese properties, (Artificial Diagenesis) pore-to-throat §!ze ratio and carbonate The tar sands and heavy oil deposits in percentage can be markedly affected Alberta and Saskatchewan represent a by diagenesis. For example, in studies. huge resource, but oil viscosities are so high in the tar sands that there is little or of reservoir sandstones it is c;mmon no conventional production. Schemes see I Itebridging~J;!orethroats.(Figure8) -and effectively doubling or tripling poreto mobilize the vast amounts of oil that to-throat size ratios. Kaollmte (FigUr~* are too deep to be mined involve heat4b) commonly forms aggregates and ing , either by steam injection or by inpore":fUllngs which are larger than the situ combustion. Mineralogical alterapore throats . When interstitial flow tion , especially of pore space minerals, velocities become high, during recovis to be expected and may be observed ery, these aggregates and individual both in laboratory tests (Perry and
172
Diagenesis
Geoscience Canada Reprint Series 4
c c
c c c
c (
c (
c c c c c c c c c c
c c
c c c
Figure 8 During steam stimulation, (A) fine smectite and (B) illite are reacted to produce (C) coarse-grained corrensite (Cor, a smectitechlorite intergrowth which in this case is not identified by X-ray diffraction), and (0) coarse-grained, euhedral analcime. The scale on each photograph is in um. All photomicrographs were taken by Alan Oldershaw.
Table 1
Problems and treatments for authigenic minerals in reservoirs. (After Almon and Davies, 1981).
Mineral
Problem
Incompatible with:
Compatible with :
Remedy
Smectite Mixed-layer Smectite-Illite Illite Kaolinite Chlorite Calcite, dolomite
swelling swelling microporosity mobile fines iron precipitate CaF2 precipitate
fresh water fresh water fresh water high flow rate 02-rich, pH 3.5 + HF
KCI, hydrocarbon KCI, hydrocarbon KCI, hydrocarbon low flow rate HCI, organic acid
HCI, HF HCI, HF HCI, HF clay stabilizers HCI, organic acid HCI
c ( c c c c c
c
c
c c: (
c c c ( (
C
c c c
173
Coarse-grained Siliciclastic Rocks
(. (
Gillott, 1979, 1982) and in actual field results(Sedimentology ResearchGroup, 1981). The impact of steam on volcaniclastic tar sands from Cold Lake is seen in Figure 9. When compared with the well-lithified, oil-free portion of a core cut after two years of steam injection, a pre-steam injection core is a pasty, unconsolidated mixture of mineral grains and tar. Detailed petrographic work demonstrates that an early, finegrained diagenetic assemblage (less than 4 ILm) of illite, smectite, zeolites Ond chlorite is replaced by coarse (4-10 ILm) smectite and large euhedral analcime crystals (20 ILm, see Figure 8). Figure 10 shows that decreases in porosity correspond to decreased kaolinite and increased smectite content in the less than 2 ILm size fraction. The reaction postulated for these mineral alterations is
G
c (
( ( ( (
o o ( c o o o c o CJ
7AI2Si 20s(OH)4 + 8Si0 2 + 2Na+ kaolinite quartz
(4)
= 2NaAl~i11030(OH)6 + 7H 20 + 2H + smectite
Figure 9 A pre-steam core from the Cold Lake tar sand is a pasty mixture of bitumen and mineral grains. Two years of cyclic steam injection and steam flooding have removed most of the oil and lithified the sand. The minerals in Figures 8c and 8d cement the individual grains together.
r......
(Sedimentology Research Group, 1981) and in conjunction with albite- and analcime-bearing reactions can be used to construct a phase diagram to outline the stability fields for analcime, kaolinite, smectite and albite. The phase diagram constructed (Figure 11) verifies the mineralogical and textural
Sm
Ka
C
;:-_-T-_...;;1? -;0
9
4? -~o
~
II
observations that smectite isa hightemperature phase compared with kaolinite. The diagram also explains the textural observation that analcime is the last-formed phase in these rocks. When injection stops, more saline formation water will invade the rocks and the phase diagram predicts that analcime will be stable at higher salinities than kaolinite and smectite at the same temperature. All the studies mentioned demonstrate that "artificial" diagenesis does occur during steam injection or insitu combustion. The studies also indicate that artificial diagenesis affects the petrophysical parameters of the rock and, presumably, the efficiency of oil recovery. SUMMARY AND CONCLUSIONS
The most obvious feature of sandstone diagenesis is the modification and general reduction of porosity. Compaction, pressure solution and authigenic cementation by mineral reactions are the p rinciple processes contributing to this porosity modification. The rate and pathway of sandstone diagenesis is strongly influenced by sediment composition, which IS In turn dependent on sedimentary and tectonic environments. Temperature also is a regulating factor, most easily recognized by the different paths of diagenesis in areas of varying geothermal gradients. Upon
Ch
4? -4;.:,0:.....,.-.:;~:......;2? -1~0 - 410
o o 0, o C l 8 C
300
1350 u
0
-
- 415
1360 -
~
'"
I/O
Q; Q; E
Q) Q)
1370 -
...\
~
I
lc, 1380 W Cl
,,
\ I
I \
1390 -
I
...
I I
, I
I
,I 1400 -
""......
,,
,
I
, I
I
I
,
- 470 I l-
I
I \
Cl
\
...
",\
,-
-c :Ii ...
200
~
100
kaolinite
w
\ I
\
~
smectite
n,
\
\
albite
w
(f
6
10 Ln aNa+ -
12
14
16
Ln aH+
- 425 I I
I \
Figure 10 This figure shows changes in porosity (tb), kaolinite (Ka), smectite (8m), illite (II) and chlorite (Ch) contents in the tar sand samples examined. Whengoing from the pre-steam to post-steam samples, a negative change means a decrease, and a positive change means an increase. The antipathetic relationship between smectite content and kaolinite content is the basis for the mineral reactions postulated. Note that increases in smectite content correspond to decreases in porosity.
Figure 11 Phase relations at constant pressure and silica acivity for kaolinite, smectite, albite and analcime show that, for a given fluid composition, smectite is a higher temperature phase than is kaolinite. In addition, at constant temperature, analcime requires higher values of the equilibrium constant (at constant pH, this could be interpreted as increased salinity) than do kaolinite or smectite.
c 174
Diagenesis
Geoscience Canada Reprint Series 4
c
C burial , sandstones display evidence of "active" diagenesis which can be modelled by mineral-solution equilibria. Much of the material required for sandstone diagenesis (especially in very quartz-rich sands) is transferred from other rock by fluid movement. Fluid movementmay. result from upward and oytward flow of compaction wate r,circulation of grollndwater in hydrodynamically complex basins or by con~ vective tJo.w...wi1hin basinal units. The propert ies of hydrocarbon reservoirs are affected by diagenesis, and can be dramatically altered by drilling fluids or enhanced recover y tech ::.---- niques. Very detailed understanding of ~ interaction between minerals , aqueous fluids, hydrocarbons and gases is required in order to maximize hydrocarbon recovery. The emphasis on more sophisticated ways of modelling fluid and mass transfer (Wood and Hewett, 1982) and dealing with the overall kinetics of processes which are governed by reactions at the mineral-solution interface (Aagaard and Heldeson, 1982)indicate the direction of physical and chemical research on sandstone diagenesis . The application of scanning transmission electron microscopy (STEM)offers the hope that our ability to deal with the composition and structure of very small mineral grains will keep pace with the theoretical data. The chemistry of mineral surfaces, the kinetics of mineral reactions and the thermod ynamic prOperties of' clay minerals and com elex .!.9 ueous SOiutions at el~vated temReratur~ pressures are not well enough understood to solve the scientific and practi- cal problems of diagenesis.
IV 07J.
REFERENCES Minerals Reactions in Sediments Aagaard , P. and Helgeson , H.C., 1982,Thermodynamic and kinetic constraint on reaction rates among minerals and aqueous solut ions. I. Theoretical considerations: American Journal of Science, v. 282, p. 237-285. Boles, J.R., 1979, Active ankerite cementation in the subsurface Eocene of southwest Texas: Contributions to Mineralogy and Petrology, v. 68, p. 13-22. Boles, J.R., 1982, Active albitization of plagioclase , Gulf Coast Tertiary: American Journal of Science, v. 282, p. 165-180.
Boles, J.R. and Franks, S.G., 1979, Clay diagenesis in Wilcox sandstones of southwest Texas: Journal of Sedlmentary Petrology, v. 49, p. 55-70. Currie, J.B. and Nwachukwu, S.O., 1974, Evidence on incipient fracture porosity in reservo ir rocks at depth : Bulletin of Canad ian Petroleum Geology, v. 22 , p.42-58. Curtis, C.D., 1978, Possible links between sandstone diagenesis and depth-related geochemical reactions in enclosing mudstones: Geological Society of London, Journal, v. 135, p. 107-117. Foscolos, A.E., Reinson, G.E. and Powell, T.G., .1982, Controls on clay-mineral authigenesis in the Viking sandstone, central Alberta . I. Shallow depths: Canadian Mineralogist, v. 20, p. 141·150. Foscolos, A.E. and Powell, T.G., 1980, Mineralogical and geochem ical transformation of clays during catagenesis and their relation to oil generation, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Society of Petroleum Geologists , Memoir 6, p. 153-172. Hower, J., Eslinger, E.V., Hower, M.E. and Perry, E.A., 1976, Mechanism of burial metamorphism of argillaceous sediments: I. Mineralogical and chemical evidence : Geological Society of America, Bullet in, v. 87, p. 725-737. Hutcheon, I., Oldershaw, A. and Ghent , E.D., 1980, Diagenesis of Cretaceous sandstones of the Kootenay Formation at Elk Valley (Southeastern Br itish Columbia) and Mt. Allan (Southwestern Alberta) : Geoch imica et Cosmochimica Acta , v. 44, p. 1425-1435. Merino, E., 1975, Diagenesis in Tertiary sandstones from Kettleman North Dome, California - II. Interstitial solutions: distribution of aqueous species at 100°C and chemical relation to diagenetic mineralogy: Geochim ica et Cosmoch imica Acta, v. 39, p. 16291645. Muffler, L.J.P.and White , D.E., 1969,Active metamorphism of Upper Cenozoic sediments in the Salton Sea geothermallield and the Salton Trough, Southeastern California: Geological Society of America, Bulletin, v. 80, p. 157·182. Nesbitt, H.W., 1980, Characterization of mineral-formation water interactions in Carbon iferous sandstones and shales of the Illinois Sedimenta ry Basin: AmericanJournaJofScience,v. 280,p. 607-630. Schmidt, G.W , 1973, Interstitial water composition and geochemistry of deep Gulf Coast shales and sandstones : American Association of Petroleum Geologists, Bullet in, v. 57, p. 321-337.
Surdam, R.C. and Boles, J.R., 1979, Diagenesis of volcanic sandstones, in Scholle, PA and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 227242.
C (
C C C
Porosity Variation with Depth
(
Hayes, J.B., 1979, Sandstone diagenesis the hole truth, in Scholle, P.A. and Schluger. P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 127-139. Galloway,WE., 1974, Deposition and diagenetic alteration of sandstone in Northeast Pacific arc-related basins: Implications for greywacke genesis: Geological Society of Amer ica, Bulletin , v. 85, p.379-390. Loucks, R.G., Dodge, M.M. and Galloway, WE., 1979, Importance of secondary leached porosity in Lower Tertiary sandstone reservo irs along the Texas Gulf Coast: Gulf Coast Association of Geological Societ ies, Transactions,v. XXIX, p.164-171. Lowry, WD ., 1956, Factors in the loss of porosity by quartzose sandstones of Virginia: Amer ican Association of Petroleum Geologists, Bulletin , v. 40, p.489-500. Maxwell , J.C ., 1964, Influence of depth, temperature and geologic age on the porosity of quartzose sandstones: American Association of Petroleum Geologists , Bulletin, v. 48, p. 697-709. Selly, R.C ., 1978, Porosity gradients in North Sea oil-bearing sandstones: Geological Society of London , Journal, v. 135, p. 119-132. Taylor, J.M., 1950, Pore-space reduction in sandstones: American Association of Petroleum Geologists, Bulletin, v. 34, p.701-716. Schmidt, V.and McDonald D.A., 1979a,The role of secondary porosity in the course of sandstone diagenesis , in Scholle, P.A. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 175-208. Schmidt, V., and McDonald, D.A., 1979b, Texture and recognition of secondary porosity in sandstones, in Scholle, P.A. and Schluger. P.R., eds., AspectsofDiagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 209-226. Stuart, C.A., 1970, Geopressures: Second Symposium on Abnormal SUbsurface Pressures, Baton Rouge , Louisiana , 121 p.
C
C C C C C C C
C C C C C
C
c
? c c c c c c c c e ( (
C
C C C C C
c c c r, c
c)
(
C ( ('
c c o o o
o o (
o ( c o o o c o o , o .
C (:.
8 C (
,'--.
r:
Coarse-grained Siliciclastic Rocks
Van Elsberg, J.N., 1978,A new approach to sediment diagenesis: Bulletin of Canadian Petroleum Geology, v. 26, p. 57-86. Wilson, H.H., 1977, "Frozen-in" hydrocarbon accumulations or diagenetic traps - exploration targets: American Association of Petroleum Geologists, Bulletin, v. 40, p. 483-491. Pressure Solution and Quartz Overgrowths DeBoer, R.B., 1977, On the thermodynamics of pressure solution-interaction between chemical and mechanical forces: Geochimicaet Cosmochimica Acta, v. 41, p.249-256. DeBoer, R.B., Nagtegaal, P.J.C. and Duyvis, E.M., 1977, Pressure solution experiments on quartz sand: Geochimica et Cosmochimica Acta, v. 41, p.257-264. Pittman, E.D., 1972, Diagenesis of quartz as revealed by scanning electron microscopy: Journal of Sedimentary Petrology, v. 42, p. 507-519. Reicke, E., 1895, Ober das Gleichgewicht zwischen einem festen, homogen deformierten Kerper and einer flussigen Phase, insbesondere uber die Depression des Schmelzpunktes durch einseitige Spannung: Anals Physik, v. 54, p.731-738. Robin, P.-Y.F., 1978, Pressure solution at grain-to-grain contacts: Geochimica et Cosmochimica Acta, v. 42, p. 1383-1389. Sorby, H.C., 1863, On the direct correlation of mechanical and chemical forces: Royal Society of London, Proceedings, v. 12, p. 538-550. Spang, J.H., Oldershaw, A.E. and Stout, M.Z., 1979, Development of cleavage in the Banff Formation at Pigeon Mountain, Front Ranges, Canadian Rocky Mountains: Canadian Journal of Earth Sciences, v. 16, p. 1108-1115. Sprunt, E.S. and Nur, A., 1977, Destruction of porosity through pressure solution: Geophysics, v. 42, p. 726-741. Waugh, B., 1970, Formation of quartz overgrowths in the Penrith sandstone (Lower Permian) of Northwestern England as revealed by scanning electron microscopy: Sedimentology, v. 14, p. 309-320. Weyl, P.K., 1959, Pressure solution and the force of crystallization - a phenomenological theory: Journal of Geophysical Research, v. 64, p. 2001-2025.
175
Diagenesis and Sedimentary Environments Mankiewicz, D. and Steidtmann, J.R., 1979, Depositional environments and diagenesis of the Tensleep sandstone, Eastern Bighorn Basin, Wyoming, in Scholle, PA and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p.319-336. Siever, R., 1979, Plate tectonic controls on diagenesis: Journal of Geology, v. 87, p.127-155. Tillman, R.W and Almon, W.R., 1979, Diagenesis of Frontier Formation offshore bar sandstones, Spearhead Ranch Field, Wyoming, in Scholle, P.A. and Schluger, P.R.,eds. Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 337-378. Fluids in Basins Burruss, R.C., 1981, Hydrocarbon fluid inclusions in studies of sedimentary diagenesis, in Hollister, L.S. and Crawford, M.L., eds., Fluid Inclusions: Applications to Petrology: Mineralogical Association of Canada, Short Course Handbook, v. 6, p. 138-156. Carpenter, A.B., 1978,Origin and chemical evolution of brines in sedimentary basins: Oklahoma Geological Survey, Circular, v. 79, p. 60-77. Cassan, J.-P., Garcia Palacios, M. delC., Fritz, B. and Tardy, Y., 1981, Diagenesis of sandstone 'reservoirs as shown by petrographical and geochemical analysisofoil bearing formations in the Gabon Basin: Bulletin Centres de Recherche Exploration et Production Elf Aquitaine, v. 5, p. 113-135. Germann, F.E.E., and Ayers, w., 1942, The origin of underground carbon dioxide: Journal of Physical Chemistry, v. 46, p.61-68. Haner, J.S., 1979,Thesedimentary genesis of hydrothermal fluids, in Barnes, H.L., ed., Geochemistry of Hydrothermal Ore Deposits, Second Edition: Wiley-Interscience, New York, 789 p. Hitchon, B., 1980, Some economic aspects of water-rock interaction, in Roberts, W.H., III and Cordell, R.J., eds., Problems in Petroleum Migration: American Association of Petroleum Geologists, Studies in Geology 10, p. 109-119. Hitchon, B., Billings, G.K. and Klovan, J.E., 1971, Geochemistry and origin of formation waters in the western Canada sedimentary basin, III. Factors controlling chemical composition: Geochimica et Cosmochimica Acta, v. 35, p.567-598.
Magara, K., 1976, Water expulsion from clastic sediments during compaction Directions and Volumes: American Association of Petroleum Geologists, Bulletin, v. 60, p. 543-553. Neglia, S., 1979, Migration of fluids in sedimentary basins: American Association of Petroleum Geologists, Bulletin, v. 63, p.573-579. Toth, J., 1980, Cross-formational flow of groundwater: A mechanism for transport and accumulation of petroleum (The generalized hydraulic theory of petroleum migration), in Roberts, WH., III and Cordell, R.J., eds., Problems in Petroleum Migration: American Association of Petroleum Geologists, Studies in Geology 10, p. 121-167. Wood, J.R. and Surdam, R.C., 1979, Application of convective-diffusion models to diagenetic processes, in Scholle, P.A. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p.243-250. Wood, J.R. and Hewett, TA, 1982, Fluid convection and mass transfer in porous sandstones - a theoretical model: Geochimica et Cosmochimica Acta, v. 46, p. 1707-1713. Diagenesis and Hydrocarbon Reservoirs Almon, WR., 1981, Depositional environment and diagenesis of Permian Rotliegendes sandstones in the Dutch sector of the southeastern North Sea, in Longstaffe, F.J., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p. 119-147. Almon, WR. and Davies, D.K., 1981, Formation damage and the crystal chemistry of clays, in Longstaffe, F.J., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p. 81-103. Perry, C. and Gillott, J.E., 1979,The formation and behaviour of montmorillonite during the use of wet forward combustion in the Alberta oil sands deposits: Bulletin of Canadian Petroleum Geology, v. 27, p. 314-325. Perry, C. and Gillott, J.E., 1982,Mineralogical transformations as indicators of combustion zone temperatures during in situ combustion: Bulletin of Canadian Petroleum Geology, v. 30, p. 34-42. Pittman, E.D., 1979, Porosity, diagenesis and productive capability of sandstone reservoirs, in Scholle, P.A. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 159-173.
c 176
Sedimentology Research Group , 1981, The effects of in situ steam injection on Cold Lake oil sands: I. Mineralogical and petrographic considerations (Oldershaw, A., Hutcheon, I. and Nahnybida, C.); II. Petrophysical and petrological considerations (Hutcheon, I., Oldershaw, A..and Nahnyb ida , C.): Bulletin of Canadian Petroleum Geology, v. 29, p.447-478. Thomas , J.B., 1981, Classification and diagenesis of clay minerals in tight gas sandstones : Case stud ies in which clay mineral properties are crucial to dril ling fluid selection, formation evaluation and completion techniques, in Longstaffe, F.J., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p. 104-118. Wardlaw, N.C. and Cassan , J.-P., 1979, Oil recovery efficiency and the rock-pore properties of some sandstone reser voirs: Bulletin of Canadian Petroleum Geology, v. 27, p. 117-138. Wilson , M.D. and Pittman, E.D., 1977, Auth igenic clays in sandstones: Recognition and influence on reservoir properties and paleoenvironmental analysis: Journal of Sedimentary Petrology, v. 47, p.3-31.
Originally published in Geoscience Canada v. 10 Number 1 (March 1983)
Diagenesis
Geoscience Canada Reprint Series 4
(
C C C
C C C (
C
C C C C C C C C (
C C C C C C C C C C C
C C C
c e ( (
C C C ( ( (
CJ C C C
o c c c c ( c o o o o o
(
C 6
177
Catagenesis of Argillaceous Sedimentary Rocks
Catagenesis of Argillaceous Sedimentary Rocks A. E. Foscolos Geological Survey of Canada Institute of Sedimentary and Petroleum Geology 3303 - 33rd Street N.W Calgary, Alberta T2L 2A 7 and Schoolof~ineralResources
Engineering Technical University of Crete Chania 73132 Crete, Greece INTRODUCTION Argillaceous sediments undergo progressive transformation upon burial beneath succeeding sediments. These transformations affect the water content and the mineralogical , inorganic and organic components of the pelitic rocks. The combined changes in organic or mineral matter resulting from pressure and temperature conditions in the subsurface that are much different from those at deposition are attributed to catagenesis. Compaction and gravitational displacement of water was the first dewatering mechanism recognized (Hedberg, 1936). Power (1967) recognized that dewatering also occurred during the conversion of smectite to illite. He designated these two processes of dewatering as States I and II, respectively (Figure 1). In 1969, Burst used statistics to correlate clay dehydration with petroleum occurrences by plotting the difference between the depth of the first clay dehydration, which occurs when 25% of smectite is converted to illite, and the production depth from 5,368 locations in the Gulf Coast area of the United States. He found it significant that although dehydration depths of the
sediments ranged from 4,000 to 10,000 feet, the hydrocarbon production depths were distributed about a mean of 1,500 feet above the theoretical first clay dehydration level. Finally, the same investigator correlated his three stages of sediment dehydration with the flushing of hydrocarbons from the source rocks, which he called the "Gulf Coast Fluid Redistr ibution Model " (Figure 2). Thus, gravitational displacement squeezes out water from the pore throats by the rearrangement of sedimentary grains while alteration of the 2:1 layer silicate promotes the sloughing of water layers from their surfaces. The latter was thought to be instrumental in the dissolution of hydrocarbons, thereby aiding primary oil migration. Dunnoyer de Seconzac (1970) reviewed the mineral transformation that takes place during the stepwise burial of shales and concurred that smectite undergoes conversion to interstratified illite-smectite and finally to illite. Perry and Hower (1972) proposed a two-step dehydration model, instead of the onestep model suggested by Burst. In the first step, there is water expulsion engendered by the relatively rapid collapse of the smectite layers, while the second step is marked by the transition from random to ordered interlayering. This promotes a second dehydration from the layered illites. The conversion of smectite to illite through the intermediary step of a binary mixed-layered silicate is discussed further by Hower et al. (1976) and Aronson and Hower (1976).They have suggested that smectite reacts with K-feldspar and sometimes with mica , in absence of K-feldspar, to produce illite, chlorite and POWERS
BURST
(1967)
(1969)
PERRY AND HOWER
Q,
w c w Cl:
a:
L.GEO .GRAD.
-2.7S"C/10om
-2.1S"C/100m
I
I
II
o
II.
H.GEO.GRAD.
I
:I:
l-
quartz. Thus, the rnlneraloqical components in pelitic sediments adjust to the prevailing physicochemical condit ions. In 1974, Foscolos and Kodama studied the transformation of smectite to illite in the Cretaceous shales of northeastern British Columbia and showed that in order to promote the dehydrat ion of expandable layer silicates , the temperature must first reach a critical value and then alum inum must enter and substitute for silicon in the structure of the expandable clay. This isomorphic substitution upsets the crystal electroneutrality because an ion with 3 valence, aluminum, substitutes for an ion with a 4 valence, silicon, thus increasing the negative charge which originates from the oxygens and hydroxyls of the lattice. To counterbalance this negative charge, potassium must be absorbed on the clay surface . This process triggers the expulsion of calcium from the clay surface and the sloughing off of the absorbed water. Further isomorphic substitution of silicon by aluminum releases magnesium and iron into the solution. The same authors recognized also that the transformation of smectite to illite involves an intermediate state . Smectite is transformed to vermiculite and then to illite. SUbsequent work by Foscolos et al. (1976) and Powell et al. (1978) in northeastern British Columbia, the Canadian Arctic Islands and the Beaufort-Mackenzie Basin of the Canadian Northwest Territories has shown that though clay dehydration coincides with the onset of hydrocarbon generation at 0.5% Ro vitrinite reflectance , the main phase of hydrocarbon generation occurs at greater levels of burial. Quantitative
II
m
m
Ii:
:)
CIl
III
:)
m
CIl
I[
STAGE
I, rr, ETC.
Figure 1 Dewatering process of argillaceous sedimentary rocks upon burial .
c 178
Diagenesis
Geoscience Canada Reprint Series 4
(
C mineralogical and geochemical studies showed systematic decreases in amorphous inorganic components and in the quantity of layer silicateswith increasing depth (Foscolos and Powell, 1982). Recent work in diagenesis of pelitic sediments "indicates that diagenesis is a multifaceted process consisting of mineral alterations along with the dewatering processes. The latter comprises at least four mechanisms: (1) gravitational displacement; (2) water expulsion from the inorganic gels; (3) the sloughing of water from the hydrous layer silicates; and (4) the release of water from the destruction of clay minerals along with cations which migrate individually or in chelating form to the adjacent sandstones forming authigenic cements. The primary objective of this review is to discuss the alterations and genesis of minerals in shales during diagenesis along with the generation of water from the last three processes and the ir relat ion various auth igenic cements encountered in sandstones . Finally the relationship between clay catagenesis, the organic matter diagenesis and cementation of adjacent sandstones is discussed. MINERAL TRANSFORMATION IN PELITIC SEDIMENTS DURING BURIAL The most abundant minerals in shales are the silicates, where carbonates are minor components. Mineral identification in pelitic sediments reveals more or less the same components. However, quantitative estimation of individual minerals with burial depth indicates that some key components, such as smectite, kaolinite and chlorite , disappear or appear at certain depths, while others transform gradually. The nature of the most commonly encountered constituents with burial depth is, therefore discussed individually. Quartz. Quartz is ubiquitous in shales and, apparently, its absolute amount increases with burial depth (unpublished data from the Beaufort-Mackenzie and Sverdrup Basins) . This implies that either the remaining minerals are removed from the system, thus increasing its relative abundance, or that quartz is added to the system during the destruction of other minerals. This is
accomplished by removing the other cations either by forming new minerals in situ which are less rich in silica, or by transporting the products to adjacent sandstones, leaving silica gel behind. For example, K-feldspar can be transformed to a 2:1 layer silicate by reacting with carbon ic acid. The latter can be released from the thermal and bacterial decarboxylation of the concomitant organic matter (Tissot and Welte, 1978). This reaction yields amorphous silica as follows:
growth (Longstaffe , 1983).Oxygen isotope work can verify this process. 51 8 0 valuesfor Si0 2, derivedfrom igneousand metamorphic rocks, must be lower than that derivedfrom the conversion of amorphous silica gel to crystalline quartz. Thus, silica gel absorbed on pre-existing quartz grains should have higher 5180 values, or if one peels off the successive layers of Si02 by HF,then &1 8 0 valuesof the remaining quartz crystal should decrease progressively.
3KAISi 308 + H2C0 3 + 12H20 = KAI2(AISi3 08 ) + 6H4Si04 + K2C0 3 (1)
The most common feldspar encountered in pelitic sediments is K-feldspar. Albite and anorthite are very seldom encountered because they are easily weathered (Birkeland, 1973). In the early stages of diagenesis K-feldspar also disappears when attacked by carboxylic acids (Surdam et al., 1984; Surdam and Crossey, 1985) or from carbonic acid derived from the thermal and bacterial < 80°C, decarboxylation of organic matter. The organic origin of CO2has been reported byCurtis (1978), Schmidt and McDonald (1979), Carothers and Kharaka (1980), and Lundegard et al. (1984). The attack by organic acids on silicates should be considered as a significant factor because of its dual effect as an acid and
Another reaction which also produces amorphous silica is the transformation of smectite to illite which in diagenesis usually follows the first equation. The reaction can be written in general terms : K+ + smectite = illite/smect ite + silica + other products (2) Indeed, equation 2 yields Ca2-; Mg2-; Fe2+ and H20 (Foscolos and Kodama, 1974; Hoffman and Hower, 1979; Boles and Franks, 1979). Amorphous silica, generated from either or both reactions 1 and 2, can be re-adsorbed on existing quartz grains, inducing crystal over-
Feldspars.
C C
C C C
C C
C C C C C C C C C
C C C
C C C
C C
6.6
NON-LIQUID HYDROCARBONS 5 .1
C
0 0 0
C
.e M
C
;
E
E s:
C C
3 .4
C
~ GO
"
~
"
lD
1.7
23.3
37.7
66.5
93.2
121 .0
14 6 . 7
Temperature in CO
Figure 2 Correlation between burial depth, geothermal gradients in °C per 100metres and occurrence of hydrocarbons. (Modified after Burst, 1969).
c e c c c c (
c
c c
(
c
Catagenesis of Argillaceous Sedimentary Rocks
179
C.
r: r:
c
C.
C (
C ('.
o (
o
o o o
a chelat ing agent capable of removing the products of dissolution such as aluminum (Surdam etal., 1984). Finally, the products of feldspar dissolution are used to enhance quartz overgrowth or to activate the transformation of smectite to illite through the intermediary step of mixed layer silicates. Layer Silicates. Layer silicates are the second most abundant component of pelitic sediments after quartz and feldspars. They include smectite, illite, interstratified 2:1 layer kaolinite, and chlorite. Stepwise burial of shales and sandstones show changes in composition of hydrous layer silicates as a function of depth reflecting mainly the increase of temperature. Recent sediments are enriched in discrete minerals such as illite, kaolinite, smect ite or vermiculite while older sediments contain mixed -layered silicates and even older sediments, discrete chlorite and 1 Md illites (Grim, 1968; Weaver and Beck, 1971; Foscolos and Kodama, 1974; Foscolosetal. , 1976; Hower et al., 1976).Due to this systematic variation in composition with depth, that is temperature, hydrous layer silicates have been used as paleotemperature indicators between 80°- 300°C (Steiner, 1968;Sumi, 1969; Eslinger and Savin, 1973; McDowell and Elders, 1980). A review art icle by Eberl (1984)
~
! i
d-spac1ng in nm ~:m ~ II ~ ~
II
.... ..'" ..." ... . II
.S
-
r-
V rvV
l-
rv
C
~
r-
'V
"
r
....
V
r-
IQ
6
I
-
lilltJI
.s::
I-
.... ,
r-
VJ
V ~
k I
'"'' J5
25
:, T
.426.7
.."..
I
:: .:E "
~ I
1/\
nanometres 1.0 1.6
~
I
r
V
i'--- f-I \....
/
'-'
'v
"-- ,......
r-: '-
- --
VJ
_ 1 1271
r
relates mineral assemblages in shales , sandstones and volcanic rocks from ambient temperture to 350°C. Smectite. Smectites are stable at temperatures much higher than those encountered during diagenesis. However, the chemical composition of the interstitial solutions is the most important factor in transforming smectites to illite via the mixed-layer path . Discrete smect ites are usually abundant in the middle or late stages of diagenesis (telodiagenesis). Any occurrence of discrete smectite in deep zones of diagenesis should be considered as due to hydrothermal phenomena. Mixed-Layered Silicates. Mixed-layered silicates appear in the early stages of diagenesis. Once potassium and aluminum are present in the interstitial pore water and the temperature is between 90° -100°C, smectite is converted to illite (Foscolos et al., 1976; Powell et al., 1978). Aluminum, derived probably from the breakdown of feldspars , substitutes for silica in the tetrahedral positions, thus promoting a charge deficit within the crystal cell unit. To counterbalance the negative charge, a potassium ion is preferentially absorbed on the smectite surface by displacing absorbed calcium andlor magnesium . This implies that silicon
VI
~731.5
I
.11 27.7
."
!1493.5
2
/.-<:Z8.8
~
j2133.6
.s:: .-
,2438 .4
C-
"
I
3
.2758.4 I 13093.7 j3352.8
10
Degrees, 28 CoKa
'-"
Figure 3 d oo1 spacings of calcium saturated 0.2 ~m specimen at 50% relative humidity versus burial depth from a North Sabine H-49 well sample (Foscolos and Powell. 1980).
from within the unit cell as well as calcium and magnesium from the surface are expelled into the solution . In addition , absorbed H2 0 sloughs off from smectites . These changes are recognized by the reduction of the doo, spacing of a Ca-saturate smect ite from 1.56 nm to 1.20 nm in the mixed layer (Figure 3). One should notice the permanent imprint that these chemical changes impart on the smectite lattice in order to be identified by x-rays. Had the isomorphic substitution of Si4+ by A13+ in the crystal lattice of smectite not taken place, the doo, spacing of the clay mineral should have stayed the same, that is, 1.56nm. As a result, the amount of water loss by dehydration for one gram of smectite converted to one gram of 50% illite - 50% smectite is calculated as follows: Two facing areas of a unit cell = 2ab = 2 x 5 .25~ x 9 .20~ = 96 .6~2 where a and b are the unit cell dimens ions for smectite.
The specific planarsurfacein m2 • g-' is 96.M2x 10-20 m2 .Jl.2 Molecular weight of smectite/6.02 x 1023 mole " 96 .6 m 2 x 6.02 x 1023 mole "
_ _ _ _ _ _ _ _ = 808 m2 ' g" 720 g ' mete-i x 10.20
where 720g is the atomicweight of smectite in grams, 10-2 0 is a conversion factor from 'A2 to m2 and 6.02 x 1023 mole-t is Avogadro's number. Since there is a reduction of the doo, spacing from 1.54nm to 1.18 nm, that is 0.36 nm, then the water loss per m2'g-' of smectite when convertedto 50% illite50% smectite is 808 x 104 cm2 • go' x 0.36 X 10-7 cm = 290.88 x 10-3 cm 3 • g" . Assuming a density of 1 g ' cm-3, this water loss converts to 290.99 mg per gram of clay. This result is comparable to the experimental result of Mooney et al. (1952) who reported a loss of 270 mg of H2 0 per gram of Ca smectite when the doo , spacing is reduced from 1.56 nm to 1.20 nm, that is, by 0.36 nm. By the thermodynamic manipulation of water vapour absorption and desorption isotherms of the clay and simultaneous layer distance measurements by x-ray diffraction of the doo, spacings of smectites, Keenan et al. (1951) and Mooney et al. (1952) have calculated the required pressure to remove a monolayer of water of 0.36 nm (3.6'A)
( Diagenesis
180
thickness, as follows: -AT II = - - In PIPo
v,.M
where II = gas constant in atmospheres R = gas constant in litre-atmospheresl mole degree T = absolutetemperature(25°C = 2981<) v,.M = Molar volume of water = 0.01802 litre ' molet PIPo = relative humidity at which the doo , spacing is observed. At 50% relative humidity, or at PIPo = 0.5, the d oo , spacing of Ca-smectite is 15.6'- or 1.56 nm (Brown, 1961). Therefore, at PIPo = 0.5 d oo , spacing of Casmectite is II =
0.0821 x 298 0.01802 x 2.303 log 0.5
= 941.02 atm . or 941 atm x 1.013 bar/atm. = 953.25 bars.
From this calculation, it is obvious that the second clay dewatering process requires much more energy, especially if the mineral is vermiculite. To remove the last monolayers of water which is absorbed on the clay surface, different free energies and pressures will be required, depending on the type of expandable 2: 1 layer silicate and the ion adsorbed on its clay surface. If it is smectite, the required free energy to release the last monolayer of water is much less than that of vermiculite. The explanation lies in two factors: first, the amount of charge deficit originating from the clay mineral is larger in vermiculites than smectites, and secondly, the seat ofthe charge is in both octahedral and tetrahedral positions, the latter being very close to the clay surface. This renders very strong attractive forces on the last monolayer of water. The magnitude of the force can be calculated from Coulomb's law:
At 4% relative humidity, PIPo = 0.04, the doc, spacing of Ca vermiculite is 12.0'- or 1.2 nm, therefore: 0.0821 x 298 n = 0.0802 x 2.3031090.04
= 4066.92alm.
As a result, to remove the second layer of water 4066.92 atm . or 4119.79 bars (4066.92 atm x 1.013 bar/atm) of pressure is required. For 0.25 nm thickness, van Olphen (1977) has calculated the pressure of removal of the second layer of water at 4000 bars. Using the partial free energy difference, .tof, we can measure the work in ergs -go' of H 2 0 , wh ich is required to free the second layer of water from the clay surface. This is given by -AT
~
=
M
x 2.303 log PIPo
where R = the gas constant(8.316 x 10 erg ' mole-i . degree K or 8.316 joules) T = the temperature in Kelvin (273°C) the molecular weight of water M (18.02 gm· mote-t) PIPo = the relative humidity. Thus, the respective free energy between free water and the second water layer which is absorbed on the clay surface is -412.05 x 107 ergs · g-' (-412.05 joules). Since one calorie is 4.184x 107 ergs, the partial free energy can be calculated as -98.48 cal, go'. 7
=
Geoscience Canada Reprint Series 4
passing research in northeastern British Columbia, the Beaufort-Mackenzie and the Sverdrup Basins in Canada , has shown that the transformation of smectite to 2:1 mixed-layer silicates passes through the vermiculitic step. Thus, the mixed-layered system cons ists of iIIitesmectite-vermiculite rather than the binary system observed by Perry and Hower (1972). The implication of having a ternary system with vermiculite as an intermediary in the transformation of smectite to illite is that the last layer absorbed water on the clay surface is very difficult to expel within the range of oil generation and therefore it should not be considered in the oil migration mechanism, at least for the wells studied in Canada. Perry and Hower (1972) reported a second clay dehydration from the pelitic sediments of the Gulf Coast at temperatures around 140°C, while in the Sverdrup and BeaufortMackenzie Basins the second dehydration step was not encountered at temperatures close to 150°C. The author believes that the difference is probably due to the type of smectites encountered in the different basins.
where F is the attractive force, Q is the charge of the particles in d-J,DKI.utn m esu, D is the dielectric con~ ~ § ~ stant of water, and r is the distance between the origin of charges and the centre of .::;:::~::.. water molecules. It is obvious that in vermicvalue is large W:~:t=':'H . ulites the and the r value is small. As a result, even at extremely low Ca <2j11rtlOl'tented PIPo relative humidity val- _ ,",",,,"""•.H ues water is still adsorbed on the vermiculite surfaces, C. '2~,""""''' making the desorption of __ .t ........ the last monolayer of water C. '2~,"orie"," very difficult. The exact free specunenat 5!5O'C energy or work requires a very precise knowledge of the PIPo value at which the ~~::.::,e:.. d oo, spacing of vermiculites becomes 9.6'-, or 0.96 nm. So far this value has not ":'~""~;'H been determined, but it is in the order of tens of thousands of atmospheres. A ..:::.:::.;:'~'.':.. good discussion on this sub ject is presented by van Olphen (1954, 1965) and Kittrick (1969a, b). Work by Foscolos and Kodama (1974), Figure 4 d oo1 spacings of calcium and potassium saturated < O.2p.m oriented specimen under various x-ray Foscolos et al. (1976) and conditions from a North Sabine H-49 well sample. Powell et al. (1978), encom-
a
c c C
C C C C
C C C C C C C C
C C C C C
C C C
C C C C (
c..
c
c
c
e (
l
c c
c
( (
Catagenesis of Argillaceous Sedimentary Rocks
181
(
c' C' ( .
C·
r
c o (
C
o r
(
C r
r: r:
r
It is possible that trioctahedral smectites behave differently than dioctahedral smectites, the latter being closer to a vermiculitic structure (Kishk, 1967). Todate, dioctahedral expandable clays have been encountered in all samples from the Sverdrup and Beaufort-Mackenzie Basins. IlIite-vermiculite-chlorite-smectite mixed layers are very often encountered at deeper burial depths (Powell et al., 1978). To identify a ternary or quaternary system of mixed-layer silicates the clay fraction must be subjected to a series of treatments with two different cations absorbed on the clay surfaces. By knowing the adsorbed cation on the clay surface and the physicochemical conditions under which the sample is subjected during x-ray analysis , the doo1 spacing can define the kind of expandable layer silicates, the components and their weighted percentages (Brown , 1961). Under given physical conditions, a binary mixture of illite-smectite with Ca2+ absorbed on its surface will show specific d oo1 spacings. K+ -saturated illite-smectite will achieve similar doo1 spacings under the same physical conditions. However, a ternary system of clays consisting of illite-smectite-vermiculite with Ca2+ absorbed on its surface will show d oo1 spacings (Figure 4). Chemical analysis of the calc ium-saturated clays can be used to confirm the results of the x-ray analysis . The CaO contents are used as a measure of the cation exchange capacity (C.E.C.) which is the charge of the clay surface, while the K20 contents measure the illitic component of the mixed layers. From these two results, an estimation of the expandable component can be made . Consequently, the chemical analysis can check the information obtained from the x-ray patterns concerning the composition of the mixed layers. Illite. Illite is not a particular mineral. The term is used to designate a group of clay minerals having a mica-type structure. Illites exist in 1M, 1Md, 2M and 3Tforms. The latter is a rare type . 1 Md illite, the disordered form , is the most common type in sedimentary basins , and as the temperature rises during catagenesis the reaction goes from 1Md to 2M. The methods for studyi ng these transformations by x-ray techniques and their significance in sedimentary rocks are reported by Maxwell and Hower (1967) and Velde and Hower (1963).
The crystallinity of illite, that is, the morphology of the 1 nm peak, has been used by Weaver (1960, 1961) and by Kubler (1964) as a diagenetic indicator. This indicator, which is quantified by measuring the ratio of heights at 1.0nm and 1.5nm orthe width ofthe 1 nm peak at half height, is valid only at the late stages of diagenesis or early metamorphism. In the early stages of diagenesis the rearrangement of ions in the interlayering spacing ofthe 2:1 layer silicates and the experimental condition under which the sample is x-rayed renders its value questionable. A comparative scheme using all these indiators for diagenesis is presented by Foscolos and Stott (1975) and Foscolos et al. (1976). Incipient metamorphism (anchizone) and low greenschist facies from the marginal zones of the Swedish Caledonides, Swiss Alps and VenezuelanAndes were determined on the basis of illite "crystallinity" (Kisch, 1980;Frey, 1970;Kubler et al., 1974; Shaga, 1977), while the lattic parameter b, of dioctahedral iIIite/muscovite has been used for the characterization of PIT gradients in incipient metamorphism (Padan et et., 1982). Kaolinite. Kaolinite is abundant in the early and middle stages of diagenesis, while usually absent in the last stages (Dunnoyer de Seconzac, 1970; Foscolos and Powell, 1980; Hower, 1981; Boles, 1981). It seems that once the chloride salt solutions have been expelled from pelitic sediments, the pH value of the system becomes more alkaline. With temperatures rising above 100°C, kaolinite is dissolved ortransformed to other minerals. As the ratio of [K+ JI [H+] in the pore solution increase, kaolinite is iIIitized (Hemley, 1959). For a [K + JI [W ] equal to 106, kaolinite is transformed to illite at 100°C while , at a ratio of [K+ J/[H+ J equal to 10a, a temperature of 200°C is needed to achieve the same conversion . Dunnoyer de Seconzac (1969) demonstrated the instability of kaolinite in seawater above 200°C. Another reaction where kaolinite is transformed at elevated temperatures is reported by Hutcheon et al. (1980). At high temperatures, kaolinite reacts with dolomite and silica to produce a magnesium chlorite, calcite and CO2, However, if Fe(OHb is present, an ironmagnesium chlorite can be produced, a product very often encountered in diagenesis.
Chlorite. Diagenetic chlorite is not detected in pelitic sediments in the early or middle stages of diagenesis in either the Sverdrup or the Beaufort-Mackenzie Basins or in northeastern British Columbia (Foscolos and Stott, 1975; Foscolos et al., 1976; Foscolos and Powell, 1980). The possible explanation lies in the presence of CO2 derived from thermal, and bacterial decarboxylation of organ ic matter or carboxylic acids generated from the thermocatalytic degradation of kerogen (Tissot and Welte, 1978; Carothers and Kharaka, 1978, 1980). These render the pore fluids acidic. Under low pH values, the chelating action of monofunctional and difunctional carboxylic acids complex Ala+ and Fea+
and at the temperature of 80°C chlorite can not be formed and therefore detected at the early stages of shale diagenesis. The instablity of chlorite in acid media and temperatures around 80°100°C is used as a technique for removing it from samples where kaolinite and chlorite are present. This facilitates mineral identification by x-ray diffraction (Brown, 1960). In late diagenesis, where amorphous silica , aluminum, iron and magnesium are available, chlorite an be formed from the aluminum , silica, iron and magnesium, as proposed by Almon and Davies (1979), where 2AI(OH) 4Mg2+(aq) + Fe2+(aq)
=
Fe2+ M9 4AI2SiaO,o(OH)s(solid)
+ 2H 20(I) + 8H+ or as proposed by Hutcheon et al. (1980), where 5CaMg(CO a)2 + AI2Si20s + Si02 + 2H20 = dolomite
kaolinite
quartz
M9sAI2Sia01o(OH)s + 5CaCOa + 5C02 chlorite calcite or, as proposed by Boles and Franks (1979), where 3.5Fe 2+
+ 3.5Mg2+ + 92H20 + 3AI2Si20s(OH)4 =
kaolinite Fe a.sMga.sAI6.oSi6.1 02o(OH)16
chlorite.
r:
Diagenesis
182
Geoscience Canada Reprint Series 4
r r
However, in most cases, chlorite in the late stages of diagenes is of pelitic sediments seems to derive from amorphous aluminum , iron and magnesium ions which are incorporated into the ternary systemof2:1Iayersilicates. Theconversion of chloritic integrades to discrete chlorites with depth was documented in samples from Drake Point 0-68 and Cape Norem A-80 wells in the Sverdrup Basin and in samples from Cretaceous shales of northeastern British Columbia (Figure 5). Based upon the presence of carbonic acid, derived from the maturation of the organic matter at the early stages, the overall reaction can be summarized as follows:
H+ + expandable clays
interchange reaction ....
time expandable clays heat .... time .... chloritized (2:1 layer silicates) heat.... .... chlorites .
water released between ambient temperature and 100°C is approximately 15%, while the water given between 100°C and 150°C is 7% (Jackson, 1956). This implies that prior to first clay dehydration, 1.5 g of H20 will be released from 10g of amorphous material to the pore space, while 0.7 g will be released after the first clay dehydration. This amount of water is roughly equal to that given off from 10g of smectite clay when it transformed to 50% illite and 50% smectite. On the basis of the previouscalculat ions, this amountsto 1.35g of H20. Thus, the amount of pure or saltfree water released in the pore system prior to, and during, the first dehydration of clays from both the amorphous material and the smectite transformation process creates an over-pressure zone. Statistical data indicate that there is a good correlation between the overpressured zone and the sonic transittime in shales (van Elsberg, 1978). Amorphous inorganic gels, besides supplying water to the system, also provide cementing agents to either the pelitic sediments or the adjacent sandstones through the process of mineral
C
C C C C C C C C C C C C
C C C
C C C
B
A
The absorption of protons by clays and their interchange with octahedral aluminum, iron and magnesium have been discussed by Foscolos (1964) and Foscolos and Barshad (1969). AMORPHOUS INORGANIC GELS Amorphous material, or gels, occur either by initial deposition or during the destruction of silicates such as feldspars, through the action of carbon ic acid and organic acids generated from kerogen degradation, or the transformation of smectite to illite. The colloidal portion of silica, aluminum and iron can be extracted and measured following the techniques of Aquillera and Jackson (1953) and that of Jackson (1965). Foscolos and Powell (1980) have determined free iron, silicon and aluminum in six wells and one formation in the Sverdrup and Beaufort-Mackenzie Basins of the Northwest Territories. The results indicate that amorphous material, which in some cases amounts to over 35% , decreases with burial depth. If one accepts that the amorphous inorganic gels have a composition similar to allophane, as proposed by Ross and Kerr (1934), then the amount of
formation. Under acid or neutral pH values amorphous silica probably is retained in shale, while amorphous aluminum being complexed by organic acids, probablydiffunctional carboxylic acids as suggested by Surdam et a/. (1984) or as hydroxyaluminum ion, migrates to form a monomer with amorphous silica. The latter is available in sandstones during this diagenesis. These monomers, as proposed by Siffert(1967), can be considered as precursors of authigenic booklets of kaolinite and hairy illites encountered in the sandstones (Figures 6 and 7). Such monomers can then link to similar monomers to form tetrahedral and octahedral sheets. The formation of an octahedral sheet involves the rearrange ment of OH ions and AI ions, while a tetrahedral sheet requires the removal of water from the OH group linked to silica ions. From Figures 6 and 7, it is obvious that the ratio of silica/aluminum determines the kind of amorphous monomer gel to be formed and, therefore, the kind of layer silicate. However, to precipitate clay minerals from solution, aluminum must be present in six-
d-spacing in nm ~ ~
i ~ a~ii
731.5
..... ..
ss
§
..,;
14".5
~-
d-spacing in ~ ~ ~ ~ ~~ l!
./
)\"'"" J
1\
I\.
112&1
I
/ )
I\.
I\..
I\.
I......
......
I\.
I\, ./
......
Il
l-
/
. . I
'"
"
I
20 15 _Zltol
e
a
U ......
I...... r.- wi
-.
co
~
l!
-A.
1127.7
_
e
I-'
-
v\ -
-
\
I--'
,.....f-
-
I\.
f
~
!
J
\
\
I\.
)\. ) ~ ) )
-\~ - ' I\.,. I
I
C C
C C C C C
c
.....
_J
c
\
)
L
f-'"
f-
§ ..10 § §...
.,.;
t-.
-
_.
DID
ss
\ )
II
10
Figure 5 X-ray diffraction patterns of oriented clay specimens from North Sabine H-49 well after heating at 550°C for 30 minutes (Fosc%s and Powell, 1979). (A) 0.2-2.0 Itm fraction. (8) <0.2Itm fraction.
C
e ( c C l (
C C C
l l
Catagenesis of Argillaceous Sedimentary Rocks
183
r:
r
r
c
o c
c r:
o
c
o o ,.c
fold co-ordination in orderto form a gibbsitic layer. Linares and Huertas (1970) and Surdam et al. (1984) have shown that organic acids complex aluminum in six-fold co-ordination , and that with pH change, aluminum hydroxide is formed and oriented into a gibbsitic layer. The latter absorbs silica tetrahedra to form kaolinite. Therefore , organic compounds derived from the diagenesis of organic matter in source rocks act similarly to induce synthesis and authigenesis of many different clay minerals in either shales or sandstones. Finally,iron should migrate from shale to sandstone in a similar way as aluminum that is, chelated by organic acids.
Carbonates. Calcite is very seldom encountered in pelitic sediments as a product of diagenesis. In the early stages of diagenesis, CO2 favours the dissolution of calcite through the reaction CaC0 3 + H20 + CO2 = Ca(HC03)2 Once Ca(HC0:J2 has encountered a neutral pH, authigenic calcite is formed. This usually occurs in sandstones adjacent to shales. Calcite cement is also formed when K~ derived from feldspar dissolution substitutes absorbed Ca2+ or Mg2+ on the smectite surfaces, thus indicating the transformation of smectite to interstratified smectite-illite along with the expelled water (Foscolos and
Kodama , 1974). The exchanged calcium moves with HCO; , derived in the early stages of diagenesis from the thermal and bacterial decarboxylation of organic matter, to the adjacent sandstone, precipitating authigenic calcite. Carbonate cements , dissolved total carbonates and c5 13 C values of CO2 in natural gas yield evidence for the organic origin of CO 2 (Boles , 1978 ; Carothers and Kharaka, 1980 ; Lundegard et et., 1984). Therefore, calcite cement is encountered in sandstones during the early stages of diagenesis (eodiagenesis). As transformation of smectite to illite proceeds with burial depth, pore fluids are enriched mainly
H
r
~><-I-
• I Octahedral - I Tetrahedral sheet sheet
r Figure 6
Formation of authigenic kaolinite (Siffert, 1967).
HO 10 HO 'Si-OH +5 [Al(OH)]2+-+ HO/ r
sheet
Figure 7
Formation of authigenic 2:1 layer silicate (Siffert, 1967).
Tetrahedra sheet Octahedral sheet
Diagenesis
184
Geoscience Canada Reprint Series 4
( ( (
with magnesium, and to a lesser extent, with iron. These enriched pore fluids migrate to sandstones along with carbonic acid to form dolomite cement or to react with calcite and form dolomite as follows: 2CaC03 + Mg2+ = CaMg(C03 k + Ql2+ . When the concentration of iron in the expelled pore fluids increases, ferroan dolomite and finally ankerite (CaMgo.5Feo.5(C03)2) is generated as an authigenic cement. Ankerite is very seldom encountered in shales. It seems that as diagenesis proceeds the pH of the pelitic sediment becomes alkaline because K-clays, in the absence of buffering salts in the pore fluids, raise the pH of the rock. This induces iron to precipitate as Fe(OHb probably on the 2:1 clay mineral surfaces. This process eventually leads to chloritization of the 2:1 layer silicates and the formation of iron-rich sedimentary chlorites. If authigenic kaolinite and quartz are present, then iron, magnesium and lor ankerite may react to produce iron-rich chlorite, calcite and CO 2, as documented by Muffler and White (1969) and Hutcheon et al. (1980).
RELATIONSHIP BETWEEN CLAY CATAGENESIS, ORGANIC MATTER DIAGENESIS AND CEMENTATION The relationship between clay catagenesis and organic matter diagenesis has been put forward by Powell et al. (1978) and Foscolos and Powell (1979) in their study of the Sverdrup Basin. The results are illustrated in Figure 8 and show that amorphous inorganic matter loses water and decreases in concentration at a burial depth about 600 m above the first dehydration step of smectite. The first clay dehydration step coincides with 0.5% Ro vitrinite reflectance and occurs at about 1500 m above the main phase of oil generation. The second dehydration takes place below the depth of the oil-generating zone, since vermiculite is a major component in the mixed layers (Powell et al., 1978). These results suggest that the depths of clay dehydration and the loss of water from the amorphous aluminosilicate gels are not coincident with the depths of oil generation, and it is difficult to envisage a role for water derived by this process in oil migration.
HYDROCARBON GENERATION
Reflectance
percent R o max.
SANDSTONES
SHALES
ORGANIC MATTER VITRINITE
An important aspect of diagenesis, however, is the destruction of 2: 1 layer silicates, as reported by Foscolos and Powell (1980). Although smectite is transformed to illite via vermiculitization, the absolute amount of 2:1 layer silicates decreases with burial depth. The destruction of silicates might have implications because absorption sites, where organic matter can be absorbed, are destroyed. Therefore, hydrocarbons might be moving away from pelitic sediments because the seats of absorption are destroyed. There is also a distinct possibility that once the absorbed sites are reduced, that is, once the cation exchange capacity is lowered, the absorptive sites are reduced and polar organic liquids are released to the pore systems. The effects, if any, of the destruction of the 2: 1 layer silicate with burial depth to the migration of hydrocarbons and to cementation of adjacent sandstone is a new area of research worth probing. Crystal-lattice-water derived from the breakdown of the layer silicates ensures an ongoing sediment dehydra-
Intensity
INORGANIC GEOCHEMISTRY
MINERALOGY
0.2 -2 p
<0.2 p.m fraction
AUTHIGENIC MINERALS SILICATES ,CARBONATES
Samacllte
> 30~
Kaolinite
emactlte
Lon of amorphous
IIl1to
FIRST CLAY DEHYDRATION
.
< 20'4 smactho
iE Ii~"
-.. 20'" vermiculite
> 55'4 IUlto incr••• lng
z:
aII
...
Isomorphous substhutlon
Intera'ratlfled clays
2
IIIltle.tlon
3
Inters'ratlfled clay. ~
zw
8>~
o
:i
.a
>
!;
E
Incorporation of
.Ii
amorphoua matt.r
::0
~
4
In mixed lay.r to form chlorlt. component.
of A1 3+for 814+ In
lncr•••ing IlIItlaotlon Interstratlfl.d clays
~
:lc
."
Disapp••renee of kaoUnite
Lo•• of ad.orbed .at.r Incr•••• In K20
I Decr•••• In c.tlon
a: w
Form.tlon of chlorlt.
o
from chlorlt. Intagrade=
ill
e z ;;; c w a: o
w
15
o
: ::
.:.•..: ::.. -:..,..: v, :••: • • '. ".
..•...:...-.~-,..•::.• ....•..\.'
5
C
C C C C
C C C C C C C C C C C C
C
Chlorite In,.grad•• Inorganic matter
" 51 II
C
C
m fraction
Intera.ratlfted clays
< 50'" Illite
..51 .
C
C C ...,25" vermiculite
0.5 -8 -----
C
!
SECOND CLAY DEHYDRATION 18
Figure 8 Relation between diagenesis (catagenesis) ofshales andthe occurrence ofauthigenic minerals in sandstones (Fosco/os and Powell, 1979).
C C C C C C (
e
c c
C l
C C
C C
c (
o (
( ( (
c o c o
o (
J ) J
r r
r:
:; l' (
Catagenesis of Argillaceous Sedimentary Rocks
tion. For example, in the North Sabine H-49 well, it was calculated that water derived prior to first clay dehydration from all reactions involving clay minerals (absorbed water, cavity water and crystal-lattice-water) amounts to 3.8%, while the water loss between two dehydration steps, that is, within the hydrocarbon generating zone, amounts to 3.7% (Foscolos and Powell, 1980). In addition to water, the dissolution of the amorphous inorganic matter and the destruction of the layer silicates releases silica, aluminum, iron, calcium, magnesium and potassium to the pore water. If these elements remain in place, they may precipitate to form part of the existing mineral, such as quartz, or form authigenic clays, such as chlorites. However, the net decrease of the layer silicates with depth suggests that some of the cations may migrate with compaction water into more porous and permeable horizons, forming authigenic clays in sandstone reservoirs (Curtis, 1978;Foscolos and Powell, 1979). In such cases, both permeability and porosity of shales and adjacent sandstones are affected by formation of authigenic cements both before and during hydrocarbon generation and migration. The importance of organic acids released during diagenesis from various types of kerogen in relation to ion migration from shales to sandstones and to the development of secondary porosity in reservoir rock is discussed in detail by Gautier et al. (1985) and Eglinton et al. (in press). The economic significance of applying the diagenetic modelling techniques with regard to the search for hydrocarbons and in evaluating reservoir properties (porosity and permeability) of targeted sandstone intervals prior to drilling is discussed in detail by Surdam and Crossey (1985) and Surdam et al. (1986)and summarized by Surdam and Crossey (1987).
CONCLUSIONS The following conclusions concerning the catagenesis of shales can be made: (1) The first stage in catagenesis is marked by the accumulation of amorphous inorganic components which occur either by initial deposition andlor during the destruction of silicates. This stage takes place prior to the onset of the first clay dehydration of the 2:1 layer silicates.
(2) The firstclay dehydration isattributed to the conversion of smectite to mixed layer silicates and occurs prior to the onset of hydrocarbon generation from the organic matter. Dehydration is attributed to isomorphic substitution of silicon ions in the clay structure and the ensuing absorption of potassium ions and displacement of calcium and magnesium ions from the clay surface. (3) The first dewatering of the clays marks the onset of the destruction of the 2:1 layer silicates, as well as kaolinite. Thus, there is a net decrease in concentrations as the clays undergo structural transformations. (4) The mixed layer clays comprise a ternary system of smectite-vermiculiteillite. Vermiculite appears to be an intermediary mineral in the conversion of smectite to illite. (5) The second dewatering of the clays occurs after oil has become cracked to gas. The high temperature requirement is attributed to the presence of a Cavermiculite component inthe mixed layer clays. (6) After the first dewatering of the clays, iron hydroxide precipitates on 2:1 layer silicates to form in the less than 0.2JA, fraction iron-rich chlorite integrades and in the 2-0.2JA, fraction iron-rich chlorites. (7) With burial, pelitic sediments release water not only upon compaction but also from the crystallization of amorphous material, the dehydration of the expandable layer silicates and the destruction of the layer silicates. (8) The destruction of feldspars, the transformation of the layer silicates and the production of organic acids during the diagenesis of organic matter provide inorganic material and chelating agents for the transportation of cations in adjacent sandstones. These cations, depending on their concentration, pH, temperature and other physicochemical conditions, form carbonates or authigenic silicates thus cementing the pore throats of sandstone reservoir rocks.
185
REFERENCES Diagenesis Almon, W.R. and Davies, D.K., 1979, Regional diagenetic trends in the Lower Cretaceous Muddy Sandstone, Powder River Basin, in Scholle, PA and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p.379-400. Aronson, J.L. and Hower, J., 1976, Mechanism of burial metamorphism of argillaceous sediments: 2. Radiogenic argon evidence: Geological Society of America, Bulletin, v. 87, p. 738-744. Boles, J.R., 1978, Active ankerite cementation in the subsurface Eocene of Southwest Texas: Contributions to Mineralogy and Petrology, v. 68, p. 13-22. Boles, J.R., 1981, Clay diagenesis and effects on sandstone cementation (case histories from Gulf Coast Tertiary), in Longstaffe, F.J., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p. 148-168. Boles, J.R. and Franks, S.G., 1979, Clay diagenesis in Wilcox sandstones of southwest Texas: Journal of Sedimentary Petrology, v. 49, p. 55-70. Burst, J.F., 1969, Diagenesis of Gulf Coast clayey sediments and its possible relation to petroleum migration: American Association of Petroleum Geologists, Bulletin, v. 53, p. 73-93. Carothers, W.W. and Kharaka, Y.K., 1978, Aliphatic acid anions in oil field waters implications for origin of natural gas: American Association of Petroleum Geologists, BUlletin, v. 62, p. 2441-2453. Carothers, WW. and Kharaka, Y.K., 1980, Stable carbon isotopes of HC03 - in oil field waters - implication for the origin of CO 2 : Geochimicaet CosmochimicaActa, v. 44, p. 323-332. Curtis, C.D., 1978, Possible links between sandstone diagenesis and depth-related geochemical reactions occurring in enclosing sandstones: Journal of the Geological Society of London, v. 135, p. 107-118. Dunnoyerde Seconzac, G., 1969, Les Mineraux argileux dans la diagenese: Passage au Metamorphisme, These Universite du Strasburg, Memoire, Service Carte Geologique d'Alsace-Lorraine, No. 29, 3200 p. Dunnoyer de Seconzac, G., 1970, The transformation of clay minerals during diagenesis and low grade metamorphism. A review: Sedimentology, v. 15, p. 281-340.
( 186
Diagenesis
Geoscience Canada Reprint Series 4
(
( Eberl, D., 1989, Clay mineral formation and transformation in rocks and soils, in Fowden, L., et al. , ed., Clay Minerals: their st ructure, behaviour and use: Royal Society of London, Philosophical Transactions, v. 311A, p. 241-257. Eglinton, T.I., Curtis, C.D. and Rowland, S.J., 1987, Generation of water-soluble organic acids from kerogen during hydrous pyrolysis; implications for porosity development: Mineralogical Magazine, v. 51, p. 495-503. Foscolos, A.E. and Kodama, H., 1974, Diagenesis of clay minerals from Lower Cretaceous shales of northeastern British Columbia: Clay and Clay Minerals, v. 22, p. 319-335. Foscolos, A.E. and Stott, D.F., 1975,Degree of diagenesis, stratigraphic correlations and potential sediment sources of Lower Cretaceous shale of northeastern British Columbia: GeologicalSurveyofCanada, Bulletin 250, p. 1-46. Foscolos, A.E. and Powell, T.G., 1979,Catagenesis in shales and occurrence of authigenic clays in sandstones, North Sabine H-49 well , Canadian Arctic Islands : Canadian Journal of Earth Science, v. 16, p. 1309-1314. Foscolos, A.E. and Powell, T.G., 1980, Mineralogical and geochemical transformation of clays during catagenesis and their relation to oil generation, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Society of Petroleum Geologists, Memoir 6, p. 153-171. Foscolos, A.E., Powell, T.G. and Gunther, P.R., 1976, The use of clay minerals, inorganic and organic geochemical indicators for evaluating the degree of diagenesis and oil generating potential of shales: Geochimica et Cosmochimica Acta, v. 40, p. 953-960 . Foscolos, A.E., Reinson, G.E. and Powell, T.G., 1982, Controls of clay mineral authigenesis in the Viking sandstones, Central Alberta: I. Shallow depths: Canadian Mineralogist, v. 20, p. 141-150. Frey, M., 1970, The step from diagenesis to metamophism in pelitic rocks during Alpine orogenesis : Sedimentology, v. 15, p. 261-279. Gautier, D.L., Kharaka, Y.K. and Surdam, R.C., 1985,Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralogists, Short Course No. 17, 279 p. Hedberg, H.D., 1936, Gravitational compaction of clays and shales: American Journal of Science, 5th series, v.31, p.241·287.
Hoffman, J. and Hower, J., 1979,Clay minerai assemblages as low grade geothermometers . Application to the thrust faulted disturbed belt of Montana, U.S.A., in Scholle, PA and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 55-80. Hower, J., 1981, Shale diagenesis, in Longstaffe, F.J., ed., Clays and the Resource Geologist: Mineralogical Association of Canada, Short Course Handbook, v. 7, p.60-80. Hower, J., Eslinger, E.V., Hower, M.E. and Perry, E.A., 1976, Mechanism of burial metamorphism of argillaceous sediments: I. Mineralogical and chemical evidence: Geological Society of America, Bulletin, v. 87, p. 725-737. Hutcheon, I., Oldershaw, A. and Ghent, E.D., 1980, Diagenesis of Cretaceous sandstones of the Kootenay Formation at Elk Valley (southern British Columbia) and Mt. Allen (southwestern Alberta): Geochimica et Cosmochimica Acta, v. 44, p. 1425-1435. Kubler, B., 1964, Lesargiles, indicateursde metamorphism: Revue de I'lnstitut Francals du Petrole, v. 19, p. 1093-1112. Longstaffe, F.J., 1983, Stable isotope studies of diagenesis in clastic rocks: Geoscience Canada, v. 10, p. 43-58. Lundegard, P.D., Land, L.S. and Galloway, W.E., 1984, Problem of secondary porosity: Frio Formation (Oligocene) , Texas Gulf Coast: Geology, v. 12, p.299-402. Maxwell, D.T. and Hower, J., 1967, Highgrade diagenesis and low-grade metamorphism of illite in the Precambrian belt series : American Mineralogist, v. 52, p. 843·857. Muffler, L.J.P.and White, D.E., 1969, Active metamorphism of Upper Cenozoic sediment in the Salton Sea geothermal field and Salton trough, southeastern California: Geological Society of America, Bulletin, v. 80, p. 157-182. Padan, A., Kisch, H.J. and Shagam, R., 1982, Use of lattice paramter b, of dioctahedral illite/muscovite for the characterization of PIT gradients of incipient metamorphism: Contributions to Mineralogy and Petrology, v. 79, p. 85-95. Perry, E. and Hower, J., 1972, Late stage dehydration in deeply buried pelitic sediments: American Association of Petroleum Geologists, Bulletin , v.56, p. 2013-1021, Powell, T.G., Foscotos, A.E., Gunther, P.R. and Snowdon, L.R., 1978, Diagenesis of organic matter and fine clay minerals: a comparative study: Geochimica et Cosmochimica Acta, v. 42, p. 1181-1197.
Power, M.C., 1967, Fluid release mechanism in compact ing mar ine and mudrocks and their importance in oil exploration : American Association of Petroleum Geologists, Bulletin, v. 51, p. 1240-1253. Schmidt, Y. and McDonald, DA, 1979,Texture and recognition of secondary porosity in sandstones, in Scholle, P.A. and Schluger, P.R.,eds., AspectsofDiagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication No. 26, p. 209-225. Surdam, R.C. and Crossey, L.J., 1985, Organic-inorganic reactions during progressive burial: Key to porosity/permeability enhancement and/or preservation: Royal Society of London, Philosophical Transactions, v. 315A, p.135-156. Surdam, R.C., Crossey, L.J., Hagen, E.S. and Heasler, H.P., 1986, Application of Basin Analysis to Diagenetic Modeling: American Association of Petroleum Geologists, Bulletin, v. 70, p. 653. Surdam, R.C. and Crossey, L.J., 1987, Integrated diagenetic modeling: A processoriented approach for clastic systems: Annual Review of Earth and Planetary Sciences, v. 15, p. 141-170. Tissot, B. and Welte, D.H., 1978,Petroleum Formation and Occurrence : SpringerVerlag, 538 p. van Elsberg, J., 1978, A new approach to sediment diagenesis . Part 1. An observed relation between sonic transittime and depth in the Tertiary sediment of the Mackenzie Delta : A potential petroleum exploratory tool. Part II. A revised concept of sediment diagenesis: Bulletin of Canadian Petroleum Geology, v. 28, p. 57-86. Velde, B. and Hower, J., 1963, Petrological significance of illite polymorphism in Paleozoic sediment rocks : American Mineralogist, v. 48, p. 1239-1254. Mineral Chemistry and Physical Chemistry
Aquillera, N.H. and Jackson , M.L., 1953, Iron oxide removal from soils and clays: Soil Science Society of America, Proceedings, v. 17, p. 359-364. Eslinger, E.V. and S. Savin, 1973, Mineralogy and oxygen isotope geochemistry of hydrothermally altered rocks of the Chake-Broadlands, New Zealand geothermal area: American Journal of Science, v. 273, p. 240-267. Foscolos, A.E ., 1964, Factors affecting the replaceability of octahedral Mg ion of various soil minerals with H ions, M.Sc. Thesis, University of California, Berkeley, California.
C C C C C C C C C
C C C
C C C C C C
C C C C C C
C C C C
C C
c e
c ( c l ( (
c
c
r:
Catagenesis of Argillaceous Sedimentary Rocks
187
r
o
c (
c o CJ o o c o o o o o o (
(
0:
o o o o o o o o Cd C ~.
'-- r
Fosolos, A.E. and Barshad, I., 1969, Equilibrium constants between both freshly prepared and aged H-montmorillonites and chloride salt solutions: Soil Science Society of America, Proceedings, v. 33, p.242-247. Hemely, J.J., 1959, Some mineralogical equilibria in the system (K 20-AI 20 3 Si0 2-H 2 ) : American Journal of Science, v. 257, p. 241-270. Jackson, M.L., 1956, Soil chemical analyses: Advanced course: Mimeograph published by the author, Department of Soil Science, University of Wisconisn, Madison, Wisconsin. Jackson, M.L., 1965, Free oxides, hydroxides and amorphous aluminosilicates, in Black, C.A., ed., Methods of Soil Analysis, Part I: American Society of Agriculture, p. 478-603. Keenan, A.G., Mooney, R.w. and Wood, L.A., 1951, The relation between exchangeable ions and water adsorption on kaolinite: Journal of Physics and Colloid Chemistry, v. 55, p. 1462-1474. Kittrick, J.A., 1969a, Interlayer forces in montmorillonite and vermiculite: Soil Science Society of America, Proceedings, v. 33, p. 217-222. Kittrick, J.A., 1969b, Quantitative evaluation of the strong-force model for expansion and contraction of vermiculite: Soil Science Society of America, Proceedings, v. 33, p. 223-225. Mooney, R.W., Keenan, A.G. and Wood, L.A., 1952, Adsorption of water vapour by montmorillonite. I. heat of desorption and application of BET theory: Journal of the American Chemical Society, v. 74, p. 1367-1371. Siftert, B., 1967, Some reactions of silica in solutions: Formation of clay: Israel Program for Scientific Translations, Jerusalem, 100 p.
van Olphen, H., 1954, Interlayer forces in bentonite: Clays and Clay Minerals, v. 2, p.418-438. van Olphen, H., 1965, Thermodynamics of interlayer absorption of water in clays. I. - sodium vermiculites: Journal of Colloid Science, v. 20, p. 822-9837. van Olphen, H., 1977, An Introduction to Clay Colloid Chemistry: John Wiley and Sons, A Wiley-Interscience publication, 317 p. Mineralogy
Birkeland, P.W., 1973, Pedology, Weathering and Geomorphological Research: Oxford University Press, 285 p. Brown, G., 1961, X-ray Identification and Crystalline Structure of Clay Minerals, Second Edition: Mineralogical Society of London, 544 p. Grim, R.E., 1968, Clay Mineralogy: McGraw-Hili Book Co., New York, 596 p. Kisch, H.J., 1980, Incipient metamorphism of Cambro-Silurian clastic rocks from .the Jantland Supergroup, central Scandinavian Caledonides, western Sweden: illite crystallinity and vitrinite reflectance: Geological Society of London, Journal, v. 137, p. 271-288. Kishk, F.M., 1967, Chemical and physical properties of soil vermiculite clays as related to their origin, Ph.D. Thesis, University of California, Berkeley, California, 175 p. Kubler, N., Martini, J. and Vuagnat, M., 1974, Very low grade metamorphism in the Western Alps: Schweizerische Mineralogische und Petrographische Mitteilungen, v. 54, p. 461-469. Linares, J. and Huertas, F., 1970, Kaolinite, synthesis at room temperature: Science, v. 171, p. 896-897.
McDowell, S.D. and Elders, W.A., 1980, Authigenic layer silicate minerals in borehole Elmore 1, Salton Sea geothermal field, California, U.S.A.: Contributions to Mineralogy and Petrology, v. 74, p. 293-310. Ross, C.S. and Kerr, P.F., 1934, Halloysite and allophane: United States Geological Survey, Professional Paper 185, p. 135-148. Shagan, R., 1977, Stratigraphic models for the northern Venezuelan Andes, in Espejo, C.A., et a/., eds., Tema II, Petrologia y Geochronologia; Tema III, Geodinamica (Sismologia, Tectonica y Geologia Estructural): Congreso Geologico Venezolana, Memoir 5, p.855-877. Steiner, A., 1968, Clay minerals in hydrothermally altered rocks at Wairakei, New Zealand: Clays and Clay Minerals, v. 16, p. 193-213. Sumi, D., 1969, Zonal distribution of clay minerals in the Matsukawa geothermal area, Japan: International Clay Conference, Japan, Proceedings, p.501-512. Weaver, C.E., 1960, Possible uses of clay minerals in search for oil: American Association of Petroleum Geologists, BUlletin, v. 44, p. 1505-1578. Weaver, C.E.,1961, Clay minerals ofOuachita structural belt and adjacent foreland, in The Ouachita System: Bureau of Economic Geology, Austin, Texas, p.144-160.
Originally published in Geoscience Canada v. 11 Number 2 (June 1984) Revised 1988
c
188
Diagenesis
Geoscience Canada Reprint Series 4
c c c c
c c c c c c c c c c c c ( c c c
c
c c c c c
C E Scalenohedral calcite crystal coated by pyrobitumen (arrows 1 and 2). Note further that pyrobitumen bridges intercr ystalline spaces (arrow 3) and is segmented by hairline cracks. Oil-wet rather than water-wet conditions appear to have been present in this porous system as is indicated by the low contact angles between the pyrobitumen and the calcite crystal faces . Beaverhill Lake reservoir sample. Photograph courtesy of F. Krause and S. Sayegh (University of Calgary and Petroleum Recovery Institute, respectively).
c
c c c e c (
c c C l l
c c c c o c o c o o o o o o o o o o o o
o o [
o o o o
o o o o
8 C
Chemistry and Diagenesis of Organic Matter
Chemistry and Diagenesis of Organic Matter in Sediments and Fossil Fuels M.A. Barnes W.C. Barnes A.M. Bustin Department of Geological Sciences University of British Columbia 6339 Stores Road Vancouver, British Columbia V6T 2B4 INTRODUCTION Organic matter occurs in almost all sedimentary rocks. In some rocks , such as coal and oil shale, organic matter is a major component, but, in most sedimentary rocks, it occurs in minor or trace amounts (Figure 1). The economic importance of organic matter, however, far outweighs its relative abundance. Organic matter, in one form or another, comprises the fossil fuel resources. The diagenesis of organic matter is thus of particular importance in understanding the generation and migration of oil and gas, in evaluating source rocks, and in estimating levels of maturation of oil shales and coals. Organic matter plays an important role in the transport of metal complexes (J.F. Barker, 1982),the formation of strata-bound and stratiform ore deposits (Anderson and Macqueen, 1982,1988; Macqueen and Powell, 1983) and has a significant influence on the diagenesis of other minerals (Powell et al., 1978;Schmidt and McDonald, 1979; Foscolos and Powell, 1980). While the diagenesis of organic matter is less understood and more complicated than that of the common minerals, significant progress has been made. It is now possible to predict the diagenetic state of organic matter if the burial and thermal history ofthe strata is known. The degree of diagenesis of organic matter has also become an accepted geothermometerto which other diagenetic reactions can be compared (Bostick, 1979).
189
In this chapter, we describe organic diagenesis from both petrographic and geochemical perspectives. The petrographic study of organic matter evolved from coal petrology, which has strongly influenced the terminology used. In a similar way, petroleum geochemistry has provided much of the terminology and techn iques of organic geochemistry. The nomenclature of these fields is complex, sometimes inconsistent and, to most geologists, unfamiliar. In this paper, we have attempted to simplify the terminology, hopefully without a major loss of information. Also, we provide a background to the geochemistry and petrography of organic matter and trace its progressive diagenesis, which generally accompanies increasing depth of burial. Our companion paper (this volume, p. 205-226) delineates some of the methods of quantifying diagenesis and evaluates predict ive models. Because of the breadth of organic diagenesis we can only present an overview of the subject here, and thus have avoided both controversial topics and those we consider less important. The reference list is essential for those readers who wish to pursue the subject in greater detail. General Considerations. Diagenesis of organic matter begins early and proceeds rapidly under surficial conditions. As a result, the terminology of diagenesis , which was largely developed for more slowly changing inorganic materials, requires some modification. Some of the changes take place even during sedimentation, and
are thus "pre-diagenetic " by traditional definitions. For example, it is commonly estimated that, under normal oceanic conditions, no more than one percent of the phytoplanktic organic matter produced in the marine euphotic zone survives microbial decomposition and zooplankton grazing to reach the sea floor unaltered (Wakeham et al., 1980; Knauer and Martin, 1981). Diagenetic changes that take place in organic matter at shallow depths below the sediment-water interface are largely biochemical , and occur as a result of the metabolism of fungi, bacteria and other micro-organisms. At greater sediment depths, microbial activity is much slower, and geochemical parameters more familiar to geologists (temperature , time) become dominant. The diagenesis of organic material is progressive and irreversible with few exceptions . The progress ive diagenesis of organic matter has been considered in terms of different stages. Tissot and Welte (1984) and many others (see Durand, 1980)refer to three stages of organic diagenesis: dia genesis, catagenesis and metagenesis. For the purpose of this chapter, eogenesis (Choquette and Pray,1970) is used to refer to those biological , physical and chemical changes in organic matter that occur at temperatures less than those required for significant cracking of hydrocarbons (Figure 2). As such, our eogenetic stage is equivalent to the diagenetic stage of Tissot at al. (1974) and we introduce the term solelyto prevent confusion between diagenesis
/
coal. oil shales, oil. gas, living organic matter bituminous rocks
mudrocks
and
s iltstones
Figure 1 Occurrence of organic matter in sediments as a percentage of mass of organic carbon. Hydrocarbons dissolved in water are not included. (Modified from Vassoyevich (1973) and Vassoyevich at al. (1976) in Bostick (1979)).
c
Diagenesis
190
-
Geoscience Canada Reprint Series 4
c c
---
(
Sea Water Bacterial Oxidation
Te/ogene.'.
Bacterial Sulphate Reduction Fermentation CH4 cO 2
Catagene.'.
lliopolyme rs
Figure 2 (above) Schematic representation of the terminology commonly used in describ ing the stages of diagenesis of organic matter (Tissot and Welte, 19 78 and others). The terms " eogenetlc ': " mesogenetic " and "telogenetic" were originally usedby Choquette and Pray(1970) to refer to the stages of diagenesis of carbonate rocks . The eogenetic zone includes reactions that are, mainly, biochem ica l. The eogenetic stages shown are modified from Curt is (1977). Catagenetic and metagenetic reactions are chem ica l, in response to the thermal history of the strata . Telogenesis results from contact with meteoric waters contain ing aerobic micro-organisms and atmospheric oxygen.
Unalter ed molecules
Monomers
rn
Prese rved
Ci5 w z w
\t:\ \\)~
(!)
o
Random po lyme rizatio n and cond ensat ion
w
Random , het ero g_ous geopo lYmef s
Ro- O.5 -
Figure 3 (right) Diagenetic fate of organic matter in sediments. (After Tissot and Welte, 1984).
-
-- -- -
-
rn
Ci5 w
z
w -e (!)
I-
< o
Ro - 2
METAGENESIS
c
"""""'Wf-=~===:....j
Gas + Condensa tes
-- -- -
c c c c c c c c c c c c c c c c c c c c c c ( c c c c c e c c c c c c c
(
c C' c c o c ( (
C' C
o c
c o
o c
(
(
l
Chemistry and Diagenesis of Organic Matter
as a stage of diagenesis and diagenesis as conventionally used by geologists. Diagenetic changes occurring at greater depths (and thus higher temperatures) in the lithosphere are referred to sequentially as catagenesis and metagenesis (Figure2). Diagenetic reactions resulting from organic material coming into contact wi th meteoric water or atmospheric oxygen following burial are referred to as telogenesis (Choquette and Pray, 1970). The different diagenetic stages do not have welldefined boundaries, but rather are gradational, which, to some extent, results from the heterogeneous composition of organic matter. Eogenetic, catagenetic and metagenetic stages are summarized in Figure 3. The processes and products of the diagenesis of organic matter can be described in kinetic terms. Because the compounds formed are metastable, and because the system is open, enabling migration of generated hydrocarbons, equilibrium, if ever established, is continually displaced (Durand, 1980). If diagenesis proceeds into what is generally considered the metamorphic realm , the organic residue may ultimately be transformed into graphite (Wedeking and Hayes, 1983). The results of diagenetic studies of organic matter are significantly affected by the methods of sample preparation, analytical procedures and equipment used. As a result data must be interpreted in terms of what fraction of the total organic matter present in the sample was "seen" by the method selected. Organic matter in rocks occurs as gases, liquids and solid particles; they differ in their physical and chemical properties, and in their accessibility to extraction by organic solvents, or to stripping by mechanical processes in the case of gases. Extractable organics are analyzed by a broad array of chemical and spectrometric methods. The analytical data obtained are used (1) in modern environments to monitor the fate of organic matter from natural biogenic sources and industrial pollutants; (2) to " fingerprint" oils and source rocks for oil-source rock correlations; (3) to assess maturation in coals; and (4) to assess the effects of maturation, migration and biodegradation which occur during the diagenetic conversion of biological lipids to petroleum. Solid organic matter tends to be finely disseminated
as discrete sedimented organics or precipitates. Although it is a minor component of rocks, it forms the major part of the organic matter present. With the exception of coals, it requires concentration before petrographic analysis by treatment with acids, bases and organic solvents to remove silicates and carbonates, or extractable organics in the case of pollen preparations (Karr 1978a,b, 1979). Sources of Organic Matter. Precursor organic material varies considerably in its chemistry, depending on its source and depositional environment. Unlike simple minerals or even solid solution series, organic matter consists of a complex array of compounds. These compounds mainly fall into four or five general classes that react differently, particularly during eogenesis (Figure 4). Prior to the Devonian, photosynthetic phytoplankton and bacteria were the principal sources of organic matter. At present, marine organisms contribute approximately 60% of the total organic productivity, with terrestrial plants making up most of the rest. Aquatic organisms contribute mainly proteins, lipids and , in the phytoplankton , carbohydrates; in addition, higher plants contribute resins, waxes, Iignins and carbohydrates in the form of cellulose (Figure 4). Lipids, which are major contributors to petroleum, contain fewer heteroatoms (nitrogen, oxygen and sulphur) than do proteins and carbohydrates. Thus, petroleum tends to be rich in carbon and hydrogen, and poor in heteroatoms. Rates of primary production in aquat ic environments are affected by the amounts of light and nutrients presen t. Most photosynthesis takes place in the upper 60-80 m of the water column , with rates being highest where abundant nutrients are present, such as in areas of upwelling and near river mouths. Open marine waters have estimated organic productivities that average 50g C ' m-2a-' (grams carbon per square metre per year); coastal waters average about twice that value, except on the western margins of continents, where values as great as 300g C'm- 2a- ! occur (Krey, 1970). Preservation is more important than productivity, however, in determining the organic content of sediments.
191
EOGENESIS The term eogenesis covers all low temperature diagenesis of organic matter ; eogenetic reactions are partly biochemical, arising from metabolic processes occurring within organisms, and partly abiotic chemical reactions which incorporate metabolic products released to the sediments. The effects of temperature and pressure are subord inate , and eogenetic reactions cannot be simulated by simple pyrolysis (Tissot and Welte, 1984). Preservation of organic matter is a function of the oxygen content of the environment. Estimates of the preservation of primary organic matter in the surficial sediments of marine environments average about 0.1 % (Menzel and Ryther, 1970). Preservation in subaerial env ironments is even less, as chemical oxidation and aerobic microbial decomposition are favoured by the high oxygen content of air as compared to only a few mL· L-' dissolved in water. Regions of rapid terrestrial sediment deposition, such as deltas and continental slopes and rises (Dow, 1978), may have relatively low organic contents because of dilution by inorganic sediments; however, the preservation of th is organ ic matter is high because of limited diffusion of oxygen into the sediments (Ibach, 1982). Terrigenous organic components are better preserved than those from aquatic sources (Meyers et al., 1984). The maximum preservation of primary organic matter, about 4%, occurs in anoxic environments such as the Black Sea (Deuser, 1971), as anaerobic microbial processes provide less energy, are slower and result in less complete decomposition of initial organic compounds . Although dissolved oxygen is lacking, other oxidizing agents such as sulphate are abundant in marine sediment pore waters and are utilized during bacterial sulphate reduction. As noted, rapid sedimentation limits exposure to oxygen and thus favours the preservation of organic matter. Areas where organic matter tends to be preserved and concentrated include cont inental marginal environments such as shelves, slopes and rises, marine landlocked basins, and lakes. Peats accumulate in regions of slow continuous subsidence with water tables at or near the surface, low pH (3 to 5) and low Eh. Most paralic coals
( 192
Diagenesis
were formed in swamps developed on deltas, interdeltaic plains , coastal plains and in back barrier regions . Proteins, carbohydrates (sugars, cellulose and chitin), lipids (fats, waxes and steryl esters) and lignins are the principal biopolymers contributed by organisms to sediments (Figure 4). However, these original biopolymers are rapidly depolymerized to their monomers by microbial processes occurring during and shortly after sedimentation. As a result, the dominant organic compounds found in sediments are geopolymers (humic compounds and kerogen) which arise from random chemical recombination of monomers released by these micro-organisms. During biosynthesis of the orig inal biopolymers by organisms, their monomers are linked together in bonds PROTEINS
-
o H H U I I
0
I
H
I
--
H
I
R
Pot.,,,*,
o I
I
h......".
- C H -C-C-OO
I
' I II
RHO
CH,-ICH,).C-O-cH,
_
HO-CH,
.'
HO-C-C-C-C-C-QH
I HI HI 0II
I
,
HO-CH, CH,-ICH,ln- C-O - CH,
acid
Fa«" acid.
I I I I N,'
o Ii
NH -C-C-C-C-C-C-oo
H
H
H H
o
II
Fatty acid
Wax (.at.,)
0
GJycerol
CH,-ICH, ln-C -00
CH,- {CH' )n- C- O- ICH' )n- CH,
, I I I I I II H
'n
(Glutamic ~
H
---
I I
3 CH -(CH ) - COOH+ HO- CH
CH,-ICH,l,,-~-O-r
......
II I I j'
---
a primary cycle that occurs in both the water column and the sediment, and a secondary cycle that occurs in sediments and sedimentary rocks (Figure3). The primary cycle, which is by far the more rapid, is the biochemical cycle operated mainly by microbial metabolism, which hydrolyzes the various biopolymers (Trudinger and Swaine, 1979; Bolin, 1983;Krumbein and Swart, 1983). The primary cycle operates on a total pool estimated at 2.7-3.0 x 101 6 g organic carbon, with a turnover half-life ranging from days to tens of years. The secondary, geochemical, cycle affects a pool of organic carbon which is about 2000 times larger,but hasa tumover half-lifeof several million years (Welte, 1970). As a result of the primary cycle, the major constituents of organisms - proteins, carbohydrates, lipids and lignins -
LIPIDS
NH,
"'C-C-N-C-C-N'"
I,
through the elimination of water; the resulting biopolymers are sensitive to hydrolytic cleavage with the re-incorporation of water by either chemical or microbial processes (Figure 4). Eogenetic changes , arising from microbial and chemical processes, profoundly affect (1) the physical structure and chemical stability oforganic matter, which can subsequently become fossil fuel after burial and catagenesis; (2)the fate of organic pollutants arising from humankind's activit ies; (3) the formation of humic compounds which form metalcomplexes and act as metal transport agents in groundwater systems; and (4) Eh, pH, the precipitation of carbonates and the formation of sulphides in sedimentary environments (Gautier, 1985). Two carbon cycles affect the constitution of organic matter:
Geoscience Canada Reprint Series 4
+ HO-ICH')n-CH, CARBOHYDRATES
o Monom...
"'oAL-( EovO~ -'
H-C=O hydrolytit
~oAo·
OH
U • CH,-ICH')n- C-OH
hydrol.""
CH,OO
lHIIuco..
Fatty ac id
I I HO-C-H I H-C-OO I H-C-OH I H-C-OH
+
o
~c/"O'
H,COC+IH
CH,OO
HO-C-H
I I H-C-OH I
LIGNIN CH,OH
CH,OO
CH
CH
CH
CH
CH
I
H-C-OO
H,-C-OH
C C
C C C C C C C C C C C C C C
C
C
#0
H-C-N-C-CH,
H
C
C C
H-C=O
I
C C C C C C
C
Slaryl ••'.,
HO
I
C
C
II
CH,-ICH,'n-C-O
H,-C-OH
"'oAL-( Eovo~OCH~.
( (
c..... age
01
.tMrbDNla
II
CH
I II
-eH,OH
I
II
~~, ","4~. ¢ OH
OH
00
SIn.",1 alcohol
Figure 4 (above) Variation in huminite-vitrinite reflectance in oil (Ro %), level of diagenesis (coal rank) and typical molecular structures at different diagenetic levels. The aromatic clusters are shown in plan view and show increases in ordering and cluster size with increases in coal rank. (Modified, in part, from TeichmOl/er and TeichmOl/er, 1982).
c
e (
c
c c c c
c
c r c (. r:
r
r r (
o
o o
c
o
o o o o o o
o o o o o
Chemistry and Diagenesis of Organic Matter
comprise less than 20% of the organic matter present in sediments and are much less significant in older sedimentary rocks. Primary cycle hydrolysis produces monomeric amino acids and sugars that (1) provide energy for bacteria, leading to the formation of CO2 , H2 0 , CH4 , NH3 and other simple inorganic products; (2)are incorporated into new biopolymers in microbial cell walls ; and (3)are chemically condensed and randomly recombined to form the irregularly organized geopolymers known as humic compounds (Figure 3). The term humic compounds includes the humic and fulvic acids, the humins and kerogen, all of which are defined simply on the basis of their solubility and molecular weight. Because they form by random recombination, their structures are variable, depending on the monomers available and the diagenetic conditions at the time of their formation. Humic compounds increase in molecular weight and decrease in solubility in progressing from fulvic acids through humic acids and humins to kerogen. The humic and fulvic acids are defined on the basis of their solubility in bases and acids (Schnitzer and Khan, 1972; Schnitzer, 1975; Aiken et a/., 1985). Fulvic acids are soluble in both acids and bases and range in molecular weight from 700 to 10,000 (Rashid and King, 1969). Humic acids are base soluble polymers which range in molecular weight from 10,000 to 100,000 and can beupto 10nm in size. Theyaresolublein acids above pH 2; at pH 2 or less, they are insoluble and precipitate. With increasing cross linkage and loss of acidic functional groups (carboxyl and phenolic hydroxyl groups), the humic and fulvic acids lose their base solubility and form an insoluble humin! kerogen fraction. The term humin is used for the base-insoluble fraction of soils and young sediments. Humins are partly hydrolyzable; they can re-incorporate water and release 15-40% of their weight as monomers or lower molecular weight polymers during demineralization with hydrofluoric acid, the standard method used in sample preparation to remove silicates. In contrast to the humin fraction , kerogens are high molecular weight geopolymers which are stable to hydrolysis with HF and are insoluble in organic solvents, acids and bases. While some kerogen
formation occurs through loss of functional groups and cross linkage formation early in the diagenetic history of organic matter (Philp and Calvin,1976), mostforms later,reaching a peak during thermal maturation at the catagenetic stage. Kerogen can also arise directly from monomers without an intervening humic stage, especially under anoxic conditions (Huc and Durand, 1977; Debyser et al., 1977). Microbial metabolism in the water column and the upper few centimetres of the sediment is commonly aerobic and produces carbon dioxide and water as end products (Sundby et a/., 1983). In areas of high productivity, aerobic processes can deplete the oxygen content of the water, leading to the development of anoxic conditions. Anaerobic metabolism, on the other hand, allows more of the initial biological compounds to survive; it also leads to the reduction of sulphate to sulphide, the reduction of ferric iron, and the production of methane and ammonia. In the water column, these processes are favoured by restricted water circulation arising from salinity or temperature stratification (Powell, 1984). Anaerobic processes are themselves limited by a lack of nutrients and the toxicity of their end products . Relative tothevery labile proteins and carbohydrates, lipids and Iignins are more resistant to primary cycle metabolism. Depolymerization of Iignins requires cleavage of chemically stable ether bonds. Many of the lipids are preserved in stable cell wall residues of micro-organisms or are associated with water-insoluble waxes and cutin from higher plants. As a result, lignins and lipids tend to survive and are concentrated along with spores, pollen, cutin, chitin and plant resins in the humin! kerogen fraction. This is especially true in anoxic environments, which lack oxygen-requiring Iignolytic fungi (Debyser et al., 1978). In peat bogs, the concentration of humic acids also tends to decrease the activity of decomposers. This is partly the result of the low pH (3 to 5) and partly because the humic compounds are themselves relatively immune to decomposition because of the presence of phenolic groups, which inhibit microbial activity. The irregularity of the linking bonds in humic compounds and their bulky three-dimensional structures make them
193
a poor fit for templating by the microbial enzymes which catalyze their hydrolysis and decomposition. These diagenetic processes not only produce energy for the microbial populations, but also release large quantities of new compounds which significantly alter the Eh, pH and ionic composition of pore waters in the sediments. Eh decreases abruptly from the oxic to the anoxic zone. The pH increases slightly, with the reduction of sulphate, ferric iron and manganic manganese, production of ammonia and sulphide, and increasing alkalinity, leading to the precipitation of carbonates and to the dissolution of amorphous silica and, to a lesser extent, of quartz (Suess, 1979). Iron (and copper, lead and zinc if present) may be precipitated as sulphides. The complex role organic matter plays in the sulphur cycle and the formation of sulphide minerals has been reviewed by Jtlrgensen (1983). Those amino acids and sugars which were released by hydrolysis of proteins and carbohydrates, but not utilized in the energy cycles of micro-organisms, can recombine through abiotic chemical processes to form the geopolymers found in sediments and sedimentary rocks. The boundary between eogenesis and catagenesis is gradational, although it is generally considered to be coincident with the boundary between sub-bituminous and high volatile bituminous coals, which corresponds to a vitrinite reflectance of about 0.5% (Figure 5). Base-soluble humic compounds are usually negligible, although they are found in some sub-bituminous coals. Thermal history, a combination of time and temperature, plays a significant role in the formation of high-volatile bituminous coals. Otherwise coal would remain at a peat or lignite level of diagenesis, as is the case for the Lower Carboniferous lignites of the Moscow Basin (TeichmQller and TeichmQller, 1982, p. 58). Similar processes control the eogenesis of organic matter whether it is disseminated in sediments or concentrated in peats, which can become coals, or in algal muds, which can become oil shales. Hydrocarbon generation during eogenesis is restricted to biogenic gas which originates from anaerobic decay of organic matter (Oremland and Kvenvolden, 1981). Relative to gas generated during catagenesis, biogenic gas
Diagenesis
194
% Ro
COAL RANK
Geoscience Canada Reprint Series 4
(,,oil (H
0.15
Peat
(right) Chemical structures of proteins , carbohydrates, lipids and lignins. Figure 5
~"'
1
0.25
Lignite
0.38
!
Sub-b ituminous
0.50
High-volat ile bituminous
1.10
Medium-vo lat ile bituminous
1.45
Low-volatile b ituminous
~
,
+
~
Semi-anthracite
2.50
Anthracite
10
~
Maceral groups (Eogenesis)
Maceral groups
o
~o~ ... 0
0"
m 05.90'W
$~ll
' C' r i c h
Lipid rich components
Lignin, carbohydrates
+
Huminite
+
Inertinite
~
Inertinite
Liptinite
~
Liptinite
0
OllOH
Ii
!
Graph ite
'H'rich
OrIgIn
-.
OIl
1.90
~
OIl
Ol/OH
II
("tt
V itrinite
Char
~
~
(Catagenesis, Metagenesis)
c c c c c c c c c c c c c c c c c c c c c c c c c c c C E (
C
C C
c e c
40
).1m
(
c Figure 6 Terminology of organic petrography and representative photomicrographs of the main maceral groups at different diagenetic levels. Photomicrographs are oil immersion, reflected light images. Liptin ite group : (a) alginite I Ro 0.3%) ;(b) cutinite (blue light excitation) I Ro 0.50%) ; (c) bituminite-sporinite IRo 1.0% ). Huminite-vitrinite group: (d) huminite I Ro 0.30%]; (e) vitrinite IRo 1.2%] ; (f) vitrinite IR o4.0%). Inertinite group : (g) inertinite and semi-inertinite I Ro O. 60% ) ; (h) inertinite I e; 1.2%] ; (i) inertin ite I Ro4. 0%) . Thecorrelation between vitrinite reflectance (Ro %) and diagenetic level is shown in Figures 3 and 7.
(
(
C C
C
r: \.
c c G C (J
C r
o
r:
c
o o o o
o r:
v
o
Chemistry and Diagenesis of Organic Matter
is primarily methane (up to 99%) and carbon dioxide (0-8%), with minor amountsof heavier gases (Claypool and Kvenvolden, 1983). Locally, biogenic gas may form significant accumulations, especially as gas hydrates (Bily and Dick, 1974; Hitchon, 1974; Kvenvolden and McMenamin, 1980;Kvenvolden et a/., 1983). Other hydrocarbons present at the eogenetic stage of diagenesis are inherited from organic matter. Those changes in organic matter occurring during eogenesis which are evident in optical microscopy are most noticeable in the huminite fraction (Figure 6). This fraction originates from humic acids and forms vitrinite during catagenesis. The huminite passes through a stage referred to as gelification (TeichmOller, in Stach et el; 1982, p. 228) in which it swells and its reflectivity increases (from 0.15%to 0.50% oil immersion); it becomes slightly darker and increasingly opaque in transmitted light. The lipid-rich fraction (Iiptinite) shows little or no change in incident light and becomes progressively darker in transmitted light, changing from nearly colourless to light brown. Under ultraviolet irradation the peak fluorescence intensity shifts from 500 nm wavelength (at deposition) to between
Type 1
195
560 and 580 nm. Carbon-rich components, such as semi-inertinite and inertinite, which originate from fires (charcoal) or fungal attack, show no visible change. All organic components become physically compacted with increased burial, and previously visible cells and tissues become progressively less distinct.
CATAGENESIS With increased burial organic matter undergoes progressive diagenesis in response to increased temperature and time. Increased pressure has little diagenetic effect on organic matter, other than compaction; it may actually retard diagenesis, as suggested in some experimental studies (TeichmOlier and TeichmOller, 1982; Sajg6 et al., 1986). Kerogen makes up the great bulk of organic matter present at the catagenetic stage. Kerogen is an irregular geopolymer formed by randomchemical recombination of organic compounds released by microbial decomposition of proteins, carbohydrates and lipids. As such, its structure isvariable and cannot be predicted in the same way that the structures of biopolymers such as proteins or carbohydrates can be predicted from their constituents. Kerogens are
Liptinite macerals
1.5
o o
o ...as CJ
1.0
E
o as
o.....
0.5
_C
f;.:;.. \: } A 1"i,i,i,II B
::L
Inertinite r,
o
O+-----,------r-----r------r-----'
o
0.1
0.2
0.3
0.4
0.5
O/C atomic ratio Figure 7 "van Krevelen plot" of diagenesis of kerogen types and coal macerals: A, immature - field ofhumic acid occurrence overlaps region of immature Types 1 and 2 kerogens; S, mature - boundary represents region of maximum oil generation; C, overmature; Rm , iso-evolution as measured by vitrinite reflectance; ., sites where coal macerals coincide with kerogen types; as diagenesis increases in the mature kerogen, coalmaceralsplotat slightly10werOlC values. (After Durand, 1980; van Krevelen, 1961; Tissot and Welte, 1984).
commonly characterized by the use of bulk parameters, such as the HIC and OIC atomic ratios, obtained from elemental analyses and expressed as a "van Krevelen diagram" (Figure 7). On the basis of these ratios, three or four types of kerogen have been defined which appear to follow distinct diagenetic pathways. This type of diagram was originally used to characterize coals and their coalification pathways (van Krevelen, 1961), but was later applied to kerogens (Mciver, 1967; Durand et al., 1972). A similar plot of hydrocarbon index versus oxygen index is obtained from measurements of the volumes of hydrocarbon gases and carbon dioxide released during the pyrolysis of rock samples containing kerogen (Espitalie et a/., 1977). Type 1 kerogen has high initial HIC and low OIC ratios. The small amount of oxygen present occurs mainly in ester bonds (Robineta/., 1977).Upt080%byweight of immature type 1 kerogens can be volatilized during pyrolysis (Tissot and Welte, 1984). The source material is mainly algal, or a combination of algal lipids and waxes from higher plants, which have been selectively concentrated by microbial reworking and enriched by bacterial lipids. It forms part of the Iiptinite macerals in coal, and is found in Scottish torbanites and algal bog-head coals (Claret et a/., 1981; Powell and Snowdon, 1983; Tissot and Welte, 1984). Examples of type 1 kerogens in fresh to brackish water lacustrine source rocks include the Eocene Green River "oil shales", the Early Cretaceous of the Songliao and Jinquan Basins of China and the Late Triassic Ordos Basin, also in China (Powell, 1986). Type 1 enriched source rocks produce oils with> 20% wax at maturation levels above 0.7% Ro ' Type 2 kerogen has intermediate initial HIC and OIC ratios, and larger contributions by aromatic and carboxylic acid groups than the type 1 kerogens. Ester bonds are abundant, as are medium-length aliphatic hydrocarbon chains and naphthenic rings. Both algal and higher plant sources appear to contribute to type 2 kerogens (Iiptinite coal macerals).Organic sulphur is present in heterocyclic groups and in sulphide bonds. Up to 60% of immature kerogen can be volatilized during pyrolysis (Tissot and Welte, 1984). Type 2 kerogens commonly are found in reduced
c 196
Diagenesis
marine strata and have been the source for major oil and gas deposits. Examples of type 2 kerogens formed in moderately reducing clastic sediments are found in the Sverdrup Basin, in Upper Devonian, Lower Carboniferous and Upper Cretaceous strata of Alberta, in the Jeanne d'Arc sub-basin east of Newfoundland (Powell and Snowdon, 1983) and in some parts of the North Sea region (Thomas et al., 1985).Examples of type 2 kerogens formed in highly reducing marine evaporites or carbonates are found in the Permian Phosphoria Formation ofthe northern United States and in Middle Devonian reefs of northern Alberta and northeastern British Columbia. Sulphur-rich type 2 kerogens also develop in saline lacustrine sedimentary sequences which had a high initial sulphate content. Heavy oils with a high asphaltene and sulphur content (up to 12%) are generated at somewhat lower levels of maturity (Ro = 0.5%); an example occurs in the Eocene to Oligocene of the Jianghan Basin of China (Powell, 1986). Mixed
Geoscience Canada Reprint Series 4
kerogen types are common. For example, marine sourced organic matter with some contribution by terrestrial higher plants produces a kerogen that falls between types 1 and 2 because of the increased OIC ratio, as in the shales of the Viking Formation (Mciver, 1967). Type3 kerogen has an initial HIC ratio less than 1,an initial OIC ratio of 0.2 to 0.3, and is derived dominantly from terrestrial plants. It consists mainly of aromatic groups formed from lignins and humic compounds, and represents the huminitelvitrinite fractions of coal petrography. Aerobically oxidized semiinertinite and inertinite macerals may also be present. Oxygen is present as carboxylic acids, ketones and ethers rather than in ester bonds (Robin et al., 1977). Lesser amounts of aliphatic groups may be present in liptinite and resinite macerals, depending on the preservation of resins and lipids from higher plant and algal sources. Deposition is mainly in deltaic and other thick, rapidly accumulating continental margin sediments; as a result, relatively
unaltered higher plant compounds can also accumulate, as they are protected by rapid sedimentation from aerobic degradation. The proportions of resinite, liptinite and vitrinite in terrestrially derived kerogen, however, strongly control both the level of thermal alteration necessary for petroleum generation and the relative proportions of oil, wet gas and dry gas produced (Snowdon and Powell, 1982;Powelland Snowdon, 1983). Liptinite- and resinite-rich kerogens can be sources of both oil and gas, whereas vitrinite-rich kerogens are mainly gas sources. Examples of type 3 kerogens include those of the Lower Mannville shales of Alberta, the Mahakam Delta of Indonesia (Durand and Oudin, 1980) and the Niger Delta. Resinite-rich type 3 kerogens are found in the Beaufort-Mackenzie Basin; liptinite-rich but resinite-poor type 3 kerogens occur on the Labrador Shelf (Powell and Snowdon, 1983). Catagenesis takes place at moderate temperatures (50°-150°C) and pressures (30 to 100 or 150 MPa), corre-
n-Alkanes
~Ro
-•.... •at
~
(Biogenic)
GEOCHEMICAL FOSSILS
0
lU
co
Hydrocarbons generated
r- ,,~~:
c:
.
CD ZCD
ale
eO eN
-
-O.5~
•e 0
N
c:
..•..
0
at
II
CJ
-1.35~
.... _.e
~CI:3
..•
cil
-
15
25
25
35
35
0
0
6
6
Aromatics
00 U
13
33
i:J
13
33
-~~LJ 15
25
35
Number of Carbon atoms
0
6
Number of Rings
t
CD
Q,
13
>-
s:
U
III CD
e
#
I
33
Number of Carbon atoms
2.0~
>0.-e ...
II
~
_C.J ~ -~ ~ 15
-•..
Cycloalkanes
OCl:3 -4-5~
Figure 8 Hydrocarbon generation: changes in chemical structure and quantity with increased burial depth (after Tissot and Welte, 1984). Ro 0/0, reflectance of vitrinite in oil. Upper profile: biogenic fingerprint (eogenesis); middle profile: dilution of biogenic fingerprint by hydrocarbons generated in the oil zone; lower profile: loss of biogenic fingerprint and chain shortening of hydrocarbons with cracking of oil to form wet gas at lower end of oil zone.
(
C C C
C (
C C C C
C C
C C C C C C C C C C .C C C C C ~
C C C
C
c
e (
c
c c c c c
r (
r C C G
r (
(
C
o o c o o o o o o
o o o o o o o r
o 8
Chemistry and Diagenesis of Organic Matter
sponding to the zone of bituminous and anthracite coal, and of oil and wet gas generation. At the end of catagenesis, aliphatic chains have been cleaved from kerogens, and the size of aromatic units has increased through chemical condensation, with the development of significant ordering (Figure 5). Coals progressively increase in rank from subbituminous to anthracite during catagenesis. Changes in chemical structure of hydrocarbons and the quantities generated with increasing burial depth are summarized in Figure 8. Hydrocarbons contributed to the sediment by precursor organisms are called biogenic; geochemical fossils, on the other hand, are compounds, unknown in living organisms, which are generated during diagenesis but which still retain a "fingerprint" characteristic of biogenic sources. Biogenic hydrocarbons can include methane and small amounts of ethane and ethylene from bacteria (Oremland and Kvenvolden, 1981),C 1 6 C 1 8 alkanes from plankton, and long chain hydrocarbons (C 1 2-C3 S ) from higher plant waxes . Biogenic hydrocarbons in the immature zone are characterized by normal alkanes with a significant odd-carbon preference, by acyclic (open chain) branched hydrocarbons, by cycloalkanes with 2 to 5 rings, and by terpenes, some of which are partly to fully aromatized during early eogenesis and form geochemical fossils (Wakeham et et., 1980; Barnes and Barnes, 1983). The characteristic fingerprints for biogenic hydrocarbons and those generated during eogenesis are shown in Figure 8 (upper profile). As burial increases, hydrocarbons generated by catagenetic processes dilute precursor compounds and there is a gradual loss of the biogenic fingerprint as oil generation progresses (Figure 8, middle profile). The level of catagenesis and its relation to hydrocarbon generation is of particular importance in defining petroleum potential. The levels of catagenesis corresponding to the oil birth line and the oil death line have been used to define an "oil window" (Figure 8). The oil birth line is the level of catagenesis which initiates generation of medium to low molecular weight (liquid) hydrocarbons from kerogen and corresponds to a vitrinite reflectance of approximately 0.5% (Dow, 1977).The amounts of poly-
aromatic compounds formed during eogenesis are relatively small, and are extensively diluted in petroleum by those aromatic compounds generated during catagenesis . The oil death line is the level of diagenesis at which both kerogen and previously formed liquid hydrocarbons are cracked to create low molecular weight hydrocarbons (Figure 8, lower profile). It corresponds to a vitrinite reflectance of about 1.35% (Dow, 1977). Significant amounts of gas are generated from kerogen throughout the oil window and continue to be produced to much higher levels of diagenesis. At low levels of catagenesis wet gas is generated; as the level of catagenesis increases the proportion of methane increases, through further cracking. By the onset of metagenesis, only meth ane (dry gas) remains . The amount, timing and composition of hydrocarbons generated depend on the kerogen type as well as on the level of diagenesis (Rogers, 1980; Mon nier et a/., 1983). The hydrocarbon generation model shown in Figure 8 applies mainly to type 2 kerogens formed in moderately reducing sediments. The expanded hydrocarbon generation model in Figure 9, proposed by Powell and Snowdon (1983), shows how kerogen types 2, 3 and 4 generate different proportions of oil and gas at different levels of diagenesis. The oil birth line occurs at a somewhat lower level of maturation (Ro = 0.4%) for both resinite-rich type 3 kerogen and type 2 kerogen formed in highly reducing carbonate or evaporite environments. In carbonate environments, low iron contents inhibit the formation of pyrite , leading to the development of S-rich type 2 kerogens; as a result, heavy S-rich and asphaltic oils are generated from lacustrine or marine sources at somewhat lower levels of maturation. Non-marine oils generated from liptinite-rich type 1 and type 3 kerogens, however, are formed at somewhat higher levels of thermal maturation (Ro = 0.6-0.7%). The maximum gas generation for type 2 kerogen occurs with the cracking of liquid hydrocar bons at levels of thermal maturation greater than Ro = 1.35%. Significant amounts of gas are generated, however, at lower levels of maturation in liptinite- and resinite-rich type 3 kerogen (Ro > 0.55%)(Monnier eta/. , 1983). The liquid hydrocarbon potential decreases
197
and the gas potential increases with increasing vitrinite content. Type 4, a vitrinite-rich terrigenous kerogen with HIC < 0.8, is primarily a gas source. Throughout catagenesis, kerogen undergoes noticeable changes in petrographic appearance (Figure 6). With increasing levels of catagenesis, the vitrinite component (huminite of the eogenetic stage) changes relatively uniformly (Figure 6d,e,f) . In incident light, reflectance increases (from 0.5% to about2.0% - oil immersion); in transmitted light vitrinite becomes darker and increasingly opaque. Cell structure and plant tissues which are readily visible in the early stages of eogenesis (Figure 6d) become progressively more difficult to recognize without etching with acids. Liptinite undergoes a rapid change in appearance (Figure sa.b.c) and physical properties. The trend of these changes is discontinuous. A discrete "jump" marks the transition from eogenesis to catagenesis. Bitumen, which includes both hydrocarbons and compounds with functional groups containing N, 0 and S, is generated . The kerogen residue which remains after bitumen generation shows an abrupt, noticeable decrease in volatiles and a corresponding relative increase in carbon content. A second "jump" occurs with the formation of low volatile bituminous coal. In sediments with the potential for oil generation, this level of catagenesis corresponds to the oil death line; at this second jump, lipitinite macerals undergo another abrupt reduction in hydrogen and volatiles. In reflected light the liptinite becomes progressively "lighter" grey; by the second jump it is indistinguishable in reflected light from vitrinite without etching. With ultraviolet light excitation the Iiptinites show a progressive increase in maximum spectral fluorescence wavelength with increasing diagenetic level, and total fluorescence intensity progressively declines. In transmitted light the liptinite, initially pale yellow, becomes progressively darker and eventually black and opaque. The semi-inertinite and inertinite components are initially rich in carbon and strongly aromatized (Telchmuller, in Stach et aI., 1982)at the time of deposition and thus undergo little additional alteration. The degree of later alteration depends on the initial degree of charring. The reflectivity of semi-inertinite and inertinite in
c 198
Diagenesis
Geoscience Canada Reprint Series 4
C C C
Vitrinite reflectance. %Ro 0.6.
0.4
0.8
1.0
1.2
(
1.4
C
--RESINITE-RICH
Gas-naphthenic condensate
/.
H/C 0.8-0.95
light naptii'hen'lC ? oil .
------------Cl:-----:::=========:;;====Ii;IE:=-en
ca
.......
~
..._-
.~ --_ Q)
L1PTIN1TE-RICH .~ RESINITE-POOR rn
Gas
H/C 0.8-0.95
dry gas
waxy on
--=:::::
VITRINITE-RICH H/C <0.8
Gas
/
drygas
HYDROCARBON GENERATION MODEL - TERRESTRIAL ORGANIC MATTER
c c c c c c c c c c c c c
c
Vitrinite reflectance, %Ro 0.6
0.4
(Carbonate? Fe-poo: C\I
Cl
w
.2 c:
~
MOD. EUXIN1C ENVIR.: H/C >1.1 (Clastic Fe-rich)
1.0
1.2
1.4
;<
HIGHLYEUXINIC ENVIR.: H/C >1.1
n,
0.8
c c c c c
c
heavy S-rich & asphaltic oil intermediate
c
c c c c c
i"
.......- - -
e (
- . Dissolution in gas phase
..... Cracking
HYDROCARBON GENERATION MODEL - MARINE ORGANIC MATTER Figure 9
Hydrocarbon generation for kerogen types 2, 3 and 4. (After Powell and Snowdon , 1983).
c ( c c
C
l
C
(
c
Chemistry and Diagenesis of Organic Matter
199
(
( (
o
c (
( (
o
o o o (
c o c o l o ( o o
incident light is initially high ; only at advanced levels of diagenesis does any increase in reflectivity occur (Figure 6g,h,i). In transmitted light semiinertinite and inertinite are opaque throughout diagenesis. METAGENESIS
Metagenesis refers to the stage of organic diagenesis at which crystalline ordering of the organic matter begins (Tissot and Welte, 1984) . Aromatic nuclei increase in size, forming clusters, and C-C bonds are broken , generating methane. Aliphatic molecules that survived catagenesis and previously generated hydrocarbons are cracked to dry gas. At the onset of metagenesis the vitrinite component of the coal has a reflectance of about 2%, which progressively increases with diagenesis to about 4% or 5%, where the strata are generally considered to be in the realm of metamorphism. The liptin ite component of the organics (type 1 kerogen) and, at later stages, the inertinite component become visually and chemically indistinct from vitrinite. TELOGENESIS
Organic compounds in rocks can be altered by late stage processes resulting from contact with meteoric waters of low salinity and with atmospheric oxygen (Figure 2). Two distinct processes appear to be active (Bailey and others, 1973b).Water washing results from the
solution of hydrocarbons in ground waters, leading to a depletion of the lighter compounds, as the solubility of normal hydrocarbons in water is inversely proportional to their chain length. The second process, bacterial degradation, is more important and can lead to the total loss of normal alkanes and isoprenoids, leaving a heavy oil residue composed of cycloalkanes and aromatics (Deroo and Powell, 1978). The heavy oil belt of Alberta is estimated to contain 142billion m3 of oil in a 1000 km arc from the Peace River in westcentral Alberta to Lloydrninstsr in western Saskatchewan. Similarities in the distribution of cycloalkanes, aromatics and benzothiophenes were used as evidence for the relationship of oil sands and heavy oils (L1oydminster) with conventional oils (Lower Cretaceous Mannville reservoirs) (Deroo et al. , 1974; Deroo and Powell, 1978). Selective removal of normal alkanes and isoprenoids on going from non-degraded oils to biodegraded heavy oils is consistent with patterns observed in incubation studies on the microbial degradation of a range of undegraded oil and oil sand samples (RUbinstein et al., 1977; Rubinstein and Strausz, 1978;Crawford et al., 1978). Winters et al. (1983) proposed hydrous pyrolysis , where source rocks are heated in liquid water for 1 to 5 days at temperatures below the critical point [<374°Cj,asatechniquefordistinguishing immature from biodegraded
o
(
c Figure 10 Photomicrograph of oxidized vitrinite in a low-volatile bituminous coal. Oxidation halos occur on periphery of vitrinite particle.
'r"
oils. Significant differences in the CPI, the amount of oil,the hydrocarbon chain length and the relative distribut ion of isoprenoids and triterpanes are apparent in gas chromatograms of a solvent extract of immature Kimmeridgian oil shale, Kimmeridgian crude oil, the crude oil microbially degraded in the laboratory and oils of different maturity generated from the immature shale by hydrous pyrolysis. While the immature extract looks like a mildly degraded oil, the crude oil and oils generated by hydrous pyrolysis ofthe immature shale show a closer match in distributions of normal hydrocarbons, triterpanes and isoprenoids. Microbial degradation of petroleum has long been recognized in the generation of sulphur in the cap rocks of salt domes (Thode et al., 1954; Krouse and McCready, 1979) where sulphate from anhydrite is used by anaerobes such as Desulfovibrio to oxidize petroleum fractions, leading to the formation of sulphide , part of which is subsequently oxidized to elemental sulphur. Such processes have been implicated in the degradation of oils in many areas, such as the Bell .Greek field, Montana (Winters and Williams , 1969)and the northern Williston Basin of Saskatchewan (Bailey et al., 1973a,b, 1974). In shallowdepth groundwaters , where sufficient dissolved oxygen may be present, aerobic bacteria can also degrade petroleum. In a study of biodegradation with aerobic cultures, Bailey et al. (1973a) found that isoprenoids were attacked atter the removal of n-alkanes; lower ring napthenes and low molecular weight aromatics were attacked at the same time as lighter normal paraffins, prior to the removal of heavier paraffins. Similar processes affect kerogen that is exposed to the atmosphere or to oxygenated groundwaters. Non-aromatic groups are preferentially removed, leading to an increase in the aromatic fraction of the kerogen (van Krevelen and Schuyer, 1957). Acidic functional groups such as carboxyl, carbonyl and phenolic hydroxyl are formed , eventually leading to the production of humic acids. Such processes are analogous to retrograde metamorphism in metamorphic rocks. If oxidation is intense it may be distinguished petrographically by the development of darker or lighter oxidation haloes on the margins of kerogen part icles (Figure 10;Bustin et al., 1985).
200
In coals oxidation results in a marked decrease in coal quality. Coals exposed to atmospheric oxygen or meteoric water have lower than normal caloric values, form poor cokes or are noncoking, and have reduced flotation properties (hydrophobicity). CONCLUSION Organic diagenesis is progressive and irreversible with few exceptions. It begins early and proceeds rapidly under surficial conditions so that even in the most favourable environments, only a small fraction of the organic matter is incorporated into the sediment. During eogenesis, micro-organisms metabolize most ofthe organic matterto CO2 and H20 in oxygenated sediments and to CH4 , CO2 , H2 and H20 under anoxic conditions. The remaining organic matter forms humic and fulvic acids. With increasing cross-linkage and loss of functional groups, these acids decrease in solubility and form part of the kerogen fraction. Similar processes control eogenetic reactions whether the organic matter is disseminated in sediments or concentrated in peat. The catagenetic and metagenetic stages of diagenesis result from increased temperature and time accompanying burial ofthe sediment. All types of kerogen progressively evolve toward pure carbon with a decrease in hydrogen and oxygen in response to the breakage of aliphatic chains and some cyclic compounds and the accompanying generation of bitumen, including hydrocarbons. By the metagenetic stage of diagenesis, crystalline ordering of aromatic compounds occurs and any aliphatic molecules that survived catagenesis, as well as previously generated wet gas and oil, are cracked to dry gas. With diagenesis the petrographic constituents of the organic matter (macerals), originally distinct chemically and morphologically, become progressively similar. During eogenesis the lipid-rich components (Iiptinite) alter little except during extreme oxidation. The huminite components swell with gelification and their reflectivity in incident light increases from 0.15% to about 0.5%. Duringthecatagenetic and metagenetic stages the vitrinite (huminite of eogenesis) macerals progressively increase in reflectivity in incident light, reaching 2.0% atthe end of catagenesis
Diagenesis
and about 4% or 5% at the end of metagenesis and onset of metamorphism. During catagenesis and metagenesis, the liptinite component becomes progressively darker and opaque in transmitted light. Inertinite, which originates from charcoal, undergoes little change during the early stages of diagenesis and only at advanced stages of metagenesis does its reflectivity in incident light increase. Semi-inertinite macerals, which are transitional in composition between vitrinite and intertinite, increase in reflectivity with diagenesis. The increase in reflectivity depends on the amount of carbonization that occurred prior to deposition. Hydrocarbons are generated from organic matter throughout diagenesis. During eogenesis biogenic methane forms as a result of anaerobic decay of organic matter. During catagenesis wet gas and/or oil is generated, depending on the type of kerogen, in response to increased temperature and time accompanying burial. By the metagenetic stage of diagenesis only dry gas is generated by cracking of kerogen and earlier formed hydrocarbons. Coal progressively increases in rank during diagenesis from peat through anthracite. The most noticeable changes in coal accompanying increased rank are the loss of moisture, oxygen and hydrogen and the corresponding increase in fixed carbon. Telogenesis can affect both organic matter and generated hydrocarbons. Contact with meteoric water of low salinity results in leaching of the lighter molecules and promotes bacterial degradation of petroleum. Oxidation of kerogen mayalso occur, resulting in preferential removal of the non-aromatic fractions and formation of humic acids. Petrographically, oxidation is marked by the occurrence of oxidation haloes and micro-fissures in the kerogen particles. REFERENCES General Reviews on Organic Compounds in Sediments and Rocks Barker, C., 1982, Organic Geochemistry in Petroleum Exploration: American Association of Petroleum Geologists, Education course notes 10. Barnes, M.A. and Barnes, W.C., 1978,Organic compounds in lake sediments, in Lerman, A., ed., Lakes: Chemistry, Geology, Physics: Springer, New York, p. 127-152.
Geoscience Canada Reprint Series 4
Brooks, J., 1980, ed., Organic Maturation Studies and Fossil Fuel Exploration: Academic Press, New York, 441 p. (Includes 25 papers covering diagenesis of organic matter, classification, chemistry and techniques). Bustin, M., Cameron, A.R., Grieve, D.A. and Kalkreuth, W.D., 1985, Coal Petrology: its principles, methods, and applications, Second Edition: Geological Association of Canada, Short Course Notes Volume 3. Durand, B., 1980, ed., Kerogen: Editions Technip, Paris, 519 p. (15papers on the petrography, chemistry and diagenesis of kerogen). Eglinton, G. and Murphy, M.T.J., 1969, eds., Organic Geochemistry: methods and results: Springer, New York, 828 p. Hunt, J.M., 1979, Petroleum Geochemistry and Geology: W.H. Freeman, San Francisco, 617 p. Ibach, L.E.J., 1982, Relationship between sedimentation rate and total organic carbon content in ancient marine sediments: American Association of Petroleum Geologists Bulletin, v. 66, p. 177-188. Karr, C., 1978a, ed., Analytical Methods for Coal and Coal Products, v. 1: Academic Press, Toronto, 580 p. (19 papers on the physical and chemical properties of coal and coal products and methods for their analysis). Karr, C., 1978b, ed., Analytical Methods for Coal and Coal Products, v. 2: Academic Press, Toronto, 669 p. (17 papers covering the structures of coal, coke, their combustion products and mineraI matter). Karr, C., 1979, ed., Analytical Methods for Coal and Coal Products, v. 3: Academic Press, Toronto, 641 p. (18 papers on coal gases and waste products, including analytical procedures). Larsen, G. and Chilingar, G.V.,1979,Diagenesis of Sediments and Rocks, in Larsen, G. and Chilingar, G.V., eds., Diagenesis in Sediments and Sedimentary Rocks: Elsevier, Amsterdam, p. 1-29. Robert, P., 1981, Classification of organic matter by means of fluorescence: application to hydrocarbon source rocks: Coal Geology, v. 1, p. 101-137. Robert, P., 1985, Histoire geothermique et diagenese organique: Bulletin des Centres de Recherches Exploration-Production Elf-Aquitaine, Memoire 8, 345 p. Stach, E., Mackowsky, M.-Th., Teichmuller, M., Taylor, G.H., Chandra, D. and Teichmuller, R., 1982, Stach's Textbook of Coal Petrology: Borntrager, Berlin, 535 p. Tissot, B.P.and Welte, D.H., 1984,Petroleum Formation and Occurrence: a new approach to oil and gas exploration, Second Edition: Springer, Berlin, 538 p. Trudinger, P.A. and Swaine, D.J., 1979, eds., Biogeochemical Cycling of Minerai-forming Elements: Elsevier, Amsterdam, 612 p.
c c c c
c
c (
c
c c c c c c c c c c c c c c c c c c c C G C C C
C C E C. ( (
c. c c c c
c (
c. c. c c~ c (
( (
c o o o o o
c r
Chemistry and Diagenesis of Organic Matter
Sources of Organic Matter in Sediments Cohen, A.D . and Spackman , W., 1980, Phytogenic organic sediments and sedimentary environments in the Everglades-mangrove complex of Florida: Part 3. The alteration of plant material in peats and the origin of coal macerals : Palaeontographica , v. 172, ser . B., p.125-149. Demaison, G.J. andMoore, G.T., 1980, Anoxic environmentsand oil source bed genesis: AmericanAssociationof PetroleumGeologists, Bulletin, v. 64, p. 1174·1209. Dow, w.G., 1978,Petroleum source beds on continental slopes and rises : American Association of Petroleum Geologists, Bulletin, v. 62, p. 1584-1606. Krey,J., 1970,Die Urproduction des Meeres, in Dietrich, G., ed., Erforschung des Meeres:Umschau, Frankfurt, p. 183-195. Meyers, PA, Leenheer, M.J., Eadie, B.J. and Maule , S.J., 1984, Organic geochemistry of suspended and settling particulate matter in Lake Michigan: Geochimica et Cosmochim ica Acta, v. 48, p. 443-452 . Styan, W.and Bustin, R.M., 1983, Petrography of some Fraser River Delta peat deposits: coal maceral and microlithotype precursors in temperate climate peats: Coal Geology, v. 2, p. 321·370. Trudinger, P.A., Swaine, D.J. and Skyring, G.w., 1979, Biogeochemical cycling of elements - general cons iderations, in Trudinger, P.A. and Swaine , D.J., eds., Biogeochemical Cycling of Mineralforming Elements: Elsevier, Amsterdam, p. 1-27. Waples, D.W., 1983, Reappraisal of anoxia and organic richness , with emphasis on Cretaceous of North America: American Association of Petroleum Geologists, Bulletin, v. 67, p. 963-978. Welte, D.H., 1970,Organischer Kohlenstoff und die Entwicklung der Photosynthese auf der Erde : Naturwissenschaften, v. 57, p. 17-23. Eogenesis of Organic Matter
(
(
Aiken , G.R., McKnight, D.M ., Wershaw, R.L. and MacCarthy, P., 1985, eds. , Humic Substances in Soil , Sediment, and Water: Wiley, New York, 692 p. Alpern , B., 1980, Petrographie du kerogen, in Durand, B., ed., Kerogen: Editions Technip, Paris, p. 339-371. Anderson, G.M. and Macqueen, R.W., 1982, Mississ ippi Valley type lead -zinc deposits : Geoscience Canada, v. 9, p. 108-117. Baker, E.w., 1969, Porphyr ins , in Eglinton, G. and Murphy, M.T.J., eds., Organic Geochemistry: methods and results: Springer, New York, p. 464-497.
Barker, J.F., 1982,The potential importance of radionuclide or metal complexation and transport by organic matter in groundwaters , in Geotechnical Research: Proceedings of the 7th Nuclear Fuel Waste Management information meeting, p. 66-110. Barnes, MA and Barnes, W.C., 1983, Oxic and anoxic diagenesis of diterpenes in lacustrine sediments, in Bjor~y, M., Albrecht,P.,Cornford,C.,deGroot,K., Eglinton , G., Galimov, E., Leythaeuser, D., Pelet, A., RullkCitter, J. and Speers, G., eds., Advancesin Organic Geochemistry 1981: Wiley, Chichester, p, 289-298. Bolin, B., 1983,The Carbon Cycle, in Bolin, B. and Cook, R.B., eds., 1983,The Major Biogeochemical Cycles and Their Interactions: Scientific Committee on Problems of the Environment [SCOPE]: Wiley, New York, p. 41-46. Bustin , A.M., Hills, L.V. and Gunther, P.R., 1977, Implications of coalification levels , Eureka Sound Formation, northeastern Arctic Canada: Canadian Journal of Earth Sciences, v. 14, p. 1588-1597. Claypool , G.E. and Kvenvolden, K.A., 1983, Methane and other hydrocarbon gases in marine sediment: Annual Review of Earth and Planetary Sciences, v. 11, p.299-327. Debyser, Y. , Pelet, A. and Dastillung, M., 1977, Geoch imie organique de sed iments marins recents : Mer Noire, Baltique, Atlantique (Mauritanie), in Campos, R. and Goni, J., eds., Advances in Organic Geochemistry 1975: Enadimsa , Madrid, p. 289-320. Debyser, Y.,Gadel, E, Leblond, C. and Martinez , M.J., 1978, Etude des composes humiques des kerogenes et de la fraction hydrolysable dans les sediments, in Geochim ie organique des sediments marins protonds Orgon II. Atlantique, Nord-Est Bresil : Paris, Editions du CNRS, p. 339·354. Deuser, W.G., 1971, Organic-carbon budget of the Black Sea: Deep-Sea Research, v. 18, p. 995-1004. Gautier, D.L., 1985, Interpretation of early diagenesis in ancient marine sediments, in Gautier, D.L., Kharaka, Y.K. and Surdam, A.C., eds., Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralogists, Short Course Notes, v. 17, p.6-78. Hitchon, B., 1974, Occurrence of natural gas hydratesin sedimentarybasins,in Kaplan, I.R., ed., Natural Gases in Marine Sediments: Plenum, New York, p. 195-225. Huc, A.Y. and B.M. Durand, 1977, Occurrence and significance of humic acids in ancient sediments: Fuel, v. 56, p. 73-80. J(lrgensen, B.B., 1983, The microbial sulphur cycle, in Krumbein, W.E., ed., Microbial Geochemistry: Blackwell, Oxford, p. 91-124.
201
Knauer, G.A. and Mart in, J.G., 1981, Phosphorus-cadmium cycling in northeast Pacific waters: Journal of Marine Research, v. 39, p. 65-76. Krouse , H.R. and McCready, R.G.L., 1979, Reductive reactions in the sulfur cycle, in Trudinger, P.A. and Swaine, D.J., eds., Biogeochemical Cycling of MineralForming Elements : Elsevier, Amsterdam, p. 315-368. Krumbein, W.E. and Swart, P.K., 1983, The microbial carbon cycle , in Krumbe in, W.E ., ed., Microb ial Geochem istry: Blackwell , Oxford, p. 5-62. Kvenvolden, K.A. and McMenamin, M.A., 1980, Natural Gas Hydrates: a review of their geologic occurrence : United States Geological Survey Circular 825. Kvenvolden, K.A., Barnard, L.A., Brooks, J.M. and Wiesenburg, D.A., 1983, Geochemistry of natural gas hydrates in oceanic sediment, in Bjor~y, M., Albrecht , P. , Cornford, C., de Groot, K., EgJinton, G., Galimov, E., Leythaeuser, D., Pelel, A., RullkCitter, J. and Speers, G., eds. , Advances in Organic Geochemistry 1981: Wiley, Chichester, p.422-430. Mackenzie , A.S., Patience, R.L. and Maxwell, J.R., 1981, Molecular changes and the maturation of sedimentary organic matter, in Atkinson , G. and Zuckerman, J., eds., Origin and Chemistry of Petroleum: Pergamon, Oxford, p. 1-31. Macqueen , R.W. and Powell, T.G., 1983, Organic geochemistry of the Pine Point lead-zinc ore field and region, Northwest Territories , Canada: Economic Geology, v. 78, p. 1-25. Menzel, D.W. and Ryther, J.H., 1970, Distribution and cycling of organic matter in the oceans, in Hood, DW., ed., Organic Matter in Natural Waters: University of Alaska , Institute of Marine Science, Occasional Publication 1,p. 31-54. Oremland, R.S~ and Kvenvolden,K.A., 1981, Microbial formation of ethane in anoxic estuarine sediments : Appl ied and Environmental Microbiology, v.42, p.122-129. Philp , R.P. and Calvin , M., 1976, Kerogen structures in recently-deposited algal mats at Laguna Mormona, Baja California: a model system for the determ ination of kerogen structures in ancient sediments , in Nriagu, J.O., ed., Environmental Biogeochemistry, v.1 : Ann Arbor Science Publishers, Ann Arbor, p. 131-148. Powell, T.G., 1984, Some aspects of the hydrocarbon geochemistry of a Middle Devonian barrier-reef complex, western Canada ,in Palacas, J.G., ed., Petroleum Geochemistry and Source Rock Potential of Carbonate Rocks: American Association of Petroleum Geologists, Studies in Geology 18, p. 45-61.
202
Aashid, M.A. and King, L.H., 1969, Molecularweight distribution measurements on humic and fulv ic acid fractions from marine clays on the Scotian Shelf: Geochimica et Cosmochimica Acta, v. 33, p. 147-151, Schnitzer, M., 1975,Chemical, spectroscopic and thermal methodsfor the classification andcharacterizationofhumicsubstances, in Povoledo, D. and Golterman, H.L.,eds., Humic Substances: their structure and function in the biosphere: Centre for Agricultural Publishing and Documentation, Wageningen, p. 293-310. Schnitzer, M. and Khan, S.U., 1972, Humic Substances in the Environment: Marcel Dekker, New York, 327 p. Suess, E., 1979, Mineral phases formed in anox ic sediments by microbial decomposition of organic matter : Geochimica et Cosmochimica Acta, v. 43, p.339-352. Sundby, B., Bouchard, G., Lebel, J. and Silverberg , N., 1983, Aates of organic matter oxidation and carbon transport in early diagenesis of marine sediments, in BjorGly, M., Albrecht, P.,Cornford, C., de Groot , K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, A., Rullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 350-354. Ting, ET.C., 1977, Microscopical investigation of the transformation (diagenesis) from peat to lignite : Journal of Microscopy, v. 109, p. 75-83. Tissot, B., Durand, B., Espitalie, J . and Combaz, A., 1974, Influence of the nature and diagenesis of organic matter in formation of petroleum : American Association of Petroleum Geologists, Bulletin, v. 58, p. 499-506. Wakeham, S.G., Schaffner,C. andGiger,W., 1980, Polycyclic aromatic hydrocarbons in Aecent lake sediments - II. Compounds derived from biogenicprecursors during early diagenesis: Geochimica et Cosmochimica Acta, v. 44, p. 415-429. catagenesis and Metagenesis
Bostick, N.H., 1979, Microscopic measurement of the level of catagenesis of solid organic matter in sedimentary rocks to aid exploration for petroleum and to determine former burial temperatures - a review, in Scholle, P.A. and Schluger, P.A., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 17-43. Claret , J., Jandine, S. and Aobert, P., 1981, Le d iversite des roches meres petrolleres: aspects geologiques et implications economiques a partir de quatre exemples: Bulletin des Centres de Aecherches Exploration-Production Elf-Aquitaine, v. 5, p. 383-417.
Diagenesis
Cook, A., Hutton, A. and Kantsler, A., 1980, Oil shales: Scientific Australian, v. 4, no. 4, p. 6-14. Cooper, B.S. and Murchison , D.G., 1969, Organic geochemistry of coal, in Eglinton, G. and Murphy, M.T.J., eds., Organic Geochemistry: methods and results: Springer, New York, p. 699-726. Creaney, S., 1980, The organic petrology of the Upper Cretaceous Boundary Creek Formation, Beaufort-Mackenzie Basin: Bulletin of Canadian Petroleum Geology, v. 28, p. 112-129. Curtis, C.D., 1977, Sedimentary geochemistry: env ironments and processes dominated by involvement of an aqueous phase: Aoyal Society of London , Philosophical Transactions, v. 32A, p. 353-372. Deroo, G . , Powell , T.G . , Tissot , B. , McCrossan, R.G. and Hacquebard, P.A., 1977, The origin and migration of petroleum in the Western Canadian Sedimentary Basin: a geochemical and thermal maturat ion study: Geolog ical Survey of Canada , Bulletin 262. Dow, W., 1977, Kerogen studies and geological interpretations: Journal of Geochemical Exploration, v. 7, p. 79-99. Durand, B. and Oudin, J.L., 1980, Exemple de migration des hydrocarbures dans une ser le deltaique: Ie Delta de la Mahakam, Kalimantan, lndonesie: 10th World Petroleum Congress, Proceedings, v. 2, p. 3-11, Durand, B., Espitalie, J., Nlcatse, G. and Combaz, A., 1972, Etude de la matiere organique insoluble (kerogene) des argiles du Toarcien du Bassin de Paris: J. Etude par les precedes optiques, analyse elementaire, etude en microscopie et diffraction electroniques: Aevue des Institute Fran9aise du Petrole: v.27, p.865-884. Espttalle, J., Laporte, J.L., Madec , M., Marquis, F., Leplat, P., Paulet , J. and Boutefeu, A., 1977, Methode rapide de characterisation des roches meres, de leur potentiel patroller et de leur degre d'evolution: Aevue des Institute Francatss du Petrole , v. 32, p. 23-42. Foscolos, A.E. and Powell , T.G., 1980, Mineralogical and geochemical transformations of clays during catagenesis and their relation to oil generation, in Miall , A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Soc iety of Petroleum Geology, Memoir 6, p. 153-172. Hacquebard, P.A., 1975, Pre- and postdeformational coalification and its significance for oil and gas exploration, in Alpern , B., ed., Petrographie organique et potential petrolier: Paris, Editions du CNAS, p. 225-241.
Geoscience Canada Reprint Series 4
Hacquebard, P.A. and Donaldson , J.A ., 1970, Coal metamorph ism and hydrocarbon potential in the Upper Paleozoic of the Atlantic Provinces: Canadian Journal of Earth Sciences, v. 7, p. 1139-1163. Hood, A. and Castano, J.R. , 1974, Organic metamorph ism: its relationship to petroleum generation and applicat ion to studie of auth igen ic minerals : United Nations ESCAp, CCOp' Technical Bulletin 8, p. 87-118. Kalkreuth, W., 1982, Aankand petrographic composition of selected Jurassic-Lower Cretaceous coals of British Columbia, Canada: Bulletin of Canadian Petroleum Geology, v. 30, p. 112-139. Legall, ED., Barnes, C.R. and Macqueen, R.W., 1981, Thermal maturation , burial history and hotspot deve lopment, Paleozoic strata of southern OntarioQuebec, from conodont and acritarch colour alteration studies : Bulletin of Canadian Petroleum Geology, v.29, p.492-539. Mciver, A.D., 1967, Composition of kerogen - clue to its role in the origin of petroleum: 7th World Petroleum Congress, Proceedings, v. 2, p. 25-36. McWhae, J.A.H., Elie, A., Laughton, K.C. and Gunther, P.R., 1980, Stratigraphy and petroleum prospects of the Labrador shelf: Bulletin of Canadian Petroleum Geology, v. 28, p. 460-488. Milner, C.W.D., Aogers, M.A. and Evans, C.R., 1977, Petroleum transformations in reservoirs : Journal of Geochemical Exploration, v. 7, p. 101-153. Monnier, F., Powell , T.G. and Snowdon, L.A., 1983, Qualitative and quantitative aspects of gas generation during maturation of sed imentary organic matter. Examples from Canadian frontier basins, in Bjor~y, M., Albrecht , P., Cornford, C., de Groot, K., Eglinton , G., Galimov, E. , Leythaeuser, D., Pelet, A., Aullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 487-495. Ogunyami, 0., Hesse, A. and Heroux, Y., 1980, Pre-orogenic and synorogenic diagenesis and anchimetamorph ism in Lower Paleozoic continental margin sequences of the northern Appalachians in and around Quebec City, Canada: Bulletin of Canadian Petroleum Geology, v. 28, p. 559-577. Powell, T.G., 1978, An assessment of the hydrocarbon source rock potential ofthe Canad ian Arctic Islands: Geological Survey of Canada , Paper 78-12. Powell, T.G., 1982, Petroleum geochemistry of the Verrill Canyon Formation: a source for Scotian Shelf hydrocarbons: Bulletin of Canadian Petroleum Geology, v. 30, p. 167-179.
c c c c c ( (
(
C C (
C C
C C
C C
C C C C
C C C C
C C
C C
C C
C C E (
C ( (
C
C C
C
C' (
c
c (
c (
(' (
(, (
o
r
c
o ( o c o o (,
o o
r.
( , r
,....
Chemistry and Diagenesis of Organic Matter
Powell, T.G., 1986, Petroleum geochemistry and depositional setting of lacustrine source rocks: Marine and Petroleum Geology, v. 3, p. 200·219. Powell, T.G., Foscolos, A.E., Gunther, P.R. and Snowdon, L.R., 1978, Diagenesis of organic matter and fine clay minerals : a comparative study: Geochimica et Cosmochimica Acta, v. 42, p. 1181-1198. Powell,T.G.and Snowdon, L.R., 1979,Geochemistry of crude oils and condensates from the Scotian Basin, offshore eastern Canada: Bulletin of Canadian Petroleum Geology, v. 27, p. 453·466. Powell, T.G. and Snowdon , L.R. , 1980,Geochemical controls on hydrocarbon gen· eration in Canadian sedimentary basins, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Society of Petroleum Geologists, Memoir 6, p. 421-446. Powell, T.G. and Snowdon, L.R., 1983, A composite hydrocarbon generation model: implications for evaluation of basins for oil and gas: Erdol und KohleErdgas-Petrochemie, v. 36, p. 163-170. Purcell , L.P., Rashid, M.A. and Hardy, LA., 1979, Geochemical characteristics of sedimentary rocks in Scotian Basin : American Association of Petroleum Geologists, Bulletin, v. 63, p. 87-105. Rashid, M.A., 1978, The influence of a salt dome on the diagenesis of organic matter in the Jeanne d'Arc subbasin of the northeast Grand Banks of Newfoundland: Organic Geochemistry, v. 1,p. 67-
77. Rash id, M.A ., 1979, Pristane-phytane ratios in relation to source and diagenesis of ancient sediments from the labrador shelf: Chemical Geology, v. 25, p.109-122. Rashid, M.A., Purcell, L.P. and Hardy, LA., 1980, Source rock potential for oil and gas of the east Newfoundland and labrador shelf areas, in Miall, A.D., ed., Facts and Principles of World Petroleum Occurrence: Canadian Society of Petroleum Geologists, Memoir 6, p. 589-607. Robin, PL, Rouxhet , P.G. and Durand, B., 1977, Characterisation des kerogenes et de leur evolution par spectroscopie infrarouge, in Campos, R. and Goni, J., eds., Advances in Organic Geochemistry 1975:Enadimsa, Madrid, p. 693·716. Rogers, M.A., 1980, Application of organic facies concepts to hydrocarbon source rock evaluation : Tenth World Petroleum Congress, Proceed ings, v. 2, p. 23·30 . Sajg6, C., McEvoy, J., Wolff, G.A. and Horvath, Z.A. , 1986, Influence of temperature and pressure on maturation - I. Preliminary report , in leythaeuser, D. and Rullkotter, D., eds ., Advances in Organic Geochemistry 1985, p.331-338.
Sawatzky, H., George , A.E., Banerjee, R.C ., Sm iley, G.T. and Montgomery, D.S., 1977, Maturation studies on Canadian east coast oils: CANMET Report 77·42. Snowdon, L.R. and Powell, T.G., 1982, Immature oil and condensate : modification of hydrocarbon generation model forterrestrial organic matter : American Association of Petroleum Geologists, Bul· letin , v. 66, p. 775-788. Thomas , B.M ., Mtblle r·Pedersen, P., Whitaker, M.F. and Shaw, N.D., 1985, Organic facies and hydrocarbon dis tributions in the Norweg ian North Sea, in Geochemistry in the Exploration ofthe Norwegian shelf: Norwegian Petroleum Society, p. 3-26. Tissot, B.P., Bard, J.F. and Espitalie, J., 1980, Principal factors controlling the timing of petroleum generation , in Miall , A.D., ed., Facts and Principles of World Petroleum Occurrence : Canadian Society of Petroleum Geologists , Memoir 6, p. 143-152. van Krevelen, D.W., 1961, Coal: Elsevier, Amsterdam, 514 p. van Krevelen, OW. and Schuyer, J., 1957, Coal Science: aspects of coal constltution : Elsevier, Amsterdam, 352 p. Wedeking, K.W.and Hayes, J.M., 1983,Carbonization of Precambrian kerogens, in Bjortby, M., Albrecht, P., Cornford, C., de Groot , K., Eglinton, G., Galimov, E., leythaeuser, D., Pelet, R., Hullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 546-553. Te/ogenesis
Bailey, N.J.L., Jobson, A.M . and Rogers , M.A., 1973a, Bacterial degradation of crude oil: comparison offield and expertmental data : Chemical Geology, v. 11, p.203-221. Bailey, N.J.L. , Krouse, H.R., Evans, C.R . and Rogers, M.A., 1973b, Alteration of crude oil by waters and bacteria: evidence from geochemical and isotope stud ies: American Association of Petroleum Geologists, Bulletin, v. 57, p. 1276-1290. Bailey, N.J.L., Evans, C.R. and Milner, C.W.O., 1974, Applying petroleum geochemistry to search for oil: examples from Western Canada Basin: American Association of Petroleum Geologists , Bulletin, v. 58, p. 2284·2294. Bustin, R.M., 1982, The effect of shearing on the quality of some coals in the southeastern Canadian Cordillera: Canadian Institute of Mining and Metallurgy, Bulletin, v. 75, p. 76-83. Chandra , D., 1962, Reflectance and microstructure of weathered coals: Fuel, v. 41, p.185·193.
203
Crawford, R.J., Spyckerelle, C. and Westlake, D.W.S., 1978,Biodegradation of oil reservoirs , in Strausz, O.P. and lown , E.M., eds ., Oil Sand and Oil Shale Chemistry : Verlag Chemie, New York, p. 163-176. Deroo, G., Tissot, B., McCrossan, R.G. and Der,E , 1974, Geochemistry ofthe heavy oils of Alberta , in Hills, L.V., ed., Oil Sands: fuel of the future: Canadian Society of Petroleum Geologists, Memoir 3, p, 148·167. Deroo, G. and Powell, T.G., 1978, The oil sanddepositsof Alberta : their origin and geochem ical history, in Strausz, O.P. and lown , E.M., eds., Oil Sand and Oil Shale Chemistry: Verlag Chemie, New York, p. 11·32. Evans, C.R., Rogers, M.A. and Bailey, N.J.L., 1971, Evolution and alteration of petroleum in western Canada: Chemical Geology, v. 8, p. 147·170. Krouse, H.R. and McCready, R.G.L., 1979, Reductive reactions in the sulfur cycle, in Trudinger, P.A. and Swaine, D.J., eds., Biogeochemical Cycling of Mineralforming Elements : Elsev ier, Amsterdam, p. 315-368. Milner, C.W.O., Rogers, M.A. and Evans, C.R., 1977, Petroleum transformations in reservoirs: Journal of Geochemical Exploration , v. 7, p. 101-153. Montgomery, D.S., Clugston, D.M., George, A.E., Smiley, G.T. and Sawatzky, H., 1974, Investigation of oils in the Western Canada tar belt, in Hills, L.V., ed., Oil Sands: fuel of the future: Canadian Society of Petroleum Geology, Memoir 3, p.168·183. Rubinstein, I., Strausz, O.P., Spyckerelle, C., Crawford, R.J. and Westlake , D.W.S., 1977, The origin of the oil sand bitumens of Alberta: a chemical and a microbiological simulation study: Geochimica et Cosmochimica Acta , v. 41, p. 1341-1353. Rubinstein , I. and Strausz, O.P., 1978, The biodegradation of crude oils: the origin of the Alberta oil sands, in Strausz, O.P. and lown, E.M., eds., Oil Sand and Oil Shale Chemistry: Verlag Chemie, New York, p. 177·189. Thode, H.G., Wanless, R.K. and Wallouch, R., 1954, The origin of native sulphur deposits from isotope fractionation studies [Texas-louisiana] : Geochimica et Cosmochimica Acta, v. 5, p.286-298. Winters , J.C. and Williams , J.A., 1969, Microbiological alteration of petroleum in the reservoir: Petroleum Transformations in Geologic Environments Symposium, American Chemical Society, Division of Petroleum Chemistry, Paper PETR 86, p. E22-E31.
204
Winters, J.C., Williams, J.A. and Lewan , M.D., 1983, A laboratory study of petroleum generation by hydrous pyrolysis, in Bjonby, M., Albrecht, P.,Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotler, J. and Speers, G. , eds ., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 524-533. Other References Cited in Text
Bily, C. and Dick, J.W.L., 1974, Naturally occurring gas hydrates inthe Mackenzie Delta, N.W.T.: Bulletin of Canadian Petroleum Geology, v. 22, p. 340-352. Choquette, P.W. and Pray, L.C., 1970, Geological nomenclature and classification of porosity in sedimentary carbonates: American Association of Petroleum Geologists, Bullet in, v. 54, p. 207-250. Schmidt. V. and McDonald , D.A., 1979, The role of secondary porosity in the course of sandstone diagenesis, in Scholle, P. and Schluger, P.R., eds., Aspects of Diagenesis : Society of Economic Paleontologists and Mineralogists , Special Publication 26, p. 175-207. Teichmiiller, M. and Teichmiiller, R., 1982, The geological basis of coal formation , in Stach, E., Mackowsky, M.-Th., Teichmuller, M., Taylor,G.H., Chandra , D. and Teichmiiller, R., Stach 's Textbook of Coal Petrology: Borntrager, Berlin , p.5-86.
Originally published in Geoscience Canada v. 11 Number 3 (September 1984) Revised 1988
Diagenesis
Geoscience Canada Reprint Series 4
c c c c c c (
r
c c c c c c c
c c c c c c
c
c ( c c c c c c c c c c e (
c
c C \.
C
(
c c
Determining Levels of Organic Diagenesis
c (
( (
( (
(
o c o o o
o (
c o
c o o
o (
C l
C
Determining Levels of Organic Diagenesis in Sediments and Fossil Fuels A.M. Bustin M.A. Barnes W.C. Barnes Department of Geological Sciences University of British Columbia 6339 Stores Road Vancouver, British Columbia V6T 2B4 INTRODUCTION In an accompanying paper (this volume, p. 189-204), we outlined the types of organic matter, their progressive diagenetic transformations and concomitant chemical reactions and products. Inthis second part, we review some of the more accepted methods of quantifying levels of diagenesis, correlate these different methods with one another, and discuss the application of numerical models in predicting diagenetic state and basin history. Because of both the complexity of kerogen and its importance in the petroleum and coal industries, considerable research has been directed toward quantifying levels of kerogen diagenesis (Durand, 1980) . Organic matter, unlike mineral assemblages, rarely undergoes retrograde diagenes is. The highest level of diagenesis reached is the level observed, regardless of whether the strata have been uplifted and exposedto lowertemperatures andpressures or to different groundwater chemistry. The only exception results from the introduction of oxidizing agents such as molecular oxygen or sulphate, which enable the bacterial degradation and oxidation of organic matter. Inasmuch as the level of organic diagenesis reached has been shown to be almost entirely a product of temperature and time, it follows that if the level of diagenesis is known, the thermal history ofthe
205
strata may be inferred. In the following sections we summarize the literature on quantification and modelling of organic diagenesis. Recent reviews and books that present the techniques in greater detail include Heroux et a/. (1979), Brooks (1980), Durand (1980), Staplin et al. (1982), Bustin et a/. (1985), Robert (1985) , and Tissot and Welte (1985). Some of the more widely applied methods are outlined below. METHODS OF QUANTIFYING ORGANIC DIAGENESIS Both petrographic and chemical methods methods have been used to quantify organic diagenesis (Figure 1;Tables 1 and 2). The petrographic methods have been developed mainly by coal petrologists and include vitrinite reflectance, liptin ite fluorescence, and kerogen colour and translucency. Vitrinite Reflectance. Vitrinite reflectance is not only the most widely used but also the most reliable and precise method of quantifying organic diagenesis . It provides a standard scale against which other diagenetic indicators can be compared. The term " vit rinite reflectance" refers to the amount of light reflected from a polished surface of vitrinite , a common coal maceral. Vitrinite reflectance measurements are made with a reflected light microscope equipped with a stabilized light source, photometer and digital voltmeter for recording. A narrow band filter centered on 546 nm is placed between the objective lens and the photometer. The optical properties required of the . microscope , and necessary accessories, are outlined in Bustin et al. (1983). Reflectance is related to the refractive and absorptive indices of the material by Beer's equation:
(/L - /Lo)2 + /Lk2 (/L + /Lo)2 + /L2k2
where /L and k are the refractive and absorptive indices, respectively, of the vitrinite, and /Lo is the refractive index of the immersion oil. In practice , this relationship is seldom if ever used, but a standard of known reflectance is compared to the vitrinite. With increasing levels of diagenesis (and aromatization) of organic matter, vitrinite reflectance increases regularly from a minimum of 0 .15% (0.2% according to
Middleton, 1982) to greater than 5.5% at a level equivalent to the lower greenschist metamorphic facies. As diagenesis increases, vitrinite becomes progressively more anisotropic and thus the standard deviation of measurements increases. Typically, the mean maximum vitrinite reflectance for each grain is recorded, which requires using polarized light and rotation of the microscope stage. Random (unpolar ized light) reflectance is much quicker to measure and has a high correlation coefficient with maximum reflectance (England and Bustin, 1986). The mean is typically determined from measurements on 50 or 100 particles. Vitrinite is derived from cellulose, lignin and tannins, and is an abundant constituent of continental sedimentary rocks (Stach et al., 1982). Even clastic marine rocks contain sufficient fine dispersed vitrinite (phytoclasts of Bostick, 1979) to enable analysis after concentration by demineralization (using HCI followed by HF) or light liquid methods. Thus, vitrinite reflectance is applicable to most sedimentary rocks, the principal exceptions being coarse grained, permeable rocks in which the percolation of oxygenated groundwater may lead to the oxidation of the vitrin ite, carbonates , evaporites and bitumen impregnated rocks. Vitrinite reflectance offers a number of advantages as a measure of organic diagenesis: (1) it is a broadly accepted and standardized procedure (Ting, 1978; Davis, 1978; ASTM, 1980); (2) vitrinite is sensitive to minor changes in the level of diagenesis and matures in a regular manner; (3) it is a microscopic method, so determinations are always made on the same physical component; (4) vitrinite is a common constituent of many rocks or can readily be concentrated and isolated; and (5) the method is applicable throughout the ent ire range of diagenesis from recent sediments to metamorphic rocks of lower greenschist facies. Some of the limitations of the technique are (1) recycled vitrinite may be present, so that care must be taken to discriminate between separate populations ; (2) oxidation may reduce vitrinite reflectance (Bustin , 1982) or, in rare instances, increase reflectance (Marchioni, 1983); and (3) vitrinite is rare in Devonian and older strata because of the absence of higher plants. Recent
( 206
Diagenesis
studies have demonstrated the potential utility of reflectance measurements of organic constituents in addition to vitrinite. The reflectance of solid bitumen has been described and compared to vitrinite reflectance by Jacob et al. (1985), and Ogunyami et a/. (1980) have described the reflectance of pyrobitumen. The reflectance of graptolites, chitinozoans and acritarchs have also been shown to increase with increased levels of organic diagenesis (Teich muller, 1978; Goodarzi, 1985), although additional studies are required to define their correlat ion with other maturation indices. Fluorescence Microscopy. Fluorescence microscopy is based on the premise that the liptinite component of organic matter, derived from algae ,
I
MET AGENESIS !!:~I
llJ
en
:;!. I An t hracite
.~
;fa
I !" ' I ~1lJ
•IlJ - l
..
-
01
0
.
.
0
"'H
- -ww ..- - -
..
b
en
b
no fluorescence
no fluorescence
dark brown-black
0'
i'
......
n
e.>
a.
b
black 1--01
·1·
:>
e.>
:..
·1·
e.>
e.>
b
0. very dark grey brown
black
...
01:1.
e.>
..
conalderable .catt.r about lin••
...
01 0 ()
.... 0
0
...
01 0 0
0
Ba••d on:
[(C 23-C31) odd +
o
0
~
in
0
0
....0
""
MOISTURE (ash Iree) Hydrogen (daf)
!
~
0
c..0
01
"
VITRINITE RE FLECT (R"m811 ...) 1100
SPORINITE (» FLUOR., max~
700
SPORINITE FLUOR . ., "~ Spectral ~ 3 quotient
:>1lJ
", .. IE
0
:>
-
0'
0
CD",I
I
SPORE COLOUR and TAl
-
in
-
.. e.>
01 0
-...
.. 0
.ca
a. ~c
:::
g:
"'ll
> ..... > 0 :0
~
m m
-l
CONODONT ..... ct~ COLOUR and • 0'0 AL TERATlO N INDEX IEIlJ (CAl) it ..... 0 _ct 0.5 :t ATOMIC "'~ ".,,0 1.0 0 RATIO ct'" 1.1l ct
:0 (J)
-'"
(II
0
~ > r-
.....
<
",ct
pale yellow
0
"'ll -l
0
- cor
.. gg
;: '"!.
V1
'"
HYDROCARBON GENERATION
0
...,
'"
:>
. ..
3
II
(/)
,.
G)
OZ
"'I:)
PYROLYSIS mm
:DO
00
0
1.0 1.0 1.4 1.2 1.0
0
:D
..... .... 0 W
0'"
I:):t
..... mm "" z~
CARBON PREFERENC E INDEX (CPI)
en
-i
:D
-<
Correlation of major organic maturation indices. The correlations shown are based on vitrinite reflectance. (Modified from (1) Barnes et al ., 1984; (2,3,4,5,7) Teichmiiller and Teichmiiller in Stach et aI., 1982, p . 45 ,47; (6) Dow, 1977; (8,9) van Gijzel, 1979 and Teichmiiller and Durand, 1983; (10) Jones and Edison, 1978; (11) Epstein et al ., 1977; (12) Durand and Monin, 1980; (13) Espitalie e t aI., 1977b; (14) Allan and Douglas, 1977). . Figure 1
C
C C C C C C C C
C C C C C C C C C C C C C C C C
CALORIFIC VALUE
01
(
..., w
....
C C C
(
VOLATILE MATTER COAL (def)
ODD
~_. (C25-~3)] !l:
2(C24 -C 32)even
ct
r r
.....
COAL RANK
Early Gas
O'!! ~ 1lJ 0 :>
-
!
0
01
0
0
c0
....
I
:
... H 2O
·1·... 1
Actua l valu •• ahow
C»
~
0
"ll
C
e.>
01
very pale brown I.
brown to dark brown
·1·
a.0
Co
0
--------., .'" O'~ ~CD
dar k brown
0
0
I I
17,000 kJ/kg
.
-
---b
01
;
23,000 kJ/kg
- - - -Oil and Gas - - - - - - -
IlJCD 1lJ_
a
I I I (j I
0
0
!
r-
ifj
01
29,000 kJ/kg
DIAGENETIC LEV EL
'I
0
,
G):E
Dry Gas
c:>
I I
...
rapid shift to higher wavelengths and loss of fluorescent intensity occurs at a leve l of diagenesis equivalent to a vitrinite reflectance of 0.8% to 0.9% (TeichmOller and Durand, 1983). Above a level of diagenesis equivalent to a vitrinite reflectance of 1.2% , fluores cent intensi ties are so low as to be unmeasurable . Quantifying the fluorescent properties of liptinite entails exciting the sample with ultraviolet light and recording the intensities ofthe fluorescent colours emitted in the visible range between 400 and 700 nm. The wavelength of maximum intensity progressively shifts with increasing levels of diagenesis and can be correlated with vitrinite reflectance. An alternative method now widely employed is measurement of the
EOGENESIS
I
High vol. bltum .
36,000 kJ/kg 01
c.>
I
e.>
0
0.
b
o
00 ~.
,.
.
01
I
,. ,=';,~rb-b""m,~ >
Medium volat ile bituminous
36,000 kJ/kg
,0.
spores , pollen , resin and cut in, autofluo resces when irradiated by ultraviolet light. The intensity and fluorescent colours of the Iiptinite, although not well understood , are thought to result from excitation of 11" electrons in conjugated systems (Cook, 1980). Fluorescent intensity and colour vary with the type of Iiptinite and the level of diagenesis. Studies by Ottenjahn et al. (1975), van Gijzel (1979), TeichmOller and Durand (1983) and others have demonstrated that with increasing level of diagenesis the fluorescen t intensity decreases and there is a shift in colour from lower to higher wavelengths. Relative to vitrinite reflectance the changes in fluorescence are somewhat irregular. At a reflecta nce between 0.35% and 0.5% the fluorescence changes rapidly. The most
CATAGENESIS
ge'-
act
Geosc ience Canada Reprint Series 4
(
e c l
c C C C C C
c c
207
Determining Levels of Organic Diagenesis
( (
( ( (
( ( ( ( (
c o o o
o c c
o o (
c
o o c.
r
red/green ratio, Q (Stach et al. , 1982): Q = relative intensity at 650 nm relative intensity at 500 nm The use of fluorescence microscopy to quantify diagenesis has both advantages and disadvantages. In samples where vitrinite is absent, it provides an alternative to vitrinite reflectance, and in samples from rocks of low diagenetic level (R o < 0.3%), the vitrin ite reflectance is low whereas fluorescent intensity is high. In addition, measurements are actually performed on that part of the kerogen that can generate liquid hydrocarbons rather than on the vitrinite, which is not generally considered to be a source of liquid hydrocarbons and which may be reworked. The disadvantages of fluorescence microscopy include (1) the method is only applicable to rocks of low diagenetic level; (2) there are no widely accepted standards; Table 1 Class Optical
Organic Geochemistry
Kerogen Co/ouration. With increasing levels of organic diagenesis the colour of kerogen observed
in transmitted white light changes progress ively. Such studies originated with palynologists, who observed changes in the colour of pollen and spores from translucent and nearly colourless to greenish yellow, yellow, amber, brown and finally to black and opaque with increasing diagenetic level (Gutjahr, 1966; Correia, 1967; Staplin, 1969). An arbitrary numeric scale has been applied to these colour changes (Staplin, 1982) and has been standardized by use of a series of reference samples. One widely adopted scale is the "thermal alteration index" (TAl) proposed by Jones and Edison (1978; Figure 2). The use of kerogen colour as an index of diagenesis has the advantage that only standard palynological preparation techniques and microscopic observation are required. Palynomorphs occur in most fine-grained sedimentary rocks so that recovery of sufficient material for analysis is seldom a problem. The tech-
Relative value of organic diagenetic indicators.
Technique - vitrinite reflectance
Analysis Required - microscopic
Precision - excellent (standard)
- Fluorescence
- microscopic
- Palynomorph color - Conodont color
- microscopic - microscop ic
- moderate (requires further study) - moderate - moderate
- Extractable organic matter
- extraction-gas chromotography - extraction-gas chromotography
- Carbon preference index
- "10 Paraffins , aromatics, & polar compounds - Glc profiles paraffins
- GC-MS profiles biological markers - Light hydrocarbon analysis
(
- Pyrolysis
l
- Ultimate analysis
6
- Proximate analysis Physiochemical - Infrared Spectroscopy
l
(3) for interlaboratory correlation of results corrections must be made for differences in microscope optics and sample background ; (4) with prolonged exposure to ultraviolet radiation, fluorescent intensity may either decline or increase; and (5) the precision of the method is less than that for vitrinite reflectance measurements. For quantitative fluorescence micro scopy, a microscope equipped with a photometer, similar to that described for reflectance measurements, is required. In addition, a stabilized high pressure mercury lamp , ultraviolet filter, barrier filter (to prevent scattered ultraviolet radiation from entering the eye) and variable grating filter (to enable intensity measurements to be made on individual wavelengths) are required (Robert, 1985).
- Electron Paramagnetic Resonance - C,H,S Isotopes - Caloric value
- moderate (?) - moderate
Range of Application - complete range of diagenesis - up to RD = 1.2"10 - up to R. = 2.4"10 - complete range of diagenesis - to above zone of wet gas generation - diagenesis to zone of hydrocarbon generation where CPI-1
- ext. -Iiq. chromatography - ext. -gas chromatography (Glc) - ext. -gas chromatographymass spectrometry - extraction or head-space gas samples -Glc - heating and/or extraction and gas chromotography - separation and various chemical analysis - heating in-different atmospheres - separation and infrared spectrometry - separation and EPR - separation and massspectrometry - separation and calorimetry
- good within stratigraphic sequence see Table 2
- diagenesis to zone of hydrocarbon cracking see Table 2
- poor
- complete range
- moderate at low levels of diagenesis - good at low levels of diagenesis - good at low levels of diagenesis - moderate-insufficient data
- complete range
- insufficient data
- unknown
- insufficient data
- up to Ro
- moderate-good at low levels of diagenes is
- complete range
- complete range - complete range - unknown
-
1.5
208
Diagenesis
Geoscience Canada Reprint Series 4
c Table 2
C
Changes in biological markers with increasing diagenesis.
C --_~ Increasing maturity, as measured by increasing vitrinite reflectance or burial depth
_
( Bi910gical Configuration Steranes Ring alteration:
Geochemical Fossil
5,14,17aH(20R)-sterane (see Fig. 4: 2a-e, h, i)
5aH,14~H,17~H(20R)-sterane2
5,14 ,17aH(20R )-sterane (Fig . 4: 2a-e , h, i)
13~H,17aH(20R)-rearranged sterane
5,14,17aH-sterane (Fig. 4 : 2g) Monoaromatic steranes (Fig. 4 : 9g) (Fig. 4: 4f) Monoaromatic diasteranes (Fig . 4 : 7g)
Monoaromatic steranes (Fig. 4: 9g)2 Triaromatic steranes (Fig. 4 : 10g) (Fig. 4 : 5f) Triaromatic diasteranes2
(Fig. 4: 3a-e, h, i)
(diasterane)I.2.3 (Fig. 4: 6a-e, h, i)
(Fig . 4 : 8g)
Changes in Ratio (with increasing maturation)
(
C
(20R)-sterane: ~~/aa increases (Mackenzie et aI., 1980, 1983; McKirdy et el., 1983; Seifert and Moldowan, 1981, 1986; Philp, 1985) «20R)-rearranged sterane)l5a-sterane increases (Pym et 81., 1975; Seifert and Moldowan, 1978; Seifert et al., 1983) MAI5a-steranes increases (Seifert et aI., 1983) TNMA increases (Seifert et el; 1983) (Mackenzie et al., 1982; 1983a; 1983b) TN(MA + TA) increases (Mackenzie et el., 1981)
C C
Side chain isomerization: 5,14,17aH(20R)-sterane' (Fig. 4: 2a-1)
5,14,17aH(20S)-sterane2 (Fig. 4: 2]-1)
5,14 ,17aH(20R)(24S)-sterane (Fig. 4: 2b, d)
5,14,17aH(20R)(24R)-sterane2.3 (Fig. 4 : 2c , e)
13~H,17aH(20R)-diasterane'
13~':l,17al-!(20S)-diasterane2
(Fig. 4: 6a)
(Fig. 4: 6J)
2OS120R increases in aaa-steranes (Mackenzie et el., 1980,1981,1982, 1983a; McKirdy etal., 1983; Philp , 1985; Seifert and Moldowan, 1978, 1981, 1986; Seifert et aI., 1983) 24R'24S increases in aa(20R)-steranes (Mackenzie et aI., 198O,1981,1983a) 20S120R increases in ~-diasteranes (Mackenzie ot al., 1980; Pym et al., 1975)
Hopanes Ring alteration: 17~H,21 ~H(22R)-hopane
(Fig. 4: 11m-q)
17aH,21 ~H(22R)-hopane (Fig. 4: 12m-q)
~~-hopanes/(total hopanes) decreases to 0 (Mackenzie
et
aI., 1980, 1981; Philp, 1985; Seifert and Moldowan, 1980, 1986)
17~H,21 aH(22R)-moretane
(Rg. 4 : 13m-q) 17aH-22,29,30-trisnorhopane2 (Tm shown in Rg. 4: 14m)
17aH,21 ~H(22R)-hopane (Fig. 4: tzm-q) 18aH-Cv trisnor-neohopane 1 (Ts sHown in Fig . 4: 15m)
~-moretane/a~-hopanedecreases (Mackenzie et al., 1980, 1981; Seifert and Moldowan, 1980 , 1986) C a TmfTs decreases2 (Seifert and Moldowan, 1978, 1980, 1981, 1986; McKirdy et al., 1983; Seifert et 81., 1983)
Side chain isomerization: 17aH,21 ~H(?2R)~hopanefor C31 - C35 (Fig. 4. 12q)
17aH,21 ~H(22S)-hopane (Fig. 4: 12r)
22S/22R increases in a~hopanes (Mackenzie at al., 1980, 1981,1983a; McKirdy etal., 1983; Seifert etal., 1979; Seifert and Moldowan, 1980)
Porphyrins Free base C21- C 32 DPEP (Fig. 5: 17) C2B- C32 Ni-DPEP C27 - CM VO-DPEP'
C a - C 32 ETIO porphyrin (Fig. 5: 18) Ca - C 32 Ni·ETIO C a - CM VO-ETIO
DPEPIETIO decreases (Mackenzie at al., 1980, 1981; Baker and Louda ,1983, 1986; Barwise and Park, 1983) Ni-DPEPlNi-ETIO decreases VO-DPEPNO-ETIO decreases and Cv - CM shifts to C 30 - CSC)
Isoprenoids (6R,10S) pristane (Fig. 5: 22)
(6R,10R) pristane + (6S, 1OS) pristane (Fig. 5: 23)
C'S (3R,7R)-isoprenoid acid
CIS (3S,7R) + (3R,7S) acids
(Fig. 5 : 26) C16 (4R,8R)-isoprenoid acid (Fig. 5: 28)
(Fig . 5: 27) C I I (4S,8R) + (4R,8S) acids (Fig . 5 : 29)
(RS pristane)/(total pristane) decreases to 50% (final mixture: RS :RR :SS = 2: 1: 1) (Mackenzie et al., 1980, 1981, 1983a; Volkman and Maxwell, 1986) RR'(total isoprenoid acids) decreases to 0 as acids isomerize and then decarboxylate to hydrocarbons RR'(total isoprenoid acids) decreases to 0 as acids isomerize and then decarboxylate to hydrocarbons
I Source parameter for oiVsource rock correlation 2 Maturation parameter 3 Migration parameter • VO :0 vanadyl
C C C
C C (
C C C
C C C C C C C C C C C
C C
c
e l C C
c c (
c
c'
Determining Levels of Organic Diagenesis
209
C'
c (
r
c
r c
c c' c o
c c c
nique is applicable to levels of diagenesis equivalent to vitrinite reflectances from 0.15% to about 2.4%, and thus covers the range of petroleum generation and cracking. The major disadvantage of the method is that it is subjective and only semi-quantitative. Palynomorph colour is influenced by particle thickness and initial pigmentation . Standardization of sample processing, microscope light source and reference samples is required. Reworking of kerogen from older rocks or caving of boreholes may complicate the analysis. The only equipment necessary for TAl determination is a transmitted light microscope and a series of reference samples. In an attempt to refine the method, Staplin (1982) suggested the
useof a light source illuminating at 3200 K and a "daylight" filter. An alternative method of determining palynomorph colour is the measurement of their translucency. Translucency measurements are made on specific taxa; at least 30 measurements must be made and averaged (Grayson, 1975). In addition to palynomorphs, other forms of kerogen display progressive changes in colouration with increasing diagenesis. Of particular importance are colour changes in conodonts. Conodonts, although composed mainly of apatite, contain organic matter which changes sequentially from pale yellow to black (Epstein et a/., 1977; Harris, 1979)with increasing levels of diagenesis, in a manner analogous to palyno-
morphs. At the highest temperatures of diagenesis or metamorphism, the colour may change to an opaque white, if the partial pressure of oxygen is high enough to permit oxidation ofthe graphite (Harris, 1979). The progressive colour changes havebeendivided intoeight visually determined levels, giving aconodont alteration index (CAl)scale. CAlcan be correlated with vitrinite reflectances from Ro<0.8% to Ro>3.6% . Conodonts can be recovered from carbonates or shales by acid treatment and heavy liquid or magnetic separation methods (Harris, 1979). The conodont alteration index has the advantage that the method can be used over a wide diagenetic range (up to and including the greenschist metamorphic facies),
o o
r:
r:
Figure 2 Examples ofkerogen, illustrating the progressive change in colour with increasing level ofdiagenesis. Samples were prepared using standard palynological techniques (Hel followed by HF) by Dr. G.E. Rouse. Photographs were taken in transmitted light. (A) Sheet of leaf epidermis, on the left, with stomata and cuticle (colourless; TAl = 1). On the right is a reworked smooth-walled spore (yellow ; TAl = 2). Magnification 600 x . (B) Fragments of leaf, showing cell outlines. TAl = 2; magnification 1000 x . (C) Mass of amorphous granular plant debris, slightly darker than (8). TAl = 2; magnification 1000 x . (D) Smooth-walled spore with yellow-brown colour. TAl = 2.5; magnification 1000 x . (E) Fragment of waxy leaf cuticle, amber-brown. TAl = 2.5; magnification 1000 x . (F) Leaf fragment with brownish-yellow colour. TAl 2.75; magnification 1000 x . (G) Leaf cuticle showing reddish-brown colour. TAl = 3.0; magnification 1000 x . (H) Mesozoic pollen grain with dark brown wall. TAl = 3.8; magnification 1000 x. (I) Same pollen type as (H), but completely carbonized. TAl = 4; magnification 1000 x .
=
c Diagenesis
210
Geoscience Canada Reprint Series 4
C C C C C C C C C C
100 1.5
cr::
~" I
N
I
N ....
II) N
N
~
•
cr. >:r: a. cr. 0 a. a. w a.
1.0
W Z
< a. o :r:
0 5
~.
50
0
~
O-+--=---.----r--r--.----, 0.2
A
0 .4
0
0 .6
1.0
0
B
VI TRINITE REFLECTANCE (%Ro)
VITR INITE REFLECTANCE (%Ro )
C C C C
0.61 0.54 3 .3 3. ·1
Pannonian Basin (Hod) 15Ma f3 = 2 .0
z
0
....
8
-e !:::!
cr. w
3 .0
2 .8
Z
Q
....
-e cr.
N
70· 7'·
W
Z
Z
< a. 0 :r:
< cr.
W .... CIJ
AROMATIZA TION
100
..... ...
I/)
-
20
CP CP
0
~
w cr.
....E E
30 500
'.-,
I
,J:J
::l
I/)
900
40
.... -e
0
50
80 20
40
80
80
PORPHYRINS
100
W
a.
:!:
....W
W
E
:J
cr.
0
,J:J
AROMA TIZA TION
0
1 3. 2
0
C
0
DIHYDROPORPHYRINS 80 20 0 10 0 80 40
:r: ....0..
C
0
C
-
C
North Sea 100Ma
0
0
C C
70·
0
..
G
~
~
W
.....
C C
w
:!: 0
:!: 0
~ I/)
.:
0:
....:!:
Type III Kerogen Bituminous Coal
2.0
X
w
0 ~
C C C
zW
cr. :r: 1.0
.... z
-e W
0.5
C C
:r:
a.
..J
>-
....:r:w
0.0 0 .0
:!:
C C C C C
1.5
w
Z
(
0.5
1.0
1.5
2.0
2.5
MEAN VITRINITE REFLECTANCE (%Ro)
F
Figure 3 Changes in biolog ical markers with increasing diage nesis. (A) Changes in side-chain isomeriz ation at C-22 for the C3 1 a(3-hopanes in Jurassic Kimmeridg e oil shales (Douglas et aI., 1983). (8) Decrease in ratio of DPEP to total porphyrins in gilsonite with increasing reflectance (Barwise and Park, 1983). (C) Changes in side-chain isomer ization atC·20 for the C2 9steranes for Pannon ian Basin sediments relati ve to the shif t from mono- to triaromatic steranes. The Pannonian Basin is assumed to have formed by rapid extension 15 million years ago . Solid dots are observed values, with depths shown in km; unde rlined values were calculated tram Mackenzie and McKenzie (1983; extension tactor ((3) '" 2.0). (0) Changes in side-chain isomerization at G-22 for the C3 2 hopanes from North Sea sediments relative to the shif t from mono- to triaromatic steranes; theoretical curves tor two extens ion factors ((3 '" 1.2 and 1.5) are shown; soliddots are observedvalues; under lined values at inter vals of 5°C were calculated from the model (Mackenzie and McKenzie, 1983). (E) Aromatization of dihydroporphyrins to DPEP with increasing depth and temperature in Black Sea sediments (Baker and Louda, 1983). (F) Methylphenanthrene index first rises and then falls with increasing reflectance; dashed lines represent one standard deviation (Radke and Welte, 1983).
c e (
c c
c
<-
c
c. c
Determining Levels of Organic Diagenesis
211
r ( (
r
and conodonts can be recovered from pre-Devonian rocks where vitr inite is rare or absent. Conodonts are, however, restricted in their stratigraphic range from the Cambrian to the Triassic, and the technique, like palynomorph colouratlon (TAl), lacks the refinement, precision and acceptance of vitrinite reflectance as a diagenetic indicator. GEOCHEMICAL METHODS Organic geochemistry has also been used to determine degree of organic diagenesis. A variety of geochemical methods have been developed to isolate and investigate the soluble and insoluble fractions of organic matter in sediments and sedimentary rocks . Petroleum geochemists (Durand, 1980, p. 24; Waples, 1981, p. 2) define bitumen as the fraction soluble in organic solvents. Kerogen is defined by some as all organic matter that is not soluble in organic solvents (Durand, 1980, p. 27; Waples, 1981, p. 20) and by others as organic matter wh ich is insoluble in both organic solvents and in aqueous bases (Tissot and Welte, 1984). As a great many methods have been developed, only the most widely used wiJI be discussed. Extractable Organic Compounds. The chemical composition of sedimentary organic matter, including petroleum, reflects both source material ("biological marker" compounds) and the level of diagenesis. Ultrasonication or soxhlet extraction with organic solvents (such as chloroform/methanol, benzene/methanol or dichloromethane) is used to isolate extractable organic matter (EOM), the quantity of which is used as a bulk parameter. Because of the complexity of the extract and the large number of organic compounds present, extensive chem ical separations are necessary to isolate and identify the specific organic compounds which are used as molecular parameters . Column chromatography, thinlayer chromatography and adduction as clathrates are used prior to analysis by high performance liquid chromatography (HPLC), gas-liquid chromatography (GLC) or on-line gas chromatographymass spectrometry (GC-MS). The small
amounts of the compounds present, coupled with the extensive chemical separation procedures employed, can lead to a loss of material, oxidation or a chemical change in the individual compounds. Phthalates, which are used as plasticizers, and naphthenic compounds, from greases used in driJIing, are common contam inants (Powell , 1978). Biological markers are especially useful in oil-source rock correlation. Although the isolation and ident ification of individual organic compounds are both more expensive and time consuming than measurement of vitrinite reflectance, they too can be used for assessment of the degree of diagenesis , particularly at very low and intermediate levels of maturity. Biological Markers. The terms "biological marker" (Eglinton, 1969) and "geochemical fossil " (Tissot and Welte, 1984) have been used to refer to organic compounds in sediments which reflect the chemistry of the contributing organism, the effect of depositional environment and the effect of burial and subsequent maturation. The carbon skeletons ofthe compou nds show little or no alteration from the precursor forms; however, the effects of eogenesis and catagenesis are reflected in changes which may include (1)an alteration in stereochemistry; (2) a loss or change in functional groups; (3) reduction and saturation; or (4) aromatization . Changes in the distribution and structure of biological markers are used (1) to correlate oils with their source rocks; (2)to evaluate the degree of maturation; and (3) to identify migration or biodegradation that may have occurred during the diagenetic conversion of biological lipids to petroleum (Meinschein, 1983; see references in Table 2). Aecent in-depth reviews of biomarkers, their precursors and diagenesis, and their use as indicators of source, maturation, migration and biodegradation in petroleum exploration have been published by Johns (1986), Sohn (1986), Mackenzie (1984) and Philp (1985: includes 373 mass spectra indexed by molecular weight and formula). Biomarkers in coal are reviewed by Chaffee et at. (1986).
Generally, thermal alteration of biological markers increases with burial depth and with increasing vitrin ite reflectance values (Figures 3a and 3b). Commonly used are a combination of parameters determined for compounds whose structures differ in thermal stability. This permits a more sensitive discrimination of degrees of thermal maturation between samples in a stratigraphic sequence than is possible using reflectance data alone . Because of their widespread occurrence, structural specificity and stability, steranes, hopanes, porphyrins, and isoprenoids are commonly used as biological markers. Eogenetic and catagenetic changes that result in ring alteration and side chain isomerization are summarized in Table 2 and Figures 4, 5 and 6. During early eogenesis, sterols [Figure 4: compound 1,side chain (A) = a-e, h, i] are converted by biologically mediated processes in the sediment to the corresponding stanols and sterenes, which form steranes (Figure 4: 2, A = a-e, h, i); the aaa-steranes retain their original biological configuration [5, 14, 17aH (20R)] in both the ring and side chain (de Leeuw and Baas, 1986). Depending on the biological precursor, there can be some variation in the stereochemistry of the alkyl side chain, particularly in marine sources (Scheuer, 1973, 1978), but the more common stereochem istry has the R configuration at C-20 (Figure 4: A = a-f) and a mixture of R (Figure 4: A = c.e) and S (Figure 4: A = b, d) at C-24. The Rand S stereochemistry is defined in Figure 7. These are the dominant forms in immature sediments. With increasing maturation , there is a shift to thermally more stable geochemical fossils. Isomerization of the alkyl side chain occurs with the shift from S to R at C-24, followed by R to S at C-20. Alteration of the ring occurs with the formation of 5aH, 14,BH, 17,BH steranes (Figure 4: 3). Aromatization occurs with the formation of monoaromatics (Figure 4: 4, 7, 9) followed by triaromatics (Figure 4: 5, 8, 10). An increase in rearranged steranes [diasteranes (Figure 4: 6)] relative to nonrearranged steranes (Figure 4: 2) reflects both source and SUbsequent mi-
Figure 4 (next page) Eogenetic and catagenetic changes in steranes andhopanes. Note the use ofboth the Roman font A and the italic font Ron this diagram. A refers to alkylside chains, whereas Rand S refer to the stereochemistry ofthe alkyl side chains. See Figure 7 for definition of Rand S. The ring systems lie approximately in the plane of the page. The tilde symbol (- ) indicates the point of attachment of A groups to the ring. Solid triangles indicate groups above the plane of the rings (in front of the page); broken triangles indicate groups below the plane of the rings (behind the page) . If an "An is not shown, the group is a methyl. A solid circle indicates a (3hydrogen above the plane of the rings; an open circle indicates an ex hydrogen below the plane of the rings ; ,r' indicates undefined stereochemistry.
..
I\)
I\)
STERANES
Biological Configuration
R
2.
1.
SaH, 14, 17{3H
EOGENESIS
~ HO
•
~
R = a-e, h, i
STEROL
R=
CAT AGENESIS ....0
•
~o
Ring
'''~.'''~c. ~,
./ ~. ':'
C2~~ /,~.
.
~-91-,
~. ~~~~1-;..
120R 241 a.
C 26
R
'?~
~,~
C 27
Geochemical Fossil
R
R
6.
'" ~ R R e. ~ S R k.
•
CATAGENESIS
a
yJyg. ~.
~.~~,~( 1
I h.
R CH3
CH3
..
r'V"ll. H
HOPANES
10.
o
~. (I)
C 27
R=H
C 26
CH 3
n.
C2~
-CH 2-CH 3
c.
~
11.
m.
(I)
UI
(ii '
C 30
-CHCCH 3 )2
HOPANE p.
q. 22R
Ring
R=m-q
- CHI
n C II _ S
n : 0 -
4
CATAGENESts
~
~(CH2)n
~t-C)
CD'
::J
g
"to
13.
r.
~~(CH1)
r
'O\'O
f<,+-
~
14 . R
III
~
- CH J
::0
~ C 27 17aH
Tm
S' .... ~ ii)'
UI ~
nf'nnrn
n~nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
r. (
Determining Levels of Organ ic Diagenesis
213
r
c
r:
PORPHYRINS
( (
c (
..
EOGENESIS
CATAGENESIS
•
25· - 40· C
( 18 . DPEP FREE BASE C 28 - 3 2
ETIO FREE BASE C 28 - 3 2
ICHELATE
CHLOROPHYLL
•
Ni 2+ - - -..... ~
Free C 28 -32 Ni DPEP
Free C 28 - 3 2 Ni ETJO
•
- C 32 - 36 Ni DPEP and ETIO
S2~
CATAGENESIS Rele ••• ot:
zo!!
IU.
~~
8s.ci
hound DPEP - - - - . . .
EOGENESIS
c ~
Bound C 27 - 34 ETIO
... -
5~
C 32 - 36 Ni DPEP and ETIO
' ;.
Bound VO DPEP - - - -... Bound C 27 - 34 VO ETIO
:< -
C 30 - SO VO DPEP and ETtO
'1I 20.
PHYTANE
z o
.
-e
N
'"
IU
:I
o
'"
RS : RR : $S
2 : 1 : 1
22. PRISTANE
COOH
COOH
25 .
24. PHYTANIC ACID
z
..
o (
-COOH
N
B
IU~
:I
~coo"-+I J ('6 ISOPRENOID ACID
28 .
I
:
~
ISOPRENOID HYDROCARBONS
0::
Cl:
~COOH
Figure 5 Eogenetic and ca tagene tic changes in porphyrins and isopreno ids . Symbols are defined in Figure 4.
214
Diagenesis
Geoscience Canada Reprint Series 4
c c (
gration and maturation effects (McKirdy et et., 1983; Seifert and Moldowan , 1983, 1986); diasteranes are thought to arise from acid catalyzed ionic rearrangement of aa-steranes or sterols and occur more commonly in oil shales than in carbonate-sourced oils (McKirdy et af., 1983). Table 2 and Figure 4 summarize some of the parameters commonly used in oil-source rock correlations and as measures of thermal maturation , migration and biodegradation. Seifert and Moldowan (1981)have defined a maturity index for non-migrated oils (Biomarker I) and a migration index (Biomarker II) from a first order kinetic plot of the side chain isomerization in C2 9 steranes (Figure 4: 2112f) relative to ring alteration (Figure 4: 3f12f). The hopanoic acids and hydrocarbons occur in cyanobacteria (bluegreen algae) and bacteria and are commonly used as indicators of microbial sources in sediments (Ourisson et a/., 1979). The ISIS-hopanes (17,sH, 21,sH; Figure 4: 11, R = m through q) with the R configuration at C-22 (R = q) are the more common biological isomers; lesser amounts of (ja-moretanes occur (17(jH, 21aH; Figure 4: 13, m through q). With increasing thermal maturity, both the (ja-moretanes and the (j(j-hopanes undergo ring alteration and form the a(j-hopanes (Figure 4: 12, m through q); alteration of (j(j-hopanes to a(j-hopanes occurs prior to the zone of oil generation (Mackenzie et et., 1980; Seifert and Moldowan, 1980, 1986). A decrease in the Tm/Ts ratio can be used as a measure of increasing maturity for oils or source rocks derived from a common organic source (Seifert and Moldowan, 1978,1986; Philp, 1985). The 17aH-C2 7 hopane (Tm, Figure 4: 14) shares the same precursors as other hopanes and may be an acid-catalyzed diagenetic
product; it is thermally less stable than Ts, a neohopane. The 18aH-C 2 7 trisnorneohopane (Ts, Figure 4: 15) has a different precursor and is more source specific. Thermal isomerization of the alkyl side chain in C 3 1 to C3 5 hopanes occurs with the shift in stereochemistry from R (Figure 4: R = q) to S (R = r) at C-22. The sensitivity of the SIR ratio for the a(j-C 3 1 hopanes to changes in vitrinite reflectance in the region Ro 0.4 through 0.6 is shown in Figure 3: a. Chemical , microbial and biodegrading processes, differences in biogenic sources, and migration can affect the concentrations of biomarkers commonly used to assess maturity. As a result, these indicators of maturity must be used with care; several different maturity ratios are necessary to confirm a trend. An increase in a(j-hopanes is usually taken as a measure of increasing maturity; however, 17aH, 21(jH extended hopanes with 22R in the side chain have been found in lichens, fungi and sphaghum peat (Philp, 1985). Increases in monoaromatic steranes, diasteranes, and (j(j-steranes relativetothe aa-steranes are all used as measures of increasing maturity. The ~3,5 steradienes can give rise to "1\' -ring monoaromatic steranes in shallow or immature sediments and have formed "C"-ring monoaromatics under laboratory simulation; ~5 sterenes may give rise to either "C"-ring monoaromatics or, under acid conditions, to the rearranged diasterenes. Recent hypersaline sediments have been shown to contain relatively uncommon sterols with a double bond in the ~7 position. Ten Haven et a/. (1985, 1986) have suggested that the alteration and hydrogenation of these ~ 7 sterols and of extended hop-17(21)-enes may lead to the formation of side-chain isomerized
a(j(j -steranes and extended a(jhopanes in immature hypersaline sediments, provid ing another route in addition to the thermal maturation of aaasteranes or (j(j-hopanes. Biodegradation removes regular aa-steranes preferentially to diasteranes and diasteranes containing a side chain with 20S preferentially to those with a side chain containing 20R . The source rock is depleted and migrated oil is enriched in the ,s(j-(20R)-steranes relative to the aa-(20R)-steranes due to the greater mobility of the (j,s-(20R)-steranes. Mackenzie and McKenzie (1983) and Mackenzie (1984) have used side-chain isomerization in steranes and hopanes, and the shift from monoaromatic to triaromatic steranes, to model the thermal, subs idence and uplift histories of stretched basins. Their ideas are discussed in the section on numerical methods. Early in eogenesis, porphyrins arising from chlorophyll (Figure 5: 16) lose magnesium and the phytol side chain (Figure 5: 19); phytol in turn may be oxidized to the corresponding acid (Figure 5: 24), reduced to the C20 isoprenoid phytane (Figure 5: 20) or shortened to pristane (Figure 5: 21) in the guts of zooplankton. In each case, the original biological configuration is retained. With thermal maturation, both pristane and phytane isomerize from RS (Figure 5: 20 and 22) to a mixture ofthe RS, RR and SS isomers (Figure 5: 21 and 23). Similar thermal isomerization of the biological configuration to an isomeric mixture
H3C,~H
A
R2-- R,
R Clockwise
H 3C,~H
A
R,-R 2 S
Anliclockwise
C (
C
C ( (
C
C C C C (
C
C C C C C C C C C C C € C C C C
PHENANTHRENES
CATAGENESIS
•
Figure 6
(
Catagenetic changes in phenanthrenes.
Figure 7 Rand S stereochemistry for a tetrahedral carbon atom with four different groups attached. Alkyl groups R, and R2/ie in theplane of the page; themethyl group lies in front of the page and hydrogen behind. Group R, is larger than group R2> which in turn is larger than a methyl group. Size is defined by the total atomic number of the attached groups. Thus , -COOH > -CH(CH:J2 > -CH2CH3 -CH3 • In R stereochemistry, group size decreases in a clockwise direction; in S stereochemistry, it decreases in an anticlockwise direction.
c
e
c
c c c c (
C
c
c c
Determining Levels of Organic Diagenesis
215
(
c (
(
(
( (
C'
o o o o o o o c o ( o o c c o o o o
(
c (
8 (
'-..-
,-
occurs in the isoprenoid acids (Figure 5: 24 goes to 25; 26 goes to 27; 28 goes to 29); with increasing burial depth, the C'5 and C'6 acids (Figure 5: 26 and 28) disappear as they decarboxylate to the corresponding hydrocarbons. Other sources of isoprenoids include carotenoid pigments, archaebacterial cell wall lipids and bacteriochlorophyll (Volkman and Maxwell, 1986). After the successive loss of magnesium and of phytol and other functional groups from the porphyrin ring, hydrogen is lost and the free base is aromatized under anoxic conditions to DPEP (Figure 5: 17) with a range in carbon number from C2 8 to C3 2 . The aromatization of the porphyrin ring system with the formation of DPEP takes place in the temperature range of 30°-60°C for samples taken from the 8IackSea(Figure3E). 8akerand Louda (1983, 1986) also suggest that rechelation with Ni to form Ni-DPEP occurs between 30 0C and 60°C. DPEP is thermally altered to the ETIO form (Figure 5: 18) with the loss of ring V. As a result, increasing thermal maturation is evident and can be measured by decreases in the DPEP/ETIO ratio , which occurfirstforthe free bases and then for their Ni chelates. The decrease in the DPEP/ETIO ratio for porphyrins isolated from gilsonite is shown in Figure 38. 8arwise and Park (1983) suggest porphyrins are a sensitive index to maturation in the region Ro 0.4 to 0.6, and can be used in algal rich samples that commonly lack vitrinite particles. Finally, kerogen-bound DPEP and its vanadyl chelates are released , and subsequently altered to the ETIO form; with both the addition and loss of alkyl side chains, a series of C3 0 to Csaporphyrins is produced. A methylphenanthrene index (MPI) has been developed to measure increases in coal rank and thermal maturation in sediments (Rad ke et a/. , 1982a,b; Radke and Welte, 1983). Methylphenanthrenes (MP) and naphthalenes, and dimethylphenanthrenes are extracted and separated by liquid chromatography. The quantity and distribution of methyl homologues of phenanthrene (P) and naphthalene in the C'5+ soluble organic fraction are strongly controlled by changes in rank. At Ro = 0.9, the Iiptinite fluorescence ratio (0) shows a sharp increase; a corresponding shift occurs in the structure of the
methyl homologues. Methyl groups adjacent to the ring junction [1-methylphenanthrene (Figure 6: 30) and 9ethylphenanthrene (Figure 6: 31)] are thermally less stable than methyl groups in non-adjacent positions [2-methylphenanthrene (Figure 6: 32) and 3-methylphenanthrene (Figure 6: 33)], which are SUbject to less steric hindrance. TwoMPI indexes are defined: MPI 1 = 1.5
(2-MP
+ 3-MP)
(P + 1-MP + 9-MP) MPI 2 = 3
(2-MP) (P + 1·MP + 9-MP)
A strong positive correlation with reflectance is observed for MPI 1, which increases within the oil window, but then decreases abruptly with the onset of the wet gas zone below (Figure 3: F). A "calculated reflectance" for coal and kerogen extracts can be derived from the MPI (Radke and Welte, 1983): ForRo <1.35%: Rc = 0.6(MPI1) + 0.4 For Ro > 1.35%: a, = -0.6(MPI1) + 2.3 The "calculated reflectance" shows a linear correlation with Ro and makes it possible to compare the thermal maturity of sediments with their extractable organic content. Pyrolysis. Pyrolytic methods have recently been developed that mimic in the laboratory the natural generation of petroleum. Saxby et a/. (1986)heated samples of a Permian torbanite (algal kerogen) and a Tertiary lignite under nitrogen in closed tubes, raising the temperature 1°C per weel< from an in itial temperature of 100°C over a period of six years. The samples had previously been extracted with chloroform and demineralized with HCI and HF, but, unlike hydrous pyrolysis experiments, they were dry. Samples were removed and analyzed at one-year intervals. Vitr inite reflectances in the lignite samples increased in a regular manner from initial values of about 0.3 to 2.8 at the final temperature of 400°C. NMR spectra on the same samples were typical of bituminous coals after heating for two years (200°C) through four years (300°C), but resembled anthracites in the samples heated to higher temperatures. The lignite samples did not generate significant quantities of oil at any temperature, but oil
resembling natural crudes and containing a smooth distribution of n-alkanes up to about n-C 3 0 was produced from the torbanite, with 35% by weight being converted to oil by 300°C. As with hydrous pyrolysis, olefins were not produced, and it was established that water is not required forthe formation of petroleum under laboratory conditions if very slow rates of heating are used. An important result of this study is that products very similar to natural crude oils were produced in only a few years, but at temperatures considerably higher than for natural crudes. Thus, very long periods of time may not be as important in organic maturation as was formerly thought, a conclusion supported by the recent development of oil fields with Pleistocene source rocks. Hydrous Pyrolysis. In hydrous pyrolysis, petroleum generation is simulated in the laboratory by heating source rocks in a liquid water phase for 1-5 days at temperatures less than the critical point of water (374°C);in a closed system pressuresof 20-35 MPa develop (Winters et et., 1983;Lewan, 1985).Three fractions are recovered: (1) an expelled oil phase floating on the water surface; (2) a benzene-methanol soluble bitumen recovered by soxhlet extraction of the crushed sample and (3) kerogen isolated after HF-HCI demineralization of the crushed sample. The expelled oil contains a smooth distribution of n-alkanes with the production of few alkenes or volatiles. Based on the relative proportions of kerogen, bitumen and expelled oil generated during hydrous pyrolysis, four stages of petroleum generation have been defined (Lewan, 1985): 1. Pre-oil stage: the original immature sample prior to heating; 2. Incipient oil stage: bitumen increases to a maximum, with a corresponding decrease in kerogen, but little production of expelled oil; 3. Primary oil generation: bitumen decreases; expelled oil reaches a maximum; 4. Post-oil stage: cracking and production of pyrobitumen . The end of the incipient oil stage and the stage of primary oil generation are defined by a maximum followed by a decrease in the amounts of bitumen and expelled oil, respectively.Hydrous pyrolysis is currently a very active subject.
216
Diagenesis
Geoscience Canada Reprint Series 4
c c (
There are many recent investigations assessing the advantages and limitations of hydrous pyrolysis as a means of modelling the effects of thermal maturation on steroids (Lewan et et., 1986), stable carbon isotopes (Lewan, 1983), exchange reactions with heavy water (Hoering, 1984), biomarker isomerization and mineral catalysis (Eglinton et et., 1986),and on the generation of aromatic compounds (Rowland et al., 1986), hopanoids (McEvoy and Giger, 1986) and cycloalkanes (Fowler et et., 1986). Total Extractable Organic Matter. The ratio of the total amount of extractable organic matter (EOM) to total organic carbon (TOC) is related to the level of diagenesis attained by sedimentary rocks (Tissot and Welte , 1984). Although the EOM/TOC ratio is high in immature surface sediments, it decreases rapidly during early eogenesis as humic substances form , then increases during catagenesis with the formation of hydrocarbons from the cracking of kerogen. At higher diagenetic levels, the loss of aliphatic, alicyclic and low molecular weight aromatics and the formation of the relatively less soluble large, multi-ring aromatic compounds results in the progressive decline of EOM/TOC . The method is limited by the dependence of the EOM/TOC ratio not only on the level of diagenesis, but also on the type of organic matter present. In addition, the ratio will decrease during catagenesis if hydrocarbons migrate out of their source rocks. Tocompensate for some of these problems, the ratios of aromatics to saturated hydrocarbons and of aromatics to total EOM have been used (Albrecht et a/., 1976); both ratios increase with increasing levels of diagenesis, but are affected by the type of kerogen present. These ratios are most significant if used in a sequence of strata of constant kerogen type, where values from different stratigraphic levels, or from a given stratigraphic unit down into or across a basin, can be compared. Carbon Preference Index. Carbon preference index (CPt) is the ratio of odd carbon number to even carbon number n-alkanes in extractable organics. During eogenesis CPI is controlled by the alkane ratio present in precursor organisms (Barnes and Barnes,1978). While most higher plants
and algae have a large odd CPI, some marine sponges, freshwater aquatic plants, ferns, lycopods, fungi, yeasts and bacteria havesmalloddcarbon preferences. Hydrocarbons from the cuticles of higher plants rangefrom n-C 2 3 to n-C 3S' with n-C 29 or n-C a, dominant; ferns, lycopods, fungi, yeasts and bacteria commonly have a broader carbon range. Algae have majorcomponents at n-C'71 n-C 17: , (heptadecane with one double bond), n-C,s or n-C'3 (Gelpi et al., 1970). Fungi have n-C 2 9 dominant, while yeasts and bacteria show both unimodal and bimodal distributions with both n-C,s and n-C 2 7 as dominant hydrocarbons (Weete, 1974; Bird and Lynch, 1974). During eogenesis,the CPI approaches unity in hydrocarbons altered by the aerobic and anaerobic decay of algae (Cranwell, 1976). Cranwell suggested that both an odd-even index near unity and an increase in hydrocarbons in the region of n-C,s to n-C 2 6 are indicators of microbial alteration. Brooks et a/., (1976) found that extraction method can influence carbon preference index. Hydrocarbons released by a simple solvent extraction at a nearly neutral pH had a large even CPI, reflecting higher plant sources, whereas those bound into microbial cell walls and released only by acid hydrolysis of sediments have much smaller odd-carbon preferences, reflecting bacterial alteration. As catagenesis progresses thermal degradation of kerogen generates alkanes with no oddeven preference, so the CPI of extractable hydrocarbons approaches unity. Thus, CPI may be used as an indicator of maturation (Figure 1). Light Hydrocarbon Analysis. The abundance and type of light hydrocarbons, defined as having between one and eight carbon atoms, have been found to reflect the level of diagenesis. As living organisms produce few light hydrocarbons except for methane (Oremland et al. , 1982),most hydrocarbons in the C2 through Cs range reflect the diagenetic level of the kerogen in the rocks. Light hydrocarbons are commonly divided into two groups by petroleum chemists : C, through C4 gases from head space gases or drill cuttings, and gasoline range (Cs through Cs) compounds from cores and cuttings. At low diagenetic levels (eogenesis), the light gases consist mainly of biogenic methane and gasoline range hydrocar-
bans are minor or absent (Leythaeuser et al., 1979). At the diagenetic level of oil generation, the light gases contain a great many compounds in the C2 through C4 range, whereas at the metagenic level methane is again the dominant gas (Figure 1) and gasoline range hydrocarbons are minor. Bailey et al. (1974) used such methods in their evaluation of maturation in western Canada. The ratio of C2 through C4 gasesto total gases (C, through C4 ) has also been found to indicate diagenetic level and was used in maturation studies in the Canadian Arctic (Snowdonand Roy, 1975; Monnier et aI., 1983); a number of other ratio methods for gasoline range hydrocarbons have also been used to assess maturity and migration (Thompson, 1979; Snowdon and Powell,1982) . Light gases are extracted from cuttings or core samples by heating, treatment with acid or mechanical disaggregation in a blender, followed by collection of the gases from the head space above the sample and analysis by gas chromatography. As a measure of diagenesis, light hydrocarbon analysis has the advantages that both extraction and analysis are rapid, requiring only simple techniques, and can be applied over a wide diagenetic range. The method lacks precision, however, and is influenced by migration. The best resultsare obtained when a sequence of strata are examined. GEOCHEMISTRY OF KEROGEN Pyrolysis. Pyrolysis involves heating a sample at a pre-selected rate in an inert atmosphere and monitoring the type and amount of gases evolved. A number of methods of determining diagenetic level from pyrolysis products have been proposed. One such method is based on the Rock-Eval~ instrument (Espitalie et al. , 1977a,b, 1985; Peters, 1986), which is essentially a furnace with an on-line gas chromatograph. The analysis simulates maturation (diagenesis) by progressively heating rock samples to temperatures up to 550°C. Three types of gases have been recognized, based on the temperatures at which they are evolved: (S,) hydrocarbons already present in the sample, which are evolved below 300°C; (S2) hydrocarbons generated by thermal cracking
c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c e c ( (
l <.
c C
Determining Levels of Organic Diagenesis
217
l (
C
c' ( (
c (
c
o c
o o o o o o
between 300°C and 550°C; and (S3) CO2 (Figure 8). The level of diagenesis is obtained from the Sl/S2 ratio and the temperature at which maximum generation of S2occurs. A plot, analogous to a "van Krevelen" diagram (atomic H/C versus O/C; Barnes et et. , 1984), is obtained byusingthe "hydrogen index" (S2/TOC) versus the "oxygen index" (S3/TOC) and can be used to estimate both the kerogen type and the degree of maturation. The production index, defined as S/(Sl + SJ, increases with depth as the amount of free hydrocarbons (Sl) increases . The production index thus increases with the level of diagenesis. The advantages of the Rock-Eval method are that no sample preparation is required and that level of diagenesis, oil and gas potential and hydrocarbons already present are all determined. Disadvantages include problems associated with mixed kerogen types of varying chemistry and with the presence of carbonates, which evolve CO2 during pyrolysis. Ultimate Analysis. "Ultimate" analysis involves the determination of the elemental C, Hand 0
(
T max (increases with degree of diagenesis)
c (
l
o (
o o o o c
content of rocks , commonly by use of a C-H-N analyzer. H/C and O/C ratios are plotted against one another to give a "van Krevelen" diagram which can then be used to determine both kerogen type and level of diagenesis. During eogenesis, oxygen is lost as CO2 and H20 , which results in a decrease in the O/C ratio with little corresponding change in the H/C ratio; catagenesis results in a rapid loss of hydrogen in the form of hydrocarbons, causing a decrease in the HIC ratio. The advantage of ultimate analysis is that it provides both a measure of diagenetic level and kerogen type. The disadvantages are that kerogen must be separated before analysis by using HCI and HCIIHF digestion of mineral matter and, less commonly, by specific gravity or hand-picking methods . Mixed kerogen types also give results that are difficult to interpret, and the method lacks accuracy. Infrared Spectroscopy. The infrared spectroscopy of kerogen enables the quantification of the main functional groups present. As the level of diagenesis increases, there is a progressive loss of carboxyl and carbonyl
Generated Hydrocarbons
Distilled Hydrocarbons
CO 2 from Organic Matter
C' INCREASING TEMPERATURE
l
82
TOC = Hydrogen Index
8 1/8 1 + 8 2
=
Production Potential
83
TOC = Oxygen Index
~ ex Level 82
of Diagenesis
Figure 8 Rock-Eva" pyrolysis products. The three gas pulses shown can be used to characterize the kerogen type present, the level of organic diagenesis and the type of hydrocarbons the kerogen will yield. (Modified from Espitalie at aI., 1977a; and Miller, 1982).
groups, a loss of saturated hydrocarbons, the formation and removal of aromatic C-H groups, and the evolution of hydroxyl and ether groups (Rouxhet et al., 1980). Quantification of these groups gives a measure of the level of diagenesis attained and of kerogen type , as the initial proportions of these groups is a function of the precursors organisms. The advantage of the method is that sample preparation, apart from isolation of the kerogen, is simple. The kerogen is separated and then pressed into a KBr pellet which is used in the infrared spectrophotometer. Only a few hundred milligrams of kerogen are required for the analysis. The disadvantages of the method are that it is only semi-quantitative and is not widely used. Electron Paramagnetic Resonance. Electron paramagnetic resonance (EPR) studies of kerogen assess the presence of free radicals (polyatomic moieties with unpaired electrons). All kerogens give EPRspectra that depend on the type of kerogen and the level of diagenesis. With increased diagenetic level the number of free radicals first increases to a maximum and then declines because of recombinations that occur as the mean distance between the radical groups decreases. The diagenetic level atwhich free radicalsare at a maximum varies with the type of kerogen. An advantage of EPR is that it is non-destructive, but the technique has several disadvantages. The kerogen must be separated carefully from the rest of the rock; impurities such astrace metals affect the spectra. Also, the type of kerogen affects its EPR spectrum , and the EPR signal decreasesover long periods of geologic time and in postmature rocks as free radicals combine (Brooks, 1980). Carbon Isotope Studies. The stable carbon isotope composition reflects both the initial c5 13C value of the organic source material (DeNiro, 1983) and the isotopic fractionation that takes place during diagenesis (Hayes, 1983). Higher plants utilizing the C3 pathway (Calvin cycle), which constitute the bulk of terrestrial organic matter, have mean c5 13C values of -27 permil (%0), with a spread from about -23%0 to -34%0 (Galimov, 1985; Schidlowski, 1986). Most organic matter formed in marine and lacustrine environments derives from the eukaryotic algae, which have
c
Diagenesis
218
013C values which mainly range from around -18%0 to -31%0, with a mean of about -22%0, significantly heavier than that of higher plants. The cyanobacteria, which are the principal formers of stromatolites, are even heavier, with a mean of around -16%0. Heterotrophs are isotopically similar to the organic substrates they use. Many lignites show considerable variation in the isotopic composition of their macrolithotypes, but, at higher coalification levels, coals lack this variability, suggesting that isotopic homogenization takes place during coalification (Schoell, 1984). Petroleum source rocks commonly show considerable regional isotopic variation, which is generally thought to reflect variations in terrestrial and aquatic inputs. During thermal maturation there is a preferential breaking of 12C_12C bonds with respect to 12C_13C bonds, so that the remaining kerogen is enriched in 13C whereas the expelled liquids and gases have more 12C. As thermal maturation continues, the reservoir of isotopically light methyl groups is depleted and more of the heavier methyl groups are subject to cracking. As a result, the 013C of methane generated at higher levels of diagenesis increases, and the difference in isotopic composition between the gases and the original kerogen decreases. The carbon isotope trend toward lighter (more negative) values for gas relative to oil, which in turn is lighter than source rock kerogen, is widely used for oil-source rock correlation. Proximate Analysis. Proximate analysis determines the fixed carbon, ash, volatile and moisture contents of kerogen containing rocks. The technique has been used most extensively for the analysis of coals, but can also be used on kerogen separated from other rocks. With an increasing level of diagenesis there is an increase in fixed carbon and a corresponding decrease in volatiles and moisture in the organic matter. Proximate analysis involves heating the sample to a low temperature (about 110°C) to determine moisture, heating to intermediate temperatures in an inert atmosphere to determine volatile content, followed by heating to high temperatures in a stream of oxygen to oxidize all remaining organic matter so that the residual ash may be determined. Fixed
magnetic resonance, differential thermal analysis, fluorescence spectra of extractable organics and thermogravimetric analysis,havebeenusedin maturation studies. The applicability of these methods has not been fully evaluated, however, and they are not widely used. Correlation between diagenetic indicators. The correlation between the main diagenetic indicators discussed in this review is shown in Table 1 and Figure 1. All relationships are based on established or estimated correlations with vitrinite reflectance.
carbon is determined by difference. The method has been used for many years for analysis of coal and is widely accepted, well documented and standardized (ASTM 03172-73, 1979). The disadvantages of the method are that the technique is useful only for organic rich rocks, such as coals, and for kerogen separates. Kerogen type affects the results and,while this is not a significant factor in coals, which are generally composed mainly of type 3 kerogen, other organic rocks commonly contain a mixture of kerogen types, leading to results that are difficult to interpret. Caloric Value. The caloric (heating) value of organic matter increases at higher levels of diagenesis in response to decreased moisture and volatile content and increased relative amounts of fixed carbon. Caloric value is generally determined with a bomb calorimeter; it is commonly used as a technique for determining coal rank, but can be used to determine the diagenetic level of any organic rich rock. The method has the advantage that it is standardized (ASTM 02015-77, 1974) and has been widely applied to coals. Disadvantages of the method include the necessity for correcting for mineral matter in the rocks and that kerogen type influences caloric value, a significant factor for rocks other than coal. Other Methods. A number of other techniques, such as nuclear magnetic resonance, proton VITRINITE
.
Ro ci '"
-o
ci
Geoscience Canada Reprint Series 4
PREDICTIVE DIAGENETIC MODELS
Models to predict the levels of organic diagenesis attained with depth were initially developed to determine the occurrence of the liquid (oil) window in time and space (Landes, 1967) or the degree of coalification (Karweil, 1956). Subsequent studies have shown that the level of organic diagenesis, when interpreted in the context of burial history, can be used to interpret the thermal history of strata. Early studies assigned particular temperatures to the liquid window. It is now widely recognized (but see Neruchev in Bostick, 1979) that the thermal history (both temperature and time) must be considered. Numerical models and graphical techniques have been developed to predict organic maturation from an assumed thermal history or to determine the thermal history from measured levels of organic diagenesis.
r-,
CD
ci
ci
3
:
240
2~
120
~
1..1::/
v: V
I I l/
DC ~
« ffi
/1
~
w
._-
~ 80
-",
.......... 40
io--'..... !-'
V
V
o 0 ..; W"i
-0" %RoMa
0
N
o
0 0
00
M
...
on
-IL
Vh
~
160
e,
0
.,;
I
200
....
N
SCALE IN MILLION YEARS
I
cP
0
~
I.;
V
/ .....11 V /' .....1.1
V
t%:~
V
~I':
v.
:,...-
i.;' ~ ~~
V
~
i> ~
",0
// 1/ ~ I
~
/
~
/ ),1 /
1/ ..... '" V
. / ...... ~ ./
......
~
o Figure 9 Correlation between temperature, time and level of organic diagenesis (mean maximum vitrinite reflectance, RaJ. The dashed line shows an example in which organic matter exposed to temperatures of 100°C for 10 Ma would have an expected level of maturation equivalent to an Ro of 0.56%. (Modified from Bustin et aI., 1977).
c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c e c (
l
l C L
c./ C C' ( . r:
(
(
( ,~
J
0 0 0 0
e,..., ~
0
Determining Levels of Organic Diagenesis
Models of organic diagenesis fall into two broad groups, numerical models and comparative models. Numerical models are based on reaction kinetics, whereas comparative models utilize established relationships between diagenetic level and known thermal history in one set of strata to predict either the thermal history or levels of diagenesis in another set of strata where the diagenetic levels have been measured or which are known to have had a similar thermal history. Comparative models can also be used to extrapolate measured levels of diagenesis to higher or lower strata. Numerical Models. Graphical methods. Models developed to predict levels of organic diagenesis during a predetermined or assumed thermal history are based on first-order reaction kinetics and thus approximate the Arrhenius equation (Huck and Karweil, 1955):
K
0 0 r: L-
r-..
0 0 D 0 r:
'-" r:
'-'
0
C r r
, \...... r:
..... r:
'---' ,'-, ''-,
"-
= A e(·E/RT),
where K = reaction rate, A = a frequency factor, E = activation energy, R = gas constant, and T = absolute temperature. Huck and Karweil (1955) assumed an activation energy of 35 kJ 'mol-' for the entire range of coalification (for a complete discussion and solution of the Arrhenius equation, see Welte and Yukler, 1980). Karweil (1956) combined empirical rank data (diagenetic level) and theoretical calculations to relate the level of coal diagenesis to thermal history (Figure 9). This relationship between rank and thermal history has SUbsequently been modified (Bostick, 1973)to better fit established relationships. Bostick (1973)suggested that the activation energy used by Huck and Karweil (1955) predicts too fast a reaction rate for coalification at low levels of diagenesis (Ra < 0.6) and too slow a rate at higher levels. Bostick (1973, 1979) adapted Karweil's (1956) method to regions with varied thermal histories. Castano and Sparks (1974) and Hood et al. (1975) proposed a graphical relationship between thermal history and an arbitrary scale of diagenesis to define a "Level of Organic Maturation" (LOM). They further suggested the use of an "effective heating time", which they defined as the time during which the rock is within 15°C of its maximum temperature, in predicting the level of dia-
219
genesis. Shibaoka and Bennett (1977) used a graphical technique similar to that of Hood et al. (1975), but related thermal history directly to vitrinite reflectance rather than to LOM. Numerical calculations. From a consideration of first-order reaction kinetics (Arrhenius equation) and from empirical data on activation energies, Lopatin (1971; see also Lopatin and Bostick, 1974; Waples, 1980; Cohen, 1981) showed thatthe rate of maturation doubles with every 1Q°C increase in temperature (an exponential relation ship) but increases only linearly with time. Lopatin's method calculates interval maturation by integrating the length of time spent by the stratum in each 1QoC temperature interval. The total maturation (diagenesis) is the sum of the interval maturation: "mu
TTl =
2:
(dT.){r·),
."",
where TTl = time-temperature index; nm a x = highest 10°C interval; nm ln = lowest 1Q°C interval; ~Tn = time (Ma) spent by strata in each 1QoC interval; and r = 2, assuming rank doubles for each 1O°Crise in temperature. Lopatin (1971) set n = 1 for the 1QO°C to 110 0C interval. At lower temperatures, n is zero or negative, and, at higher temperatures, n is greater than 1. Lopatin's method enables periods of uplift and subsidence to be considered, so that the entire thermal history is used in calculating the maximum level of diagenesis attained by a stratigraphic horizon that was deposited at time = O. For calculations by computer, the following generalized relationship is more suitable (McKenzie, 1981):
.
TTl =
f
2(T(11 -'0.5Y'O dt
10
where T(t) is temperature (0C) as a function of time (t) from the time of deposition (to)to the present (tp ) . Using the data presented by Waples (1980), the following relationship between TTl and mean maximum vitrinite reflectance may be defined: Ro = -0.10528 log TTl + 0.20647
(log TTI)2 + 0.5011 (r2 = 0.99) . Ro is not defined by this equation if TTl values are less than 3. An example of Lopatin's method is shown in Figure 10.
Many recent studies have attempted to refine the numerical models of organic diagenesis by consideration of such factors as kerogen type and by use of empirically determined activation energies and appropriate frequency factors (Tissot and Espltalie, 1975; Tissot and Welte, 1985). These approaches facilitate evaluation of not only the degree of diagenesis, but also the proportion of the kerogen which was transformed to hydrocarbons. Tissot and Welte (1985) havedescribed the model in detail, and Chiaramonte and Novelli (1986) have applied the techniques to modelling maturation of source rocks in northern Italy. Using numerical models, it is possible to predict the approximate level of maturation from the thermal history of the strata (Figure 10). Predicting the thermal history from a single maturation level is difficult becausemany combinations of time and temperature will yield the same maturation level. Where measured maturation levels are available through a stratigraphic succession, however,they can be compared to calculated maturation gradients (rather than a single point) to infer the thermal history. This method involves calculating maturation gradients using a range of geothermal gradients or depths of burial and matching the calculated curves with the measured curves (Figure 11). Using this method, it is possible to estimate paleogeothermal gradients (or paleo-heat flow) irrespective of the absolute level of maturation. Such a method has been used by Bustin (1984) to estimate paleo-heat flow in the Bowser Basin of northern British Columbia and by England and Bustin (1986b)to interpret the effects of thrust faulting on organic maturation in the southeastern Canadian Cordlllera. Other Canadian studies utilizing Lopatin-type models include Kalkreuth and McMechan (1984)and Hughes and Cameron (1985) in the Cordillera, and Bustin (1985) in the Arctic Archipelago. Kerogenisthe product of manychemical reactions, which act at different rates on a complex mixture of organic compounds. Its reflectance (and fluorescence) is a physical property which develops asthe sum of all these temperature-dependent reactions, and thus is affected by differences in source materials and their reaction rates as well as by temperature history (Jones et
( 220
Diagenesis
Geoscience Canada Reprint Series 4
(
c aI., 1984). As a result, calibration of vitrinite reflectance scales is empirical, and differs from basin to basin. By selecting specific chemical reactions as a basis for modelling, Mackenzie and McKenzie (1983) developed a model which reflects kinetic effects of thermal diagenesis in stretched basins, but is independent of variations in depositional history or in organic precursors. An intramolecular chemical change, which is temperature dependent, but independent of the source or of other chemical reactions, is measured for single chemical compounds. Side-chain isomerization in steranes and hopanes and the shift from monoaromatic (M] to triaromatic steranes [T] , are used to monitorthe subsidence and thermal history of stretched basins in the North Sea and the Pannonian Basin (Hungary), and to model subsidence and subsequent uplift in the Lower Saxony Basin (West Germany) and the Paris Basin (France). Three reactions which occur prior to and during the early stages of oil formation were selected: (1)the shift from R to S at C-20 in the C2 9 sterane (Figure 4: 2f goes to 21); (2) the shift from R to S at C-22 in the 0I{3-C 3 2 hopane (Figure 4: 12q goes to 12r); and (3) the shift from the 501 or 5{3 C29 monoaromatic (Figure 4: 4f) to the C2 8 triaromatic steranes (Figure 4: 5f) . These three reactions fit the following assumptions reasonably well: (1) initial concentrations of thermal products are small or negligible in immature sediments; (2) thermal alteration spans the oil window; (3) a single reactant gives rise to a single product in a unimolecu-
.e
~
G
oS
,-,_-
:I: l-
e,
W
Q
A
.-
.--
lar first-order reaction which follows Arrhenius' Law; (4) the reactions are free of matrix effects; if mineral catalysis occurs, the number of active sites is assumed to be large relative to the concentration of reactant and the reaction is pseudo first order. While the sidechain isomerizations are reversible, the aromatization is irreversible. Activation energies (E), and frequency factors (A) are derived from Arrhenius plots of the logarithm of the rate constants versus the reciprocal of the borehole temperatures. Rate constants (k) and 0, the ratio of the forward to reverse reaction for side-chain isomerizations are derived from the ratio of products to reactants measured by GC·MS for samples from the North Sea and the Pannonian Basin . Kinetic parameters are estimated as: A = 1.8 X 101 4 s-1, E = 200 kJ· mot-t for the aromatization; A = 0.006s-1,E = 91 kJ'moJ-1,O = 1.174for side-cnaln isomerization in steranes; A = 0.016s", E = 91 kJ . rnol-', 0 = 1.564 for side-chain isomerization in hopanes . Temperature was calculated from equations forthe variation in temperature in a subsiding basin (McKenzie, 1978; Mackenzie and McKenzie, 1983) with
heat flux , time , depth, the quartz-elay content and estimates of the extension factor ({3) from sediment thickness (Sclater and Christie, 1980; Wood, 1981; McKenzie, 1981). Kinetic parameters for the chemical reactions are independent of basin history while calculated temperatures are dependent on basin age and extension. These values are used to generate theoretical curves for isomerization relative to aromatization from equations for the change in rates with temperature and time in a subsiding basin. The fit of observed values for the extent of reaction and borehole temperatures or depth are shown for the North Sea (Figure 3D) and the Pannonian Basin of Hungary (Figure 3C). Plots of hopane relative to sterane isomerization are independent of the basin thermal history. Mackenzie and McKenzie's model has the advantage that it is based on kinetics that measure rates for known single reactions rather than on the unknown multiple reactions that cause changes in vitrinite reflectance. By using pairs of reactions, it is possible to determine uniquely the entire time-temperature history of a basin, inclUding not only its development, but also subsequent uplift after extens ion. As a
B Interval (oC) 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 110.0 100.0 90.0 80.0 70.0 60.0
30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 110.0 100.0 90.0 80.0 70.0 60.0 50.0
....
Interval TTl
2-8 2-7
2-6 2-5
2-4 2-3 2-2 2-1 2-0 2-0 2-1 2-2 2-3 2-4 2-5
0.04 0.08 0.16 0.31 0.63 1.25 2.50 5.00 10.00 10.00 5.00 2.50 1.25 0.63 0.31
Cumulative TTl 0.0 0.1 0.3 0.6 1.2 2.5 5.0 10.0 20.0 30.0 35.0 37.5 38.7 39.3 39.6
Time (Ma B.P.)
Hydrocarbon Generation
140.00 130.00 120.00 110.00 100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.0
Total Maturation = 39.6 TTl
-. Log,. TTl
Equivalent Reflectance
C C C C C C C
C C C C C C C C C
C C C C C C
C C C
C (
C
C (
c
e (
= 0.86 Ro %
Figure 10 Lopatin's method of predicting organic maturation. (A) shows the burial history of a hypothetical rock that was deposited at 140Ma, buried to 3000m by 60 Ma, followed by uplift to 1000m by time = O. (B) (right) shows the method ofincrem.entaHy calculating maturation using Lopatin 's formula (see text) for the burial history shown in (A). With a constant geothermal gradient of 30°C per km and a surface temperature of 20°C, the present maturation of the rock would be 39. 6 TTl or O. 86% Ro• If the oil window is assumed to occur between 0.500/0and 1.35% Ro (Dow, 1977), then the rock entered the oil window at a depth of about 2600m at 72 Ma and still lies within the oil window.
c (
C L
C C
(
c c C'
r.....
C' C
c r c c c c
221
Determining Levels of Organic Diagenesis
result, the time of hydrocarbon generation may be determined and compared with the time of formation of suitable traps, creating a powerful tool for petroleum exploration. Comparative Methods. Comparative methods have been used to establish either levels of organic diagenesis if thermal history is known or assumed, or thermal history if the level of organic diagenesis is known. These methods are based on the premise that a given thermal history will result in particular levels of diagenesis, and have been used mainly to compare geothermal gradients in different areas. Hacquebard and Donaldson (1974), for example, compared the coalification
o
(diagenetic) gradients of strata from the Canadian Rocky Mountains and Foothills with the gradients established in the Peel area of the Netherlands (see also Nurkowski, 1984). Bustin et al. (1977) noted the similarity of coalification gradients of Tertiary strata from the Canadian Arctic with those of the Tejon area of California reported by Castano and Sparks (1974). Because of the similar burial histories of strata in both areas, Bustin et al. (1977)assumed that the paleogeothermal gradient in the Arctic was similar to the known geothermal gradient of the Tejon area. Similar methods were used by Buntebarth (1978) in the upper Rhine graben and by Barker and Elders (1979) in the
REFLECTANCE
o o
0.2
0.3
0.4
0.6
0.8 1.0
1.5
2.0
o o o r:
o c r: o D·
o o
1
E
.::J:.
J:
2
l-
e.. W
Increasing Gradient (calculated)
C
r
0,
cr ,
3
4....L.
--:-_--..:L.-
-.L._ _
~
Figure 11 M~thod of predicting paleogeothermal gradients or paleo-heat flow based on organic maturation. Maturation gradients are calculated iteratively using a Lopatin-type method with varying temperatures, subsidence rates, etc. until the slope (in metres per log cycle) of the calculated gradient is the same as the measured gradient. In the example shown, the pa/eogeothermal gradient would be between 20°C and 30°C. The method requires knowledge of the burial history of the strata.
Cerro Prieto geothermal field. In both studies, known temperatures and coalification levels from boreholes were used to develop general relationships between thermal history and coalification levels so that the thermal histories of adjacent areas can be interpreted. On the assumption that changes in level of diagenesis with depth are predictable, a number of studies have extrapolated measured diagenetic gradients to higher or lower stratigraphic levels. This method was applied by Hacquebard (1977)to calculate the amount of overburden that had been removed by erosion from the Mannville coals in southern Alberta. Hacquebard extrapolated the moisture content ofthe coals at the surface to that prior to burial, based on assumed changes in moisture content with depth. SUMMARY Quantification and modelling of organic diagnesis have proven to be useful in assessing petroleum source rocks, coal rank and thermal history of strata. A large number of methods have been utilized to quantify organic diagenesis. Almost all involve detailed and generally expensive analytical procedures and require isolation or at least concentration of either the kerogen or bitumen prior to analysis. At present, the most widely accepted and precise method methods are petrographic: vitrinite reflectance, fluorescence and kerogen colour. Of these, fluorescence microscopy and kerogen colour are only useful for a limited range of diagenesis (less than 1.2% Ra ) and vitrinite reflectance, although useful for the entire range of diagenesis, has its greatest utility only in Carboniferous and younger strata. A large number of geochemical methods have been developed.Ofthese ultimate analysis, proximate analysis and caloric value are best suited to coal, where the relative abundanceof kerogen is high. One of the most useful methods of evaluating finely dispersed organic matter is by the pyrolysis method proposed by Espatalie et al. (1977a,b) which provides an indication of diagenetic level and kerogen type without requiring concentration of the organics. A number of geochemical techniques, such as carbon preference index (CPI), total extractable organics and light hydrocarbon analysis, are influenced by the type of organic matter present and by
c 222
Diagenesis
Geoscience Canada Reprint Series 4
(
C migration and thus are most useful for resolving the relative degree of diagenesis in astratigraphic succession . While isolation and identification of individual organic compounds are more expensive and time consuming than measurement of vitrinite reflectance, biological markers can be used for a more sensitive assessment of diagenesis, especially at low and intermediate levels of maturity. In addition, they are now commonly used in oil-source rock correlation and as measures of thermal maturation, migration and biodegradation. Geochemical techniques which have not been totally assessed as to their ut ility as diagenetic indicators include infrared spectroscopy, electron paramagnetic resonance and nuclear magnetic resonance. Such methods require highly specialized equipment that is not readily available outside major research centres. Models of organic diagenesis have been developed to predict levels of organic diagenesis and have also been used to interpret thermal history of strata where the level of diagenesis is known. Although the accuracy of the models has not been fully tested , it is now possible to predict diagenetic levels and thermal history which compare favourably with measured values. One of the most useful methods is that initially proposed by Lopatin (1971), which predicts that the level of diagenesis increases exponentially with temperature and linearly with time .
REFERENCES General Barnes, M.A., Barnes, W.C. and Bustin, A.M ., 1984, Chemistry and evolution of organic matter: Geoscience Canada, v. 11, p. 103-114. Brooks, J., 1980, ed., Organic Maturation Studies and Fossil Fuel Exploration: Academic Press, Toronto, 441 p. 22 papers covering organic maturation and kerogen description. Brooks, J. and Welte, D., 1984, eds ., Advances in Petroleum Chemistry, v. 1: Academ ic Press, Toronto, 344 p. Bustin, A.M. , Cameron, A.A., Grieve, D.A. and Kalkreuth, W.O., 1985, Coal Petrology, its principles, methods and applications: Geological Association of Can ada , Short Course Notes Volume 3, 273 p. A description of methods and applications of kerogen analysis.
Durand , B., 1980, ed., Kerogen : insoluble organic matter from sedimentary rocks : Editions Technip, Paris . 15papers on the petrography, chemistry and diagenesis of kerogen. Heroux, Y. , Chagnon, A. and Bertrand, R., 1979,Compilation and correlation of major thermal maturation indicators: American Association of Petroleum Geologists, Bulletin, v. 63, p. 2128-2144. Johns, R.B., 1986, ed., Biological Markers in the Sedimentary Record: Elsevier, New York, 364 p. 10 papers on precursors, diagenesis and use with fossil fuels. Meinschein, W.G., 1983, ed., Organic Geochemistry of Contemporaneous and Ancient Sediments: Society of Economic Paleontologists and Mineralogists, Great Lakes Section , Short Course notes, 143 p. Philp, A.P., 1985, Fossil Fuel Biomarkers: applications and spectra: Elsevier, New York, 294 p. Includes 373 mass spectra . Robert, P., 1985, Historie geothermique et diagenese organique: Bulletin des Centres Recherches Exploration-Production Elf-Aquitaine, Memoire 8, 346 p. Sohn, M.L. , 1986, ed., Organic Marine Geochemistry: American Chemical Society, Symposium Series 305, 428 p. Stach, E., Mackowsky, M.-Th., Teichmiiller, M., Taylor, G.H., Chandra, D. and Teichmuller, A. , 1982, Stach 's Textbook of Coal Petrology: Gebriider Borntrager, Berlin, 535 p. Staplin, F.L., Dow, W.G., Milner, C.W.O., O'Conner, 0.1., Pocock, SA, van Gijzel , J.P., Welte, D.H. and Yukler, M.A ., 1982, How to Assess Organic Maturation and Paleotemperatures: Society of Economic Paleontologists and Mineralogists, Short Course No. 7, 289 p. Tissot, B.P.and Welte, D.H., 1984, Petroleum Formation and Occurrence, Second Edition: Springer-Verlag, New York, 699 p. Waples, D., 1981, Organic Geochemistry for Exploration Geologists: Burgess Publishing Company, 151 p.
Vitrinite, Graptolite, Chitinozoan and Bitumen Reflectance American Society for Testing and Materials (ASTM) , 1980, Microscopical determination of the reflectance of the organic components in a polished specimen of coal: American Society for Testing and Materials, Philadelphia, Standard 02798-79, 4 p. Bostick, N.H ., 1979, Microscopic measurement of the level of catagenesis of solid organic matterin sedimentary rocks to aid exploration for petroleum and to determine former burial temperaturesa review, in Scholle, P. and Schluger, P. A.,
eds. , Aspects of Diagenesis : Society of Econom ic Paleontologists and Mineralogists, Spec ial Publication No. 26, p.17-43 . Bustin, A.M., 1982, The effect of shearing on the quality of some coals in the southeastern Canadian Cordillera: Canadian Institute of Mining and Metallurgy, Bulletin, v. 75, p. 76-83 . Davis, A., 1978, The reflectance of coal, in Karr, C., Jr., Analytical Methods for Coal and Coal Products: Academic Press, New York, v. 1, p. 27-81. England, T.D.J. and Bustin, A.M., 1986a, Thermal maturation of the Western Canadian Sedimentary Basin south of the Red Deer River - I: Alberta Plains: Bulletin of Canadian Petroleum Geology, v. 34, p. 71-90. Goodarzi, F., 1985, Reflected light microscopy of chitinozoan fragments: Marine and Petroleum Geology, v. 2, p. 72-78. Jacob , H., Hiltmann, W., Wehner, H., Raschka, H. and Weiser, T., 1985, Disperse, feste Erdolbitumina als Migrations - und Maturitatsind ikatoren im Rahmender Erdol-/Erdgas Prospektion . Eine Modellstudie in NW-Deutschland: Deutsche Gesellschaft fur Mineralolaissenschatt und Kohlechemie E.V., Projekt 267. Jones,J .M., Davis, A., Cook, A.C., Murchinson, D.G. and Scott, E., 1984, Provincialism and correlation between some properties of vitrinite: Journal of Coal Geology, v. 3, p. 315·331. Marchioni, D.L., 1983, The detection of weathering in coal by petrographic, rheolog ic and chemical methods : International Journal of Coal Geology, v. 2, p.231-259. Ogunyami, 0., Hesse, R. and Heroux, Y., 1980, Pre-orogenic and synorogenic diagenesis and anchimetamorphism in Lower Paleozoic continental margin sequences of the northern Appalachians in and around Quebec;City, Canada : Bulletin of Canadian Petroleum Geology, v. 28, p. 559-577. Teichmuller, M., 1978, Nachweis von Graptolithen periderm in geschieferten Gesteinen mit Hille Kohlenpetrologisches methoden: Neues Jahrbuch fUr Mineralogie, Geologie und Palaeontologie, Monatshefte 7, p. 430-477.
Fluorescence Cook, A.C., 1980, Optical techniques forthe examination of organic matter in oil shales, in Cook, A.C. and Kantsler, A., eds., Oil Shale Petrology Workshop: Keiraville Kop iers, Wollangong, Australia, p. 1-15.
C C
C C C C C C C C C C C C C C
C C
C C C C C C C C C C C
C
e (
C
C C l
C (
C
c c
r
Determining Levels of Organic Diagenesis
223
r>
'--
c
c c C· (
C
c
c c
Ottenjann, K., TeichmQller, M. and Wolf, M., 1975, Spectral fluorescence measurements of sporinites in reflected light and their applicability for coalification studies, in Alpern, B., ed., Petrographie organique et potential petrollsr: C.N.R.S., Paris, p. 49-65. Robert, P., 1981, Classification of organic matter by means of fluorescence: application to hydrocarbon source rocks: International Journal of Coal Geology, v. 1, p. 101-137. TeichmQller, M. and Durand, B., 1983, Fluorescence microscopical rank studies on Iiptinites and vitrinites in peat and coals, and comparison with results ofthe RockEval pyrolysis: International Journal of Coal Petrology, v. 2, p. 197-320. van Gijzel, P., 1979, Manual of the techniques and some geological applications of fluorescence microscopy: American Association of Stratigraphic Palynologists, 12th Annual Meeting, Dallas, 55 p.
Conodont Colour
Epstein, A.G., Epstein, J.B. and Harris, L.D., 1977, Conodont Color Alterationan index to metamorphism: United States Geological Survey, Professional Paper 995, 27 p. Harris, A.G., 1979, Conodont color alteration, an organo-mineral metamorphic index, and its application to Appalachian Basin geology, in Scholle, P. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication No. 26, p. 3-16. Perry, WJ., Wardlaw, B.R., Bostick, N.H. and Maughan, K., 1983, Structure, burial history, and petroleum potential of frontal thrust belt and adjacent foreland, southwest Montana: American Association of Petroleum Geologists, Bulletin, v. 67, p. 725-743. Geochemical Methods
Kerogen Colour
o O·
o
r
o o r. C 8
Correia, M., 1967, Relations possibles entre l'etat de conservation des elements figures de la matiere organique (micro fossils palynoplanctonologiques) et l'exlstence de gisement d'hydrocarbons: Revue de I'lnstitute Franc;:ais du Petrole, v. 22, p. 1283-1306. Grayson, J.F., 1975,Relationship of palynomorph translucency to carbon and hydrocarbons in clastic sediments, in Alpern, B., ed., Petrographie de la matiere organique des sediments, relations avec la paleotemperature et Ie potentiel petroller: C.N.R.S., Paris, p.261-273. Gutjahr, C.C.M., 1966,Carbonization measurements of pollen-grains and spores and their application: Leidse Geologische Mededelingen, v. 38, p. 1-29. Jones, R.W. and Edison, T.A., 1978, Microscopic observations of kerogen related to geochemical parameters with emphasis on thermal maturation, in Oltz, D.F., ed., Low Temperature Metamorphism of Kerogen and Clay Minerals: Society of Economic Paleontologists and Mineralogists, Pacific Section, Los Angeles, p. 1-12. Staplin, F.L., 1969, Sedimentary organic matter, organic metamorphism, and oil and gas occurrences: Bulletin of Canadian Petroleum Geology, v. 17, p. 47-56. Staplin, F.L., 1982, Determination of thermal alteration index from color of exinite (pollen, spores), in Staplin, F.L., Dow, WG., Milner, C.W.D., O'Conner, D.I., Pocock, S.A.J., van Gijzel, P., Welte, D.H. and Yukler, M.A., eds., How to Assess Organic Maturation and Paleotemperatures: Society of Economic Paleontologists and Mineralogists, Short Course 7, p. 7-11.
Albrecht, P., Vandenbroucke, M. and Mandengue, M., 1976,Geochemical studies on the organic matter from the Douala Basin (Cameroon) - I. Evolution of the extractable organic matter and the formation of petroleum: Geochimica et Cosmochimica Acta, v. 40, p. 791-799. American Society for Testing and Materials (ASTM), 1974, Test for gross calorific value of solid fuels by adiabatic bomb calorimeter: American Society for Testing and Materials, Philadelphia, Standard D2015-77. American Society for Testing and Materials (ASTM), 1979, Proximate analysis of . coal and coke: American Society for Testing and Materials, Philadelphia, Standard D3172-73. Allan, J. and Douglas, A.G., 1977, Variations in the content and distribution of n-alkanes in a series of carboniferous vitrinites and sporinites of bituminous rank: Geochimica et Cosmochimica Acta, v. 41, p. 1223-1230. Bailey, N.J.L., Evans, C.R. and Miller, C.WD., 1974, Applying petroleum geochemistry to search forcil: example from Western Canada Basin: American Association of Petroleum Geologists, Bulletin, v. 58, p. 2284-2294. Baker, E.w. and Louda, J.W, 1983,Thermal aspects of chlorophyll geochemistry, in Bjortby, M., Albrecht, P.,Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 401-421. Baker, E.w. and Louda, J.w., 1986, Porphyrins in the geological record, in Johns, R.B., ed., Biological Markers in the Sedimentary Record: Elsevier, New York, p.125-225.
Barker, C., 1974, Pyrolysis techniques for source-rock evaluation: American Association of Petroleum Geologists, Bulletin, v. 58, p. 2284-2294. Barnes, M.A. and Barnes, WC., 1978, Organic compounds in lake sediments, in Lerman, A., ed., Lakes: chemistry, geology and physics: Springer, New York, p. 127-152. Barwise, A.J.G. and Park, P.J.D., 1983,Petroporphyrin fingerprinting as a geochemical marker, in Bjortby, M., Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Hullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 668-674. Bird, C.W and Lynch, J.M., 1974, Formation of hydrocarbons by micro-organisms: Chemical Society Reviews, v.3, p.309-328. Brooks, P.W., Eglinton, G., Gaskell, S.J., McHugh, D.J., Maxwell, J.R. and Philp, R.P., 1976, Lipids of Recent sediments, part I: Straight-chain hydrocarbons and carboxylic acids of some temperate lacustrine and sub-tropical lagoonall tidal flat sediments: Chemical Geology, v. 18, p. 21-38. Chaffee, A.L., Hoover, D.S., Johns, R.B. and Schweighardt, F.K., 1986, Biological markers extractable from coal, in Johns, R.B., ed., 1986, Biological Markers in the Sedimentary Record: Elsevier, New York, p. 311-345. Cranwell, P.A., 1976, Decomposition of aquatic biota and sediment formation: compounds in detritus resulting from microbial attack on the alga Ceratium hirundinella: Freshwater Biology, v. 6, p.41-48. de Leeuw, J.W and Baas, M., 1986, Earlystage diagenesis of steroids, in Johns, R.B., ed., Biological Markers in the Sedimentary Record: Elsevier, New York, p. 101-123. Deniro, M.J., 1983,Distributionofthestable isotopes of carbon, nitrogen, oxygen and hydrogen among plants, in Meinschein, W.G., ed., Organic Geochemistry of Contemporaneous and Ancient Sediments: Society of Economic Paleontologists and Mineralogists, Great Lakes Section, Short Course notes, p. 4-1-4-27. Douglas, A.G., Hall, P.B. and Solli, H., 1983, Comparative organic geochemistry of some European oil shales, in Miknis, F.P. and McKay, J.F., eds., Geochemistry and Chemistry of Oil Shales: American Chemical Society, Symposium Series 230, p. 59-84. Durand, B. and Monin, J.C., 1980, Elemental analysisof kerogens(C, H, 0, N, S, Fe),in Durand, B., ed., Kerogen: insoluble organic matter from sedimentary rocks: Editions Technip, Paris, p. 113-142.
c 224
Diagenesis
Geoscience Canada Reprint Series 4
(
C Eglinton, G., 1969, Organic geochemistry: the organic chemist 's approach,in Eglinton, G. and Murphy, M.T.J., eds., 1969, Organic Geochemistry: methods and results: Springer-Verlag, New York, p.828. Eglinton, T.I., Rowland, S.J., Curtis, C.D. and Douglas, A.G., 1986, Kerogen-minerai reactions at raised temperatures in the presence of water: Organic Geochemistry, v. 10, p. 1041-1052. Espitalie, J., Deroo, G. and Marguis, F., 1985, La pyrolyse Rock-Eval et ses applications: Revue de I'lnstitute Francals du Petrole, v. 40, p. 563-579. Espitalie, J., Laporte, J.L., Madec, M., Marquis, F., Leplat, P., Paulet, J. and Boutefeu, A., 1977a, Methode rapide de characterisation des roches meres, de leur potential petrolier et de leur degre d'evolutlon: Revue de I'lnstitut Francais du Petrole, v. 32, p. 23-42. Espitalie, J., Madec, M., Tissot , B., Mennig, J.J. and Leplat, P., 1977b, Source rock characterization method for petroleum exploration: 9th Offshore Technology Conference, Proceedings, OTC#2935, p.439-444. Fowler,M.G., Abolins, P.and Douglas, A.G., 1986, Monocyclic alkanes in Ordovician organic matter: Organic Geochemistry, v. 10, p. 815-823. Galimov, E.M., 1985, The Biological Fractionation of Isotopes : Academic Press, Toronto, 261 p. Gelpi, E., Schneider, H., Mann, J. and Oro. J., 1970, Hydrocarbons of geochemical significance in microscopic algae: Phytochemistry, v. 9, p. 603-612. Hayes, J.M., 1983, Practice and principles of isotopic measurements inorganic geochemistry, in Meinschein, W.G., ed., Organic Geochemistry of Contemporaneous and Ancient Sediments : Society of Economic Paleontologists and Mineralogists, Great Lakes Section , Short Course notes. p. 5-1-5-31. Haering, T.C., 1984, Thermal reactions of kerogen with added water, heavy water and pure organic substances : Organic Geochemistry, v. 5, p. 267-278. Huang, W.-Y., and Meinschein, W.G., 1979, Sterols as ecological indicators: Geochimica et Cosmochimica Acta, v. 43, p.739-745. Lewan. M.D. , 1983, Effects on thermal maturation on stable organic carbon isotopes as deterined by hydrous pyrolysis of Woodford Shale: Geochimica et Cosmochimica Acta, v. 47, p. 1473-1474. Lewan, M.D., 1985,Evaluation of petroleum generation by hydrous pyrolysis experimentation: Royal Society of London, Philosophical Transactions, v.315A , p.123-134.
Lewan, M.D., Bjor0y, M. and Dolcater, D.L.. 1986, Effects of thermal maturation on steroid hydrocarbons by hydrous pyrolysis of Phosphoria retort shale: Geochi mica et Cosmochimica Acta, v. 50, p. 1979-2. Leythaeuser, D., Schaefer, R.G. and Weiner, B., 1979, Generation of low molecular weight hydrocarbons from organ ic matter in source beds as a function of temperature and facies: Chemical Geology, v. 25, p. 95-108. Mackenzie, A.S., 1984, Applications of biological markers in petroleum geochemistry, in Brooks, J. and Welte, D., eds., Advances in Petroleum Chemistry: Academic Press, Toronto, v. 1, p. 115-214. Mackenzie, A.S., Hoffmann, C.F. and Maxwell,J.R., 1981, Molecular parameters of maturation in the Toarcian shales, Paris Basin, France- III. Changesin aromatic steroid hydrocarbons: Geochimica et Cosmochimica Acta, v. 45, p. 1345-1355. Mackenzie, A.S., Lamb, N.A. and Maxwell . J.R., 1982, Steroid hydrocarbons and the thermal history of sediments: Nature, v. 295, p. 223-226. Mackenzie, A.S. and McKenzie, D., 1983, Isomerization and aromatization of hydrocarbons in sedimentary basins formed by extension: Geological Magazine , v. 120, p. 417-528. Mackenzie, A.S., Patience, R.L., Maxwell. J.R., Vandenbroucke, M. and Durand. B., 1980, Molecular parameters of maturation in the Toarcian shales, Paris Basin, France - I. Changes in configuration of acyclic isoprenoid alkanes, steranes and triterpanes: Geochimica et CosmochimicaActa, v. 44, p. 1709-1721. Mackenzie, A.S., Patience, R.L. and Maxwell, J.R., 1981, Molecular changes and the maturation of sedimentary organic matter, in Atkinson, G. and Zuckermann, J.J. , eds., Origin and Chemistry of Petroleum: Pergamon, Oxford, p. 1-31. Mackenzie, A.S., Quirke, J.M.E. and Maxwell, J.R., 1980, Molecular parameters of maturation in the Toarcian shales, Paris Basin, France - II. Evolution of metalloporphyrins, in Douglas, A.G. and Maxwell , J.R., eds., Advances in Organic Geochemistry 1979: Pergamon, Oxford, p. 239-248. Mackenzie, A.S., Ren-Wei, L., Maxwell, J.R., Moldowan, J.M. and Seifert, W.K., 1983, Molecular measurements of thermal maturation of Cretaceous shale from the Overthrust Belt, Wyoming, USA, in Bjorey, M.• Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 496-503.
McEvoy, J. and Giger, W., 1986, Origin of hydrocarbons in Triassic Serpiano oil shales - hopanoids: Organic Geochemistry, v. 10, p. 943-949. McKenzie, D., Mackenzie, A.S., Maxwell, J.R. and Sajgo, C.S., 1983, Isornerlzation and aromatization of hydrocarbons in stretched sedimentary basins: Nature, v. 301, p. 504-506. McKirdy, D.M., Aldridge, A.K. and Ypma, P.J.M., 1983, A geochemical comparison of some crude oils from pre-Ordovician carbonate rocks, in Bjorey, M., Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotter, J. and Speers, G., eds., Advancesin Organic Geochemistry 1981: Wiley, Chichester, p. 99-10? Milner,C.w.D., 1982,Geochemical analysis of sedimentary organic matter and interpretation of maturation and source potential, in Staplin , F.L., Dow. W.G., Milner, C.W.D., O'Conner, D.1., Pocock, S.A.J., van Gijzel , P., Welte, D.H. and YukJer, M.A., eds., How to Assess Organic Maturation and Paleotemperatures: Society of Economic Paleontologists and Mineralogists, Short Course 7, p.217·252. Monnier, F., Powell . T.G. and Snowdon, L.R., 1983, Qualitative and quantitative aspects of gas generation during maturation of sedimentary organic matter examples from Canadian frontier basins,;n Bjorey, M., Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullk6tter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 487-495. Oremland, R.S., Marsh. L. and Desmarais, D.J., 1982,Methanogenesis in Big Soda Lake, Nevada - an alkaline, moderately hypersaline, desert lake: Applied and Environmental Microbiology, v. 43, p. 462-468. Ourisson, G., Albrecht , P. and Rohmer, M., 1979,Paleochemistry and biochemistry of a group of natural products: Pure and Applied Chemistry, v. 51, p. 709-729. Peters, K.E., 1986, Guidelines for evaluating petroleum source rocks using programmed pyrolysis: American Association of Petroleum Geologists, Bulletin, v. 70, p. 318-329. Peters, K.E., Rohrbach, B.G. and Kaplan. I.R., 1981, Carbon and hydrogen stable isotope variations in kerogen during laboratory-simulated thermal maturation: American Association of Petroleum Geologists, Bullet in, V. 65, p. 501-508. Powell, T.G., 1978, An Assessment of the Hydrocarbon Source Rock Potential of the Canadian Arctic Islands: Geological Survey of Canada, Paper 78-12, 82 p. Pym,J.G ., Ray,J.E., Smith, G.W.and Whitehead, E.V., 1975, Petroleum triterpane fingerprinting of crude oils: Analytical Chemistry, v. 47, p. 1617·1622.
C
C
C C
C C C C
C C C C
C C C
C C C C
C C
C C C C C C
C C (
c e c c c c (
c ( c
c
c
c r c.
o
r
(J
C (
l C
o o o o o
o o o o
c o c
c
o o o o
o o r. 8·
r.
Determining Levels of Organic Diagenesis
Radke, M. and Welte, D.H., 1983, The methylphenanthrene index (MPI) - a maturity parameter based on aromatic hydrocarbons, in Bjorl,6y, M., Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Hullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 504-512. Radke, M., Welte, D.H. and Willsch, H., 1982a, Geochemical study on a well in the Western Canada Basin - relation of the aromatic distribution pattern to maturity of organic matter: Geochimica et Cosmochimica Acta, v. 46, p. 1-10. Radke, M., Willsch, H., Leythaeuser, D. and TeichmQller, M., 1982b, Aromatic components of coal- relation of distribution pattern to rank: Geochimica et Cosmochimica Acta, v. 46, p. 1831-1868. Rouxhet, P.G., Robin, P.L. and Nicaise, G., 1980, Characterization of kerogens and of their evolution by infrared spectroscopy, in Durand, B., ed., Kerogen: insoluble organic matter from sedimentary rocks: Editions Technip, Paris, p.163-190. Rowland, S.J., Aareskjold, K., Xuemin, G. and Douglas, A.G., 1986, Hydrous pyrolysis of sediments - composition and proportions of aromatic hydrocarbons in pyrolysates: Organic Geochemistry, v. 10, p. 1033-1040. -Saxby, J.D., Bennett, A.J.R., Corcoran, J.F., Lambert, D.E. and Riley, K.W., 1986, Petroleum generation - simulation over six years of hydrocarbon formation from torbanite and brown coal in a subsiding basin: Organic Geochemistry, v. 9, p. 69-81. Scheuer, P.J., 1973, ed., Chemistry of Marine Natural Products, Volume 1:Academic Press, Toronto, 201 p. Scheuer, P.J., 1978, ed., Chemistry of Marine Natural Products, Volume 2: Academic Press, Toronto, 308 p. Schidlowski, M., 1986, 13C/12C ratios as indicators of biogenicity, in Johns, R.B., ed., Biological Markers in the Sedimentary Record: Elsevier, Amsterdam, p.347-361. Schoell, M., 1984, Stable isotopes in petroleum research, in Brooks, J. and Welte, D., eds., Advances in Petroleum Chemistry: Academic Press, Toronto, v. 1, p.215-245. Seifert, W.K., Carlson, R.M.K. and Moldowan, J.M., 1983, Geomimetic synthesis, structure assignment, and geochemical correlation applications of monoaromatized petroleum steroids, in Bjorl,6y, M., Albrecht, P.,Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 710-724.
225
Seifert, W.K. and Moldowan, J.M., 1978, Applications of steranes, terpanes and monoaromatics to the maturation, migration and source of crude oils: Geochimica et Cosmochimica Acta, v. 42. p. 77-95. Seifert, W.K. and Moldowan, J.M., 1979, The effect of biodegradation on steranes and terpanes in crude oils: Geochimica et Cosmochimica Acta, v. 43, p. 111-126. Seifert, W.K. and Moldowan, J.M., 1980, The effect of thermal stress on source rock quality as measured by hopane stereochemistry, in Douglas, A.G. and Maxwell, J.R., eds., Advances in Organic Geochemistry 1979: Pergamon, Oxford, p. 229-237. Seifert, W.K. and Moldowan, J.M., 1981, Paleoreconstruction by biological markers: Geochimica et CosmochimicaActa, v. 45, p. 783-794. Seifert, W.K. and Moldowan, J.M., 1983, Use of biological markers in petroleum exploration, in Meinschein, W.G., ed., Organic Geochemistry of Contemporaneous and Ancient Sediments: Society of Economic Paleontologists and Mineralogists, Great Lakes Section, Short Course notes, p. 3-1-3-31. Seifert, W.K. and Moldowan, J.M., 1986, Use of biological markers in petroleum exploration, in Johns, R.B., ed., 1986, Biological Markers in the Sedimentary Record: Elsevier, New York, p. 261-290. Seifert, W.K., Moldowan, J.M. and Jones, R.W., 1980, Application of biological marker chemistry to petroleum exploration: 10th World Petroleum Congress, Bucharest, Romania, September 1979, Proceedings, Heyden, London, Paper SP-8, p. 425-440. Snowdon, L.R. and Powell, T.G., 1982, Immature oil and condensate - modification of hydrocarbon generation model for terrestrial organic matter: American Association of Petroleum Geologists, Bulletin, v. 66, p. 775-788. Snowdon, L.R. and Roy, K.J., 1975, Regional organic metamorphism in the Mesozoic strata of the Sverdrup Basin: Bulletin of Canadian Petroleum Geology, v. 23, p. 131-148. ten Haven, H.L., de Leeuw, J.W., Peakman, T.M.andMaxwell,J.R.,1986,Anomalies in steroid and hopanoid maturity indices: Geochimica et Cosmochimica Acta, v. 50, p. 853-855. ten Haven, H.L., de Leeuw, J.W. and Schenck, P.A., 1985, Organic geochemical studies of a Messinian evaporitic basin, northern Apennines (Italy) - 1. Hydrocarbon biological markers for a hypersaline environment: Geochimica et Cosmochimica Acta, v. 49, p. 2181-2191. Thompson, K.F.M., 1979, Light hydrocarbons in subsurface sediments: Geochimica et Cosmochimica Acta, v. 43, p.657-672.
Tissot, B. and Welte, D.H., 1984, Petroleum Formation and Occurrence: a new approach to oil and gas exploration, Second Edition: Springer-Verlag, Berlin, 538 p. Volkman, J.K. and Maxwell, J.R., 1986, Acyclic isoprenoids as biological markers, in Johns, R.B., ed., Biological Markers in the Sedimentary Record: Elsevier, New York, p. 1-42. Weete, J.D., 1974,Aliphatic hydrocarbons, in Weete, J.D., ed., Fungal Lipid Biochemistry: Plenum, New York,p. 39-65. Winters, J.C., Williams, J.A. and Lewan, M.D., 1983, A laboratory study of petroleum generation by hydrous pyrolysls, in Bjorl,6y, M., Albrecht, P.,Cornford, C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Rullkotter, J. and Speers, G., eds., Advances in Organic Geochemistry 1981: Wiley, Chichester, p. 524-533. See for diagram of apparatus. Predictive Diagenetic Models General
Bostick, N.H., 1979, Microscopic measurement of the level of catagenesis of solid organic matter in sedimentary rocks to aid in exploration for petroleum and to determine former burial temperatures - a review, in Scholle, P.A. and Schluger, P.R., eds., Aspects of Diagenesis: Society of Economic Paleontologists and Mineralogists, Special Publication 26, p. 17-43. Dutcher, R.R., Hacquebard, P.A., Schopf, J.M. and Simon, J.A., 1974, eds., Carbonaceous Materials as Indicators of Metamorphism: Geological Society of America, Special Paper 153, 108 p. Karweil, J., 1956, Die metamorphose der Kohlen vern Standpunkt der physikalischen Chemie: Deutsches Geologischen Gesellschaft Zeitschrifte, v. 107, p.132-139. Landes, K.K., 1967, Eometamorphism and oil and gas in time and space: American Association of Petroleum Geologists, Bulletin, v. 51, p. 828-841. Numerical Models
Bostick, N.H., 1973, Time as a factor in thermal metamorphism of phytoclasts (coaly particles): 7th Congress Internationale de la Stratigraphie et Geologie Carbonifere, Krefeld, 1971, Compte Rendu, v. 2, p. 183-193. Bustin, R.M., 1983, Heating during thrust faulting - friction or fiction?: Tectonophysics, v. 95, p. 305-328. Bustin, R.M., 1984, Coalification levels and their significance in the Groundhog coalfield, northcentral British Columbia: Coal Geology, v. 4, p. 21-44.
226
Bustin, A.M., 1986, Maturity of Late Cretaceous and Tertiary coal measures, Canadian Arctic Archipelago: Coal Geology, v. 6, p. 71-106. Bustin, R.M., Hills, L.V. and Gunther, P.A., 1977, Implication of coalification levels, Eureka Sound Formation, northeastern Arctic Canada: Canadian Journal of Earth Sciences, v. 14, p. 1588-1597. Castano, J.R. and Sparks, D.M., 1974, Interpretation of vitrinite reflectance measurements in sedimentary rocks and determination of burial history using vitrinite reflectance and authigenic minerals, in Dutcher, R.R., Hacquebard, P.A., Schopf, J.M. and Simon, J.A., eds., Carbonaceous Materials as Indicators of Metamorphism: Geological Society of America, Special Paper 153, p. 31-52. Chiaromonte, M.A. and Novelli, L., 1986, Organic matter maturity in northern Italy - some determining agents, in Leythaeuser, D. and Rullkotter, J., eds., Advances in Organic Geochemistry 1985, pt. 2, p. 281-290. Cohen, C.A., 1981, Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration - discussion: American Association of Petroleum Geologists, Bulletin, v. 65, p. 1647-1648. Dow, w.G., 1977, Kerogen studies and geological interpretations: Journal of Geochemical Exploration, v. 7, p. 79-99. England, T.D.J. and Bustin, A.M., 1986b, Effect ofthrust faulting on organic maturation in the southeastern Canadian Cordillera, in Leythaeuser, D. and Rullkotter, J., eds., Advances in Organic Geochemistry 1985, p. 609-616. Hood, A., Gutjahr, C.C.M. and Heacock, A.L., 1975, Organic metamorphism and the generation of petroleum: American Association of Petroleum Geologists, Bulletin, v. 59, p. 986-996. Huck, G. and Karweil, J., 1955,Physikalischchemische Probleme der Inkohlung: Brennstoff-Chemie, v. 36, p. 1-11Hughes, J.D. and Cameron, A.R., 1985, Lithology, depositional setting and coal rank-depth relationships in the JurassicCretaceous Kootenay Group at Mount Allan, Cascade Coal Basin: Geological Survey of Canada, Paper 81-11, 41 p. Kalkreuth, W. and McMech..an,-M., 1984, Regional pattern orttlermal maturation as determined from coal-rank studies, Rocky Mountain Foothills and Front Ranges north of Grande Cache, Alberta - implications for petroleum exploration: Bulletin of Canadian Petroleum Geology, v. 32, p. 249-271Lopatin, N.V., 1971, Temperature and geologic time as factors in coalification: Akademie Nauk SSSR, Ser. Geol., Izvestiya, no. 3, p. 95-108. [in Russian, translated by N.H. Bostick).
Diagenesis
Lopatin, N.Y. and Bostick, N.H., 1974, The geologic factors in coal catagenesis: Illinois Geological Survey, Reprint 1974Q, 15 p. [Originally published 1973, in Russian: Priroda organicheskogo veshchestva sovremennykh i iskopaemykh osadkov: Nauka Press). McKenzie, D.P., 1978,Some remarks on the development of sedimentary basins: Earth and Planetary Science Letters, v. 40, p. 25-32. McKenzie, D.P., 1981, The variation of temperature with time and hydrocarbon maturation in sedimentary basins formed by extension: Earth and Planetary Science Letters, v. 55, p. 87-98. Middleton, M.F.,1982,Tectonic history from vitrinite reflectance: Royal Astronomical Society, Geophysical Journal, v. 68, p. 121-132. Sclater, J.G. and Christie, P.A.F., 1980,Continental stretching - an explanation of the post-mid-Cretaceous subsidence of the central North Sea basin: Journal of Geophysical Research, v.85, p. 3711-3739. Shibaoka, M. and Bennett, A.J.R., 1977, Patterns of diagenesis in some Australian sedimentary basins: Australian Petroleum Exploration Association Journal, v. 17, p. 58-63. Tissot, B.P. and Welte, D.H., 1984, Part V, Chapter 4: Geochemical modelling: A quantitative approach to the evaluation of oil and gas prospects, in Tissot, B.P. and Welte, D.H., Petroleum Formation and Occurrence, Second Edition: Springer-Verlag, New York, p. 583-609. Tissot, B.P. and Espitalie, J., 1975, L'evolution thermique de la matiere organique des sediments: Applications d'une simulation mathematique: Revue de L'lnstitute Francals du Petrole, v.30, p.743-777. Waples, DW., 1980, Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration: American Association of Petroleum Geologists, Bulletin, v. 64, p. 916-926. Welte, D.H. and Yukler, A., 1980,Evolution of sedimentary basinsfrom the standpointof petroleum origin and accumulation - an approach for a quantitative basin study: Organic Geochemistry, v. 2, p. 1-8. Wright, N.J.R., 1980,Time, temperature and organic maturation - the evolution of rank within a sedimentary pile: Journal of Petroleum Geology, v. 2, p. 411-425. Yorath, C.J. and Hyndman, R.D., 1983,Subsidence and thermal history of Queen Charlotte Basin: Canadian Journal of Earth Sciences, v. 20, p. 135-159. Wood, R.J., 1981, The subsidence history of Conoco well 15/30-1 , central North Sea: Earth and Planetary Science Letters, v. 54, p. 306-312.
Geoscience Canada Reprint Series 4
Comparative Models Barker, C.F.and Elders, W.A.,1979,Vitrinite reflectance geothermometry in the Cerro Prieto geothermal field, Baja California, Mexico: Geothermal Resources Council, Transactions, v. 3, p. 27-30. Buntebarth, G., 1978, The degree of metamorphism of organic matter in sedimentary rocks as a paleo-geothermometer, applied to the Upper Rhine Graben: Paleogeophysics, v. 117, p. 83-91. Bustin, R.M., Hills, L.V. and Gunther, P.R., 1977, Implication of coalification levels, Eureka Sound Formation, northeastern Arctic Canada: Canadian Journal of Earth Sciences, v. 14, p. 1588-1597. Castano, J.R. and Sparks, D.M., 1974, Interpretation of vitrinite reflectance measurements in sedimentary rocks and determination of burial history using vitrinite reflectance and authigenic minerals, in Dutcher, R.A., Hacquebard, P.A.,Schopf, J.M. and Simon, J.A., eds., Carbonaceous Materials as Indicators of Metamorphism: Geological Society of America, Special Paper 153, p. 31-52. Hacquebard, P.A., 1977, Rank of coal as an index of metamorphism for oil and gas in Alberta: in The Origin and Migration of Petroleum in western Canadian Sedimentary Basins: Geological Survey of Canada, Bulletin 262, p. 11-23. Hacquebard, P.A. and Donaldson, J.R., 1970, Coal metamorphism and hydrocarbon potential in the upper Paleozoic of the Atlantic provinces, Canada: Canadian Journal of Earth Sciences, v. 7, p. 1139-1163. Hacquebard, P.A. and Donaldson, J.R., 1974, Rank studies of coals in the Rocky Mountains and inner foothills belt, Canada, in Dutcher, R.A., Hacquebard, P.A., Schopf, J.M. and Simon, J.A., eds., Carbonaceous Materials as Indicators of Metamorphism: Geological Society of America, Special Paper 153, p. 75-94. Nurkowski, J.R., 1984, Coal quality, coal rank variation and its relation to reconstructed overburden, Upper Cretaceous and Tertiary plains coals, Alberta, Canada: American Association of Petroleum Geologists, Bulletin, v. 68, p. 285-295. Suggate, R.P., 1956, Depth-volatile relations in coalfields: Geological Magazine, v. 93, p. 201-217.
Originally publlshed in Geoscience Canada v. 12 Number 1 (March 1985) Revised 1988
c c c c c c c c c c c c c c c c c c c c c c c c c
c c
(
c c
c
c
c
e
c c c
C l C C
C
c c c
Diagenesis of Biogenic Siliceous Sediments
c
r (
('
c
c c c o o o c o c o o o o
c o o o o c C
6 ( (
L
C l. (
Origin of Chert: Diagenesis of Biogenic Siliceous Sediments Reinhard Hesse Department of Geological Sciences McGill University 3450 University St. Montreal, Quebec H3A 2A 7 INTRODUCTION Highly silicified sedimentary rocks are generally called chert, but are also known under generic or local names such as radiolarite, porcelanite, flint, jasper, Iydite, or novaculite. They may contain 95% or more silica. The origin of chert has been much debated in the past. Systematic studies in the last decade have advanced our knowledge of chert genesis significantly, and have shed light on many facets of the socalled "chert-problem". Thethree basic aspects of the chert-problem, now understood in principle, are: (1) the source of the silica, (2) the depositional environment of siliceous sediments and (3) the diagenesis of cherts and chertbearing rocks. A historical perspective will be included in this review in order to give the reader some idea of the developments that led to our present understanding of silica diagenesis and to the problems on which current research is focussed. Although the emphasis of this review is on diagenesis, it will start with a brief outline of the first two aspects of the former chert-problem because a discussion of silica diagenesis is not possible without frequent reference to these other subjects. Terminology and Concentration Units for Dissolved and Solid Si02 Phases and Fine-grained Siliceous Sediments. To facilitate communication, a few of the terms and concentration units used for the variousdissolvedand solid speciesof Si02 , as well as for fine-grained siliceous sediments, will be defined at the outset.
Disso/vedsilicon(orsimply silicon) isthe term used for the undissociated monomeric silicic acid Si(OH)4 which is the dominant species of dissolved Si0 2 in most natural waters. Its anhydrous equivalentSi02 hasa molecularweightof about 60. One ppm Si0 2 by weight (or 1 mg'kg-') therefore corresponds to 16.6 microgram-atoms Si0 2 per litre (p.gat-L", the concentration unit commonly used in the oceanographic literature) or 16.6p.M where p.M stands for micromole per litre.This, inturn, isequivalentto 0.467 mg' L-' Si. Conversely, 1 mM (= 1 millimola-L") corresponds to 60 ppm Si02 • In its various solid phases, Si02 will be referred to as silica. These include amorphous silica or opal-A, cristobalite-tridymite or opal-CT, low-temperature cristobalite or opal-C, low-temperature tridymite, chalcedony, quartzine, lutecite and quartz. Opal-CT lepispheres are small spheres less than 5 microns in diameter of intergrown cristobaliteltridymite blades precipitated into sediment pores. Magadiite [NaSi 70,3(OHb' 3H 20j is a silica precursor mineral in lake sediments. Physical and mineralogical characteristics of these species are discussed in appropriate sections of this and the following paper. Biogenic fine-grained siliceous sediments are called siliceous oozes. The term "siliceous mud" instead of ooze is used for deposits which have a substantial detrital terrigenous component. Diatom ooze, radiolarian ooze and (sponge) spicule ooze are distinguished according to the main groups of organisms contributing siliceous shells to pelagic (or lacustrine) sediments. Their lithified equivalents are diatomite, radiolarite and spiculite. The rock names porcelanite and chert are used to denote fine-grained siliceous sedimentary rocks according to their textural and physical properties. "Porce/anite" refers to a porous rock with dull or matte lustre similar to unglazed porcelain, whereas "chert" denotes a dense vitreous, hard and brittle rock (Bramlette, 1946). The difference between the two is related to detrital content, especially clay content, which is higher (25-50%) in porcelanite than in chert (0-25%, Jones and Murchey, 1986). Siliceous rocks with more than 50% clay are called siliceous claystone, mudstone, or argillite. Lydite is a dark-grey to black variety of argillaceous chert or siliceous argillite. Local names for various chert types include flint, jasper, and nova-
227
Flint is a name widely applied to chert nodules in the Upper Cretaceous chalk of northwestern Europe(Schmid, 1986), which have been used in artifacts and tools made by early man (arrowheads, knives, drills, etc.). The term is synonymous with firestone or hornstone and has been commonly applied for more than a thousand years to any very hard and dense rock with conchoidal fracture surfaces (Bates and Jackson, 1980). Generally, jasper is a red, brown or yellow variety of chert (also green or black) that derives its colour from ironoxide impurities. It may be associated with iron-ores or soil-forming processes. Novaculite is very dense, Iightcoloured quartz-chert from the Ouachita Mountains in Arkansas and Oklahoma, which has undergone high-grade diagenesis and low- to intermediategrade metamorphism. The word is derived from the Latin word for razor (novacu/a) and has been formerly used in England for rocks that served as whetstones or razor hones (Goldstein, 1959).
SOURCES OF NON-DETRITAL SILICA IN SILICEOUS SEDIMENTS There are three main sources of nondetrital silica in sediments: (i) siliceous tests and skeletal elements of organisms, (ii) weathering solutions in semiarid climates, and (iii) silicon supplied in solution by hydrothermal-volcanic systems. The predominant source of Phane,rozoic cherts is biogenic, although there also exist hydrothermal-volcanogenic deposits or low-temperature chemogenic deposits (in alkaline lakes or arid to semi-arid soils). In the Precambrian, the cherty variety of banded iron formations represents a widespread type of chert almost unknown from younger geologic periods whose origin is still being debated. Low-temperature chemogenic and lor hydrothermal-volcanogenic deposits are the dominant Precambrian types of chert. This review focusses on the diagenesis of biogenic siliceous sediments. Inorganic and replacement cherts are dealt with in the following article (Hesse, this volume, p. 253-275). The total present-day silica production by marine organisms such as diatoms, radiolarians, silicoflagellates or siliceous sponges is estimated at 2.5 x 10'6 grams per annum (g-a-'). This is roughly 25 times the input of silica totheoceans(10.8 x 10'4g' a-'), towhich
c 228
rivers contribute 4.3, pore-water reflux a potentiaI5.?, submarine weathering 0.8, and submarine volcanism 0.05 x 10'4 g' a-1, according to Heath (1974). More recent calculations by Edmond et al. (1979a) attribute an additional 1.9 x 10'4 g' a" to hydrothermal influx, although this still ,would not balance input and output. The greater part of the output is primary production by diatoms which appear at the beginning of the marine food chain. The discrepancy between the figures for input and output, however, is not a real difference because of internal oceanic recycling which involves dissolution of siliceous tests and upwelling of the dissolved silica (Figure 1). According to estimates by different authors (e.g., Calvert, 1968, Berger, 1970, Hurd, 1973) between 90% and 99% of the silica extracted from surface seawater by shell-producing organisms (in the form of solid amorphous silica) redissolve before burial. Dissolution starts after death of the organisms when the tests settle through the water column, and continues while they rest on the sea floor, and particularly after burial in the sediment. A fraction of the silica dissolved during burial, however, returns to the ocean by means of diffusion . This is evident from the strong silica concentration gradient in the upper 20 cm of the sediment. De Master (1981)includes the figure for pore-water reflux quoted above (5.7 x 10'4 g' a-') with internal oceanic recycling. This is the diffusional flux across the sediment! water interface. Most of the dissolution during burial, however, provides the silicon required for diagenetic reprecipitation in the sediment, the mechanisms of which are the central topic of this review.
Diagenesis
global oceanic circulation system and occurs (i) along equatorial divergences, (ii) in the vicinity of subpolar convergences, (iii) along the west coasts of continents, and (iv) in certain marginal seas where surface currents diverge or are driven offshore. In the Pacific Ocean, regions of upwelling correspond to 5 major belts of silica accumulation on the sea floor (Figure 2): an equatorial belt (north of the equator), the sub-Antarctic belt (north of the Antarctic convergence), the sub-Arctic belt (south of the Arctic convergence), the Gulf of California (continental west coast! marginal sea upwelling) and the Okhotsk and Bering Seas (marginal sea setting). In low-latitude areas of high primary surface production , pelagic carbonates are deposited together with siliceous tests leading to mixed calcareous-siliceous oozes. Pure siliceous ooze or high-silica ooze-wi!! only accumulate in areas where dilutionby other components is inhibited or decreased. This is the case below the calcite compensation level (CCl), e.g., in the equatorial belt of siliceous oozes to the north ofthe mixed calcareous-siliceous belt. (An equivalent belt south of the equator does not exist, because the young , shallow oceanic crust has not yet subsided below the CCl). Temperature affects silica solubility more than pressure. Silica is therefore considerably more soluble in surface waters than at great
Geoscience Canada Reprint Series 4
depth (Figure 3). In the deepsea below 1 or 2 km water depth, temperature decreases only atew degrees Celsius . The temperature effect thus becomes very smaU-and can be offset by the effect of incr.!!asing pressure which raises silica solubility. Pressure-corrected curves for silica solubility (Figure 3) therefore have a minimum at about 1.5 km depth, below which solubility increases slightly. However, an equivalent of the CCl for silica has not been observed in the oceans. From the foregoing it follows that silica dissolution is predominant in the upper 1000 m of the water column, in contrast with the dissolution of pelagic carbonates. In the Gulf of California, siliceous muds accumulate despite the supply of terrigenous detritus to the gulf, because the terrigenous input is seasonally and locally suppressed by the prevailing wind and current patterns (Calvert, 1966a,b). Actual concentrations of dissolved silicon in the ocean range from a fraction of a part per million to 10-15 ppm dissolved silicon (or up to 160-250 rrucromoles- L-'). That is, they are considerably lower than the equilibrium solubility of amorphous silica, which is between 60 and 130 ppm at 25°C and a pH < 9. Silica-secreting marine microorganisms thus must extract dissolved silicon from seawater at concentrations well below the thermodynamic equilibrium solubility of opal-A. Thermodynamically this is not possible. The
C C C C
c c c c c c c c c c c c c c c c. ( c c c c c c c
VOLCANIC INPU
DEPOSITIONAL ENVIRONMENTS OF SILICEOUS SEDIMENTS
Igla",hydroth 5Ol,.tcl
~
8
Biogenic Siliceous Oozes on the Modern Ocean Floors. Biogenic siliceous sediments occur both in deep- and shallow-water depositional environments. Siliceous ooze is a typical pelagic sediment which accumulates on the deep sea floor in regions where silica production in surface waters is high. This is the case in zones of oceanic upwelling, where nutrients dissolved at some depths in the oceans are returned to the surface. Upwelling is a characteristic phenomenon of the
(
'".
halm rolYSIS
' ,' Of
boson
C L
C
e
C c..
c. ( Figure 1 Global dissolved silicon budget of the oceans. (From Riech and modified after Heath, 1974). All numbers are in
10' 3 g
Si02 per annum.
11.
Rad, 1979;
l C C C
(
(
c
c
c· ( (
( ( (
c
o o o o o o o o c
Diagenesis of Biogenic Siliceous Sediments
organisms must therefore be capable of raising the concentration to saturation under the catalyzing influence of enzymatic reactions. Concentrations are lowest in surface seawater because of the extraction by organisms. Concentration increases with water depth because of the dissolution of siliceous tests settling through the water column, leading to a mid-depth maximum . A subsequent decrease at greate r depths is caused by the renewal of bottom
229
waters from silicon-poor surface -water sources in polar regions, especially in the North Atlantic. Bottom maxima in dissolved silicon, on the other hand, are associated with belts of siliceous sedlments, particularly d iatom oozes (Edmond et sl., 1979b), with Antarctic bottom water, and with areas of hydrothermal activity. Cherts in Pelagic Stratigraphy. The zonal distribution of pelagic sediments described for the Pacific, when
combined with plate motions and the thermal subsidence of ageing oceanic crust , gives rise to a distinct stratigraphic succession of pelagic sediments known as "plate stratigraphy" (Berger and Winterer, 1974) or "pelagic stratigraphy " (Hesse et et., 1974). It is typically developed on Cretaceous Cenozoic crust of the major oceans and consists of (from bottom to top): midocean ridge basalt overlain by pelagic carbonates, then cherts and siliceous
o
o o o o ( o o o o o 8 ~Iio ••nic cokar.ou.
~Aby..al clay
§
Voicano ••nic
I]Jlio.onic silic.ou.
DHemipolagic
§
Turltidit••
Figure 2 Distribution of pelagic sediment types in the bottom sediments of the Pacific Ocean . (From Hesse et aI., 1974). Note that siliceous sediments south of the A/eutian and Kurile Islands are mixed with volcanogenic material.
r
230
Diagenesis
oozes and brown abyssal clay (Figure 4). Calcareous ooze is generally the first pelagic sediment deposited on newly created oceanic crust, at least since late Jurassic time. This is due to the elevation of young oceanic crust at midocean ridge crests, which normally projects above the CCL. With age-related subsidence, the crust will eventually pass below the CCl, leading to the deposition of siliceous ooze and/or brown abyssal clay. The sequence may be duplicated, if the northwestward motion of the Pacific plate carries it under the equator, where the CCl is sufficiently depressed to intercept the sea floor for a second time (Figure 4). An equatorial carbonate/siliceous ooze may thus accumulate on the older ridgecrest carbonate and adjacent off-ridge siliceous ooze and/or brown clay. On top of this, more brown clay is added during the residence of the crust in the North Pacific. Upon approaching the West and North Pacific island arc/deep-sea trench systems, the oceanic crustal section which carries this pelagic stratigraphy, collects an ultimate cap of hemipelagic and volcanogenic sediments before it is destroyed in asubduction zone. During subduction, some of the pelagic sediments may be scraped from the downgoing crustal slab and incorporated by tectonic accretion into an imbricated tectonic wedge under the trench-slope and fore-arc basin. In this way, some siliceous deposits escape subduction and may ultimately be returned to the surface by tectonic uplift in
orogenic belts, wherethey may occur as parts of ophiolite sequences (see next paragraph). Ophiolites, the "Steinmann trinity", and Chert in Mountain Belts. In the plate-stratigraphic succession, siliceous sediments typically occupy an intermediate position between pelagic carbonate and brown abyssal clay. In mountain belts, the association of radiolarian cherts in an ordered sequence with other pelagic sediments, and with mafic and ultramafic rocks of the ophiolite suite, hasbeen recognized for nearly a century. In the Alps, it was called the "Steinmann trinity", after its discoverer Gustav Steinmann (1906, 1927). It was taken asa suite of rocks diagnostic for "eugeosynclines", whose equivalence with oceanic crust wasonly recognized with the advent of the plate tectonics concept (e.g., Chipping, 1971). Ophiolites and the Steinmann trinity are thus today identified with the remnants of oceanic crust and mantle in mountain belts (Figure 5). The "petrogenetic problem", which Wenk (1949)
Geoscience Canada Reprint Series 4
saw in the association of ultrasiliceous sediments (radiolarites) with ultrabasic igneous rocks is no longer a problem. The silica of the siliceous pelagic sediments is not derived directly from silicaundersaturated basic and ultrabasic rocks by submarine exhalative volcanism, as previously thought (asrecently as 1971: see Gibson and Towe, 1971). Rather, it is supplied through upwelling from the oceanic silica reservoir and biogenic precipitation in surface waters, as discussed above. The "geotectonic significance" of chert, where it forms part of an ophiolite sequence, is not affected by this re-interpretation. However, since the distribution of siliceous sediments and cherts is controlled by (paleo-)oceanographic conditions, they are therefore not restricted to oceanic crust. It is no surprise to find ancient cherts underlain by continental crust as, for example in the Apennines (McBride and Folk, 1979), where they occur next to ophiolitic cherts (Folk and McBride, 1978). It should also be mentioned that in the Alpine-Mediterranean
'Snow 'Valley Rift 'snow '
,
Equator
N
.
Line
. .. .
", '
00
.. ..
::
- ••• _.:::
o
0
0
o
0
0
Temperature (OC)
5
/
10
15
20
25
~.' , '.-:
.'
~T.M"rO'u,. I
2 Depth
(km)
J 3
1
,I
.
.
I
I ~Sol"bllll,
Plate Motion Increasing Age of Crust
Siliceous B' . P' ~ Calcareous iogeruc articles Abyssal Clay
B::;]
5
rnm
6
800
.. ..
l"nCOffeCI.dJ
950
1100
I
Siliceous Ooze Carbonate Ooze :.:..=.; Basalt C.C.L. Calcite Compensation Level s.
1250 1400 1550 1700
Solubility (/OM
s,02)
Figure 3 Solubility of biogenic silica in the CentralequatorialPacific estimatedas a function oftemperature andpressure (dotted line). (From Hurd, 1983; after Griffin, 1980).
c c c c c c c
c
c
c c c c c c c c c
c
Line o
c c c c
Figure 4 Pelagic stratigraphy model for the Pacific Ocean showing duplication of the biogenic sediment sequence due to the northwestward passage of the Pacific plate under the equatorial high-productivity belt. (From Hesse et aI., 1974). Section oriented approximately NW-5E. The calcite compensation level (CCL) acts as a "snow tine". The East Pacific Rise and other regions rising above it are covered with Iight-coloured biogenic oozes; areas below the CCL and outside the belts of siliceous ooze are characterized by red-brown abyssal clay (Figure 2).
c c c c c c c c c
c c e c l.
C (
c..
c c c
(
c c
c.
G ( (
(
C'
C· ()
o
o o o o o
o o
o ( o c o c o o o o
Diagenesis of Biogenic Siliceous Sediments
ophiolite terrains radiolarian cherts directly overlie the basalt. The sequence is the reverse of the standard Pacific sequence, in which pelagic carbonates precede the cherts. This is probably due to a much shallower CCl in pre-late Jurassic oceans, a consequence of the absence of widespread calcareous planktonic organisms in these times (Bosellini and Winterer, 1975). Although most ophiolite-associated cherts are not volcanogenic, but biogenic, in origin, as we know now,the geochemistryof some radiolarian cherts nevertheless reveals a metasomatic overprint which is due to igneous activity at a former mid-ocean ridge. Examples are the above-mentionedJurassic Mediterranean bedded cherts which directly overlie basalt of the oceanic crust. Their iron content decreases and their manganese concentration increases in the section with increasing stratigraphic distance from the basalt reflecting the effects of hydrothermal activity (Barrett, 1981). POST-SU8DUCTION DEPOSITS FLrSCH
~RSE
filE
fEUGIC DEPOSITS
(CHElITI LlKESTOKEJ
GflEfNSTOKES - SPIL ITES
ULTflAIlASI&flOCKS (OPHIOLITES)
(0)
(b) C.nhn..t.1 PI.1t
l
CD
c· ,.. I. '-... .
.,
..
'-..
(
(c) Figure 5 (a) Schematic representation of the " Steinmann trinity " (= ophiolite stratig· raphy) of ultrabasic rocks , altered basalts and pelagic sediments (radiolarian cherts and limestones); (b) origin of the ophiolite stratigraphy through ocean crustgeneration at mid-oceanic ridges; and (e) its destruction In subduction zones. (From Chipping, 1971).
Formation of Bedded Chert. Geological and sedimentological evidence suggests a deep-water origin for most ancient bedded chert formations. Many of them appear to owe their emplacement to redeposition of siliceous pelagic sediment by turbidity currents (e.g., lmoto and Saito, 1973; Nisbet and Price, 1974; Robertson, 1977; McBride and Folk, 1979) or other ocean currents. Sedimentary structures and fabrics support a redeposition mechanism, as evidenced by flute casts and other sole marks (e.g., Imoto et al., 1974), common graded bedding, parallel lamination, faint cross lamination, starved ripples, grain fabrics such as parallel alignment of sponge spicules and radiolarians, shale clasts, and sharp contacts with interbedded shales. Another interpretation of bedded cherts is (i) diagenetic segregation of silica into chert-rich beds and silicapoor clay- stones from initially homogeneous siliceous muds. Still other interpretations include (ii) variations in surface-water productivity, orland (iii) variations in the input rate of the terrigenous component. These three alternative processes, however, do not adequately explain the sharp lithologic boundaries between the chert and interbedded shale layers which should have been obliterated by the activity of burrowing organisms, if these sediments were deposited through slow pelagic accumulation. Nearly all "bedded cherts" contain shale interbeds, often of red or green colour, that comprise between 5-40% of the sequence. Insome cases, the chert beds display none of the evidence for redeposition listed above, but still have sharp bottom and top contacts. This is why some investigators prefer variations in biogenic productivity andlor diagenetic segregation as origin for these deposits (e.g., Diersche (1980) for the well-bedded Upper Jurassic radiolarites of the Eastern Alps). However, the lack of primary sedimentary structures may be due to other reasonssuch as diagenetic obliteration during recrystallization and/or uniform small grain-size. From the recent literature, a general consensus seems to emerge that the bedding in radiolarian cherts cannot be ascribed to a single process (e.g., Baltuck, 1983; Heinand Karl, 1983; Jenkyns and Winterer, 1982). Recognition of symmetrical geochemical changes
231
(from the centre of beds upward to the top and downwardto the base)andsymmetrical grading of radiolarian abundance (Mizutani and Shibata , 1983; Sano, 1983; Steinberg et el., 1983) lendnew support to the diagenetic origin of some bedded cherts. (For recent reviews of radiolarian cherts see Jenkyns and Winterer, 1982; andJones and Murchey, 1986). Shallow Water Cherts. The effects of upwelling are not restricted to open ocean environments or the continental slope. Upwelling zones may transgress from the slope onto the shelf. Biogenic siliceous sediments, therefore, are not restricted to deep-water pelagic environments, but may on-lap the continental shelves and even occur on the inner shelf, as for example, off Namibia (Calvert, 1983). Tertiary opaline claystonesof the coastal plains of the southeastern United States are ancient equivalents associated with transgressive-regressive cycles (Weaver and Wise, 1974). The Monterey Formation of California, one of the better-studied ancient chert formations on land, was deposited in a Miocene continental margin environment composed of deeper and shallower water sites (Isaacs et aI., 1983) comparable to the present California borderland.The Monterey Formation is one of the principal sources of, and an important reservoir for, California's onshore and offshore hydrocarbon accumulations, underlining the economic significance of organic-matter rich biogenic siliceous sediments (Isaacs, 1984). Biogenic silica accumulations in near-shoreenvironments may occur in close association with shallow-water carbonates and give rise to chert nodules. Shallow-water sandstones and siltstones, apart from common silica cementation, are remarkably poor in concretionary silica (Dapples, 1967), in contrast with continental deposits. The aforementioned Tertiary shallow marine examples of the Atlantic and Gulf coastal plains of the US contain silica-cemented sandstones that appear to have received the silica from the associated opafine claystones (Wermund and Moiola, 1966; Wise and Weaver, 1973; Weaver and Beck, 1977; Carver, 1980). In Paleozoic sequences, siliceous sponge spicules often abound, particularly in deep basinal
232
deposits , but have also been considered as shoreline indicators (Cavaroc and Ferm, 1968,Lane, 1981). In a number of cases, spicules provide an adequate source of silica. Their diagenetic end-products are nodular cherts or chert nodules which frequently occupy only a small fraction of the total volume of the carbonate or evaporite host rock, These nodular cherts are distinctly different from the ancient bedded cherts in the rock record and from the biogenic siliceous oozes, muds and cherts on and under the modern ocean floors. Diagenesis of the latter will be discussed in this paper. The diagenesis of the former may proceed under conditions considerably different from those of the deep-water cherts, and those of the siliceous deposits in alkaline lakes and in soils.Thediagenesis ofthese and other types of chert , including silicified wood, will be discussed in the companion paper (Hesse, this volume , p. 253-275), ending with a note on the cherty varieties of Precambrian banded iron formations. STAGES OF BURIAL DIAGENESIS IN BIOGENIC SILICEOUS SEDIMENTS During burial, biogenic siliceous sediments undergo a characteristic sequence of mineralogic transformations which may be subdivided into three stages, each defined by the predominance of one of the common low-temperature silica phases (Calvert, 1971): (1) the opal-A stage of siliceous oozes or muds; (2) the opal-CT stage of porcelanites; and (3) the microquartz or quartz stage of chert sensu stricto (quartz chert). Silica recrystallization in petrified wood follows the same sequence, whereas some cherts in carbonates and some non-biogenic cherts may follow different diagenetic pathways as discussed in the companion paper. Porcelanite and chert do not simply reflect the mineralogical difference between opal-CT and quartz as a result of progressive diagenesis. As suggested by Isaacs et a/. (1983), primary compositional differences, such as the ratio of clay minerals to silica minerals, areoften moreimportant in determining textural and physical properties on which the distinction of rock types is based. In the Monterey Formation, the most silica-rich sediments (pure diatomites) form rocks that appear macro-
Diagenesis
scopically as cherts although they may still consist entirely of opal-CT. Diatomaceous shale on the other hand, has the appearance of porcelanite, even after having reached the quartz stage. Because these rock names were originally designed for the field description of siliceous rocks (Bramlette , 1946), they will be used in this review in the sense of Isaacs et al. (1983). That is, the distinction between porcelanite and chert will be based primarily on detrital content. Often, however, porcelanite consists predominantly of opal-CT, and 2.5
A
3.0
3.5 I
4 .0 I
50
A
s------C
o
E
I
30
I
I
I
28
26
20
I
18 0 29
CUKlI
Figure 6 (A) X-ray diffractogram for opalA (broad hump at about 41\) and detrital quartz (Q) in diatomaceous shale. (8 to I) peak-sharpening and shift in position of the (101)diffraction of opal-CT with progressive burial in the Monterey Formation of California . Doublet peaks at 4.11\ and 4.31\ for cristobalite (Cr) and tridymite (Tr), respec tively. Numbers in brackets: cristobalited(101)ofopal-CTinangstroms. (J) Diffractogram of quartz chert. (Modified from Murata and Larson, 1975, fig, 5).
Geoscience Canada Reprint Series 4
chert predominantly of (micro-)quartz. For this reason, Calvert (1971) first suggested a nomenclature of siliceous rocks strictly based on mineralogy, a usage largely followed by the DeepSea Drilling Project (e.g., Riech and v. Rad, 1979). However, for the alternative descriptive approach, see Pisciotto (1981 ). The characteristics of the different silica phases and the diagenetic transitions from one phase to the other will be described in the following paragraph. Important questions to be discussed are: What are the physicochemical mechanisms involved in the transformations? Are the transformations of opal-A to opal-CT and of opal-CT to (micro-) quartz exclusively due to dissolution-reprecipitation reactions as favoured by recent findings, or is there any evidence for the solid state reactions originally suggested by Ernst and Calvert (1969)? What are the ratecontrolling factors? In particular, how does lithology influence transformation rates? Why is an intermediate state of metastable opal-CT required in thetransition from opal-A to quartz? CHARACTERISTICS OF THE SILICA PHASES IN BIOGENIC SILICEOUS SEDIMENTS Opal-A: Physical Characteristics, Solubility, and Polymerization. The tests of siliceous microfossils as well as silica precipitated inorganically under earth surface conditions, as in hot-spring areas or desert soils, consist of an X-ray amorphous substance named opal-A (Jones and Segnit, 1971), due to its mineralogical similarity to precious opal. The X-ray pattern of this near-amorphous substance resembles glass, with a broad hump near 4~ (Figure 6A). This material is very porous and has a substantial water content (4-9% in precious opal). Opal-A dissolves easily in seawater, which is highly undersaturated with respect to opaline silica everywhere in the modern oceans. In inorganic systems, equilibrium solubility of silica depends on a number of factors besides temperature (Figure 7a), notably surface area and particle size of the solid silica phase (Figure 7b), and to a lesser extent pressure, as stated in a previous section (Figure 7c). It increases significantly with a pH greater than about 9 (Figure 7.d,
c c c c c
c c
c
c
c
c
c c c
c c c c
c c c c c c c c c C E C C C C (
e c
L
C C C
C C
C
c C' C
c' c
c ( c c c
Diagenesis of Biogenic Siliceous Sediments
233
shell wall (Hurd and Theyer, 1975). Equilibrium solubilities of the other lowtemperature silica polymorphs, o-crlstobalite and quartz, are respectively one and two orders of magnitude lower than those of opal-A (Figure 7a). However, for opal·CT no thermodynamic data exist. Silica solubilities bear great import for the diagenetic transformations of the
Williams et al., 1985, Williams and Crerar 1985).Other ions in solution can have distinct effects on dissol ution rates of solid silica. Dissolved AI and Fe adsorb to the opal-A surface and suppress its solubility and dissolution rate (lIer, 1955, 1973; Lewin, 1961). Also , ageing of the biogenic siliceous tests affects solubility through changes in surface area and crystallite size in the
c·
c
c
(b)
(0)
-1.0 ,...---r---r--,.---,----,---r--....,...~ 5000
_
-1.5
-
2000 1000 500_
tT
~
-2.0
N
eno r:
solid phases. For this reason, a somewhat closer look will be taken at the chemistry and kinetics of the dissolution and precipitation processes of silica , including polymerization . In solutions of pH < 9, dissolved silicon is present predominantly in the form of undissociated silicic acid H4Si04 (orthosilicic or monomeric silicic acid). The dissolution of solid silica can be
a.
200 Q. 100'1; 50 ~ o 20 en 10 5
-2.5
E -3.0 0'
o .. -3.5
~
-4.0
0'
o
r:
-45L...-....1--J-_'----'----'-_-'----'---=- 2 .0 400 100 200 300
Temperature (OC)
0"
220 IIEE---_ t\I
o
en
10
~
o
50
100
150
200
250
Specific surface area m~g
(d)
(c) o
r
-
-05,..,...,...,.."""'"'''''"""''................__.........,...,.-r'ln'''ft'''""'I
-; -1.0 c
;;" -I. 5 o -2.0 Cf)
-2.5
100--'-'
E 0' -3.0 .2 -3.5 n
-
~
-4.0
g-4.5 0 0.5 1.0 1.5 2.0 2.5 3.0 ' 3.5 4.0 4.5
Pressure (Kb)
Figure 7 Solubility diagrams for various silica phases as a function of (8) temperature (from Williams and Crerar, 1985; modified after Walther and Helgeson, 1977);(b) specific surface area (from Williams et al., 1985); (c) pressure (for quartz at various temperatures between 25° and BOOOC (from Williams and Crerar, 1985; modified after Walther and Helgeson, 1977); and (d) pH (from Williams and Crerar, 1985; after Volosov et al., 1972).
c 234
Diagenesis
Geoscience Canada Reprint Series 4
(
C written as Si02 (s) + H20 = H4Si04 (aq) . At equilibrium, the solubility product, Ksp , thus approximates the activity a(H4Si04 ) . In solutions of pH> 9, orthosilicic acid dissociates (Figure 7d), in a first step into H4Si04 (aq) -+ H3Si04" + H + , and for higher pH values (above 13)in a second dissociation step into H3Si04" -+ H2SiO~- + H+ . These dissociations raise the solubility significantly, because the total dis solved silicon concentration is the sum of the dissoc iated and undissociated species. In supersaturated solutions, dissolved silicon polymerizes forming first oligomers (dimers, tetramers and ring structures) and later higher-molecularweight polymers as siloxane bonds (SiO-Si) develop through combination of silanol (-Si-OH) groups:
(silanol)
(siloxane)
Williams and Crerar (1985) emphasize the fact that low-molecular-weight oligomers also contribute to the total solubility in solutions at or near opal-A saturation, contrary to the assumption in most previous experimental work that all dissolved silicon is monomeric in such solutions. Oligomers may even
persist in undersaturated solutions at temperatures below 130°C, although under such conditions they will gradually de polymerize (e.g., Krauskopf, 1956). At concentrations below opal-CT saturation, monomeric silicic acid is the only form present. On the other hand, if supersaturation persists, high-molecular-weight polymers (with molecular weights up to 10,000) form. Such polymers havecolloidal dimensions (greater than 50.8.) and may remain suspended as sols as long as pH remains relatively high and salinity low. Otherwise, they will form cross-links with neighbors and coagulate into gels (Figure 8a). Silica polymers display a negative surface charge, which results in a point of zero charge (PZC,i.e. ,the pH atwhich the residual surface chargedisappears) as low as 2±0.5 (Parks, 1965). Silica colloids thus repel each other unless the surface charge is neutralized by other ions in solution, such as metal hydroxides. Interestingly, the hydroxide most commonly usedfor silica precipitation in industrial applications, Mg(OH)2 (IIer, 1979),is also thought to be instrumental in the nucleation of opal-CT, as discussed below. The ultrastructure of precious opal shows a hexagonal or cubic closest packing of opal-A spheres that are remarkably uniform in size (lIer, 1979). Apparently, they are deposited from colloidal particle suspensions of uniform size that grow in supersaturated solutions in the absence of natural coagulants. The sphere diameter is in the appropriate size range (0.17-0.38JLm) to
diffract visible light (in the same manner crystal lattices diffract X-rays) and to produce the iridescent diffraction colours of gem opals. These spheres consist of several concentric shells of primary smaller spheres with diameters ranging from 100to 500.8. depending on the opal type (Figure 8b). Uniform size of the primary spheres apparently is achieved by the dissolution of smaller polymers and redeposition of the dissolved silica on the larger ones (Darragh et a/., 1976). Opal-CT: Ultrastructure. n The "CT in opal-CT stands for cristobalite/tridymite (Jones and Segnit, 1971) and denotes a modification of opal which has structural characteristics of both o-crlstoballte and a-tridymite. Opal-CT is thus the low-temperature form (a-form) of cristobalite/tridymite formerly called lussatite (Mallard, 1890).In opal-CT,cristobalite alternates randomly with unidimensionally disordered tridymite layers. On X-raypowder diagrams the main diffraction peak of opal-CT is a doublet at 4.1 and 4.3.8. (Figures 68 to 61). Wise et al. (1972) using the scanning electron microscope (SEM) first observed the occurrence of opal-CT in the form of small spheres less than 5/Lm in diameter in deep-sea drilling samples . These spheres, which consist of an interpenetrative growth of tiny cristobalitel tridymite blades, were called opal-CT lepispheres (Figure 9a) by Weaver and Wise (1972). On high-resolution SEM photos, lepispheres in their incipient growth stage show sets of subparallel
C C (
C
C (
C C C (
C C C C
C (
C C C (
C C
C C C
E C C C
C
c
e
ThreeDimensional Gel Networks
c
l
C (
Figure 8 (8) Polymerization behaviour of dissolved silicon as a function of pH and ionic strength of the solution. (From Williams and Crerar, 1985). (b) Formation of precious opal by deposition oflarge spheres of opa/-A of uniform size in a hexagonally or cubically closest packed layer. Large spheres consist of concentric shells of smaller primary spheres. (From Williams and Crerar, 1985; after Darragh et al., 1976).
C ( (.
Diagenesis of Biogenic Siliceous Sediments
235
r> .
c
c c
r (
c
[TOOjcrist . ----~
/
r:
~
0
0 r
0,
C
d
0 0 0 r
JI r: ~
r '-
r:
Cl '-.J
'--r: '-..-'
Figure 9 (a) Small opa/-CT lep ispheres (2-3p.min diameter) growing on euhedral calcite crystals in a cav ity in partially silicified Maestrich tian chalk, DSDP 14-144-3-2, 103·104 cm. (b) Opa/-CT blades of a "juvenile" lepisphere displaying twinning angle of 70° corresponding to the intersection angle of the faces of a cr istobalite octahedron as sho wn in (c) and (d). (From v. Rad et a l., 19 77). (e) Composite lepisphere SOp'm in diameter which resulted from coalescence ofnumerous individuallepispheres, DSDP 12-117A-2cc. (From Florke et al., 1976). (1) Sieve structure of diatom trustute. (SEM photos in Figures 9a,b,e,f courtesy U. v. Rad)
( 236
Diagenesis
opal-CT blades which penetrate each other at angles between 70 and 71 0. Individual blades are 2-51Lm long and 0.05-0.1ILm thick, and display ragged or rounded edges (Figure 9b). The penetration angle corresponds to the angle of 70°32' at which equivalent faces of a cristobalite octahedron intersect one another (Figures 9c,d). Florke et al. (1976)give a plausible crystallographic interpretation for the peculiar tw inning behavior of opal-CT in the lepispheres, which appears to be dictated by crystal growth kinetics. At more advanced growth stages, the octahedral skeletons will coalesce to form lepispheres. The diameter of individual lepispheres usually does not exceed the length of the blades (about 51Lm) of which they are composed. Larger lepispheres are almost invariably aggregate forms (Figure se: see also Carver, 1980, figs. 1C, 1D, 3A, 3B). Relics of former opal-CT
Si02 0.0
0 .2
0.4
0.6
o.e
lepispheres in Paleozoic cherts which have been portrayed in the literature (e.g., Jones and Knauth, 1979;Meyers, 1977) are invariably too large for genuine lephispheres, unless they are aggregates. Low-temperature Tridymite and Cristobalite as Separate, Distinct Mineralogic Phases. The occurrence, under special geologic conditions, of low-temperature tridymite and cristobalite as separate mineralogic phases distinct from opal-CTwas proposed by Tadaand lijima(1983). The disordered low-temperature tridymite (a-tridymite) is characterized by a relatively broad diffraction peak at 4.11 A on X-ray powder diffractograms obtained from mixtures of a-tridymite, o-cristobalite and opal-CT. On the same diffractograms, the better ordered o-cnstobalite (also called opal-C) produces a sharper diffraction peak at 4.04A. As
(mM) 1.0
1.2
1.4
1 .6
i.e
Geoscience Canada Reprint Series 4
will be shown later, both the 4.11 and 4.04A peaks have also been attributed to opal-CT at various stages of burial maturation (Murata and Larson, 1975). The justification given by Tada and lijima (1983) for distinguishing a disordered tridymite from 4.11 A-opal-CT is that it did not recrystallize to a 4.04A· phase upon heating (for 11 days to 1100°C)contrary to what was expected. In nature, this low-temperature tridymite is observed as a pore-filling cement precipitated from groundwater or hydrothermal solutions in silicic volcanic rocks. The low-temperature acristobalite is also generally associated with volcanic and silicic volcaniclastic rocks. In deep-sea sediments, it has been reported from volcanogenic sediments overlying basalt in the West Pacific (Lau Basin, Griffin et al.• 1972) and Central Pacific (Zemmels and Cook, 1973).
Si02 0 .0 0 .2 0.4 0.6 0.8
2.0
r C (
C c r-'"
e c c c c c c c c c c
(mM) 1 .0
(
1 .2 1.4 1.6 1.8 2.0
c
c
c
100
100
(
c
200
200
c
c c
300
300
E
E
400
..c
500
400
...j....l
-C 500
0.. GJ 600
...j....l
0.
0
ill 600
0
1
700
L66S493 }{ L63S471 y L63S469 z L60S458 x L48S405 ... L43S387 e L41S369 x L38S348 + L38S338 f. L31S299 o L25S241 l!l L24S238 lie
700
BOO 800
L65S485 e L58S446 x L38S346 + L38S345 A L38S343 e L38S336 I!I L25S239 't
900
1000
900
1000
Figure 10 Concentration of dissolved silicon in pore-water profiles of DSDP holes. (L, DSDP leg; S, DSDP site).
L81S552
x L72S516
c c c c c c c e (
L
C (
C C C (
(
r r C
C C C
o ( c c· c o o o o o o
o o o o I""'""
o
0,
D
Diagenesis of Biogenic Siliceous Sediments
NATURE OF THE TRANSFORMATION MECHANISMS Opal-A to Opal-CT Transformation. Only a small percentage of the plenitude of biogenic silica (opal-A) produced by planktonic organisms in the surface ocean reaches the sea floor. Only a fraction of this escapes dissolution during the first metres of burial. Even this very small proportion of solution-resistant species will ultimately undergo dissolution at greater sub-bottom depths. The effects of progressive dissolution have been documented by systematic SEM studies of diatom oozes (e.g., Hein et al., 1978).Breakage of partially dissolved diatom valves accompanies and enhances dissolution, culminating in the complete destruction of the tests. Dissolved silicon profiles from DSDP holes in the pelagic realm reflect the dissolution of solid silica particles in the subsurface. These profiles show consistent increases in dissolved silicon with sub-bottom depth, not uncommonly exceeding 1 mM (or 60 ppm) at a few 100 m below the sea floor. Concentrations may fluctuate considerably with depth, but these fluctuations are generally superimposed on an overall trend of downward increasing silica concentrations (Figure 10).In many holes, however,the gradual downward increase is followed by an abrupt decrease below a certain depth(e.g., DSDPsite495,Harrisonetal., 1982; site 570, Hesse et al., 1985) consistent with reprecipitation as opalCT. In the examples mentioned, this is substantiated by the recovery of
o o
G r \-
o
r
r: '-,'
Figure 11 Radiolarian test replaced by opal-CT,DSDP41-366-23-1, 42-44 em. (From Riech andv. Rad, 1979; SEM photo, courtesy U. v. Rad).
the first porcelanite nodules from this depth. Below this depth, a less pronounced second maximum may be observed in sufficiently deep holes, which probably corresponds to the dissolution of opal-CT and the subsequent reprecipitation as quartz described in the next section. Continued dissolution of opal-A in the sediment during burial is the result of slowly increasing temperature and pressure. Siliceous tests of diatoms and Radiolaria have large specific surface areas ranging from several tens to 450 m2'g.1 (Kastner et al., 1977),compared with 0.1 m2 • g-1 of crushed quartz in the 3 to 5 phi (25-75JLm) size range (VanLier et al., 1960). The sieve structure of the porous test walls of these organisms (Figure 9f) is only partly responsible for these large specific surface areas. As Hurd et al. (1979) have shown, it is ultimately the size of the small opal-A domains (2.5-4.5'&' diameter)in the tests which causes the high specific surface areas. Surface area can significantly affect solubility. When the surface area to volume ratio of a substance becomes large, small changes in pressure and temperature may then markedly influence its solubility. Suppression of solution inhibiting factors such as the removal of protective coatings of organic matter may further increase solubility. This explains the continuation of opal-A dissolution during burial, even though up to 99% of the original opal from the surface water may have already been dissolved during settling and initial burial. In contrast with the earlier dissolution, dissolution during sediment burial occu rs in a more or lessclosed system in which concentration levels may reach supersaturation before opal-A dissolution has gone to completion. In this case, dissolution will be interrupted by reprecipitation of a less soluble nonbiogenic opal-A called opal-A', which forms overgrowths on partially dissolved siliceous tests. (Hein et al., 1978). Crystallite size calculated from X-ray diffraction data for opal-A' is larger (20-27.&.) than for biogenic opal-A (12-16'&'). InindividualDSDPholes,opal-A" overgrowths have been found to occur only over a narrow stratigraphic range of a few metres. This indicates that they redissolve with the remaining opal-A shortly after formation (Hein et al., 1978). Opal-A' is therefore rarely
237
observed. Its presence, however, may explain some of the fluctuations seen in the concentration-depth profiles of dissolved silica (Figure 10). Confirmation of the Opal-A to Opal-CT Transformation as DissolutionReprecipitation Mechanism. The discovery of the opal-CT lepispheres with the euhedral crystal shapes of the cristobalite/tridymite blades by Wise et al. (1972), Weaver and Wise (1972) and Berger and v. Rad (1972) provided the first proof that the recrystallization of siliceous oozes to porcelanite occurs through a dissolution-reprecipitation mechanism. Lepispheres develop only where crystallization takes place in open pore spaces such as the cavities of microfossils. More commonly, a densely felted mass of opal-CTforms which impregnatesthe sediment andlor replacesother mineral phases. The latter process may involve pseudomorphic replacement of opal-A by opal-CT in radiolarian tests which perfectly preserves the shape of the shell (Figure 11). The process is still a dissolution-reprecipitation mechanism, which probably proceeds on a grain-by-grainbasiswith local precipitation immediately following dissolution. The presence of a matrix of organic matter, which is negligibly affected by dissolution, may serve as a template which helps preserve the original shape of the shell. The process may be similar to the silicification mechanism in ooids (Hesse, 1987), or to the petrifaction of wood. Where Radiolaria occur embedded in lutitic pelagic limestone, an opalCT replaced test may simply represent a cast of the former opal-A shell. Conservation of the original shape during the replacementprocesscannotbeused as evidence for a solid-state reaction in the opal-A to opal-CT transformation. Opal-CT to Quartz Transformation. The opal-CT to quartz transformation was assumed to be a solid-state mechanism by Ernst and Calvert (1969) and Heath and Moberly (1971). However, morphological evidence from scanning electron microscopy (Stein and Kirkpatrick, 1976) makes it more likely that this step is also a dissolution-reprecipitation reaction, like the opal-Ato opalCT transformation. In their hydrothermal experiments, Ernst and Calvert (1969) heated opal-CT in distilled water to temperatures of 300°, 400°, and 500°C under 2 kbar pressure. 9uartz
C
238
formed in each experiment. At 500°C the conversion was complete after 30 hours, at 400°C after 300 hours (12.5 days), and at 300°C after 5000 hours (7 months). For a given temperature the conversion rate seemed to be constant. The concentration of the reactant opalCT had decreased during the experiment in a more or less linear fashion with time. On the basis of general reactionrate laws, Ernst and Calvert concluded that the reaction was of zero order with respect to the number of participating reactants; it had to be a solid-state reaction. Stein and Kirkpatrick (1976)re-examined the reaction products of these experiments under the SEM and found fibrous and euhedral prismatic quartz crystals among the reacted material clearly requiring a dissolved phase for crystallization. A re-evaluation of the experimental data in the light of nucleation and growth theory (Christian, 1965; Nielsen, 1964)showed that a sigmoidal curve with slower reaction rates at the beginning and end of the conversion process would approximate the rate data better than the straight-line plots of Ernst and Calvert (1969). The Opal-A to Opa/-CT and Opa/-cT to Quartz Transformations as Examples of Ostwald Processes. Recognition of inorganically precipitated opal-A' by Heath (1969) and Hein et al. (1978) demonstrates that the opalA to opal-CT transformation is not a single-step process. Rather, it involves a series of dissolution-reprecipitation react ions as postulated theoretically by the model of Williams et al. (1985). This model is based on surface-area effects (Figure 7b) and is equally applicable to the opal-CT to quartz transformation. The hypothetical Si02-H 20 system used is a closed system at constant temperature, pressure, pH and consists of the pure reactants. In this system, opal-A of a diatom species with a specific surface area, say of 250 m2.g '1, (i.e ., Thalassiosira decipiens, Figure 7b), will have a solubility of about 100 ppm. Ifdissolution ofthe frustules of this species is fast relative to nucleation and growth of other new silica phases, then the solution will soon become supersaturated with respect to opal-A of a lower surface area (i.e., opal-A'). The solution is also supersaturated with respect to opal-CT of any given surface
Diagenesis
areato the left of point 3 in the diagram, as well as with respect to any quartz . The nucleation and growth ratesof opalCT are slower than those of any opal-A, and those of quartz are slower than those of opal-CT, because opal-CT and quartz are more highly ordered forms of solid silica than opal-A. Under these assumptions, an opal-A' with a surface area lower than that of opal-A will precipitate and the solution will evolve along the pathway from points 1 to 2 shown in Figure 7b. This process is called "Ostwald ripening" (lIer, 1979; Williams et al., 1985). Near point 2, the effect of surface area on opal-A solubility becomes negligible. Atthis point, the solid silica phase with the next lower solubility, t.e., opal-CT may appear. Opal-CT is preferent ially precipitated overquartz, because of its higher nucleation and growth rates. The effects of reaction kinetics are informatively displayed by the Williams et al. (1985)model. Their discussion will be carried one step further considering the relationship between the dissolution rate of opal-A and nucleation and growth rates of opal-CT. Depending on the mutual relationship of these rates, further evolution of the solution will follow pathways from points 2 to 4 either along curve " a" (fast nucleation rate) or curve "b" (slow nucleation rate). In the first case, rapid nucleation (relative to growth) leads to relatively large specific surface area because the nuclei are small and numerous. Silica concentration will drop off and the surface area of the newly formed opal-CT decrease (curve 2-4a), only when the silica removal rate exceeds the dissolution rate of the remaining opal-A. It, on the other hand, the growth rate of opal-CT exceeds the nucleation rate early in the process, then the fluid should evolve along pathway 2-4b. Analogous considerations should apply to the opal-CT to quartz transformation discussed in a subsequent section, although the structural reorganization within the opal-CT preceding this transformation is interpreted as a solid-state reaction by some authors. The relationship between specific surface area and solubility schematically shown in Figure 7b also explains why opal-CT is required as an intermediate metastable phase in the sequence of diagenetic silica transformations. Silicon solubility in equilibrium
Geoscience Canada Reprint Series 4
with opal-A of any specific surface area is too high for quartz to form directly from the dissolution of opal-A. At such high silicon concentrations the faces of any embryonic quartz crystal would be crowded by silanol groups which would not have time to be properly fitted into the crystal lattice. Quartz growth will be blocked and a less "well-ordered " phase, opal-CT, will form instead. No precise thermodynamic data exist for this phase. Quartz crystallization is only possible when the "equilibrium solubility" of opal-CT has been lowered sufficiently through Ostwald ripening. Opal-CT is a classical example for Ostwald's step ruIewhich states that the transformation of an unstable to a stable mineral phase (at earth surface conditions) often requires one or more intermediate metastable phases. This is an immediate consequence of the crystallization kinetics (Morse and Casey 1988). Ostwald's rule is the answer to the earlier question concerning the requirement of intermediate metastable phases such as opal-A' or opal-CT. However, under special circumstances (see section on absence of opal -CT) quartz may precipitate directly from solutions whose silicon was derived from opal-A dissolution but had not reached opal-A equilibrium concentration (or even the lower opalCT concentration). Loweringtheequilibriumsolubilityfor dissolved silicon through Ostwald ripening of the opal-CT phase also counterbalances a solubility increase with rising temperature during burial. This is a prerequisite for quartz precipitation which would not occur if opalCT solubility was raised substantially with temperature. A high-surface area opal-CT at 50°C has a theoretical solubilityof 130ppm.Thesolubilityof alowsurface area opal-CT at 110°Cis about 150 ppm (Fournier, 1973), only slightly higher than that of the high-surface area opal-CT at 50°C. A solution "in equilibrium" with "ordered" (low-surfacearea) opal-CTat 110°C,is therefore super-saturated only 1.5 times with respect to chalcedony or cryptocrystalline quartz. A solution "in equilibrium" with a high-surface-area (or " disordered") opal-CT at 110°C, on the other hand,would havea solubility of 290 ppm (Fournier, 1973) or nearly 3 times the equilibrium solubility of chalcedony. Sincethe growth of a new ("more highly
r
c c c c c c ( c c c
c c c c c c c c c
c
c c
c
c c
c c c c c c c e c..
l C C
C C L
C
(
C
Diagenesis of Biogenic Siliceous Sediments
239
(
C C C' (
o c ( (
C
o o a o ( o o o
ordered") silica phase such as chalcedony or quartz is favoured by low supersaturation, the " ordering" process during burial of opal-CT operates toward this end and ultimately facilitates quartz precipitation. Crystallographic Structural Changes of Opal-CT and Quartz in the Porcelanite and Quartz-chert Stages. Sharpening and shift ofthe opal-CT (101) reflection. The important finding of a dissolution-precipitation step in the opal-CT to quartz conversion by Steinand Kirkpatrick in 1976did notclose the discussion about the nature of the transformation mechanism involved in this phase change. A year before, in 1975, Murata and Larson's results of a high-precision X-ray diffraction study gave rise to new speculations about the possibility of low-temperature solid-
-
E ..........
0 200
-
state reactions during diagenesis. This study revealed characteristic changes in the main diffraction peak of opal-CT (4.1'&') with progressive burial diagenesis.ln contrast to Tada and lijima (1983), this peak is usually attributed to the (101) diffraction of a-cristobalite. In siliceous rocks of the Monterey Formation in the Temblor Range of California, a distinct shift of this peak from 4.11 to 4.04.&. is observed with increasing burial (Figure 6). (Slow scanning at a rate of 0.5°29 ' min-' permitted measurement of the peak position with a precision of ± 0.0052'&'). This decrease in the d-spacing is accompanied by a progressive sharpening of the same peak, and a gradual disappearance of the c-tridymite (001) peak (at 4.32-4.26'&'), which is replaced by the quartz (100)peak (at the opal-CT to quartz transition). Minor
Dry bu Ikdensity Radiometric '(10f) age (10 6 yr) cristabalite (angstroms) (g/cm3 ) 4.04 4.06 4.08 4.10 4.12 0.5 1.0 1.5 2.0 2.5 3.0 I I I I I I I I I 5? I I I ,I I I
, , \
8?
\
2:
()
c:::>
~
+
600 ·
c'
o o
c:::>
u-
\ Clay "shales
1000
2:
c:::>
<:..!)
UJ
,, ,, ,
\ \
1400
::c
<...:> ~
UJ
o (
,
\
I
~
a:::
\ I
1800
I
u-
\ I I
c:::> CL
c:::> ~
2200
3= c:::>
--.J UJ
2600
a:l
CL
UJ
14
I
15.3
::c ~
I I I I I I I
12?
I
, I
3000
C)
22.5
3400
26
••
I I I
Figure 12 d(101) spacing of opal-CT in porcelanite (circles) and cherts (triangles) from the Monterey Formation in Chico Martinez Creek, plotted against depth below the top of the Etchegoin Formation. (From Murata and Larson, 1975, fig. 6). Density data : crosses, diatomaceous shale; open circles, opal-CTporce/anite; half-filledcircles, porcelanite with quartz; filled circles, quartz chert; filled squares, shale; dashed line, densities of normally compacting shales.
r
secondary cristobalite peaks at 3.1 and 2.8'&' also appear with increasing burial approximately in the middle of the opalCT stage. While the shift in the d-spacing of the (101) peak is relatively rapid at the top and bottom of the opal-CT zone in the TemblorRange, it progresses very slowly in the middle stages (Figure 12). Nevertheless , it does proceed continuously over the entire burial range of the opal-CT zone. In contrast, Pisciotto (1981) found the shift to become more rapid in the Santa Maria ValleyofCalifornia with increasing burial depth, starting in the middle of the opal-CT zone. The bulk density of the sediments, on the other hand, does not show any systematic increase over this range of burial depths (730-2030 m below reference level). It stays more or less constant at an average value of 1.16 g'cm-3, but increases abruptly to 1.6g .cm-3 over an 80 m thick sediment interval in the opalCT to quartz transition zone (Figure 12). Oxygen isotope data from the same burial depths display a similar kind of behavior. There is no systematic change within the opal-CT zone (average value of o,ao = + 29.4%0 relative to SMOW), but a significant decrease by about 5%0 appears within a short distance below the porcelanite/chert boundary. Murata et al. (1977) concluded from these observations "that the structural ordering of cristobalite, manifested in changes of the d(101) spacing, occurred through reactions in the solid state" within the opal-CT stage, but involved complete dissolution-reprecipitation at the stage boundary with the quartz zone. As Isaacs at al. (1983) pointed out, however, " neither the structure of the mineral opal-CT nor the mineralogic significance of the d-spacing of opalCT are" sufficiently well understood to interpret the changes in the structure of opal-CT as an ordering process in the strict sense, although this is what most authors assume. For instance, the sharpen ing of the (101 )-cristobalite peak with increasing burial may reflect growing crystallite size of the opal-CT. The growing crystallite size in turn may result from gradual dissolution of smaller crystals and redeposition of the dissolved silicon on larger ones (an Ostwald ripening process). This may be considered a re-ordering process in a general, but not in a strict, sense. Highprecision X-ray diffractograms are sen-
c 240
Diagenesis
70
o
,
6!:-9-------J.,---'"'------:;';----~
68
•
!
67
CR Y STAL LINIT Y INDE X
sitive enough to depict such minute structural changes, whatever their physical cause, and therefore record these modifications continuously with burial. Standard oxygen mass spectrometry or pycnometry methods may not pick up small, but continuous , changes with progressive burial. Williams and Crerar (1985) and Williams et al. (1985) suggest that some of the seemingly abrupt mineralog ical or geochemical changes at the boundaries of the opal-CT stage may in fact be more gradual than they appear to be. Some of these changes occur so slowly that the less sensitive methods do not see them before a certain threshold level characteristic of a diagenetic stage boundary is reached. Effects will only be seen after a significant portion of the rock has undergone transformation. Alternatively, the oxygen-isotope composition may in fact change very little within
Geoscience Canada Reprint Series 4
the opal-CT stage because the isotopic ratio of growing larger crystals may be inherited from dissolving smaller ones. Density may also change very little because the morphology of the opal-CT blades remains essentially unchanged. These considerations would make the problem amenable to interpretations based entirely on dissolution-reprecipitation reactions. This would eliminate the need for solid-state reactions, which at the low temperatures of diagenesis are prohibitively slow. Crystallinity index of quartz. Sharpening of the (101) cristobalite peak with increasing maturation may be likened to the improvement in quartz crystallinity with progressive diagenesis described by Murata and Norman (1976). The quartz crystallin ity index of these authors is based on high-precision XRD measurements at slow scanning speeds (0.25°20min-1 ) of the high119'30'
119'3Z'30'
o ,,', h'l
o
10.0
.5
I
1km
o
4.11
"" 7.2
2.6
1.2
35'
35'
OS'
OS'
Figure 13 (a) X-ray diffractogram showing the quartz quintuplet at about 68 028, and the Parameters a and b used to define the quartz crystallinity index Cl of Murata and Norman (1976, fig. 2; for explanation see text). (b) Tracings of diffractograms showing increasing quartz crystallinity from bottom to top. Asterisk marks (212) peak at 6Z74 0 28 which is used for the determination of the index . CuK" radiation (from Murata and Norman , 1976, fig. 1).
r
c c ( (
(
C (
C C C C C C (
C C C C C C C C
C
5.8
< 1.0
(
C C C C C C C
c
e (
c 119'3Z' 30'
119'30'
Figure 14 d(101)spacing isopleths for opal-CTporcelanite in the TemblorRange southwest of Taft (California) . These provide an example of " isograd " mapp ing in low- to intermediategrade diagenetic rocks. Note apparent congruency between structural features and diagenetic trends. (From Murata and Randall, 1975).
(
c (
l. l l
c c' c' e, (
( (
('
c (
c
c co (
c.
Diagenesis of Biogenic Siliceous Sediments
angle region between 67 and 69% (Cu K", radiation). In this 20 range, moderately to well-crystallized quartz produces a quintuplet of XRD peaks. Of these the (212) peak at 67.74° measures the effectsoft he recrystallization ofcryptocrystalline to microcrystalline authigenic quartz during diagenesis and low-grade metamorphism. The quartz crystallinity index most probably is a measure of crystal size. It is defined as CI = 10Fa/b where a is peak height on the high-angle side, b is the total peak height (as shown in Figure 13a)and F is a scaling factor which varies with instrumental settings and adjusts the crystallinity index to a scale of 0 to 10. The crystallinity of authigenic quartz is characteristically poor (e.g., 2.0 to 3.2 in the Monterey Formation), even in quartz
o c o c c c C· ('
(b)
:"~lA 4.01
4.09 4.11
(
L...-
J[
cherts sensu stricto. Cretaceous quartz cherts in the West Pacific at about 400 m subsurface depth have crystalIinities generally less than 1 (using a scaling factor of 1.36, Pisciotto, 1980). Franciscan cherts show considerably better crystallinities (Figure 13b), but reach high values (above 8.0) only in metamorphosed rocks (Murata and Norman, 1976). Opal-CT d(101) spacing and quartz crystallinity as mapping tools for diagenetic grade. Variations in the d(101) parameter of opal-CT have been used to map regional maturation levels in relatively low-grade diagenetic rocks (Murata and Randall, 1975)comparable to isograde mapping in higher-grade metamorphic terrains. The contourlines of iso-d-spacings in the Monterey Formation of the Temblor Range of Cali-
QUARTZ
-'-
241
.~
A' b
~
,;,.....---J
o
c o o
( C)
("
L (
8· r Ir,
c I'
'-..-
c
Figure 15 (8) Geologic cross-section with reconstructed d(101) spacingprofiles for opa/-CT in the Temblor Range showing near-concordant relationship between folding and maturation levels. For location of section A-A' see Figure 14. (From Murata and Randall, 1975). (b) d(101) spacings of opa/-CT in surface samples along line A-A'. Region marked "quartz" indicates replacement of porcelanite (opa/-CT) by quartz-chert, where d(101) dropped below the lower limit of 4.04"- for opal-CT. (c) Schematic representation of discordant structural and maturation patterns. (From Mizutani, 1977). Dashed lines, d(101) isopleths; continuous lines, bedding. Note that the reconstructed d(101) isolines in (a) dip less steeply than the geologic structures, suggesting thatopa/-CT maturation was still in progress when tectonic deformation started. Alternatively delayed crystallographic changes might indicate lithologic control.
fornia generally conform to the geological structures (Figure 14). The lowest d-spacings (4.05 to 4.04~) straddle the axes of anticlines. However,as Mizutani (1977)noticed, the dip of the reconstructed iso-d-spacings in the cross-sections of Figure 15 is less than the structural dip. This probably indicates that structural deformation in the Temblor Range started when opalCT maturation was still in progress (Figure 15c). Alternatively, as discussed below, discordant maturation levels with respect to tectonic structures could reflect lithological effects of the host rocks (Isaacs, 1982). Keller et al. (1985) used quartz crystallinityto map diagenetic and metamorphic grade in the Ouachita Mountains of Arkansas and Oklahoma and in a contact metamorphic aureole on the Isle of Skye, Scotland. Crystallinity was determined by mean apparent crystal size measured underthe SEM. Crystal size ranged from less than 1 /Lm for cryptocrystalline, anhedral quartz in nonmetamorphic terrains to more than 100 /Lm for polygonal, triple-point euhedral quartz in tremolite, diopside and forsterite-grade rocks. The contour maps outline the regional structural trends and depict areas of subsurface igneous activity. RATE CONTROLLING FACTORS OF THE DIAGENETIC SILICA TRANSFORMATIONS Rates of the Opa/-A to Opa/-CT Transformation in Deep-sea Environments: Temperature and Time. Temperature is the dominant rate-determining factor for the diagenetic silica transformations as illustrated by the experiments of Ernst and Calvert (1969) for the opal-CT to quartz transformation (described above). This is also evident in the experiments by Kastner et al. (1977)for the opal-A to opal-CTtransformation (described below). The theoretical basis for this is the Arrhenius equation for the rate constant k of a chemical reaction k = A . exp(-Ea/RT) where Ea is the activation energy, T is temperature, R is gas constant, and A is a frequency factor. During the diagenesis of deep-sea pelagic sediments or silica-bearing shallow-water sediments, pressure effects are generally negligible compared with temperature or other effects. Pressure is generally hydro-
c 242
Diagenesis
Geoscience Canada Reprint Series 4
r
c static in these environments and varies only to a small extent over the .burial intervals of a few hundred metres in which the first silica transformations take place. What are the actual in-situ temperatures at which the opal-A conversion to opal-CT takes place? Downhole temperature measurements at DSDP sites 184 and 185 (leg 19) in the Bering Sea give temperatures of 35°-51°C at 500-600 m sub-bottom depths in the diatomaceous sediments where the bulk change occurs from opal-A to opalCT (Hein et al., 1978). In the Temblor Range of California, Murata et al. (1977) obtained oxygen isotopic temperatures for opal-CT from porcelanites of the Monterey Formation in the range of 41 0·56°C (mean 48°C). They used the fractionation factors of Clayton et al. (1972) and labeyrie (1974) assuming a /)1110 of 0%0 (SMOW) for the pore water from which the opal-Of precipitated. At a nearby locality, an interstitial water /)'B() of + 3.7%0 had to be assumed to obtain the same range of temperatures. These estimates correspond well with a temperature of 50°C calculated on the basis of geothermal gradients by the same authors for the marine sect ion of the Temblor Range where the transformation to opal-CT allegedly took place at a sub-surface depth of 700 m. A considerably lower range of temperatures (9°-27°C) was estimated by Pisciotto (1981). This range applies strictly to the top of the opal-CT zone (defined as the depth where opal-CT exceeds 5 weight percent (wt. %)) in the Monterey Formation of the Santa Maria Valley and is an absolute minimum range. It is based on detailed considerations of the thermal and subs idence history of the region. Opal-CT formation started about 2 million years (m.y.) after deposition and reached a peak after 4 to 10 m.y. (the shorter time corresponding to higher heat flow). In the pelagic realm away from the mid-ocean ridges, the transformation occurs at shallow subsurface depths of a few hundred metres, where temperature rarely exceeds 20° -30°C. At these low temperatures, the transformations are very slow and the opal-A to opal-CT conversion takes some 10 m.y. to go to completion. For temperatures well below 30°C (Figure 16), the conversion would take up to 30 or 50 m.y. The transformation can be traced regionally
on seismic reflect ion profiles because it is associated with the transition from siliceous ooze to porcelanite, which is a major regional lithification event. In the Northwest Pacific , for example, this event is marked by the occurrence of an "opaque" seismic reflector in the subsurface. It first appears at the meridian of Hawaii (at approximately 155°W, Figure 17) in mid-Tertiary siliceous sediments (15-20 m.y. old) which rest on 90 m.y. old Upper Cretaceous oceanic crust. The reflector increases in thickness westward from less than 20 m to 200-300 m in the West Pacific. As numerous drill holes have shown (e.g., DSDP legs 7, 16, 17, 20, 32, 55, 61, 62, 89), the thickness of the reflector represents the combined thickness of both siliceous rocks and pelagic limestones. The "opaque layer" can be deciphered on seismic profiles as a distinct interval because it is underlain by a "lower seismically transparent layer" consisting of older, less brittle and relatively poorly consolidated clayey sediments. The latter correspond to the lower horizon of brown abyssal clay(-stone) that formed when the northern part of the Pacific
I
TERTIARY
20
plate was still located south of the equator and had first subsided below the CCl (see section on "pelagic stratigraphy model" p. 229). The "upper opaque layer" on North Pacific seismic profiles west of Hawaii thus records adiagenetic history that required an activation time of at least 15m.y. before it produced bulk physical changes . This illustrates the importance of the factor time, which may play the major role in diagenesis, if temperatures are low. Host-rock Lithology as Rate-controlling Factor. In addition to temperature and time, there are other factors which may promote or retard the diagenesis of siliceous sediments. Since Bramlette 's (1946) classical work on the Monterey Format ion numerous exceptions from the general maturation sequence "sil iceous ooze -+ porcelanite -+ quartz chert " have been recognized . For example , in the Monterey Formation, parcelanite and/or quartz chert beds occur within the zone of non-recrystallized diatomaceous mudstone and diatomite , and chert layers occur within the porcelan ite zone. Apparently, the dia-
I
CRETACEOUS
100
60
11.0
J U R. 160m.y.
.~!:~::~::~::~::~:: t'.t-,:.... :.... :: ••: ••••:.... :....:
....................... ... ... ... ... ... ... ... ..................... ......................... ....-.... .-.... ... ... .-.... .-.. -.....
....................... ..opal... - •••:: ••:: ••.-... . .-.-CT .-.-................. _. -.-..:.:..... :..•.. -.-.-... ................. ...-.....-..... .. .-.....-.--.... - . ..:---::.....:---•........ ::..:::..:::. ..•:?'.:.••.:..•.:..•. .............. ?~
~.:
_
:I:
~ ~
_
UJ
~.:
°80
••• '
..J 000(
IX: ~
ell
!1200
II::.::'::.:.:'::.::.••:•.• . , ' . ••• " ~
I
• • • •• • • • • c
)~~~~l~~~~~
- : •• : ....
:t::-.::•• ::•• ::
. li!.: •• ::••:••••::.•::••::.•
~titk'
':'::••:'::-,:.::-,:'::-,:'::.•:'::••::'
:;::.:.: :::.:: ::.::::.::::.::::.:: ::.:: :••::r, •• • ~... ::.::_.::.::::.::::.:: •.. - -- ::.::::.: :::.:::. -.. •::.:.:.:.:.:.:·:·:6:.:-:·:.:.:·:·:.:.:-:·:.:.:·:·:.:.:·:·:.:.:-:.:.:.:.:-:.:.:-:.:.:.:.:.:.::.:-:::.:.:::.:.:::-:.:::.:.:::.:.:: :-:.::~ _
1600 m
~ opal-A _ opal-CT(porcellanites) k~~~:::}:·:] quartz (cherts) A opal-A (rare opal-eTl E quartz B opal-A. opal- CT (access. quartz) A l'OU"lI opol-CT in CSiotom. • .cs lInclion Oeeonl C opal-A, opal-CT. quartz & l'OUng opol-CT in oll•••CS o.hn I Allontte oe.onl
o opal-CT,quartz
Figure 16 Age/burial fields for occurrences of opal-A , opa/-CT and authigenic quartz in drillholes of the Deep Sea Drilling Project. (From v. Rad, 1979; after Riech and v. Rad, 1979).
C
C C C
C C C C C C C C C (
C C C C C C C C C C C C l
C C C
c
e c c
C l l
C (
C
(
c
Diagenesis of Biogenic Siliceous Sediments
243
(
c c ( ( ( (
(
o o o o o o o o o o o o
genetic reactions leading to the silica transformations proceed faster in certain horizons than in adjacent ones. In other cases, the trans ition to quartz seemed to have occurred directly from opal-A without an intermediate opal-CT stage (e.g., Lancelot , 1973). As observed during early legs of the DSDP (e.g., v. Rad and Rosch, 1972; Lancelot, 1973; Keene, 1975), lithology of the host rocks has a distinct influence on silica transformations . For example , in clayey sediment sections, silica maturation had reached the opal-C'T stage of porcelanite, whereas in accompanying calcareous sediments of the same age and thermal maturation level, quartz chert had formed. This led to speculations that, if the host rock is calcareous, opal-A might convert directly to quartz (quartz-prec ipitation hypothesis of Lancelot, 1973) without the intermediate opal-CT stage. Experiments by Kastner et a/. (1977) provided insight into the role of hostrock lithology and related solution chemistry for the opal-A to opal-CT transformation. In these hydrothermal experiments, diatom frustules and radiolarian tests were reacted with seawater or distilled water at 150°C in the 150'
120 '
o o o o o
presence of either calcareous ooze or montmorillonitic clay. The main results of these experiments can be summarized as follows : 1. In all experiments at 150°C, but not in all room-temperature experiments, corrosion of the siliceous tests took place after periods of one day to one month. 2. After 2 to 3 months, precipitation of opal-CT lepispheres was observed, but only in the experiments with seawater in the presence of calcite. In these experiments, embryonic opal-CT lepispheres could be detected after only one day. Precipitation of the lepispheres occurred preferentially on the surfaces of foraminifera. 3. In the presence of calcite, the concentration of dissolved silicon decreased compared with those without calcareous ooze, in which no opal-CT precipitation was observed. Particular attention was paid in the experiments on the effects of Mg(OHb for the precipitation process because of the known role of magnesium hydroxide as a coagulant in silica precipitation. Apparently, the presence both of Mgions and hydroxyl-groups is essential for opal-CT crystallization, as shown by further results of Kastner et a/. (1977): 180'
4. In each of the experiments in which opal-CT crystallization occurred , the concentration of Mg2+ and (OH)- was lowered in the ratio 1:2, parallel with the decrease in the concentration of dissolved silicon . 5. Newly formed embryonic opal-CT lepispheres always contained trace amounts of Mg. 6. In the absence of Mg (e.g., in experiments with art ificial seawater lacking Mg), no opal-CT crystall ization was observed during the experiment, even though calcite was present. 7. In experiments with initial borate alkalinity instead of bicarbonate alkalinity, opal-CT lepispheres formed in the same manner as in the presence of bicarbonate.This indicatesthat it isthe hydroxyl ion which is involved in the transformation and not the bicarbonate ion. 8. In experiments with montrnorillonlttc clays, dissolution of siliceous tests took place, but no reprecipitation of opal-CT occurred. However, small amounts of an Mg-rich clay formed. The results highlighted the signifi cance of a magnesium hydroxide compound (MHC). This compound, MHC, attracts silanol groups with their high negative surface charge and causes 120'
90'
Opaque layer th ickne.. 60'
~
~
~.~
20-200 meters 2oo-JOO meters >300 meters
o JO' !!l...---+---.~.&
120'
150'
Figure 17 Thickness of seismic-stratigraphic units in the Pacific . (From Wyllie, 1971; after Ewing et aI., 1968). Isopach lines for upper transparent layer in tenths of seconds reflection time (two-way travel time, i.e., each contour represents approximately 100metres).
Diagenesis
244
them to precipitate and nucleate as opal-CT lepispheres. The MHC has been postulated to be the mineral sepiolite (Kent and Kastner, 1985).OpalCT precipitation apparently is more rapid than sepiolite growth once the nuclei have formed, therefore opal-CT outgrows sepiolite . The role of calcium carbonate, which also dissolves in the process, is thought to resupply alkalinity consumed in the formation of the nuclei. Seawater supplies the necessary Mg. When the (OH)-groups have been exhausted, the rate of nucleus formation drops and the existing opal-CT embryos grow to well-developed lepispheres, provided continued opal-A dissolution supplies the required dissolved silicon. In experiments with mootmorillonitic clays, the clay minerals compete for the Mg and thus prevent
::L
0-
~
o
-,9
a
:J:
1500
-+r+ )(. +
•
Q.
500
I·
•
--+-
0_0_
50·C
1000
/~ '+
.
~~: 75·C
6000
1000
29 34
4000
27 32
2000
+----+ 00
10
100·C
30
~
<2:
-.
25 30
-+-+ - A.- 30 3000 +++ ' 000 __ - - +
20 20
?OOO
15 10
1000
. . ...
+-
+,,+-+
\0""0-0
l25·C 0-
10 0
0 A
24 21 20 17
4000
I
5000
-,9 +-
10
0
20
30
\
20 20 2000
22 10
+
Opal-A I3-Cristobalite H4S i04 Mg
0
Alkalinity
10
30
20 20
c
DAYS
~
4000
A.
.....-.-----; o'+,+-+
+
'°'°_0- 0 -
2000
c c c
c
c c (
10
OL..----'---.L..--~
-
c c c c c c c c c
c 30
75·C 10 0
c c
c
10
1)000
•
°'°'0 - 0 - - - - _ 0 -
15 10
30
,9
.-
,~--.
DAYS
A
10
A
150·C
0
20
Q•• _·T_+~...!!+_
• __ e ~ . -
I"
20
o L..-_---.1.._ _-'-_---'-' b
25 30
30
0 __
50'C
e
~+18 0 20
C'
~
(
26 20
0 ' " - 0- - 0-
-,9
32 29
-.e.- 30
'+
.-
..........--.
-A
.-.-
-A
d
31 36
c
3000
¢
:J:
28 25
0
5000
(J)
10,000
•
~
<2:
25 30 F
......--.-
'0
2000
~
~
0
0-
-,8
o
C'
.------..
b 3000
41
E E
.(J)
0
periods of 1 to 30 days (Figure 18). In all experiments , Mg and (OH) concentrationsdecreased in the 1:2ratio established previously, and opal-CT lepispheres or embryonic lepispheres formed. This confirmed that the MHC serves to nucleate the opal-CT. In experiments with low-surface area silica (i.e., Eocene Radiolaria with a specific surface area of 5.3 m2'g-') , the opal-A dissolut ion rate apparently was the rate-controlling step at temperatures of 50 0 , 75 0 , and 100°C. At these temperatures, the concentration of dissolved silicon remained below the equilibrium solubility of l3-cristobalite. Solubility data are only available for this phase which most closely approximates a " poorly ordered opal-CT". Nuclei formation and growth of opal-CT apparently kept pace with opal-A dissolution
~ <,
~
1000
the formation of the magnesium-hydroxide nuclei. Consequently, no opal-CT formed during the duration of the experiments (up to 6 months) . Opal-CT Precipitation as Rate-limiting Factor. Extending these hydrothermal experiments, Kastner and Gieskes (1983) determined the rate-limiting steps in the dissolution-reprecipitation process. The overall reaction rate is a function of the rates of dissolution, nucleation, and growth of the silica phases involved. Which one of these mechanisms will be the rate-limiting factor depends on specific chemical and physical conditions. Amorphous silica of varying specific surface area was placed in aqueous solutions of MgCI2 (0.03M) and NaHC03 (0.03M) and heated to temperatures of 50 0 , 75 0 , 1000 , 1250 , and 150°C for
Geoscience Canada Reprint Series 4
c c c c c c c c c
30 15 20 10 10
5
c c c c c e
100·C
o
10
20
30
DAYS
Figure 18 Changes in dissolved silicon, magnesium and alkalinity as a function of time and
Figure 19 Changes as in Figure 18, how-
temperature in hydrothermal experiments with Eocene Radiolaria (specific surface area of 5.3 mr- g-1). -+{3 solubility of a-cristobalite; -+A solubility of opal-A. (From Kastner and Gieskes, 1983, fig . 1).
ever for Ludox silica (specific surface area of 61.5 m 2 .g-1). (From Kastner and Gieskes, 1983, fig. 2).
c c
C r '-.
C' C C ( (
r
o o o o
Diagenesis of Biogenic Siliceous Sediments
at these temperatures . At 125° and 150°C, however, opal-CT precipitation became the rate-limiting step after a few days , because the concentration of dissolved silicon rose above the equilibrium solubility of ,s-cristobalite (Figure 18). In experiments with higher-surface area silica (i.e., Ludox silica with a specific surface area of 61.5 m2 • g"), the same behaviour was found at all temperatures studied: ,s-cristobalite solubility was rapidly exceeded; but opalA equilibrium solubility concentrations were not attained. This again indicated that opal-CT precipitation was the ratelimiting. process (Figure 19). These experiments also clearly illustrate the dominant role of temperature as ratecontrolling factor in the conversion of opal-A to opal-CT. The Opa/·CT to Quartz Transformation: Temperature and Rate. Temperature estimates range from 55° to 110°Cfor the opal-CT to quartz transformation in the Monterey Formation. These estimates are based on oxygen
245
isotope data and heat-flow considerations (Murata et a/., 1977; Murata and Larson, 1975; Pisciotto, 1981). The actual temperature at which the transformation occurs at a particular locality is a complex function of many factors, including the previous opal-CT maturation history. Pisciotto's (1981) temperature estimates for the base of the opalCT zone in the Monterey Formation (35°-61°C), are also considerably lower than those of other authors as they were for the top of this zone. As discussed previously, opal-CT precipitation is the rate-limiting factor in the opal-A to opal-CT conversion, when dissolution of immature, high-surface area opal-A is involved. Under such conditions, the concentration of silicon in solution will eventually approach the theoretical equilibrium solubility value for opalCT,which is a function ofthe surface area ofthis phase for a giventemperature (see section on Ostwald processes). Lowered silicon concentrations are also required to understand the re-
RELATIVE DETRITAL MINERAL CONTENT, WEIGHT %
70
50
30
10
w a:
::>
Opal-A
....
c, ~
w
....
Cl Z (/)
a:
u
z
Quartz
30
50
70
90
RELATIVE SILICA CONTENT, WEIGHT % Figure 20 Schematic summary of the role of host rock composition and temperature as controlling factors in the opet-cr maturation sequence. The initial d(101)spacing of opal-CThas been shown to vary inversely with detrital content (from Isaacs et aI., 1983; after Isaacs, 1982).
duced d(101)spacing of the initial opalCT inclayey host sediments. Inthe Monte· rey Formation, Isaacs (1982) observed that the first opal-CT to appear in a progressive burial sequence has a d(101) spacing which varies inversely with detrital mineral content of the host sediment (Figure 20). This observation confirms the results of Murata and larson (1975) from the Chico Martinez Creek study which showed that, at any given burial depth, the opal-CT d-spacing in (detritus-rich) porcelanite is 0.004 to 0.015,a, smaller than in associated (detritus-poor) quartz chert (Figure 12). This indicates retardation of opal·CT nucleation in clayey sediments because of the competition for MHC by clay minerals, consistent with the results of Kastner et a/. (1977). Furthermore, it indicates delayed growth of opal-CT until silicon solubilities "in equilibrium" with a lower surface area opal-Of (of a d-spacing of 4.08,a, or less) are encountered . Such reduced silicon concentrations in clay-rich host sediments can be accomplished most easily by adsorption of silica on detrital minerals (Siever and Woodford, 1973). Lowered silicon concentrations can also be achieved by the neoformation of silica-rich clay minerals, provided their nucleation and growth rates are faster than those of the silica phases that could potentially form. These and other possible scenarios are discussed in more detail by Williams et st. (1985). One would expect from the previous discussion that the detrital mineral content , while retarding the opal-A to opalCT transformation, would actuallyenhance the opal-CT to quartz transformation . The findings of Isaacs (1982) summarized in Figure 20 suggest that this is actually the case. In sediments with more than 70% detrital minerals the initial opal-CT has a d(101) spacing of 4.08,a, or less. In these sediments, the transformation to quartz should occur earlier than in sediments with less than 30% detrital minerals with an initial d(101)spacing of 4.11,a,. This is because reduction of the opal-CT d-spacing to values below 4.07,a, (4.09,a, in certain Cretaceous cherts from the West Pacific ; Pisciotto, 1980) is a prerequis ite for quartz precipitation. Such an inference, however, is at variance with the earlier observation (i.e., Lancelot, 1973) that the silica maturation sequence seems to be accelerated in calcareous host
c 246
rocks, which leads to the relatively early formation of quartz chert. The ent ire silica maturation sequence in clayey sediments is delayed by the initial retardation of the opal-A to opal-O'f transformation. The ambiguity between the conclusions of Isaacs (1982)and those of Lancelot (1973)may be resolved, if it is assumed that the delay cannot be compensated for by later acceleration due to an abbreviated opal-CT stage. Although the opal-CT stage may in fact last longer in carbonates than in clay-rich siliceous sediments, (contrary to Lancelot's (1973) assumption), this may be outweighed by a considerably earlier initiation of the transformation sequence in carbonates . The field relationship between detrital content and opal-CT transformation rates, reported by Isaacs (1982) from the Santa Barbara coast, is not observed everywhere in the Monterey Formation. Pisciotto (1981), for example, found the appearance of opal-CT as well as quartz at lower temperatures in the Santa Barbara basin to be associated with lower detritus contents. In the most recent study of the problem, the role of organic matter for the transformation reactions was investigated (Hinman, 1987). Results suggest that in organic-matter rich sediments the rate of silica diagenesis should be reduced. Organic acids released byorganic-matter maturation would dissociate and lower the pH thereby reducing carbonate alkalinity. This in turn would slow down the transformations by decreasing the silica polymerization rate. Although lowering of the pH will promote carbonate dissolution and thus buffer the pH, this effect may be nullified, if little carbonate is available or accessible for dissolution . In this case, no effect of carbonate content on the silica transformation rates may be seen. as in Isaacs' (1982) study. The initial opal-CT formed in organic-matter rich sediments should have a lower d(101) spacing because of a slower nucleation and growth rate under decreased carbonate alkalinity. In so far as organ icmatter content may be positively correlated with detrital mineral content, Hinman's (1987)results are in accordance with Isaacs' findings . However, Hinman (1987)did not find evidence that an initially lower d(101) spacing of opal-CT in organic-matter and detritus-rich sediments also accelerates the subsequent
Diagenesis
transition to quartz. In contrast, in the Santa Maria basin, quartz formation seemed to be slowed down because it occurred at higher temperatures in porcelanite than in chert. Absence of Opal-CT as an Intermediate Metastable Phase. It seems unlikely that the opal-CT stage can be bypassed entirely in the normal diagenetic silica maturation process, except under special conditions. The reasons why the diagenetic evolution path opal-A --+ opal-CT --+ quartz is observed in most situations have been outlined in the previous section on reaction kinetics and Ostwald's step-rule. Rareexceptions seem to be restricted to micro-environments such as the cavities of foraminifera (e.g., Keene, 1975) or Radiolaria in calcareous sediments. These are often filled with chalcedony, while the host rocks are calcareous opal-CT rocks or diatomaceous (opal-A) shales (Isaacs , 1982). This indicates anomalously early quartz (i.e., chal cedony) precipitation. Precipitation of the chalcedony occurred in spherical open cavities indicating limited overburden weight and very shallow burial. This chalcedony does not appear to have had an opal-CT precursor in the cavities, because elsewhere in the host rock opal-CT is still present which shows nosign thatthe transformation to quartz has started. Most likely, the chalcedony formed directly from opal-A, possibly because the concentration of dissolved silicon in the microenvironment of the cavities, for some unknown reason, remained below a-cristobalite equilibrium solubility during the early stage of opal-A dissolution. Among the mechanisms to maintain low silica concentrations in pelagic carbonates discussed by Williams et at. (1985), the early precipitation of calcium zeolites is a possibility. It seems that as of yet no specific studies have been directed at this problem . The early diagenetic stabilization of Radio/aria by chalcedony might be the key for the occurrence of Jurassic radiolarites as ribbon cherts which appear to be absent or scarce in younger, diatom-dominated siliceous sediments. Hinman (1987) suggested that direct quartz precipitation may be possible in the absence of Mg2+ , if sufficient alkalinity is available. Mg2+ ions in solution appear to have an inhibiting effect on quartz precipitation. This behaviour of
Geoscience Canada Reprint Series 4
quartz is opposite to that of opal-CT. Opal-CT crystallization is promoted in the presence of Mg2+ (and high alkalinity). Summary: The "Stability" Fields of the Various Silica Phases in Deep-sea Diagenetic Environments. The distribution of the various silica phases in oceanic sediment sequences has been summarized by Riech and v. Rad (1979) on the basis of available DSDP data (Figure 16).There is considerable overlap on this depth/age graph between the fields for the three main silica phases. Opal-A has been found in sediments as old as 85 Ma (Late Cretaceous), but only at shallow subsurface depths. With increasing depth (and temperature) its maximum survival time is shortened significantly. For example, at 1,000 m subsurface depth, opal-A is normally no longer present in sediments older than about 20 Ma. Opal-CT takes a minimum of about 10 m.y. to first appear, requiring elevated heat-flow.Otherwise its formation takes even longer. Exceptionally young opalCT has been observed by Weaver and Wise (1973) in Pliocene sediments (less than 5 Ma), where it occurred adjacent to igneous dykes and sills. On the other hand, opal-CT may still persist in sediments 100 to 120 m.y. old. Little new opal-CT forms, however, in pre-Tertiary sediments (i.e., sediments olderthan 65 Ma), because very little source material (opal-A) is left in those sediments. With age and increasing burial, opal-CT undergoes the progressive crystallographic structural changes in the d(101) spacing described earlier. No opal-CT has been found in pre-Cretaceous sediments (older than 144 Ma). In Figure 16,the quartz field overlaps the opal-CT field, and to a slight degree the opal-A field. This overlap illustrates how local variations in heat flow, pore fluid Chemistry, host-rock lithology and other factors can affect the rate of the diagenetic transformations . Quartz cherts need a minimum of 30 to 40 m.y. to form at burial depths of 500 m or more, and considerably longer at shallower depths . Genuine quartz cherts, consequently, are rare in Cenozoic pelagicsediments, but becomepredominant in Lower Cretaceous and older siliceous rocks. They are the exclusive lithology of siliceous sediments in the Paleozoic and Precambrian.
c c c
c c c c c c c c c c c
c c c c c c c c c c c c c c c c c c c
e c
c c c
c <.. (
(
(
r
Diagenesis of Biogenic Siliceous Sediments
r
c (J (
c c o c /
o c o o o o
PHYSICAL DIAGENESIS OF SILICEOUS SEDIMENTS The treatment of the diagenesis of siliceous sediments would be incomplete without a discussion of phenomena of phys ical diagenesis in cherts. A stepwise compaction history is characterist ic of siliceous sediments during burial, which contrasts with the more gradual compaction history of argillaceous sediments. This is implicit in the constant bulk density of Figure 12 for the opal-CT stage and reflects the rap id crystallographic changes associated with the phase changes at the stage boundaries. Other diagenetic features characteristic of chert formations are dykes, diapirs, and breccias that have long been described in the literature (e.g., Taliaferro, 1934).They attest to differential compaction and variable rates of silica recrystallization and cementation. In a given sequence of siliceous sediments, some layers may already be
Iithified or semi-lithified while others are not. Shaking during an earthquake may rupture the lithified layers and cause injection of unlithified sediment into the fractures from adjacent uncemented layers (see numerous illustrations in Snyder et a/., 1983), part icularly if these had been overpressured. The opal-A to opal-CT and the opal-CTto quartz transformations are dehydration reactions. They may generate excess pore pressure and cause hydraulic fracturing . The examples shown in Figure 21 are Cretaceous radiolarian porcelanites from the West Pacific east of the Marianas deep-sea trench/island arc. The dykes probably reflect seismic activity assoc iated with the Marianas subduction zone . The brittle host sediment is porcelanite consisting predominantly of opal-CT with chalcedony-filled radiolarians. The fractures and dykes contain porcelanite fragments and finegrained opal-CT sediment cemented by microquartz and chalcedony. Some of
247
the microquartz seems to have recrystallized in place from opal-CT. The folded dyke in Figure 21 a behaved as a more competent layer, while the surrounding host sediment was still undergoing compaction. Steinitz (1970) and Snyder et at. (1983)invoke similar mechanisms of dyke and vein formation. Paris et at. (1985) ascribe chert breccias and veins in the Archean Barberton greenstone belt in South Africa to the injection of hot, high-pressure hydrothermal fluids related to volcanism. Some chert breccias, however, are of normal epigenetic, synsedimentary origin, such as the ones in the Burlington Limestone of Missouri. According to Carozzi and Gerbe r (1978), these formed due to storm action on a Mississippian carbonate flat. They provide evidence for the early-diagenetic origin of some chert-nodules in limestones, which will be discussed in more detail in the accompanying paper.
J
r-... ~
o
V
o o o o C)
o r
Figure 21 (a) (left) Folded chert dyke in Lower Cretaceous radiolarian porcelanite, West Pacific . Porcelanite fragments in the dyke are cemented by microquartz and chalcedony which enclose considerable amounts of opal-CT sediment , except where the cement is very lightcoloured (especially at the dyke margins). Folding of the microdyke was caused by differential compaction. The pure chalcedonic cement at the dyke margin was probably precipitated into pore space that was generated by shrinkage of the host sediment during ongoing compaction and was probably the latest cement phase . Height of thin-section approximately 2 em. DSDP site 20-198A, sample scc .nt. (b) (right) Multiple fractures filled with injected opal-CT sediment and fragments which were cemented by microquartz and chalcedony. Marg ins and ends of the finer fractures are light coloured and consist almost entirely of microquartz or chalcedony cement which probably was the last cement. Height of section approximately 2.5 cm. DSDP site 20-198A. sample Scc, #4.
248
CONCLUSIONS The century-old chert problem has been resolved as a result of chemical oceanographic and sedimentological studies made possible largely due to drilling in the deep oceans. Siliceous pelagic sediments which coverwide areas ofthe deep-sea floor do not originate directly from submarine volcanic and hydrothermal activity. Chemical precipitation as a silica gel or silica extraction by deep-water radiolarians (Davis, 1918) do not contribute significantly to the formation of Phanerozoic siliceous sediments. The silicon supplied from various sources is involved in an ocean-wide cycle. Siliceous organisms extract silicon from nutrient-rich surface waters in regions of oceanic upwelling. A small fraction (less than 10%)of this biogenic silica is removed by sedimentation, the remainder stays in the cycle being redissolved during settling and exposure on the ocean floor. Dissolved silicon is re-used by organisms up to a 100times for shell construction. Porcelanites and cherts under the ocean floors, which represent an essential element of the pelagic stratigraphy, develop from the biogenic siliceous oozes through a two-stage diagenetic transformation involving the transition from opal-A to opal-CT and the transition from opal-CT to quartz. Both transformations are dissolution/reprecipitation processes. Radiolarian cherts of ophiolite sequences are ancient equivalents of these deep-sea cherts transferred to orogenic belts by obduction (tectonic accretion during subduction or collision) along active margins. Occurrence of these Mesozoic and older radiolarites as ribbon cherts, which seem to be absent in Cenozoic deposits, may be related to the early diagenetic precipitation of chalcedony in Radiolaria. Solving the principal aspects of the chert problem has also brought into focus some unresolved detailed questions of silica diagenesis. What is the mineralogic significance of the change in the d(101)spacing of opal-CT during progressive burial diagenesis? Why is there little or no change in the isotopic composition of opal-CT during burial although recrystallization reactions seem to occur throughout the opal-CT stage and are not restricted to the stage boundaries. The same question may be raised for the opal-A and the quartzchert stages of silica diagenesis, but
Diagenesis
does not seem to have been studied. What is the chemical significance of organic matter in affecting the rates of silica transformations? What is the role of trace elements forthe transformation rates, and what is the role of some of the major seawater constituents besides Mg and alkalinity, such as sulphate? The influence of host-rock lithology on the diagenetic silica transformations also requires further study. Results of Isaacs (1982) suggest that detrital content accelerates the opal-CT to quartz transformation while it has a retarding effect on the opal-A to opal-CT transformation (Kastner et a/., 1977). However, the overall maturation seems to be faster in carbonates than in clayey sediments (Lancelot, 1973). These are challenging questions for further research on silica diagenesis which have direct applications in petroleum exploration and basin analysis.
ACKNOWLEDGEMENTS Funding forthis review and the author's research on silica diagenesis came from the Natural Sciences and Engineering Research Council of Canada. Valuable critical comments streamlining the manuscript were provided by S. Calvert, E. McBride, A. Mucci and especiallyT. Barrett and E. Burton. U. v. Rad made available the originals for most of the photomicrographs included in this review. R. Yates helped with the photography and drafting. Their support is gratefully acknowledged.
REFERENCES General These cover the sections Introduction, Sources of non-detrital silica in siliceous sediments, Biogenic siliceous oozes, Pelagic stratigraphy, Formation of bedded chert, Shallow water chert, and Conclusions. Baltuck, M., 1983, Some sedimentary and diagenetic signatures in the formation of bedded radiolarite, in lijima, A., Hein, J.R. and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p.299-315. Barrett, T.J., 1981, Chemistry and mineralogy of Jurassic bedded chert overlying ophiolites in the North Apennines, Italy: Chemical Geology, v. 34, p. 289-317. Bates, R.L. and Jackson, J.A., 1980, Glossary of Geology, Second Edition: American Geological Institute, Falls Church, Virginia, 749 p.
Geoscience Canada Reprint Series 4
Berger, W.H., 1970, Biogenous deep-sea sediments: fractionation by deep-sea circulation: Geological Society of America, Bulletin, v. 81, p. 1385-140t Berger, W.H. and Winterer, E.L., 1974, Plate stratigraphy and the fluctuating carbonate line, in HsO, K. and Jenkyns, H.C., eds., Pelagic Sediments on Land and underthe Sea: International Association of Sedimentologists, Special Publication, v. 1, p. 11-48. Bossellini, A. and Winterer, E.L., 1975, Pelagic limestone and radiolarite of the Tethyan Mesozoic: A genetic model: Geology, v. 3, p. 279-282. Bramlette, M.N., 1946,The MontereyFormation of California and the origin of its siliceous rocks: United States Geological Survey, Professional Paper 212, 55 p. Calvert, S.E., 1966a,Accumulation of diatomaceous silica in the sediments of the Gulf of California: Geological Society of America, Bulletin, v. 77, p. 569-596. Calvert, S.E., 1966b, Origin of diatom-rich sediments from the Gulf of California: Journal of Geology, v. 74, p. 546-565. Calvert, S.E., 1968, Silica balance in the ocean and diagenesis: Nature, v. 219, p.919-920. Calvert, S.E., 1983,Sedimentarygeochemistry of silicon , in Aston, S.R., ed., Silicon Geochemistry and Biogeochemistry: Academic Press, New York, p. 143-186. Carver, R.E., 1980,Petrology of PaleoceneEocene and Miocene opaline sediments, southeastern Atlantic coastal plain: Joural of Sedimentary Petrology, v. 50, p. 569-582. Cavaroc, Wo, Jr. and Ferm, J.C., 1968, Siliceous spiculitesas shoreline indicatorsin deltaic sequences: Geological Society of America, Bulletin, v. 79, p. 263-271. Chipping, D.H., 1971, Paleoenvironmental significance of chert in the Franciscan Formation of western California: Geological Society of America, Bulletin, v. 82, p. 1707-1712. Dapples, E.C., 1967, Silica as an agent in diagenesis, in Larsen, G. and Chilingar, G.V., eds., Diagenesis in Sediments: Developments in Sedimentology, v. 8, p.323-342. DeMaster, D.J., 1981, The supply and accumulation of silica in the marine environment: Geochimica et Cosmochemica Acta, v. 45, p. 1715-1732. Diersche, V., 1980, Die Radiolarite des Oberjura im Mittelabschnitt der Nordlichen Kalkalpen: Geotektonische Forschungen, v. 58, 217 p. Edmond, J.M., Measures, C., McDuff, R.E., Chan, L.H., Collier,R., Grant, B., Gordon, L.1. and Corliss, J.B., 1979a,Ridge crest hydrothermal activity and the balancesof the major and minor elements in the ocean: The Galapagos data: Earth and Planetary Science Letters, v. 46, p. 1-18.
c c
c c c
c c c c c c c c c .C
C C C (
C C C C C
C C
C C E C C
C C
c
e <--
c C l
C
C
c
r ,r
Diagenesis of Biogenic Siliceous Sediments
r:
249
r '-
c C' (
o o (
o J
J
""\
J
r:
Edmond, J.M., Jacobs, S.S., Gordon, A.L., Mantyla, A.W. and Weiss, R.F., 1979b, Water column anomalies in dissolved silica over opaline pelagic sediments and the origin of the deep silica maximum: Journal of Geophysical Research, v. 84 (C12), p. 7809-7826. Folk, R.L.andMcBride,EE, 1978, Radiolarites and their relation to subjacent "oceanic crust" in Liguria, Italy:Journal of Sedimentary Petrology, v. 48, p. 1069-110t Garrels, R.M., 1987, A model for the deposition of the microbanded Precambrian iron formations: American Journal of Science, v. 287, p. 81-96. Gibson, T.G. and Towe, K.M., 1971, Eocene volcanism and the origin of the horizon A: Science, v. 172, p. 154-154. Goldstein, A., Jr., 1959, Cherts and novaculites of Ouachita facies, in Ireland, H.A., ed., Silica in Sediments: Society of Economic Paleontologists, Special Publication No. 7, p. 135-149. Griffin, J., 1980, The effect of pressure on the solubility of 11 silicates in seawater at 20°C, pH 8, unpublished Master of Science thesis, University of Hawaii, Honolulu. Heath, G.R., 1969, Mineralogy of Cenozoic deep-sea sediments from the equatorial Pacific Ocean: Geological Society of America, Bulletin, v. 80, p. 1997-2018. Heath, G.R., 1974, Dissolved silica and deep-sea sediments, in Hay, W.W., ed., Studies in Paleo-oceanography, Society of Economic Paleontologists and Mineralogists, Special Publication, v.20, p.77-93. Hein, J.R. and Karl, S.M., 1983, Comparisons between open-ocean and continental margin chert sequences, in Iijima, A., Hein, J.R. and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p. 25-43. Hesse, R., 1990,Silica diagenesis: Origin of inorganic and replacement cherts, in Mcllreath, LA. and Morrow, D.W., eds., Diagenesis: Geological Association of Canada, Geoscience Canada Reprint Series 4, p. 253-275. Hesse, R., Foreman, H.P., Forristall, G.Z., Heezen. B.C., Hekel, H., Hoskins, R.H., Jones, EJ.W., Kaneps, A.G., Krashenninikov, VA, MacGregor, I. and Okada, H., 1974, Walther's facies rule in pelagic realm, a large-scale example from the Mesozoic-Cenozoic Pacific: Zeitschrift der Deutschen Geologischen Gesellschaft, v. 125, p. 151-172. Hurd, D.C., 1973, Interactions of biogenic opal, sedimentand seawater in the central Equatorial Pacific: Geochimica et Cosmochimica Acta, v. 37, p. 2257-2282. Hurd, D.C.,1983,Physicaland chemical properties of siliceous skeletons, in Aston, S.R., ed., Silicon Geochemistry and Biogeochemistry:AcademicPress,NewYork, p.187-244.
Imoto, N., Shimizu, D., Shiki, T. and Yoshida, M., 1974, Sole markings observed in bedded cherts from the Tarnba belt, Southwest Japan: Kyoto University of Education, Bulletin, v. B44, p. 19-26 (in Japanese). Isaacs, C.M., 1982, Influence of rock composition on kinetics of silica phase changes in the Monterey Formation, Santa Barbara area, California: Geology, v. 10, p. 304-308. Isaacs, C. M., 1984,The Monterey - key to offshore California boom: Oil and Gas Journal, v. 82, p. 75-81. Isaacs, C.M., Pisciotto, K.A. and Garrison, R.E., 1983,Facies and diagenesis ofthe Miocene Monterey Formation, California: A summary, in lijima, A., Hein, J.R. and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p. 247-282. Jenkyns, H.C. and Winterer, EL., 1982, Palaeoceanography of Mesozoic ribbon radiolarites: Earth and Planetary Science Letters, v. 60, p. 351-375. Jones, D.L. and Murchey, B., 1986, Geologic significance of Paleozoic and Mesozoic radiolarian chert: Annual Reviews of Earth and Planetary Sciences, v. 14, p. 455-492. Kastner, M., Keene, J.B. and Gieskes, J.M., 1977, Diagenesis of siliceous oozes. I. Chemical controls on the rate of opal-A to opal-CT transformation - an experimental study: Geochimica et Cosmochimica Acta, v. 41, p. 1041-1059. Lancelot, Y.,1973,Chert and silica diagenesis in sediments from the central Pacific, in Winterer, E.L., Ewing, J.1. et et., Initial Reports ofthe Deep Sea Drilling Project, v. 17, p. 377-405: United States Government Printing Office, Washington, D.C. Lane, N.G., 1981, A nearshore sponge spicule mat from the Pennsylvanian of westcentral Indiana: Journal of Sedimentary Petrology, v. 51, p. 197-202. McBride, E.E and Folk, R.L., 1979, Features and origin of Italian Jurassic radiolarites deposited on continental crust: Journal of Sedimentary Petrology, v. 49, p. 837-868. Mizutani, S. and Shibata, K., 1983,Diagenesis of Jurassic siliceous shale in central Japan, in lijima, A., Hein, J.R. and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p. 283-298. Nisbet, E.G. and Price, I., 1974, Siliceous turbidites: bedded cherts as redeposited ocean ridge-derived sediments, in Hsu, K. andJenkyns, H.C., eds., Pelagic Sediments on Land and under the Sea: International Association of Sedimentologists, Special PUblication, v. 1, p.351-366. Robertson, A.H.E, 1977, The origin and diagenesis of cherts from Cyprus: Sedimentology, v. 24, p. 11-30.
Sano, H., 1983, Bedded cherts associated with greenstones in the Sawadani and Shimantogawa Groups, Southwest Japan, in lijima, A., Hein, J.R. and Siever, R., eds., Silicous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p. 427-440. Schmid, E, 1986, Flint stratigraphy and its relationship to archeology, in Sieveking, G. deG.and Hart, M.B.,eds., TheScientific Study of Flint and Chert: Cambridge University Press, p. 1-5. Steinberg, M., Bonnot-Coutois, C. and Tlig, S., 1983, Geochemical contribution to the understanding of bedded chert, in Iijima, A., Hein, J.R. and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p.193-210. Steinmann, G., 1906, Geologische Beobachtungen in den Alpen. II: Die Schardtsche Oberfaltungstheorie und die geologische Bedeutung der Tiefseeabsatze und der ophiolitischen Massenge steine: Berichte der Naturforschenden Gesellschaft Freiburg, v. 16, p. 18-67. Steinmann, G., 1927, Die ophiolithischen Zonen in den Mediterranen Kettengebirgen: Congres Geologique International, Comptes Rendues, 14e session, Madrid, 1926, v. 2, p. 637-668. Weaver,C.E. and Beck, K.C., 1977, Miocene of the S.E United States: a model for chemical sedimentation in a peri-marine environment: Sedimentary Geology, v. 17, p. 1-234. Weaver, EM. and Wise, SW., 1974, Opaline sediments of the southeastern coastal plain and horizon A: Biogenic origin: Science, v. 184, p. 899-90t Wenk, E, 1949, Die Assoziation von Radiolarienhornsteinen mit ophiolithischen Erstarrungsgesteinen als petrogenetisches Problem: Experientia, v. 5, p. 226-232. Wermund, E.G. and Moiola, R.J., 1966, Opal, zeolites, and clays in an Eocene neritic bar sand: Journal of Sedimentary Petrology, v. 36, p. 248-253. Wise, S.W.andWeaver, F.M.,1973,Originof cristobalite-rich Tertiary sediments in the Atlantic and Gulf Coastal Plains: Gulf Coast Association of Geological Societies, Transactions, v. 23, p. 305-323. Stages of Burial Diagenesis in Biogenic Siliceous Sediments - Characteristics of the Silica Phases in Siliceous Sediments Bramlette, M.N., 1946,The MontereyFormation of California and the origin of its siliceous rocks: United States Geological Survey, Professional Paper 212, 55 p. Calvert, S.E, 1971, Composition and origin of North Atlantic deep sea cherts: Contributions to Mineralogy and P~trology, v. 33, p. 273-288.
250
Carver, R.E., 1980, Petrology of PaleoceneEocene and Miocene opaline sediments, southeastern Atlantic Coastal Plain: Journal of Sedimentary Petrology, v. 50, p. 569-582. Darragh, P.J., Gaskin, A.J. and Sanders, J.V., 1976, Opals: Scientific American, v. 264, p. 84-94. Ernst, WG. and Calvert, S.E., 1969, An experimental study of the recrystallization of porcelanite and its bearing on the origin of some bedded cherts: American Journal of Science, v. 267A, p. 114-133. Florke, OW., Hollmann, R., v. Rad, U. and Rosch, H., 1976, Intergrowth and twinning in opal-CT lepispheres: Contributions to Mineralogy and Petrology, v. 58, p.235-242. Hurd, D.C. and Theyer, E, 1975,Changes in the physical and chemical properties of biogenic silica from the central Equatorial Pacific. 1.Solubility, specific surface area, and solution rate constants of acidcleaned samples: Analytical Methods in Oceanography: American Chemical Society, Advances in Chemistry Series, v. 147, p. 211-230. lIer, A.K., 1955, The Colloid Chemistry of Silica and Silicates: Cornell University Press, Ithaca, New York, 324 p. lIer, R.K., 1973, Effect of adsorbed alumina on the solubility of amorphous silica in water: Journal of Colloid and Interface Science, v. 43, p. 399-408. lIer, A.K., 1979, Chemistry of Silica: WileyInterscience, New York, 866 p. Isaacs, C.M., Pisciotto, K.A. and Garrison, R.E., 1983, Facies and diagenesis ofthe Miocene Monterey Formation, California: A summary, in Iijima, A., Hein, J.R. and Siever, A., eds., Siliceous Deposits in the Pacific Region. Developments in Sedimentology, v. 36, p. 247-282. Jones, D.L. and Knauth, L.P.,1979,Oxygen isotopic and petrographic evidence relevant to the origin of the Arkansas Novaculite: Journal of Sedimentary Petrology, v. 49, p. 581-597. Jones, J.B. and Segnit, E.R., 1971, The nature of opal. 1.Nomenclature and constituent phases: Journal of the Geological Society of Australia, v. 18, p. 57-68. Krauskopf, K.B., 1956, Dissolution and precipitation of silica at low temperatures: Geochimica et Cosmochimica Acta, v. 10, p. 1-26. Lewin,J., 1961, The dissolution of silica from diatom walls: Geochimica et Cosmochimica Acta, v. 21, p. 182-198. Mallard, M.E., 1890, Sur la lussatite, nouvelle variete minerale cristallisee de silice: Bulletin de la Societe franc;:aise de Mineralogie et Cristallographie, v. 13, p.63-66. Meyers, WJ., 1977, Chertification in the Mississippian Lake Valley Formation, Sacramento Mountains, New Mexico: Sedimentology, v. 24, p. 75-105.
Diagenesis
Parks, GA, 1965, The isoelectric points of solid oxides, solid hydroxides, and aqueous complex systems: Chemical Reviews, v. 65, p. 177-198. Pisciotto, K.A., 1981, Distribution, thermal histories, isotopic compositions and reflection characteristics of siliceous rocks recovered by the Deep Sea Drilling Project, in Warme, J.E., Douglas, R.G. and Winterer, E.L., eds., The Deep Sea Drilling Project: A decade of progress: Society of Economic Paleontologists and Mineralogists, Special Publication 32, p. 129-147. Riech, V. and v. Rad, U., 1979,Silica diagenesis in the Atlantic Ocean: Diagenetic potential and transformations, in Talwani, M., Hay,Wand Ryan, WB.E, eds., Deep Drilling Results in the Atlantic Ocean: Continental margins and paleoenvironment: American Geophysical Union, Maurice Ewing Series, v.3, p.315-341. Tada, A. and lijima, A., 1983, Identification of mixtures of opaline silica phases and its implication for silica diagenesis, in lijima,A., Hein,J.R. and Siever, R.,eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p.229-245. Volosov, A.G., Khodarovskiy, I.L. and Ryzhenko, B.N., 1972, Equilibria in the system Si02-H 20 at elevated temperatures along the lower three-phase curve: Geochemistry International, v. 9, p.362-377. v. Rad, U., Riech, V. and Rosch, H., 1977, Silica diagenesis in continental margin sediments off Northwest Africa, in Lancelot, Y.,Seibold, E. etel., Initial Reports of the Deep Sea Drilling Project, v. 41, p. 879-905: United States Government Printing Office, Washington, D.C. Walther, J.V. and Helgeson, H.C., 1977, Calculation of the thermodynamic properties of aqueous silica and the solubility of quartz and its polymorphs at high pressures and temperatures: American Journal of Science, v. 277, p. 1315-1351. Weaver, EM. and Wise, S.W, Jr., 1972, Ultramorphology of deep sea cristobalitic chert: Nature, v. 237, p. 56-57. Williams, L.A. and Crerar, D.A., 1985,Silica diagenesis, II. General mechanisms: Journal of Sedimentary Petrology, v. 55, p.312-321. Williams, L.A., Parks, G.A. and Crerar, D.A., 1985,Silica diagenesis, I. Solubility controls: Journal of Sedimentary Petrology, v. 55, p. 301-311. Wise, S.W.,Jr., Buie, B.E and Weaver, EM., 1972, Chemically precipitated sedimentary cristobalite and the origin of chert: Eclogae Geologicae Helvetiae, v. 65, p.157-163.
Geoscience Canada Reprint Series 4
Nature of the Diagenetic Transformation Mechanisms between Different Silica Phases and Physical Diagenesis of Siliceous Sediments Berger, W.H. and v. Rad, U., 1972, Cretaceous and Cenozoic sediments from the Atlantic Ocean, in Hayes, D.E., Pimm, A.C. et al., Initial Reports of the Deep Sea Drilling Project, v. 14, p 787-954: United States Government Printing Office, Washington, D.C. Bramlette, M.N., 1946, The Monterey Formation of California and the origin of its siliceous rocks: United States Geological Survey, Professional Paper 212, 55 p. Carozzi, AV. and Gerber, M.S., 1978,Synsedimentary chert breccia: A Mississippian tempestite: Journal of Sedimentary Petrology, v. 48, p. 705-708. Christian,J.W, 1965,Thetheoryoftransformations in metals and alloys: Pergamon Press, Oxford, 975 p. Clayton, R.N., O'Neil, J.A. and Mayeda, T.K., 1972, Oxygen isotope exchange between quartz and water: Journal of Geophysical Research, v. 77, p. 30573067. Ewing, J., Ewing, M., Aitken, T.and Ludwig, W.J., 1968, North Pacific sediment layers measured by seismic profiling, in Knopoff, L., Drake, C.L. and Hart, P.J., eds., The Crust and Upper Mantle of the Pacific Area: American Geophysical Union, Geophysical Monograph, v. 12, p.147-173. Fournier, A.O., 1973, Silica in thermal waters: Laboratory and field investigations, in Ingerson, E., ed., International Symposium on Hydrogeochemistry and Biogeochemistry, Tokyo 1970, Proceedings, v. 1 (Hydrogeochemistry), p. 122-139: Clarke Co., Washington, D.C. Harrison, WE., Hesse, R., and Gieskes, J.M., 1982, Relationship between sedimentary facies and interstitial water chemistry in slope, trench and Cocos plate sites from the Mid-America Trench transect, active margin off Guatemala, Leg 67, DSDP,in Huene, A., Aubouin, J. et al., Initial Reports of the Deep Sea Drilling Project, v. 67, p. 603·614:United States Government Printing Office, Washington, D.C. Heath, G.A., 1969, Mineralogy of Cenozoic deep-sea sediments from the equatorial Pacific Ocean: Geological Society of America, Bulletin, v. 80, p. 1997-2018. Heath, G.R. and Moberly, R., Jr., 1971, Cherts from the western Pacific, Leg 7, Deep Sea Drilling Project, in Winterer, E.L., Riedel, WR.etal., Initial Reports of the Deep Sea Drilling Project, v. 7, p. 991-1007: United States Government Printing Office, Washington, D.C.
c c c c ( c c c c c c c c c c c c c c c c c c c c c c c c c c c c c
e (
c (.
c
c c (
c c c c c
r
r~
r.
C
C r
r )
"
t
c C l
C r' L
Diagenesis of Biogenic Siliceous Sediments
Hein, J.R., Scholl, DW., Barron, J.A., Jones, M.J.andMiller,J., 1978,Diagenesisof late Cenozoicdiatomaceousdeposits and formationofthe bottomsimulating reflectorin the southern Bering Sea: Sedimentology, v. 25, p. 155-181. Hesse, R., 1987, Selective and reversible carbonate-silica replacements in Lower Cretaceous carbonate-bearing turbidites of the Eastern Alps: Sedimentology, v. 34, p. 1055-1077. Hesse, R., Lebel, J. and Gieskes, J.M., 1985, Interstitial water chemistry of gas hydrate bearing sections on the MiddleAmerica Trench Slope, Deep Sea Drilling Project, Leg84, in Auboin, J., Huene, R.V. et al., Initial Reports of the Deep Sea Drilling Project, v. 84, p. 727-737: United States Government Printing Office, Washington, D.C. Hinman, N.W., 1987, Organic and inorganic chemical controls on the rates of silica diagenesis: A comparison of a natural system with experimental results, unpublished Ph.D. thesis, University of California, San Diego, 381 p. Hurd, D.C.,Wenkham,C., Pankratz, H.S.and Fugate, J., 1979, Variable porosity in siliceous skeletons: Determination and importance: Science, v. 203, p. 1340-1343. lIer, R.K., 1979, Chemistry of Silica: WileyInterscience, New York, 866 p. Isaacs, C.M., 1982, Influence of roc.lU<..omposition on kinetics of silica phase changes in the Monterey Formation, Santa Barbara area, California: Geology, v. 10, p. 304-308. Kastner, M.andGieskes,J.M., 1983,Opal-A to opal-CT transformation: A kinetic study,in Iijima, A., Hein, J.A., and Siever, R., eds., Siliceous Deposits in the Pacific Region: Developments in Sedimentology, v. 36, p. 211-227. Kastner, M., Keene, J.B. and Gieskes, J.M., 1977, Diagenesis of siliceous oozes. I. Chemical controls on the rate of opal-A to opal-CT transformation - an experimental study: Geochimica et Cosmochimica Acta, v. 41, p. 1041-1059. Keene, J.B., 1975, Cherts and porcelanites from the North Pacific, DSDP, leg 32, in Larsen, R. and Moberly, A. et al., Initial Reports ofthe Deep Sea Drilling Project, v. 32, p. 429-507: United States Government Printing Office, Washington, D.C. Keller, W.D., Stone, C.G. and Hoersch, W.D., 1985, Textures of Paleozoic chert and novaculite in the Ouachita Mountains of Arkansas and Oklahoma and their geological significance: Geological Society of America, Bulletin, v. 96, p. 1353-1363. Kent, D.B. and Kastner, M., 1985, Mg2+ removalinthe system Mg2+ - amorphous Si02 - H20 by adsorption and Mg-hydroxysilicate precipitation: Geochimica et Cosmochimica Acta, v. 49, p. 1123-1136.
Labeyrie, L.D., 1974, New approach to surface seawater paleotemperatures using 180 /160 ratios in silica of diatom frustules: Nature, v. 248, p. 40-42. Lancelot, Y.,1973,Chert and silica diagenesis in sediments from the central Pacific, in Winterer, E.L., Ewing, J.1. et et., Initial Reports ofthe DeepSea Drilling Project, v. 17, p. 377-405: United States Government Printing Office, Washington, D.C. Mizutani, S., 1977, Progressive ordering of cristobalitic silica in the early stage of diagenesis: Contributions to Mineralogy and Petrology, v. 61, p. 129-140. Morse, J.W.and Casey,W.H., 1988,Ostwald processes and mineral paragenesis in sediments: American Journal of Science, v. 288, p. 537-560. Murata, K.J., Friedman, I. and Gleason, J.D., 1977, Oxygen isotope relations between diagenetic silica minerals in Monterey Shale, Temblor Range, California: American Journal of Science, v. 277, p. 259-272. Murata, K.J. and Larson, A.A., 1975, Diagenesis of Miocene siliceous shales, Temblor Range, California: United States Geological Survey, Journal of Research, v. 3, p. 553-566. C fo Murata, K.J. and Norman, M.B., II, 1976, An index of crystallinity for quartz: American Journal of Science, v. 276, p. 1120-1130. Nielsen, A.E., 1964, Kinetics of Precipitation: The Macmillan Co., New York, 151 p. Paris, I., Stanistreet,I.G. and Hughes, M.J., 1985, Cherts of the Barberton greenstone belt interpreted as products of submarine exhalative activity: Journal of Geology, v. 93, p. 111-129. Pisciotto, K.A., 1980,Chert and porcellanite from Deep Sea Drilling Project Site 436, Northwest Pacific, in Langseth, M., Okada, H. et el., Initial Reports of the Deep Sea Drilling Project, v. 56/57, p. 1133-1142: United States Government Printing Office, Washington, D.C. Pisciotto, K.A., 1981, Diagenetic trends in the siliceous facies of the Monterey Shale in the Santa Maria region, California: Sedimentology, v. 28, p. 547-571. Siever, A. and Woodford, N., 1973,Sorption of silica by clay minerals: Geochimica et Cosmochimica Acta, v. 37,p. 1851-1880. Snyder, W.S., Brueckner, H.K. and Schweikert, A.A., 1983,Deformational styles in the Monterey Formation and other siliceous rocks, in Isaacs, C.M. and Garrison, R., eds., Symposium Volume on Monterey Oilfields: Society of Economic Paleontologists and Mineralogists, Pacific Section, Los Angeles, p. 151-170. Stein, C.L. and Kirkpatrick, R.J., 1976, Experimental porcelanite recrystallization kinetics: A nucleation and growth model: Journal of Sedimentary Petrology, v. 46, p. 430-435.
251
Steinitz, G., 1970, Chert "dike" structures in Senonian chert beds, southern Negev, Israel: Journal of Sedimentary Petrology, v. 40, p. 1241-1254. Taliaferro, N.L., 1934, Contraction phenomena in cherts: Geological Society of America, Bulletin, v. 45, p. 189-231. VanLier, J.A., de Bruyn, P.L. and Overbeek, J.T.G., 1960, The solubility of quartz: Journal of Physical Chemistry, v. 64, p. 1675-1682. v. Rad, U., 1979,Si0 2Diagenese in Tiefseesedimenten: Geologische Rundschau, v. 68, p. 1025-1036. v. Rad, U. and Rosch, H., 1972, Mineralogy and origin of clay minerals, silica and authigenic silicates in leg 14sediments, in Hayes, D.E., Pimm, A.C. et el., Initial Reports ofthe Deep Sea Drilling Project, v. 14, p. 727-751: United States Government Printing Office, Washington, D.C. Weaver, F.M. and Wise, S.w., Jr., 1972. Ultramorphology of deep sea cristobalitic chert: Nature, v. 237, p. 56-57. Williams, L.A. and Crerar, D.A., 1985,Silica diagenesis, II. General mechanisms: Journal of Sedimentary Petrology, v. 55, p.312-321. Williams, L.A., Parks, G.A. and Crerar, D.A., 1985,Silica diagenesis, I. Solubility controls: Journal of Sedimentary Petrology, v. 55, p. 301-311. Wise, S.w., Jr., Buie, B.F.and Weaver,F.M., 1972,Chemically precipitated sedimentary cristobalite and the origin of chert: Eclogae Geologicae Helvetiae, v. 65, p. 157-163. Wyllie, P.J., 1971, The Dynamic Earth. Textbook in geosciences: Wiley-Interscience, New York, 476 p. Zemmels, I. and Cook, H.E., 1973, X-ray mineralogy from the Central Pacific Ocean, in Winterer, E.L., Ewing, J.1. et el., Initial Reports of the Deep Sea Drilling Project, v. 17, p.517-559: United States Government Printing Office, Washington, D.C.
Originally published in
Geoscience Canada v. 15 Number 3 (September 1988) Revised 1989
252
Diagenesis
Geoscience Canada Reprint Series 4
c c c c
c
c c <;;
c
c c (
c c
c
c
c c
c c c c c c
c
c c
c c
SEM photomicrograph of authigenic quartz in the Lower Silurian Whirlpool Sandstone Formation of southeastern Ontario. This quartz crystal has grown from the surface of a detrital quartz grain into a pore space. Fibrous authigenic illite coats the surface of the detrital quartz. Photograph courtesy of K.J. O'Shea (Golder Associates Ltd.).
c c c c c (
c C l
C l
C C
r. C.
Origin of Inorganic and Replacement Cherts
253
( :
C
c·
c
c
c·
c c (
(
o o o o o o o c o o c o o o c (; o o
,-
Silica Diagenesis: Origin of Inorganic and Replacement Cherts Reinhard Hesse Depanrnent of Geological Sciences McGill University 3450 University St. Montreal, Quebec H3A 2A7 INTRODUCTION Silicification is a common diagenetic phenomenon in a wide variety of originally non-siliceous sediments, and its extent ranges from minor to pervasive. Examples of partial and minor silicification include: (1) silicification of fossil wood; (2) chertification of carbonates and carbonate-bearing sandstones; and (3) replacement of evaporites. In all three examples the sourceofthe silica is predominantly biogenic. Pervasive chertification occurs on the scale of individual layers, beds or entire formations. It has been described in lacustrine, pedogenic and hydrothermalvolcanogenic environments. Examples are (4) Magadi-type cherts; (5) other lacustrine cherts; (6) silcretes; and (7) hydrothermal-volcanic cherts. In the last example chertification tends to affect entire formations. In examples (4) to (7) the source of the silica is predominantly inorganic. SILICIFICATION OF WOOD Silicification of plant remains is the most common among several preservation mechanisms for fossil wood, which include petrifaction by carbonates, sulphides, sulphates and phosphates. The sequence of silica phases and their transformations in silicified wood is essentially the same as that in biogenic siliceous oozes, and therefore forms the link between silicification of the latter (dealt with in the previous article: Hesse, this volume, p. 227-251) and non-siliceous materials. As analyses
and experimental stud ies by Drum (1968), Leo and Barghoorn (1976), and Sigleo (1978) have shown , silicification of wood is not a replacement but a permeation or void-tlllinq process. Dissolved silicon has a particular affinity for the organic molecules making up the vascular tissue, particularly lignin and cellulose, with their numerous exposed functional groups (i.e., hydroxyl groups). The mechanism of silica deposition on wood surfaces involves hydrogen bonding between the hydroxyl groups in cellulose or lignin and those in the silanol molecule. The silicic acid actually involved in the petrifaction mechanism may not be monomeric silicic acid. At the time of bond formation it may have polymerized to a low-molecular weight polysilicic acid with the capacity of forming multiple hydrogen bonds (Leo and Barghoorn, 1976). Minerai parageneses and trace element data for petrified wood samples and their host sediments indicate that silica mineralization takes place within the chemical and pH range of most surface waters. According to Sigleo (1978) clay minerals can co precipitate with silica. In order to produce silicification, the aqueous environment ought to be anoxic to prevent rapid oxidation, and should contain a moderate concentration of silica (up to 140 ppm) over an extended period of time. Ground waters that have percolated through volcanic ashes (Murata, 1940), as well as other silica-bearing meteoric or formation waters have a high potential for the fossilizat ion of wood through silicification. Although most fossil wood consists of quartz, the silicification process involves the same maturation sequence as in biogenic siliceous oozes, i.e ., opalA ..... opal-CT ..... quartz (Buurman, 1972; Leo and Barghoorn, 1976; Stein, 1982). Once silicic acid has permeated the plant cells and replaced the cell fluids, it polymerizes, precipitates and dehydrates in cell centres and openings between cell walls. These initial steps in the petrifaction reactions can occur in a matter of days to years, as suggested experimentally by Leo and Barghoorn (1976).The opal-A formed in the process preserves in remarkable detail the tissue structure seen in many petrified wood samples. According to experiments by Merrill and Spencer (1950), cellulose fibers can absorb
up to 12.5 mmoles Si0 2 per gram of airdried pulp in 24 hours . From a summary of the literature and his own work, Stein (1982)concluded that the conversion of opal-A through opal-CT to quartz in fossil wood occurs at rates comparable to those of the silica transformations in biogenic siliceous oozes. For samples that have reached the quartz stage , Stein (1982) demonstrated a progressive increase in the quartz crystallinity index with geologic age. The quartzfabric may still show a distinct relationship with the original cell structure of the tissue (Figure 1), indicating that even in the second transformation step (from opal-CT to quartz) the organic molecules of the remaining wood tissue have a controlling influence on the distribution of the grain boundaries of the precipitated quartz. This is not surprising in view of the relatively low temperatures (less than 110°C) at which the transformation is assumed to occur (by analogy with siliceous oozes). The pyrolysis results of Sigleo (1978)showed that Iignin-derived pyrolysates are still present at this step. CHERTIFICATION OF CARBONATES AND CARBONATE-BEARING SANDSTONES Much less is known about the chemical conditions and temperature of silicification of ancient carbonate rocks, because most of the evidence is based on thin section and/or geological field observations rather than on experimental or geochemical results. Pore fluid compositions are generally not available for the solutions which caused silicification in these rocks . Studies of silicified carbonates in the past have focussed on (i) description and analysis of the silica fabrics, (ii) relative timing of the silicification process with respectto diagenetic events affecting the carbonate host sediment, and (iii) general constraints for the chemical environment of silicification based on geologic evidence as well as on identification of the source(s) of the silica. These are mostly studies undertaken since the mid-1960s. Extensive earlier literature on the origin of nodular chert in carbonate rocks dating back to the last century contributes little to the foregoing problems due to the limited understanding of the solution chemistry involved and ofthediagenetic reaction sequences in carbonates and siliceous sediments.
c
254
Chertification of carbonates involves the precipitation of pore -filling silica cements as well as the replacement of carbonate by silica. Both may occur before and/or after, but probably not during carbonate cementation of the host sediment. The same relationships hold for carbonate-bearing sandstone. The processes of silicification and carbonate cementation seem to be mutually exclusive with the possible exception of dolomitization. The change-over from one to the other may occur several times in the diagenetic history of a sediment. In early diagenesis. the choice of which of the two processes will operate first. appears to depend strongly on environmental conditions (submarine versus meteoric or mixing-zonediagenesis). During later diagenetic stages hostrock composition becomes important. Silica Fabrics. The silica occurs in the form of quartz. which includes equigranular and fibrous types. Seven different recurring fabrics have been recognized (Figure 2). although some of these are more typically
Diagenesis
developed in evaporites than in carbonates. At least some of these different fabrics may have had opal-A and/or opal-CT precursors, although generally these phases are not preserved and their former presence is difficult to prove. However, wherever pore fluids became highly supersaturated with silica during burial, these metastable phases are expected to have preceded the precipitation of quartz. (a) Equigranu/ar types. (1)Thecommon microcrystalline quartz ("microquartz" of Folk and Pittman, 1971) consists of tiny crystals of quartz (generally less than 5 to 20 microns in diameter) which display an equigranular texture with pin-point extinction pattern. If the size of individual crystals is below the resolution of an ordinary light microscope, this fabric is called cryptocrystalline. (2) Megaquartz (size range of 20 to 2000 microns or larger) has a mosaic fabric similar to drusy calcite mosaics, commonly displaying a progressive increase in crystal size from margin to
Geoscience Canada Reprint Series 4
centre (Figure 2a). Such fabrics character ize void-filling cements, but with quartz they seem not to be restricted to cements. Replacement of carbonate fossils or of an earlier opal-CT phase (e.g., in fossil wood) may produce the same fabric . Replacement fabrics can be distinguished from cement fabrics such as mold infills by the presence of carbonate and/or organic matter inclusions in the former (Figure 2b). The term "megaquartz", applied toamicroscopic feature, clearly is a misnomer, regardless of how well entrenched it may be in the literature. (b) Fibrous types. (3) The most abundant variety among these is "chalcedony sensu stricto" or " chalcedonic quartz" or "length-fast chalcedony", which are the common names in use for this mineral, in which the elongation of the fibres is perpendicular to the crystallographic c-axis (or the epsilon vibration direction). Bundles of fibres commonly radiate from a single point on the substrate from which they grow. This radiating pattern is charac-
c
c c
c
c
c c
C C
C C
C C
C C C
C C C C
C C C
C C C
C C C C C
C
c e (
c
Figure 1 Photomicrographs of petrified wood. (8) Well-preserved cell-structure. (b) As in (a) but with crossedpo/arizers. Note that variations in grain size and orientation of quartz crystals are controlled by cell-structure of original tissue .
c. c c
c c (
c c c
c· c
c
c· (
(
c c o o o o o o o o o o o o o
teristic of chalcedony which typically occurs as a vOid-filling cement , e.g., in the interior of radiolarian tests . It may also form replacement fabrics (e.g., Wilson, 1966). So-called "chalcedonic overlays" (Wilson, 1966; Figure 2c) commonly line primary or secondary pores as isopachous rims of chalcedony radiating from silicified framework grains on the cavity walls. They often consist of consecutive layers which in plain polarized light produce a brownish color banding that is probably due to the presence of bands of very small inclusions. They may be followed by megaquartz, the grain size of which increases toward the pore centre, and locally by spherulitic chalcedony (Figure 2c). Parallel, non-radiating chalcedony fibers occur in "fringing" or "palisade cement" in high-grade diagenetic or low-grade metamorphic calcareous sandstones that have undergone tectonic deformation (Figure 3a). In these fringing cements, length-fast chalcedony is intimately intergrown with length-slow chalcedony. The fibres form in the pressure shadows of rigid, detrital grains, specifically sugary dolomite
M
0,
C)
w
z
...c
w
E
(
c C
D C '-. /
,.'
0
R
p
grains, in a fashion similar to pressure fringes in metamorphic rocks (Hesse, 1987). Magnesium dissolved from the dolomite may play asimilar role in nucleating this chalcedony by providing magnesium hydroxide nuclei as does the magnesium-hydroxidecompound (MHC) referred to in Hesse (1990b) for opal-CT nucleation. The direct precipitation of quartz from dissolved opal-A without an opal-CT precursor is favored by a lack of Mg according to Hinman (1987). The fringing cement, however, forms diageneticallytoo late for opal-A to be present as a silica source. (4) In "length-slow chalcedony" or quartzine (the name first proposed by Michel-Levy and Munier-Chalmas , 1892)the elongation of the fibres is parallel with the crystallographic c-axis. This fabric occurs in chert nodules replacing former evaporites, and has been used as evidence for "vanished evaporites " (Folk and Pittman, 1971). This popular view was challenged by Kastner (1980) and Keene (1983), who showed that quartzine also occurs in deep-water pelagic sediments and is not restricted to shallow-water evaporitic environments. Therefore, the pre-
H
0
mosaic chalceclonic spherulitic quartz overlay chalcedony
c
c o o o
255
Origin of Inorganic and Replacement Cherts
L
o lutecite
G
y
m icrocryst. granular qtz.
b
CII CII
U
11I
0. CII
'-
Figure 2 Microscopic quartz fabr ics. (From Wilson, 1966).(a) Mosaic drusy quartz displaying increasing crystal size toward pore centre. (b) Mosaic quartz replacing biogenic carbonate fabric, relics of which are preserved as inclusions. (c) Chalcedonic overlay lining silicified serpula tube grading inward through microquartz to mosaic megaquartz. Lower half: cnetcedonic overlays on silicified grains followed toward pore center by spherulitic chalcedony. (d) Pore-filling spherulitic chalcedony. (e) Spherulitic chalcedony containing inclusions of a replaced former aragonite shell. Lower half: Chalcedony spheres supporting silicified ooid layers. (f) Lutecite partially replacing calcitic shells. (g) Equigranular microcrystalline quartz replacing framework grains rimmed by cha/cedonic overlays.
sence of quartzine alone cannot be used to establish the former presence of evaporites that have been dissolved. Folk(1975)himself cautioned against an uncritical application of this criterion after having observed a number of examples of quartzine-replaced fossils (echinoderms, brachiopods, bryozoa) in rocks that showed no evidence of evaporitic conditions. As an alternative, he suggested Mg-rich fluids as the geochemical environment for carbonate replacement by quartzine, an idea which is supported by the more recent evidence of Kastner (1980) and Keene (1983). These authors also concluded that quartzine-precipitating flu ids ought to be enriched in sulphate. However, Brown et al. (1969) and Meyers (1977) saw no evidence for the presence of either magnesium or sulphate in fluids causing replacement of calcitic fossils by spherulitic quartzine (Figure 2e). (5) Lutecite is a fibrous quartz variety intermediate between chalcedony sensu stricto and quartzine with the fibre axis inclined about 30° to the crystallographic c-axis (Figures 2f,4; Frondel, 1962). (6) "Zebraic chalcedony" displays a beautifully banded extinction pattern under crossed polarizers (Figure 3c). This results from a systematic helical twisting of the fibre axes about the crystallographic c-axis (McBride and Folk, 1977; Frondel, 1978) as schematically illustrated in Figure 4. Although zebraic chalcedony is commonly associated with cherts replacing evaporites. it has also been described in chert formations without any evidence for the former presence of evaporites (e.g., Simonson, 1987). (7) "Microflamboyant quartz" has afabric intermediate between equant and fibrous quartz types (Milliken, 1979). Also known as "flamboyant lutecite" (Folk and Pittman, 1971) or "flamboyant spectral extinction" (Chowns and Elkins, 1974), this fabric displays undulose extinction caused by composite , fanning crystals whose individual crystal boundaries are not clearly recognizable. Undulose extinction in this case results from crystal growth in domains rather than from grain deformation. Megaquartz in evaporite-replaced chert rarely displays radially undulose extinction as a growth phenomenon as well (Milliken , 1979, fig . 3).
256
Diagenesis
Geoscience Canada Reprint Series 4
(
(
. d .
\
c· c c·
c
c. (
c (
c c c
Origin of Inorganic and Replacement Cherts
Origin of Quartz Fabrics: Void-filling Cements versus Replacements. Void-filling cement. Of the fabrics described, only one has been identified exclusively as a void-filling cement, i.e.• Wilson's (1966) "chalcedonic overlays". Where this cement is followed by mosaic megaquartz and/or spherulitic chalcedony, the latter are unquestionably also of void-filling origin (Wilson, 1966; Schmitt and Boyd, 1981). As mentioned previously, "drusy megaquartz"
in itself is not a reliable fabric indicator for a void-filling mechanism because it is also observed in quartz replacing fossil fragments. In the latter case, however, it is riddled with carbonate inclusions (Figure 2b). Length-fast chalcedony occurs predominantly(e.g., Meyers and James, 1978, p. 107), but not exclusively as a cement. Replacement. Lutecite, on the other hand, appears to be the only fabric known exclusively as a replacement,
257
although zebraic chalcedony and quartzine are predominantly replacive too. However,the occurrence of quartzine (together with chalcedony) in the fringing cements (pressure fringes), as the name says, suggests a mechanism involving crystal growth into a porespace, i.e., cementation. This pore space is opening under tectonic deformation by the detachment of cement crystals or matrix grains from rigid framework grains and is being filled in
0 0 0 0
0 CO
0 0
c 0 o 0 r ·
r: ~
r-~
~)
,.
c
r: r
r:
1..... r
'-'-' v-:
'--
Figure 3 Ca) (opposite page, upper left) Fringing cement ("" palisade cement, p) forming discontinuous rim on microcrystalline, "sugary" dolomite grain in the direction perpendicular to maximum tecton ic stress (N-S). Note absence of fringing cement in the direction parallel to maximum stress (E-W). Intergrown quartzine and chalcedony fibers are elongated NoS. Fibrous silica cement was followed (or partially replaced) by fibrous late calcite cement (e) which started to grow from the edge of the detrital dolomite. Lower Cretaceous Gault Formation, Uechtenstein. Height of picture corresponds to 0.5 mm. (b) (opposite page, upper right) Fringing cement under crossed polarizers, same thin section as in (a). Ce) (opposite page, lower left) Large fans of zebraic chalcedony radiating from sificified framework grains into large pore. Note smooth fan boundaries (arrows). Centre of pore filled by drusy megaquartz. Length of scale ber » 0.5 mm. Lower Proterozoic Sokoman Formation, Quebec. (From Simonson, 1987, fig. 7c; photomicrograph courtesy of 8.M. Simonson). Cd) (opposite page, lower right) Spherufitic chalcedony cement with isopachous growth bands in sifica-cemented eolian (?) sands from Steamboat Springs , Nevada. Note smooth interfaces between spherulitic fans (arrows). Scale bar"" 0.05 mm. (From Simonson, 1987, fig. 12c; photomicrograph courtesy of 8.M. Simonson). (e) (above left) Selective sificiflcation of ooid cortex at contact points with detrital quartz grains. Individual silicified domains in ooid rim are in optical continuity with their quartz neighbors and represent "overgrowths' ~ Lower Cretaceous Gault Formation , Grisons, Eastern Switzerland. Dimensions of ooid "" 0.43 x 0.64 mm. Cf) (above right) Silicified centre of echinoderm ossicle showing microdolomite inclusions in identical crystallographic orientation. The inclusions continue from the quartzine into the unreplaced calcite. Same thin section as (a). Long diameter of quartzine area "" .0.5 mm.
258
Diagenesis
statu nascendi by the growing fringing silica cement which thus is not a replacement. Schmitt and Boyd (1981) use the terms "delayed" and "immediate replacement" instead of the familiar terms "void-filling cement" and "replacement" to describe different silicification patterns of Permian pelecypods and brachiopods in the Park City Formation of Wyoming. With this terminology the authors stress the point that cementation of secondary pores such as molds of fossil fragments and the replacement of such fragments by another mineral are end-members of the same couple of chemical reactions, l.e., dissolution and precipitation. In the case of cementation of dissolution voids, these reactions are separated by a finite time interval, whereas in the case of replacement, they are separated by an infinitesimally brief interval. Selective replacement. Where there is evidence for replacement , the process is commonly highly selective; certain components of the carbonate host rock are replaced while others are unaffected. Meyers (1977) found the following order in terms of decreasing susceptibility to replacement of various carbonate constituents by microquartz in the Mississippian Lake Valley Formation of New Mexico: carbonate mud, bryozoans, thin-shelled brachiopods, crinoids, thick-shelled brachiopods, syntaxial calcite cement rims on crinoids. The last three components show
fi~r.
orientatioll
only partial replacement, and most commonly they are unaffected by silicification. The degree of selectivity of the silicification process apparently is a function of a number of variables including pH, dissolved silicon activity, and the extent of previous lithification of the rocks (i.e., porosity and permeability reduction prior to silicification). According to Jacka (1974), both high-magnesian calcitic (fusulinids, bryozoans, echinoderm fragments) and aragonitic skeletal elements (parts of molluscs shells) of the Getaway carbonate turbidites of the Middle Permian Cherry Canyon Formation of TexaslNew Mexico are replaced by quartzine and megaquartz, but notthe micritic carbonate matrix or intraclasts. Similar observations were made on dolomitic limestones in the Middle Devonian Columbus and Delawareformations of Ohio by Hatfield (1975). Many samples exhibit selective silicification of echinoderm fragments only, whereas in others bryozoans,brachiopodsand coralswere silicified in addition. Inthe most intensely silicified samples, the micritic carbonate matrix also had been replaced.
Geoscience Canada Reprint Series 4
Selective silicification of ooids, calcareous algae, bryozoans and echinoderm fragments occurs in carbonaterich turbidites of the Lower Cretaceous Gault Formation of the East Alps (Hesse, 1987). Calcitic or calcite-replaced shells of molluscs are not affected . The replacement quartz is megaquartz, which forms a syntaxial "overgrowth" on detrital quartz that occurs as a nucleus of or in grain contact with an ooid. Replacement is restricted to the ooid cortex in contact with the quartz grains; it does not affect the microspar calcite cement that envelopes both ooids and quartz grains (Figure 3e). Detrital carbonate clasts, both limestone and dolomite, are also unaffected. These rocks had been extensively cemented by a calcitic microspar prior to silicification. In quartz-rich turbidites of the same formation which apparently were not carbonate-cemented before silicification, the replacement pattern is quite diffferent. Here, the ooid cortices are completely replaced by micro- to cryptocrystalline quartz. The syntaxial megaquartz "overgrowths" of the calcarenites (replacing the calcareous cor-
c-ni,
~
I /
lutecite a
zebraic chalc. Figure 4 Orientation of fibre axes in fibrous silica minerals. (From Milliken, 1979).
c c c
c c c c c
c c c c c c c c c c c c c c c c
quartzine
chalcedony
c c c c c c c
b
Figure 5 (8) Zoned silicified ooid cortex with (1) outermost near-opaque cryptocrystalline zone followed inward by (2) microcrystalline zone and (3) macrocrystalline zone with concentrically arranged laminae of carbonate inclusions. The inward increasing quartz crystal size represents a replacement sequence with decreasing nucleation rate. centre of ooid.not shown. Magnification 120x. (From Choquette, 1955). (b) Zoned syntaxial overgrowth on well-rounded detrital quartz grain. Innerzone: euhedral quartz crystal with concentrically arranged laminae of carbonate inclusions which show spherulitic extinction revealing an overall radial crystallographic orientation of the original carbonate crystals. Outer inclusion-free zone. For interpretation seetext. Cortex zonation similar as in (a). Magnification 80 x. (From Choquette, 1955).
c c c e c c
c c c c c c
c
c
Origin of Inorganic and Replacement Cherts
259
r'--.r: '-.
r: '-.-.,
c r
r r:
c o o o
o
o o o o o, L
J r: ~)
'-.-
tex of a quartz-nucleus ooid) were not observed. The small grain size of the replacement quartz in the quartz-rich turbidites suggests rapid (homogeneous) nucleation. In the carbonate turbidites in contrast, detrital quartz grains served as nuclei for slow heterogeneous nucleation. This replacement fabric suggests relatively low silicon concentrations in the pore fluids and/or restricted permeability in the calcarenites so that the replacement phase could be precipitated as quartz rather than as a less well-ordered silica phase. In the quartz-rich turbidites, on the other hand, the fluids were probably highly supersaturated with respect to quartz. They were conceivably also supersaturated with respect to opal-CT, which was likely the first silica phase to be precipitated. This could explain why detrital quartz grains in these rocks exerted no influence on the nucleation of the silica precipitate. In the carbonate turbidites, homogeneous nucleation also occurred, but only in ooids with nuclei other than quartz. In that case , it produced a few megaquartz crystals in the cortex. Silicified oolites have been a preferred object of petrographic studies of silicification, because they reveal an abundance of relict fabrics that offer insight into the detailed mechanisms of the silicification process. For example, the concentric fabrics of the ooid cortex provide information concerning the direction of the advancing silicification front, which may be inward or outward. It is clearly inward from rim to nucleus in the case of incomplete silicification where innner lamellae of the ooid cortex have not been affected. It is outward from nucleus to cortex margin in the case ofthe "syntaxial overgrowth"-type of replacement (also called "undergrowth") where detrital quartz forms the ooid nucleus (Hesse, 1987). Among the known oolite formations that have been completely silicified, the UpperCambrian State College Oolite of Pennsylvania is one that has been studied since the last century and has been the subject of an early modern examination (Choquette, 1955). Ooids of the State College Oolite show a tripartite zonation of the cortex (Figure 5a) consisting of the following characteristics: (1) an outermost cryptocrystalline, mostly near-opaque rim, followed by (2) a somewhat coarser-grained, but still
microcrystalline central zone, and (3) an innermost zone of macrocrystalline quartz with concentrically arranged laminae of carbonate inclusions. This zonation results from inward progression of the replacement and reflects decreasing silicon activity or/and decreasing permeability as the ooid rim gets sealed off by silicification. The centre of the ooid may consist of: (i) a moderately coarse-grained mosaic of (mega-) quartz or (ii) a detrital nucleus, usually a quartz grain with a syntaxial overgrowth. The latter reflects outward silicification. Other descriptions of silicified oolites are, e.g., by Swett (1965) and Chanda at at. (1976, 1977). Timing of Silicification Relatille to carbonate Cementation
and Diagenesis. Chertification of carbonate host sediments apparently occurs at different times in their diagenetic history and examples of early, intermediate and late silicification have been described in the literature. Meyers (1977) and Meyers and James (1978)suggested early chertification for the Mississippian Lake Valley Formation, New Mexico. RelativetotheestabIished carbonate cement stratigraphyof these rocks, pore-filling length-fast chalcedony was precipitated before and after early carbonate cements. Part of the chalcedony appeared before (1) a first generation of non-ferroan, manganese-poor, largely non-luminescing calcite cement, whereas most postdates (2) the younger manganese-rich, brightly luminescing carbonate ce ment. All other carbonate cements are younger than the chert. The first two carbonate cements are interpreted to be due to phreatic meteoric diagenesis related to two unconformities: (i) a preMeramecian (pre-middle Visean) unconformity on top of the Lake Valley Formation and (ii) a post-Meramecian unconformity that terminates the overlying Rancheria Formation. Supporting evidence for early chertification comes from unconformity-associated chert clasts that appear to have been chertified before redeposition, and from differential compaction. Chertified crtnoidal limestones, for example, show minor compaction features, whereas compaction is intense in their nonchertified equivalents. Stylolites are absent within chert nodules but common in the surrounding carbonate host
sediment. Meyers and James (1978) estimated that in this case early chertification took place under a maximum overburden of not more than 215 m (the thickness of the Mississippian of the region) and generally probably under much less overburden, perhaps only a few metres or tens of metres. Corresponding burial temperatures would have been less than 30°C, based on an assumed mean annual surface temperature of 20°C and a geothermal gradient of 3.6°C per 100 m. Another case for early, unconformityrelated chertification has been described from the Lower Pennsylvanian Marble Falls Group of central Texas (Namy, 1974). Chert occurs in and below a thin (1 to 2 cm thick) crust of silicified oolite at the top of a 3-6 m thick oosparite/biosparite horizon just beneath a regional unconformity that extends for some distance from the north and east flanks of the Llano Uplift. Silicification decreases downward in the crust. At the top of the crust, silicification is pervasive. In the middle, ooid and pisolite cortices are replaced by microquartz, and pore spaces are filled with microquartz and/or chalcedony, but ooid and pisolite nuclei are unaffected. Near the base of the crust, ooids are no longer silicified, but the pore-filling cement is partly calcitic and partly siliceous. Deeper in the Marble Falls oolite, carbonate cementation was not interrupted by silicification and progressed toward complete elimination of porosity and recrystallization to a coarsely crystalline sparry calcite. The time of silicification can be pin-pointed relative to carbonate cementation. In the silicified ' crust, silica cement postdates a finely crystalline sparry calcite cement which fringes the ooids. Since this calcite is a typical incipient submarine carbonate cement, the subsequent silicification event was attributed to temporary emergence of the oolite shoal above sea level when it was invaded by meteoric water (Namy, 1974).Thetimelimitsof silicification can only be given in relative terms because incipient submarine carbonate cements can exist in the marine environment for up to 50 Ma or longer without progressing to complete pore cementation. This is illustrated, for example, by a pre-Eocene oolite from a West Pacific seamount, which still contains open pore spaces lined only with an incipient microspar cement (Hesse , 1973).
260
Eocene to Pleistocene fluvial conglomerates of western Texas provide textural, fabric, and isotopic evidence for the origin of unconformity-related silicification of carbonate pebbles by ground water below the water table (McBride, 1988). The ground waters became silica-enriched after percolating through Neogene volcanic ashes, and they selectively silicified the carbonate pebbles derived from uplifted Cretaceous rocks. Microquartz and chalcedony are the major silica phases; lutecite and opal-CT occur as minor phases. In some pebbles, concentric chert and carbonate bands that are conformable to the pebble shape alternate rhythmically and may represent Liesegang banding. Proper hydraulic and chemical conditions for chertification may not have persisted for more than a few million years underscoring that the chertification process was rapid (McBride, 1988). Other examples of early diagenetic chert formation havebeen documented for the Ordovician Aleman Dolomite of New Mexico (Geeslin and Chafetz, 1982), and for Silurian back- and interreef carbonates of Illinois and Indiana (Dapples, 1959). In the latter case, silicification postdates early diagenetic glauconite formation and predates dolomitization. Jacka (1974) argued for silicification in the Getaway turbidites prior to the aragonite to calcite transformation, and also prior to the high-magnesian calcite stabilization reaction as discussed below.The original structure offormerly aragonitic shells was commonly pseudomorphed in silicified portions of the fossils, but obliterated in calcitized portions. This observation suggests that the original aragonitic fabric was still in place during silicification, but had been lost prior to calcitization. Silicification preceded calcitization . In the Gault turbidites by contrast, Hesse (1987) found former aragonite shells replaced by fabric-destroying calcitic spar which showed no signs of silicification. In primary calcite grains, original fabrics were preserved, even after partial replacement by silica, suggesting silicification postdated aragonite conversion to calcite. Choquette (1955) provided compelling evidence that many of the quartzreplaced ooids in the State College Oolite underwent aragonite dissolution
Diagenesis
prior to or during silicification . The dissolution affected aragonite both in the cortex and in biogenic fragments forming the nuclei of ooids . The best evidence for the simultaneous occurrence of replacement and dissolution is provided by grains with a quartz nucleus mantled by a zoned syntaxial overgrowth: the inner zone of the overgrowth is riddled with radially oriented carbonate inclusions, the outer zone is inclusion-free (Figure 5b). The quartzovergrowth which mantled the detrital nucleus initially found the aragonite of the ooid cortex still in place and replaced it. The outer inclusion-free zone formed when the adjacent aragonite of the ooid cortex had been dissolved without a trace, not even of organic matter. Grains with eccentric quartz nuclei also indicate partial dissolution of the cortex prior to completion of silicification: dissolution of inner aragonite lamellae caused the nucleus to fall to the bottom ofthe partly silicified cortex. The geopetal fabric atthe centre of"bipartite" ooids has the samesignificance: fine-grained, crypto- or microcrystalline quartz replaces micritic carbonate that fell to the cavity bottom, but was not dissolved, and drusy megaquartz fills the upper part of the cavity. Drusy fabrics of megaquartz in the centre of ooids with inward increasing grain size and a lack of inclusions reflect complete nucleus dissolution prior to silicification. Using evidence from carbonate inclusions in authigenic quartz, Richter (1972) found that silicification postdated the aragonite to calcite transformation, but predated the high-Mg calcite stabilization. Authigenic quartz in original aragonite skeletons contained low-Mg calcite inclusions as it also did in original low-Mg calcite skeletons. In original high-Mg calcite skeletons, the carbonate inclusions in authigenic quartz showed elevated Mg contents. Microdolomite in low-magnesium calcite is interpreted as an exsolution product from former high-magnesium calcite (Lohmann and Meyers,1977).lnthe Getaway turbidites, it was found only in the silicified portions of former high-Mg shell fragments, but not in the unsilicified portions of the fossils (fusulinids, bryozoans, echinoderm plates) which are now low-Mg calcite . Jacka (1974, p. 425) inferred from this observation that Mg-exsolution from the high-Mg calcite
Geoscience Canada Reprint Series 4
took place "during ionic mobilization that accompanied solution-precipitation replacement of magnesium calcite by silica". In other words, silicification and microdolomite formation would have occurred simultaneously and prior to the recrystallization of high-Mg calcite to low-Mg calcite. In the Gault Formation, on the other hand, Hesse(1987) observed that trains of microdolomite crystals in quartzine continued with the same optical orientation and alignment into the unsilicified portions of echinodermossicles(Figure3f).This ledhimto the conclusion that silicification followed carbonate recrystallization. For the Gault Formation, other independent lines of evidence support an intermediate burial stage as the time of silicification. Review of the literature thus reveals that virtually all possible sequences occur in the relative timing of silicification and well-established· diagenetic carbonate reactions: (1) silicification before and alternating with the earliest carbonate cements (e.g., Meyers, 1977); (2) silicification after incipient pore-linings of calcitic microspar but before complete pore cementation (e.g., Namy, 1974); (3) silicification before the aragonite to calcite conversion and the high-Mg calcite stabilization reaction (e.g., Jacka, 1974); (4) silicification during and after aragonite dissolution (e.g., Choquette, 1955); (5) silicification after the aragonite conversion, but before the high-Mg calcite transformation (e.g., Richter,1972); and (6) silicification after both the aragonite and high-Mg calcite transformations (e.g., Hesse, 1987). Silica Sources in Carbonates. Most chert nodules in carbonates can be adequately explained by the redistribution during diagenesis of biogenic silica contained in the host sediment. However, it has often been questioned whether siliceous organisms are sufficiently abundant ln shallow-water carbonate sediments to provide the quantities of silica required for the formation of chert nodules (e.g., Dapples, 1959). According to Knauth (1979), no example has been reported in the literature for modern shallow-water carbonate environments with abundant siliceous organisms . For Paleozoic carbonates, on the other hand, sponge spicules are a widely quoted source of silica, particularly for the early chertification of
c c c c c c c ( c c c c c c
c c c
c (;
c c c
c c c
c c
c c c c c
c c e
c c c
c (
c (
c
c. c c,'-r: I <;
[J
C C C
o o
o o
o o
r
261
Origin of Inorganic and Replacement Cherts
lutitic limestones or dolomites (e.g., Meyers, 1977; Geeslin and Chafetz, 1982). As Knauth (1979) pointed out, nodular chert horizons may have received the silica from pore waters that had previously percolated through and leached considerable thicknesses of unlithified source sediment which now may be devoid of any silica . However, not all chert nodules in carbonates have a biogenic source. Banks (1970) discussed the example of the cherty Mississippian Leadville Limestone of Colorado which does not contain any sponge spicules or Radiolaria or their pseudomorphs. An alternative source for the silica was seen in the dissolution of quartz and other detrital silicates in a hypersaline, marginal marine environment similar to the modern Coorong lagoon. Banks also found eolian sands of the underlying (Upper Devonian) Chaffee Formation widely replaced by carbonate; this could have provided the silica for the Leadville nodular cherts by an upward advective process. Banks (1970) observed three chert generations: (1) an early generation formed in a hypersaline marine environment of high, but fluctuating, pH; (2) a later generation precipitated after initial carbonate lithification; and (3) a
latest generation of hydrothermal jasperoid deposits associated with Laramide ore deposition. For silicification during intermediate stages of burial diagenesis, pressure solution of quartz in the host sediment, together with the smectite to illite transformation in associated shales have been invoked as silica sources in addition to biogenic silica (Hesse, 1987). Knauth's Geochemical Model and Isotopic Evidence for Early Chert Formation in Umestone. Silicification of carbonate sediments requires pore solutions that are supersaturated with respect to the silica phase precipitated and undersaturated with respect to the carbonate mineral(s) dissolved. Although this general geochemical and thermodynamic requirement for chert formation by replacement is the same for deep-water pelagic sediments and shallow-water carbonates, the specific conditions under which it may be reached during early diagenesis in shallow-water carbonates may be quite differentfrom those during the somewhat later and deeper burial stages of silicification in pelagic sediments. The sequence of silica phases may also be different. Knauth's (1979) model for shallow-
------quartz ---~
+1 r
~--opal-CT ---+-__
o
o o :J
r
U
opal-A
-I
o
5
10
25
50 15 100
% sea water Figure 6 Mixing model for silicification of shallow water carbonates through meteoric water influx during early stages of emergence. (Modified from Knauth, 1979). The saturation level is the ratio of the ion activity product of the dissolved species to the equilibrium solubility of the mineral. Positive log values = supersaturation, negative log values = undersaturation. Right side of diagram shows area where mixing ofmeteoric andconnate marine watercreates a pore fluidsimultaneously undersaturated with respect to calcite andsupersaturated with respect to opa/-CT and/or quartz.
water early chert in limestone is conceived by analogy with the Dorag mixing-zone model for dolomit ization (Badiozamani, 1973).In the mixing zone between meteoric and marine pore waters in coastal areas, solutions may become undersaturated, for example, with respect to calcite. Undersaturation occurs if the starting solutions havedifferent temperatures or partial CO2 pressures, even if each of them alone is originally supersaturated. Meteoric waters percolating through coastal carbonates will become saturated rapidly with respect to calcite due to the dissolution of aragonite and Mg-calcite. Surface seawater in the tropics and subtropics is also calcite-supersaturated. The mixture of the two, however, may be undersaturated if the solutions mix in the appropriate ratio and the system is closed with respect to CO2 (Figure 6). The same carbonate-saturated meteoric waters may also become saturated with respect to opaline silica following dissolution of biogenic siliceous shells. However, upon mixing with marine connate water they will immediately become undersaturated with respect to opal-A because of the low silicon concentration in seawater. They may not become undersaturated with respect to opal-CT, however, and most probably not with respect to quartz. The mixing zone, therefore represents a geochemical environment conducive to calcite dissolution and simultaneous silica precipitation as opal-CT or quartz. This explains why in the mixing zone opal-CT or quartz can form under near-surface conditions and do not require the deeper burial (of at least a few 100m)and longer periods of time (1Q6 to 107 a) typical of deep-seacherts. Opal-CT might bethefirstsilica phaseto replace calcite, but at presentthere is no compelling evidence that it does in fact precede quartz in these shallow-water carbonates. If the solutions in the mixing zone are only supersaturated with respect to quartz, and not with respect to opal-CT, then quartz would precipitate directly, given that kinetics permit nucleation . If the system is open with respect to CO2 , calcite may not dissolve, because the solutions may not be calcite-undersaturated, but silica could still precipitate as a pore-filling cement. In other situations, dolomite may not becomesilicified while calcite would, because the
Diagenesis
262
mixing-zone solutions may be supersaturated with respect to dolomite, as suggested in the original Dorag model. The acidity ofthe pore solutions isone additional factor that may be important for the precipitation of silica in carbonate sediments. It is implicit in Knauth's model, but was already emphasized by Lovering and Patten (1962). Their work is one of the few experimental studies bearing on the problem of carbonate replacement by chert. In the meteoricmarine mixing zone, the pH of the pore solutions is largely a function of the availability of carbonic acid or the partial pressure of CO2 , Lovering and Patten (1961) tested the idea pursued in very early experiments (Church, 1862; Cox et a/., 1916) that silica-supersaturated solutions will replace carbonate minerals with silica at low temperature and pressure in the presence of (but not in the absence or at low partial pressure of) CO2 , They found that solutions supersaturated with respect to opal-A, when brought into contact with CO2 of 1 atm pressure at room temperature, would decrease in pH from original values of 6.8 to 7.1 to values of 5.7 to 6.2. They were capable not only of dissolving calcite and dolomite, but also of precipitating silica. On the other hand, solutions in contact with air at a partial pressure of CO2 of only 0.03 atm generally acquired a pH of 8.2 to 8.6. The higher pH apparently resulted from buffering by carbonates. These solutions did not precipitate silica, although some carbonate was dissolved. The difference in the partial CO2 pressures apparently was sufficient to cause silica precipitation in one case, but not in the other. It therefore seems that if acidic solutions supersaturated with respect to a given silica phase (e.g., opal-CT or quartz) come into contact with Ca or CaMg carbonates , they will dissolve carbonate and precipitate silica. Neutral or slightly basic solutions that are silicasupersaturated, on the other hand, may move for considerable distances through carbonate rocks without precipitating silica. Knauth's model is consistent with the various lines of evidence for early chert formation in shallow-water limestones enumerated in the previous sections. It deliberately does not take into account kinetic considerations. The model finds strong support from oxygen and hydrogen isotopic studies of nodular cherts in
carbonates of greatly differing ages from the Precambrian to the Cenozoic (Knauth and Epstein, 1976). On 50/0180 plots, trend-lines for different age groups of cherts are parallel with the meteoric water-line (Figure 7). This suggests that mixing of connate marine waters to various extents with meteoric waters is responsible for the isotopic signatures of the cherts which precipitated from these waters. However, the detailed results of Meyers and James (1978) and Land
Geoscience Canada Reprint Series 4
(1979) have shown that the interpretation of isotopic compositions may be quite complex. In the Mississippian Lake Valley Formation discussed previously, 01 8 0 values of early calcite cements vary very little (+ 27.5 to + 28.5%0 SMOW), whereas the silica minerals of the associated early diagenetic chert fabrics (microquartz, chalcedony, and megaquartz) span a considerably wider range of values ( + 24.3 to + 29.8%0SMOW).The chert minerals must have precipitated from waters iso-
c c c c c c c
C C C
C C C
C C
C C C
-20
C (
C C C C C C
-60 -80 -1 00
C -120
C
-140
C C
C C C 16
20
24
28
32
36
40
618 0 (%0) Figure 7 Hydrogen and oxygen isotopic trends in Phanerozoic carbonate-hosted cherts. Elongation of the trend fields parallel to the meteoric water line (set of diagonal lines) for different geologic periods supports Knauth 's (1979) mixing model for early silicification in shallow·marine carbonates. Temperatures of quartz crystallization are based on quartzwater isotope fractionation relationship 1000 In = 3.09 x 106 7-2 - 3.29, where a is the fractionation factor between quartz and water and temperature is in degrees Kelvin. (From Knauth and Epstein, 1976). Line A: Isotopic composition of cherts in equilibrium with SMOW at various temperatures.
c e ( (
C
C (.
c
c cr o c' c (
c
o o o o o o o
o o o
o o o ()
o o
Origin of Inorganic and Replacement Cherts
topically at least 5%0 lighter than those from which the early carbonate cements precipitated or at temperatures up to 40°C higher than the carbonate cements. Petrographically, the silica minerals and carbonate cements occur in close association. Isotopically lighter pore waters for silica precipitates than for the carbonate cements are in line with Knauth's model, if the former are affected by mixing with meteoric water and the latter are equivalent to SMOW or only slightly depleted. Isotopic reequilibration at higher temperatures could also explain the data, but would require an opal-CT precursor recrystallizing to quartz. This would be possible for microquartz, but difficuIt to perceive in the case of megaquartz, (and probably also of chalcedony). Chert nodules in Miocene pelagic chalks exposed on the north Jamaican coast are different in their mineralogical composition from those in downfaulted age-equivalent rocks under water on the north slope of the island: opal-C'T characterizes the submarine occurrence, which probably did not experience any meteoric diagenesis, whereas microquartz occurs in the subaerially exposed material (Land, 1979). At first glance, this seems to be a textbook illustration of Knauth's model: direct quartz-precipitation in a diagenetic environment affected by percolating meteoric water, and opal-CT preservation in young (less than 10 Ma old) pelagic sediments. However, the iso-
topic composition of the subaerially exposed cherts did not show the expected 180 _or O-depletion indicative of the involvement of meteoric water. Among the several explanations offered by Land (1979)for this unexpected finding is recrystallization of the microquartz from an opal-CT precursor, where the precursor's isotopic composition was inherited. This would provide an elegant solution , but is it feasible geologically without isotopic change? Flint Formation in the Cretaceous Chalk: A Geochemical Analysis. The "paramoudra structure" in the Upper Cretaceous chalk of England is a tubular chert structure related to bioturbation (Bromley and Ekdale, 1986) which provides a rare lucky break revealing insight into the chemistry of the carbonate/silica replacement process. It consists of a cylindrical flint nodule surrounding a chalk core which in its centre contains the pyritized, glauconitized and commonly also phosphatized vertical burrow of Bathichnus paramoudrae (Figure 8). The structure constrains the geochemical conditions under which the replacement took place in the chalk. This environment is different from the one envisaged in Knauth's (1979) model in that the replacement seems to occur under fully marine conditions and does not require mixing with fresh water. However, the sediments may have been affected by meteoric water influx at some later stage not related to the replacement
o o ()
c
[".
B
C ". r' .
(e)
II
f-f1int-+-cnented-+-f11llt ---1 chalk (dl r. .":' (bl ~ (el ..: .1
a
Figure 8 Schematic three-dimensional view of paramoudra flint structure and sample position. (8) Bathichnus sp. burrow; (b) pyrite; (c) calc ite cement; (d) disseminated pyrite and glauconite. (Modified after Clayton, 1986).
263
process. Clayton's (1986)detailed geochemical analysis of a paramoudra structure sheds light on the possible sources and migration paths of some of the ions involved in the cementation and replacement reactions. In the paramoudra structure, silica precipitationwasprecededbycarbonate cementation inthe innerchalkcore. Both carbonate cementation and the subsequent replacement by silica in the yet uncementedcarbonate aroundthechalk core are related to reactions initiated in the central burrow, as illustrated by the concentric geometry. The change from carbonate precipitation to carbonatedissolution and silica precipitation reflectsa change in pH. This resulted from bacterial sulphate reduction and associated iron-sulphide precipitation which was limited by iron availability and isa protonconsuming reaction (see e.g., Hesse, 1990a,for review). Only a small portion of the cement carbonate was provided by sulphate reduction, because the cement is isotopically heavy.Sulphate reduction produces isotopically light carbonate (013C = -20 to -25%0 POB),and less than 10% of the cement could have been derived from this source. However, this small fraction of the cement ions supplied by sulphate reduction seems to have been sufficient to tip the balance in favor of carbonate precipitation. The bulk of the carbonate and all of the Ca2 + had to be derived from dissolution of carbonate host sediment (aragonite dissolution and/or pressure solution of coccoliths during early chalk diagenesis). The decisive role of sulphate reduction for carbonate precipitation was to raise the ' pH through iron-sulphide precipitation and to supply additional carbonate alkalinity. The 013C values increase from centre to margin of the chalk core (Figure 9) reflecting a decrease in the proportion of cement present which is isotopically lighter than the host sediment. The time of carbonate cementation in the paramoudra structure is bracketed by porosities between 80 and 650/0. (These were estimated from the degree of AI dilution compared to "undiluted" AI values in the slightly cemented chalk surrounding the paramoudras ; samples Band G, respectively, Figure 9). Carbonate precipitation terminated when exhaustion of reducible iron stopped iron-sulphide precipitation. In the
c
c
C'· C r:
CJ C C ( (
o G
o
o o o r
or o o
o
L·
o
o o o o r
o D
o
'-'
v:
'-- .
........ '
Origin of Inorganic and Replacement Cherts
model, the flint bands mark the position of the redox boundary which separates an oxic (above) from an anoxic environment (below) at the time of silica precipitation. Silica precipitation thus occurs very early during diagenesis, a feature shared with Knauth's (1979) model. Burial depth at the time of silicification may have been as little as 5-10 m. The rhythmic bedding of the Chalk, exhibited by the regularlyspaced flint bands In many coastal cliff sections, may then reflect abrupt, stepwise rises of the redox-boundary related to basinwide sedimentation pulses and/or breaks. Clayton 's (1986) work for the first time provides a firm basis for an evaluation of the geochemical environment of carbonate/chert replacements in fully marine carbonates where the replacement process is not affected by meteoric diagenesis. The chalk is a shelf equivalent of deep-sea pelagic limestones which commonly also contain chert nodules probably formed under similar conditions as the flint nodules. It is not known though, whether they also occur in bands. Early diagenetic silicification in deepwater limestones without meteoric water influx has also been described by Coniglio (1987) for the Cambro-Ordovician Cow Head Group in western Newfoundland, Canada. Uncompacted silicified burrows provide evidence for very early silicification in interbedded mud-
Figure 10 Common associations of silica fabrics in chert nodules replacing evaporites: (1) randomly fibrous microcrystalline quartz; (2) quartzine spherules; (3) anhydrite inclusions; and (4) calcite. (8) Thin rim of spherulitic quartzlne followed inward by megaquartz of increasing grainsize which often contains anhydrite inclusions . (b) Thick rim of spherulitic quartzine followed inward by strongly undulose megaquartz in the central part of nodule. Megaquartz may show crudely radial or fan-like undulosity and may have a nucleus of spherulitic quartzine or microflamboyant quartz riddled with anhydrite inclusions. (c) Very thin rim of spheruJitic quartzine followed by megaquartz. (d) Marginal spherulitic quartzine followed inward by rim of palisade megaquartz and large carbonate replaced centre of nodule with isolated inclusions of quartz crystals indicating incomplete initial replacement of anhydrite by chert. (From Milliken, 1979).
265
rocks which is quite unusual in the light of the silicification mechanisms and the find ings of the DSDP for biogenic siliceous deep-sea sediments discussed in the accompanying paper (Hesse, 1990b). The silicification history of the Cow Head is protracted, since it includes later post-eompactional silica remobilization and reprecipitation. CHERT REPLACING EVAPORITES
Petrographic Characteristics and Fabrics of Chert-replaced Evaporites. Given the geochemical environments of chert formation discussed in earlier sections, it is not surprising to find chert commonly replacing evaporites. The hypersaline and generally high-pH environment of evaporite deposition permits silicon concentrations in solution to build up to high levels. Milliken (1979) summarized the petrographic features of chert replacing evaporites. This chert is typically nodular. As pointed out in the previous section, the most common silica fabric in chert-replaced evaporites is length-slow chalcedony or quartzine. However, as a single criterion this fabric is not sufficient evidence for the former presence of evaporites if all other testaments have been lost. Other quartz fabrics found in chert nodules replacing evaporites are (i) spherulitic quartzine, lutecite, and microflamboyant quartz; (ii) strongly undulose megaquartz; (iii)
§ (l) ~ (2) ~ (3)
EJ
(4)
c
fibrous quartz-overlays on euhedral megaquartz consisting of intergrown chalcedony and quartzine as well as zebraic chalcedony; and (iv) so-called "cubic quartz" crystals. These crystals generally do not represent a pseudomorphic replacement of cubic minerals, but are simple stubby crystals mostly without prism faces and often doubly terminated by one set of rhomb faces on either end. Finally, (v) anhydrite inclusions in megaquartz, quartzine and microfamboyant quartz represent the most convincing fabric evidence for evaporite replacement by chert. Milliken (1979) listed the following sequence of quartz fabrics in chert nodules replacing Mississippian evaporites of southern Kentucky and northern Tennessee (in the order of their occurrence from nodule edge to centre): (1) isolated spherules of quartzine in randomly fibrous microcrystalline quartz; (2) interlocking spherules of quartzine, lutecite, or microflamboyant quartz; (3) spherules of quartzine, lutecite, or microflamboyant quartz grading into "cubic megaquartz" ; (4) mosaics of megaquartz spherules and undulose megaquartz; (5) simple quartz crystals (normal euhedrally or "CUbically" terminated) and spherules of quartzine, microflamboyant quartz or megaquartz which occur isolated in secondary cavity fillings of calcite or dolomite (Figure 10d) indicating incomplete original
c
Diagenesis
266
evaporite replacement by chert; (6) overlays offibrous quartz (zebraic, chalcedonic, lutecitic, or quartzinic) on euhedral quartz terminations; and (7) breccias of various of the aforementioned quartz types in a secondary carbonate cement filling of the cavity. The absence of such secondary carbonate cementation leads to "rattling stones" ("Klappersteine"), in which quartzcrystals and quartzine spherules which were formerly suspended in the anhydrite fell to the bottom of the cavity (now geode) after anhydrite dissolution. None of the nodules studied by Milliken contain the entire sequence; one or more of the fabrics are usually missing and the relative importance of individual fabrics varies strongly between nodules (Figure 10). In most nodules, the rim starts with fabrics (2)or (3). If the uncemented central cavity is large, (4) may be missing. Fabrics (5)and (6) are mutually exclusive. However, the basic pattern, which has been recognized for more than a century, "is one of a rim of fibrous quartz (usually quartzine) followed inward by megaquartz.
Characteristics of the Replacement Pore Fluids. From oxygen-isotope analyses, Milliken (1979) derived temperatures and compositions of the pore fluids whi2h replaced the evaporites with chert. For the earliest microcrystalline quartz, connate marine water of ambient seawater temperature (25°-30°C) was considered the most likely fluid which had caused silicification. Ouartzine and megaquartz, which follow In the diagenetic sequence, formed at somewhat higher temperatures (up to 40°C) from mixed marine-meteoric pore waters. Zebraic chalcedony, which appeared after megaquartz, is isotopically slightly heavier and was interpreted to have precipitated from meteoric water at somewhat lower temperatures than the previous quartz types. Other well-documented examples of silicified evaporites have been published by Folk and Pittman (1971) , Siedlecka (1972, 1976), Chowns and Elkins (1974), Friedman and Shukla (1980), and Geeslin and Chafetz (1982). Relict hopper moulds in the Rocklea Dome cherts of the Pilbara Craton, Western Australia, probably preserve the oldest evidence (3300-3500 Ma) for halite precipitation on earth (Boulter and Glover, 1986).
MAGADI·TYPE CHERTS Cherts which result from the conversion of the chemically precipitated, silicarich mineral magadiite to silica form in alkaline lakes in semiarid climates (Eugster, 1967, 1969; Hay, 1968; Maglione, 1979; Rooney et al., 1969). Although related to a replacement process, these cherts are quite different in origin from most of the types hitherto discussed and come close to directly precipitated inorganic cherts.
Magadlite Precipitation. The hydrous sodium silicate magadiite NaSi70, 3(OHh· 3H20 was precipitated from alkaline waters of Lake Magadi in the East African rift valley during the Late Pleistocene, at times when the lake level was more than 10 m higher than today. Eugster (1967, 1969) first described the conversion of magadiite to silica and analyzed the chemical conditions of magadiite precipitation and conversion. Present-day Magadi brines may contain more than 30 mM dissolved silicon (or 1900 ppm SiO~. These high values are primarily obtained through evaporative concentration. Because of high pH levels, the brines appear to still be slightly undersaturated with respect to amorphous silica. A very high l pH results from the precipitation of trona (Na2Q03.e-NaHC03· 2H 20), which re moves considerable amounts of bicarbonate, thereby increasing the pH more rapidly than the concentration of dissolved silicon (Jones et a/., 1967). In a stratified lake, magadiite precipitation may periodically take place in water layers where lowering of the pH occurs without dilution. Thiscould happen near the pycnocline where methane-oxidation would produce CO 2, Eugster (1969) viewed the fine lamination in magadiite beds as varves. Magadiite also replaces host sediment as, for example, at Alkali Lake, Oregon (Rooney et a/., 1969). Here, veins at least 1 m deep are filled with magadiite precipitated by groundwater reactions.
Magadilte Conversion to Quartz Chert. The conversion of magadiite to silica through dehydration and sodium lossmay involve kenyaite (NaSi, ,02o.s(OH)4· 3H 20) as an intermediate phase, or it may directly produce a silica phase (see equations (2) to (4) in Eugster, 1969).(It rarely leads to silhydrite 3Si0 2• H20; Gude and Sheppard, 1972).The reaction is favoured by lowering of the pH or
Geoscience Canada Reprint Series 4
the sodium ion activity. Dilution accomplishes both. The conversion occurs in near-surface layers that are accessible to rain and runoff waters . The intermediate kenyaite has been observed in contact with magadiite in the field (Eugster, 1969, fig. 4). Only microcrystalline quartz, equant or fibrous, occurs in Lake Magadi cherts. The fibrous variety chalcedony may be length-fast or length-slow and occur in spherulites. It is likely that initially opal-A may be involved in the conversion, because for lowered pH values the solutions seem to become ·h!ghly supersaturated with respect to arrioephous silica. Eugster (1969) suggested tllat once some of the amorphous silica has recrystallized to microquartz, direct quartz precipitation may continue spontaneously, as long as the sodium produced is removed efficiently. However, if the solutions causing the conversion are sufficiently dilute, direct microquartz precipitation is possible. Eugster (1969) estimated that the minimum time necessary for the conversion was probably not more than a few hundred years, perhaps less, based on angular chert chips embedded in gravel beds interlayered with the cherts. Chert formations that have been interpreted as lacustrine Magadi-type cherts in the geological record include examples from the Cambrian Officer Basin of South Australia (White and Youngs, 1980),the Middle Devonian Old Red Sandstone of northern Scotland (Parnell, 1986), and Jurassic to Pleistoceneoccurrences in the western United States (Surdam et a/., 1972; Sheppard and Gude, 1986), including the wellknown Eocene Green River Formation (Eugster and Surdam, 1973). INORGANICALLY PRECIPITATED LAGOONAL AND LACUSTRINE CHERTS Direct inorganic silica precipitation in surface environments was first described by Peterson and Von der Borch (1965) from the Coorong Lagoon, the famous carbonate-precipitating locality from which primary dolomite formation was also first reported. The mechanism of direct inorganic silica precipitation which was formerly held responsible for the formation of bedded deep-sea cherts, but proven wrong for these deposits (Hesse, 1990b) occurs here In the lacustrine environment. The
c c C-
C C ·C
C C C
C C C C C C C
C C C C C C C C
C C C C C
C C
C C
C (
C
C C C
C C C
c (
C' C
c' ( c· c
c. c (
o
o o
o (
o
o o
o o
c o G
()
o o o
o c 0'
C)
l
B l ,~
Origin of Inorganic and Replacement Cherts
discovery has already been referred to in the context of silica-replaced carbonates. In the playa lakes associated with the Coorong Lagoon, silicon is dissolved from detrital silicates (mainly quartz) by alkaline lake waters (pH = 9.5 to 10.2). It is precipitated when the pH drops as a result of decaying plant matter during evaporation and ensuing hypersalinity. The initial silica phase impregnating the carbonates is opal-A. Disks of silica-impregnated sediment about 1 cm thick and 10 cm in diameter are formed from cracking of dried-up lake sediment. Opaline cherts from Lake Bogoria of the East African rift valley in Kenya happen to be associated with an alkaline lacustrine environment, but are in fact hydrothermal-volcanogeniccherts , because they were precipitated from hydrothermal hot springs (Renaut et al., 1986; Renaut and Owen, 1988). They occur only at sites of hydrothermal activity in the rift lake. Although they are diatom-rich, alternative silica sources besides hydrothermal are unlikely. These opaline cherts appear to have formed from a gelatinous siliceous precursor that began to Iithify before it was subjected to subaerial exposure. The precursor was probably not magadiite, otherwise the cherts would be expected to have been transformed to quartz. Inorganic silica precipitation in the sublacustrine environment resulted from a rapid drop in temperature and, possibly, pH. The planktonic diatom Melosira, which is abundant in these cherts, is indicative of a considerably fresher lake (alkalinity of 5-6 rneq-L-l) at the time of silica precipitation in the early to middle Holocene than today (900 meq- L'l). At the same time, the lake was deeper than at present and the sites of chert formation which are now exposed up to 1.7m above the lake level were 8-10 m under water. An ancient analogue of Coorong-type chert occurs in the Triassic rift grabens of Eastern North America. Wheeler and Textoris (1978) described such cherts from the Newark Group in the Deep River Basin of North Carolina and other Triassic rift basins . PEDOGENIC CHERTS: SILCRETES
Environments of Silcrete Formation. Another example of near-surface, inorganic silica precipitation is the formation of silcretes, a process closely
related to soil-forming and/or weathering processes . Silcretes develop under climatic and environmental conditions similarto other duricrusts (calcretes and ferricretes) and laterites. Summerfield (1983), in a study of silcretes from southern Africa, distinguished "weathering-profile silcretes" and "non-weathering-profile silcretes". The non-weathering-profile silcretes occur in the Kalahari Basin as silicified calcretes, playa sediments and other silicified evaporite-pan deposits. They represent the type of environment that is generally envisaged for silcrete formation (e.g., Smale, 1973): semi-arid climate, high evaporation rates, and highly alkaline pore-fluids that rise by capillary action and precipitate silica upon mixing with descending fluids of lower pH. The "weathering-profile silcretes" are found in the Cape coastal zone associated with deep weathering profiles up to 20 m thick. They occur under more humid climatic conditions than the first type. The soils are poorly drained with high organic activity, and pore-solutions have fluctuating pH-values which temporarily fall below 4 (Summerfield, 1983). The evidence for low pH comes from relatively high Ti02 levels (> 1%) compared to the titanium-poor (Ti0 2 < 0.2%) non-weathering-profile silcretes. The n02 occurs as iine-grained anatase within siliceous cements. Titanium is mobile as Ti(OH)4 at pH values below 5, but becomes insoluble as Ti02 at pH levels above 2 upon dehydration. As these silicified rocks have lost their aluminum, pH must have been below 4, at least temporarily or locally in micro-environments (Thiry and Millot, 1987). (Alternatively, Al 20a was removed by organic complexation). Although many silcretes are found at the surface, not all of them have formed there. Woolnough (1927), LangfordSmith and Drury (1965), and Hutton et al. (1972) considered silcretes to be surface products. However, Frankel and Kent (1937), Frankel (1952), Mountain (1952), and Bruckner (1966) presented evidence that silcretes may also form in the subsurface at depths of 20 to 30 m. This point of view was also adopted by Exon et a/. (1970), and clearly documented by Summerfield's (1983) work. Recent Australian silcrete studies are by Watts (1978) and Langford-Smith (1978).
267
Silcrete Fabrics. Summerfield (1983) distinguished four different fabric types in silcretes , similar to the five types of Smale (1973). These are: (1)grain-supported, (2)floating, (3) matrix and (4) conglomeratic fabric. In the grain-supported fabric, frame-work grains produce a self-sustaining fabric. The pore space is filled by optically continuous quartz overgrowths, or more rarely by opaline silica, chalcedony or microquartz. This variety corresponds to Smale's (1973) quartzitic type and is petrographically very similar (identical according to Smale) to sedimentary quartz arenites. Only its occurrence distinguishes this type of silcrete from the latter as surface-formed. Ouartzitic silcrete is abundant in the first occurrence of silcrete from the Arabian Gulf area described by Khalaf (1988). The floating fabric corresponds to Smale's terrazzo type. Framework grains constitute more than 5% by volume, float in the cement, and do not form a self-supporting fabric. The matrix fabric (which from a petrologist's point of view should properly be called cement fabric) contains less than 5% framework grains and includes both Smale's Albertinia and opaline or fine-grained, massive types. The conglomeratic fabric consists of abundant clasts ofthe floating or terrazzo type set in a usually brick-red to brown siliceous cement. Some clasts are very irregular and angular in shape, indicating a local derivation without significant transport. Others are remarkably rounded, either due to substantial transport or (and) chemical attack. The quartzitic or grain-supported types with optically continuous quartz overgrowths have in common that they occur topographically on elevated grounds. This possibly reflects environments of relatively low supersaturation of the Silica-precipitating fluids, i.e., supersaturation with respect to quartz, but not with respect to the less wellordered silica phases. Floating (or terrazzo) types, on the other hand, are found on flats or below the surface (Smale, 1973). Examples of Ancient Silcretes. Ancient examples of silcretes and silicified paleosols from the Eocene of the Paris Basin and the Central Massif of France are rich in titanium oxides and appear to have formed under seasonally humid climatic conditions (Thiry and
268
Millot, 1987). They occur in the lower Eocene as horizons up to 3 m thick with a distinct zonation from granular (bottom) through columnar to pseudobrecciated chert (top). Eluviation and iIIuviation features typical of soil formation are abundant. The mineralogy of the secondary (non-detrital) silica phases is vertically zoned. Opal-A prevails at the bottom, microcrystalline quartz in the middle, and euhedral megaquartz in the upper part ofthe profile (Figure 11). This suggests downward-migrating solutions of increasing silica supersaturation. As emphasized by Thiry and Millot (1987), the presence of impurity elements has an important effect on the type and size of quartz crystallizing from solution. Times of precipitation alternated with leaching periods during which clay minerals were dissolved and detrital quartz grains corroded. Early to Middle Proterozoic silcretes of the Churchill and Bear Provinces of the Canadian Shield appear to be the oldest silcretes recognized in the geologic record. They formed under arid climatic conditions, as suggested by paleoclimatic indicators of aridity in associated formations (Ross and Chiarenzelli, 1985). They are low in Ti0 2 , but titanium content alone without other independent evidence would not be a sufficient criterion for aridity in the Precambrian, as these authors point out. Titanium mobility would have been low under both arid and hum id climatic conditions, because in the absence of a vegetation cover Precambrian soil waters would not have become as acidic as under humid conditions later in the Phanerozoic. Humid-climate P.recambrian soils, especially silcretes would, therefore, also be low in Ti0 2 . Other ancient examples include halite silcretes in the Cambro-Ordovician Whitehall Formation of eastern New York State (Rubin and Friedman, 1981), and the Maestrichtian to Eocene alunitic silcretes of western Portugal (Meyer and Pena dos Reis, 1985). In the Pliocene Ogallala Formation of western Kansas weathering processes have led to the redistribution of biogenic (diatoms) and volcanogenic silica (bentonites) in the subsurface (Swineford and Franks, 1959).
Diagenesis
HYDROTHERMAL-VOLCANOGENIC CHERTS Silicification associated with hydrothermal activity is a major alteration mechanism in volcanic sequences. The process may occur at elevated temperatures and fall on the borderline between diagenesis and metasomatism/metamorphism. Davis' (1918) hypothesis that the silica in the Franciscan radiolarian cherts in California was derived from silica-r ich sea-floor springs emanating from centres of submarine volcanism gave rise to the view that chert in general was of volcanic origin. This theory prevailed for more than half a century, until disproved by evidence from modern sea-floor studies as outlined in Hesse (1990b). The problem of differentiating volcanogenic chert from other types of chert in the geologic record remains nevertheless important. However, few studies of silica deposition and silicification in modern volcanic terrains or geothermal areas seem to be available. The only detailed studies are by Rimstidt and Cole (1983)and Keith et al. (1978)in active geothermal areas on land in Nevada and Wyoming.
Environments of Volcanogenic Chert
Formation. Silicification related to hydrothermal activity occurs at different levels in volcanic edifices, both in submarine and continental environments. Three settings may be distinguished: (1) the ascent-routes of the fluids within volcanic complexes which experience STRUCTURES
SAND
Geoscience Canada Reprint Series 4
widespread and pervasive silicification; (2) isolated ponds and depressions in the median-rift valley of mid-ocean ridges where the fluids discharge and may accumulate siliceous sediments; (3) high-temperature geothermal areas on land associated with spreading lineaments, more commonly with volcanic island arcs, and also with transform fault zones or continental intraplate volcanism (hot spots). They may accumulate large amounts of siliceous deposits. These environments of silicification are typically sites of base-metal and iron-enrichment. Attention invariably has centred on the valuable metals and not on iron and silica, although the latter are much more abundantly transported components in these systems. This accounts forthe common association of the ore bodies with iron-formation and cherty siliceous strata (Hutchinson , 1982). Despite the large number of occurrences, little is therefore known in detail about the silicification process. Apparently the ore-transporting fluids which also affect silicification are strongly chlorinated, highly reduced and weakly acidic (Hutchinson, 1982).
Silicification Within Volcanic Edifices. Within the volcanic edifices, silicification occurs in alteration zones which tend to be intensely pyritized and commonly sericitized. Sericitization indicates ac idic hydrothermal solutions (Meyer and Hemley, 1967). Hydrothermal silica precipitation within volcanic sequences (type 1 above) occurs before SILICA
GRAIN SIZE ILLUVIATED 10
1
01
001mm
PSEUDO -BRECCIATED
COlUMNAR
c c ( c
c c c c c c c c c c c
c
c
c c c
c
c c c c c c c c c c
c c
c: (
GRANULAR
CLAYS
Figure 11 Schematic vertical profile through Lower Eocene silcrete horizon, Paris Basin. (From Thiry and Millot, 1987).
c c c
c (
~.
c C
( (
c (
( ( ( ( ( (
o o o
o o o o
o o o o o o o o r:
V
o C·
6 r· , '.
'-
Origin of Inorganic and Replacement Cherts
the solutions discharge on the sea floor. The hot, silica-rich solutions rise through faults and poorly cemented volcanic breccias and pillows. Upon cooling near the top of the sequence they precipitate Si02 as long as no major dilution by seawater takes place. Silica Deposition by Submarine Exhalations. The alteration zones typically mark the foot-wall alteration of ferruginous chert and other metal-rich deposits formed by exhalative volcanic activity on the sea floor (type 2 above). In the great majority of cases, the hydrothermal solutions become diluted rapidly after discharge by mixing with strongly silicaundersaturated seawater.Therefore silica precipitation does not occur. Phanerozoic siliceous deposits in the pelagic sediments overlying the basaltic oceanic crust are therefore almost exclusively biogenic and not volcanogenic in origin. Only where discharge rates are particularly high and/or discharge takes place in topographic depressions filled with stagnant brines, is silica precipitation likely to occur. This is why so few examples of modern submarine hydrothermal silica deposits have been reported from the recently discovered hydrothermal vents in the median rift valley of the mid-ocean ridge-system. The first known example of this kind was the Atlantis II Deep in the Red Sea (Bischoff, 1969). The nontronite (ironmontmorillonite) and goethite-rich hydrothermal sediment-layers in this depression contain about 20 wt.% amorphous silica which was probably chemically precipitated from solution. The FAMOUS area on the Mid-Atlantic Ridge is another site where silica- and iron-rich hydrothermal sediment was precipitated in the immediate vicinity of the vent (Lalou et al., 1977). The silica apparently co-precipitated with iron-hydroxides and oxides and was followed by Mn-oxides. As the oceanic crust migrates away from the ridge crest, hydrothermal circulation and discharge continue in large cylindrical or ellipsoidal convection cells of 5·10 km diameter, which extend outward on the ridge flanks to 80 Ma old crust (Anderson et ei., 1979; Fehn, 1986). Whereas recharge occurs through wide sediment-covered areas, discharge appears to be focussed in narrow zones along fault scarps ortopo-
graphic highs of exposed basement. In exceptional cases, hydrothermal deposits have been discovered in such localities. For example in the Lau Basin, a back-arc basin in the Southwest Pacific, hydrothermal opal and barite occur on a sediment-free basement ridge 750 m above the basin floor (Bertine and Keene, 1975). Ancient equivalents of type 2 quoted in Barrett (1979) include the siliceous sinter, ironstone and Mn lenses in a Tertiary ophiolite suite in the British Solomon Islands which occur above massive sulphide bodies and highly altered silicified lavas (Taylor, 1974). In Cyprus, Cretaceous ochres and umbres with an average 20% Si0 2 (Robertson, 1976) overlie massive sulphide orebodies and silicified, pyritebearing stockworks of the Troodos ophiolite (Constantinou and Govett, 1972; Robertson and Hudson, 1973). Carboniferous cherty ironstones at Tynagh, Ireland, have been interpreted as exhalative deposits (Russell, 1975). Other Paleozoic cherty iron-formations in the southern Urals are associated with spilitic volcanics and silicified and sericitized alteration zones pointing toward a similar exhalative origin (Ivanov, 1971). Ordovician banded ironformations in the Bathurst mining district, New Brunswick, fall into a similar category although different geotectonic setting (Troop, 1984). Finally, many large massive sulphide deposits in the Archean greenstone belts of the Canadian Shield are accompanied by intense silicification and sericitization which in all likelihood are of hydrothermal origin (Barrett, 1979). Silica Deposition in Geothermal Areas on Land. Siliceous sinter deposits in high-temperature geothermal areas on land represent the third type of volcanogenic silica accumulations. They are low-temperature surface or near-surface products of high-temperature processes in the subsurface. Present-day systems are commonly associated with geyser activity. Recent or sub-recent deposits that have been studied have undergone only moderate to minor diagenesis. The sinter deposits may form terraces around hydrothermal vents and consist of Iight-coloured (white, tan or grey), commonly banded masses of friable or dense, well-cemented amorphous silica . The hydrothermal solu-
269
tions at depth must contain sufficient dissolved silicon to become supersaturated with respect to opal-A during ascent when they cool below 100°C. In the reservoir, silicon concentration is controlled by quartz solubility. A hydrothermal solution precipitating opal-Aat 100°C must contain at least 370 ppm(or about 6 mM) dissolved silicon, which,in equilibrium with quartz, requires a reservoir temperature above 235°C (Figure 12). Foropal-A precipitation at50°C, the reservoir temperature in the subsurface must exceed 165°C to generate a high enough silicon concentration from quartz dissolution, excluding boiling as in the 100°C situation. This quartz (or silica) geothermometer (Fournier and Rowe, 1966) is widely applied in geothermal regions and its limitations were discussed by Rimstidt and Barnes (1980). Reservoirs with temperatures lower than those calculated from the quartz geothermometer are unlikely to produce siliceous sinter accumulations. The Beowawe geothermal system of north-central Nevada has a temperature range of 200°-250°C near 3000 m depth (Rimstidt and Cole, 1983). It occurs along the trace of a grabenbounding fault and feeds boiling hot springs with a temperature of 93°C which is the boiling point of water at the elevation of 1500 m. The associated silica sinter deposit is 65 m thick and 1.6 km long. It is well bedded and displays layering on the metre to millimetre scale (lamination). Mineralogically it consists dominantly of opal-Awith traces of quartz. Opal-CT is absent. Silica deposition occurs through heterogeneous nucleation on suspended solids. Homogeneous nucleation is not possible in the area near the vent where most of the precipitation is observed because the level of supersaturation of the fluid is insufficient and temperatures are too high. Agglomeration and attachment of suspended particles to the surface of the deposit is aided by rapid precipitation of an opaline silica cement once particles come in contact with one another. High turbulence of the water exiting the vent largely prevents deposition, and the particles are swept off to be deposited more distally. The Beowawe system provided important information on surface processes of silica precipitation and deposition. In another study of a mod-
c 270
Diagenesis
Geoscience Canada Reprint Series 4
c (
ern system in the Upper Geyser Basin , Yellowstone National Park, Wyoming, Keith et a/. (1978) documented the mineralogical and petrographic composition of these deposits in the subsurface. It is based on two drill-holes up to 150 m deep . One is centred on the zone of rising hot fluids and is very close to hot spr ings, the other one is a little over a 100 m away. The hot spring water has a temperature of 93°C at the surface and rises from a reservoir which is probably at a temperature above 200°C. Ground temperatures above 80°C are encountered at depths of between 6 and 15m. Silicification of the sediments is by silica cementation, especially of sinter. Devitrification of volcanic glass contributes only minor amounts of silica minerals. Hydrothermal activity in this area has probably continued for the last 12,000 years or more . Sinter forms at the surface as layers of amorphous opaline silica. These are deposited from exiting hot spring water which has a maximum silicon content of 330 ppm and becomes rapidly supersaturated upon cooling. Most deposits consist of angular fragments of primary sinter cemented by amorphous silica. At a depth of less than 8 m, o-crlstobalite (disordered opal-CT) appears at the expense of opal-A. The overlap in the depth range of these two phases is narrow. Chalcedony and occasionally quartz occur in veinlets at greater depth. The silica minerals observed in the drill holes are interpreted as a primary sequence of precipitates which reflect decreasing supersaturation of the solutions with increasing temperature, ratherthan an alteration sequence which is plausible as little burial has occurred. Alteration of volcanic glass starts with hydration (perlitization), which increases the water content by 2-5%. (Non-perlitic obsidian has 0.3% H2 0 ). Unaltered rhyolitic obsidian still persists in the center of a flow breccia at a temperature of 170°C. This is apparently the result of reduced water accessibility due to low permeability. Further alteration leads to devitrification through zeolite and smectite formation or complete dissolution. Euhedral tridymite seems to be a high-temperature devitrification product associated with lava extrusion; o-crtstoballte occurs mainly as an alteration product of obsidian filling the holes of "sieved plagioclase" pheno-
crysts ; it is associated with celadonite, montmorillonite and mordenite. The various silica phases show typical associations with particular zeolites: e-crlstobaflte with clinoptilolite; hydrothermal chalcedony or quartz with analcime. Hydrothermal quartz forms conspicuous overgrowths on clastic quartz. Cementation resulting from these alteration processes, which also include reactions other than those involving silica minerals, appears to have a selfsealing effect on the system . This is shown by significant differences in fluid pressures at the base of the 50 m thick sedimentary section in the two drill holes. PRECAMBRIAN CHERTS Cherts constitute a greater proportion of the sedimentary record in the Precambrian than in the Phanerozoic because of the abundance of chert in Precambrian (banded) iron formations (Pettijohn, 1975, p. 594). Excellent Precambrian analogues of modern environments of chert formation have been documented in literature, such as the Proterozoic silcretes in the Bear and
Churchill Provinces ofthe northwestern Canadian Shield (Ross and Chiarenzelli, 1985). However, the problem of the origin of many of the Precambrian cherts, particularly the cherty iron formations, remains unresolved. This applies to the fine-grained banded iron formations ofthe "Algoma type" as well as to the coarser-grained " Superiortype" deposits. Models proposed for the dominantly Proterozoic Banded Iron Formations (BIFs) range from organic (LaBerge, 1973) to inorganic (Cloud, 1973), and from continental-lacustrine (Magaditype, Eugster, 1969;Eugster and Chou, 1973), or continent-derived/fluviatile (Garrels, 1987) to restricted or open marine (Drever, 1974; Holland, 1973). The latter models, which involve either (i) stratified basins (or oceans), (ii) precipitation at the pycnocline or, (iii) oceanic upwelling, are particularly attractive as they invoke a reservoir of reasonable magnitude for iron and silica. As far as silica is concerned, the stratified-basin requirement is not necessary, because the silica concentration of Precambrian seawater may
c c
c c c c c c
c c
Q,
>
--.t-... m :3
c c c c c c c c c c c c
c
[1000
oU)
c c
c
c c
500
c
100
200
TEMPERA TURE,
300
°c
c c
e
c
Figure 12 " Quartz-geothermometer " for hydrothermal systems using solubility of amorphous silica (opal-A) and quartz as a function oftemperature. (From RimstidtandCole, 1983). A hydrothermalsolution precipitating opa/-A at 100°C contains 372ppm dissolvedsilicon which , in equilibrium with quartz, requires a reservoir temperature of 2380C.
(
C (
(
C C
264
chalk, the concentration of iron-bearing minerals(mainly clay minerals)is notoriously low. After the termination of sulphide precipitation, the build-up of HSand S210wered the pH to the extent that the system switched to carbonate dissolution and accompanying silica precipitation. Sulphate reduction extended a certain distance from the burrow into the chalk because of increased organic matter content and enhanced permeability in the burrow. The latter facilitated the resupply of sulphate. The burrow and its immediate surroundings therefore became an anoxic environment, whereas the host-sediment outside the paramoudra structure was generally still oxic. Sulphide diffusing outward from the sulphate-reduction zone of the paramoudra structure toward the oxic surrounding therefore would be re-oxidized bacterially lowering the pH even more. The transition from anoxic to oxic conditions thus appears to coincide with the cemented chalk/flint boundary. This is supported by the Fe and Mn distributions. Fe is enriched in the cemented chalk over the slightly-cemented carbonate host sediment because of Fe-sulphide precipitation in the chalk core. It is strongly depleted at the chalk/ flint boundary because of the termination of sulphide precipitation at that point. It shows a second maximum at some distance from this boundary (sample 3, Figure 9) revealing immobilization by oxidation of some reduced iron that escaped sulphide precipitation. Mn 2 + , which is slightly more mobile in a weakly oxidizing environment than Fe2+ • has a secondary maximum at a slightly greater distance from the boundary (sample 4) where it underwent oxidation as well. The source of silica for the paramoudra structure in the chalk does not pose a problem. Siliceous tests and skeletons (sponge spicules, radiolaria, diatoms) are abundant in the chalk. Dissolution of the opaline silica increased silicon concentration during early diagenesis beyond the solubility of opalCT, which, according to Clayton (1986), was the first diagenetic silica phase to precipitate. Today, the flint nodules consist entirely of quartz and chalcedony. The paramoudra structure has been used by Clayton (1986) as basis for a general model for the formation of bedded flint in chalk. According to the
Diagenesis
Geoscience Canada Reprint Series 4
, / -', ... 1
1---+-+-+-+-+-+-+-+-+--::7'f--+--.."'.'--+-+-+--t---j r
Feb
Ca./_-t-r-t--t- t_\ . _ t--1I---+_ -i
Mg 0.-8~-..B .... cC'"-S)",B'~ i ~~ 1000 I-A ' f-
I~. ..
1\
;e"'~I--C
·~ri""ri~~:.J
I
11
1.1\
I
f-M,P-ng,:::g:"'O_6_6_0-0,,:".6,~, I
1 \ .FF ' ~I 1000
0
1---+-+-+-+-+-+-+-+1
9 Mg .~-~:.: =p'=:
\
I
.
I
1:':;
I
~".....,Q:-:. ~~"'frn·;.::~~ ·~."1-1-:~"~.~nV-V"V"·v-·O:..... \" -N",+-:+ -+'
I , --+ I I I I K +' -·+--+_+--+e.
-ee,' '.
i'i'O,¥TI Ji 100 ~~~~~~~33~\-':~ , } ' N'
1--+-+-+-+-+-1--1--+ I-~
1---+-+-+-+-+-+-+-+ 1-0-00"'
I
W/ '0/: 0
.,
-:
~o().-o
c
c c
r :' i .1-.:'. !i~
\
c c c c c c c c c
ti: 100 \0.=70:::.0:= .,I~ I--
:-I--+-~
'~+-1:- I - + - - - i
c c c
c
c c c
, 0--0
I~
c
c o
c c
0
\, 1\ I
ABC
S'3C
o EF GH 1
I
1.80
I
/1
/t\
1
c
c
2 3 4 5 678 9 CA/4
- ".60 -
.
.
SI3C 1_1--1-
1.60 %0
-1.0
~_I"
];/
',+
I - I - I.....
S"O'l-I/
1.40 %0 1.30
S"O
-1.50 -1.60
'+ I
I
I
I
l!n::~::~Y~: "1mB -CEMENTED CHALK CORE
I
fliNT
c
c c c
c.
-
, ....
c
C
l
C
e <..
I UNCEMENTEDHOST
<..
C C
Figure 9 Trace element and carbon- and oxygen·isotope distribution across paramoudra flint structure from central burrow (left) to sample Ca/4 in the host sediment. (Modified from Clayton, 1986).
C (
C C
(
271
Origin of Inorganic and Replacement Cherts
c' eel r:
c [J
c o
r
o o o
0,
o
o o o o o c o on v o o r. o o
'-
have been considerably higher (20 to 100 ppm according to Holland and Malinin, 1979) than that of presentday ocean water. Such concentrations are plausible if there were no efficient silica-consuming organisms to counterbalance the silica input by rivers , submarine volcanism, and the pore-water reflux in Precambrian oceans. This would also explain the occurrence of Precambrian cherts in shallow to deep marine environments, for which no evidence of an evaporite association exists. An alternative origin for these cherts would be silicification by hydrothermal-volcanogenic replacement (e.g., Hughes, 1976; Simonson, 1985). Simonson (1987) has presented evidence for pervasive early diagenetic chertification in these rocks, including the silicification of carbonate ooids. Walker (1984) suggested that early diagenesis in the BIFs occurred under suboxic conditions. CONCLUSIONS Review of the literature on inorganic and replacement cherts shows that diagenetic pathways, phase transformations, reaction rates , timing and burial depths at which silicification reactions occur in these rocks may be quite different from those in biogenic siliceous sediments in the open oceans discussed in the previous paper (Hesse, this volume, p. 227-251). In certain cases, however, e.g., in silicified wood, diagenetic phenomena may be similar or identical to those in pelagic siliceous sediments. The formation of nodular cherts in limestones is instructive concerning the timing of silicification reactions relative to well-established carbonate-diagenetic reactions. Virtually all theoretically possible sequences occur in nature including very early chert formation at very shallow burial depths. The latter situation is very different from deep-sea siliceous sediments, in which the silica phase transformations take millions or tens of millions of years and considerable burial depths to start. Field evidence has also been presented for early chert formation in ancient argillaceous continental margin sediments. This evidence remains controversial, however, because it is difficultto reconcile it with experimental results suggesting slow reaction rates in argillaceous sediments. There is conclusive
evidence that the well-established sequence of silica transformations in pelagic siliceous sediments from opal-A to opal-CT to microquartz may not be adhered to in many nodular cherts in limestones. The geochemical environment and conditions of carbonate replacement by silica remain poorly known in detail. Therefore, the differences between silica diagenesis in limestones on the one hand and diatomites and radiolarites on the other are incompletely understood. They are explained only in terms of Knauth 's (1979) general mixing model for shallow-water carbonates which awaits testing in a suitable ancient or modern example. An exception is Clayton's (1986) study of flint formation in the English Chalk. The chalk is a shelf equivalent to deep-sea pelagic limestones, which commonly also contain chert nodules probably formed under similar conditions as the flint nodules. The question of why nodules or concretions grow exactlywhere they do and not elsewhere is an unresolved dilemma concerning nodular or concretionary rocks. Clayton 's (1986) model attempts some answers to this question for the flint nodule bands implying basin-wide sedimentation pulses that might be related to sea-level changes or other exogenic forces. In other cases, nodule distribution is probably strictly stochastic. The point of initial nucleation may represent only a slight perturbation of the geochemical environment. Average nodule spacing may be controlled by diffusional distances. The variety of rock types and materials undergoing silica-replacement is impressive and the extent ranges from minor to pervasive. Fossil wood, Phanerozoic carbonates and carbonatebearing sandstones and evaporites commonly undergo partial or minor silicification. However, examples of completely silicified beds , members or formations are known . In Phanerozoic examples, the source ofthe silica forthe replacement is predominantly biogenic. Lacustrine , pedogenic and hydrothermal-volcanogenic cherts are predominantly inorganic in origin and have the tendency for pervasive replacement which, in volcanic edifices in particular, may affect entire formations. Direct chemical silica precipitation from aqueous solution which was thought to be the main chert-forming process at
the beginning of the century has only been described from lacustrine, pedogenic and hydrothermal-volcanogenic environments. Hydrothermal silicification processes in and near modern vent systems on mid-ocean ridges have been studied little so far because of inaccessibility, but they offer a prospective area of study for future ocean drilling. Much work, including experimental study, remains in order to elucidate the chemical conditions of chertification in carbonates. Answering the remaining unsolved questions on the origin of Precambrian cherty iron-formations will potentially also shed light on the composition of seawater in Precambrian times. ACKNOWLEDGEMENTS Funding for the author's diagenesis research was provided by the Natural Sciences and Engineering Research Council of Canada (Grant A7368). A slightly modified version of this paper was first published in Earth Science Reviews(v. 26, p. 253-284). Permission to reprint the article was granted by Elsevier. S. Calvert, E. McBride and T. Barrett read earlier drafts of the manuscript and suggested substantial improvements. Reviewers for Earth Science Reviews were D. Houseknecht, G. de V. Klein and Anonymous, whose comments have been largely incorporated into the manuscript. The originals for Figures 3c and 3d were provided by B. Simonson. R. Yates helped with the drafting and photography. Fortheirsupport and assistance the author wishes to express his gratitude to these colleagues who gave freely of their time , particularly T. Barrett. REFERENCES Silicification of Fossil Wood Buurman, P., 1972, Mineralization of fossil wood: Scripta Geologica , v. 12, p. 1-43. Drum, R.W., 1968, Silicification of Betula wood tissue in vitro: Science, v. 161, p.175·176. Hesse, R., 1990b, Origin of chert: Diagenesis of biogenic siliceous sediments, in Mcllreath, I.A. and Morrow, DW., eds., Diagenesis : Geological Association of Canada, Geoscience Canada Reprint Series 4, p. 227-251, Leo, R.F. and Barghoorn , E.S., 1976,Silicification of wood: Harvard University, Botanical Museum Leaflets , v. 25,47 p.
( 272
Merrill, A.C. and Spencer, R.w., 1950,Sorption of sodium silicates and silica sols by cellulose fibers: Industrial and Engineering Chemistry, v. 42, p. 744-747. Murata, K.J., 1940, Volcanic ash as a source of silica for the silicification of wood: American Journal of Science, v. 238, p.586-596 . Sigleo, A.C., 1978,Organic geochemistry of silicified wood, Petrified Forest National Park, Arizona: Geochimica et Cosmochimica Acta, v. 42, p. 1397-1405. Stein, C.L., 1982, Silica recrystallization in petrified wood: Journal of Sedimentary Petrology, v. 52, p. 1277-1282. Chertificatlon of Carbonates and Carbonate-bearing Sandstones
Badiozamani, K., 1973, The Dorag dolomitization model - Application to the Middle Ordovician of Wisconsin: Journal of Sedimentary Petrology, v. 43 , p.965-984. Banks, N.G., 1970, Nature and origin or early and late cherts in the Leadville Limestone , Colorado: Geological Society of America, Bulletin, v.81, p. 3033-3048 . Bromley, A.G. and Ekdale, A.A., 1986, Flint and fabric in the European chalk, in Sieveking, G. de G. and Hart , M.B ., eds., The Scientific Study of Flint and Chert: Cambridge University Press, p. 71-82. Brown, G., Catt, J.A., Hollyer, S.E. and Oilier, C.D., 1969, Partial silicification of chalk fossils fromtheChiiterns: Geological Magazine, v. 106, p. 583-586. Chanda, S.K ., Bhattacharyya , A. and Sarkar, S., 1976, Early diagenetic chert nodules in Bhander Limestone, Maihar, Satna District, Madhya Pradesh, India: Journal of Geology, v. 84, p. 213-224. Chanda, S.K. , Bhattacharyya , A. and Sarkar, S., 1977, Deformation of ooids by compaction in the Precambrian Bhander Limestone , India: Implications for lithification: Geological Society of America, Bulletin, v. 88, p. 1577-1585. Choquette, P.W., 1955, A petrographic study of the "State College" siliceous oolite: Journal of Geology, v. 63 , p. 337-347. Chowns, T.M. and Elkins, J.E., 1974, The origin of quartz geodes and caulifower cherts through the silicification of anhydrite nodules: Journal of Sedimentary Petrology, v. 44, p. 885-903. Church, A.H., 1862, Observations on silica: Journal of the Chemical Society, v. 15, p.l07-110. Clayton, C.J., 1986, The chemical environment of flint formation in Upper Cretaceouschalks, in Sieveking, G. de G. and Hart, M.B., eds., The Scientific Study of Flint and Chert: Cambridge University Press, p. 43-54.
Diagenesis
Geoscience Canada Reprint Series 4
C ( (
Coniglio , M., 1987, Biogenic chert in the Cow Head Group (Cambro-Ordovician), western Newfoundland: Sedimentology, v. 34, p. 813-823. Cox, G.H., Dean, R.S. and Gottschalk, V.H., 1916, Studies on the origin of Missouri chert and zinc ores: Missouri University School of Mines and Metallurgy, Bulletin, Technical Series, v. 3, no. 2, 34 p. Dapples, E.C.,1959, The behaviour of silica in diagenesis, in Ireland, H.A., ed., Silica in Sediments: Society of Economic Paleontologists and Mineralogists, Special Publication No. 7, p. 36-54. Folk, R.L., 1975, Third-party reply to Hatfield : Discussion of Jacka A.D., 1974, Fossils by length-slow chalcedony and associated dolomitization: Journal of Sedimentary Petrology, v. 45, p. 952. Folk, R.L. and Pittman , J.S., 1971, Lengthslow chalcedony : a new testament for vanished evaporites : Journal of Sedimentary Petrology, v. 41, p. 1045-1058. Frondel, C., 1962, Dana's System of Mineralogy, v. 3: The Silica Minerals: Wiley, New York, 334 p. Frondel, C., 1978, Characters of quartz fibers: American Mineralogist, v. 63, p. 17-27. Geeslin , J.H. and Chafetz, H.S., 1982, Ordovician Aleman ribbon cherts: An example of silicification prior to carbonate lithification: Journal of Sedimentary Petrology, v. 52, p. 1283-1293. Hatfield, C.B., 1975,Aeplacementoffossils by length-slow chalcedony and associated dolomitization: Discussion: Journal of Sedimentary Petrology, v. 45, p.951-952. Henderson, R.A., 1984, Diagenetic growth of euhedral megaquartz in the skeleton of a stromatoporoid: Journal of Sedimentary Petrology, v. 54 , p. 1138-1146. Hesse, R., 1973,Diagenesis of a seamount oolite from the West Pacific, Leg 20, DSDP, in Heezen, B.C., MacGregor, LD. et a/.: Initial Aeports of the Deep Sea Dr illing Project , v. 20 , p .363-387 : United States Government Printing Office, Washington, D.C. Hesse, R., 1987, Selective and reversible carbonate-silica replacements in Lower Cretaceous carbonate-bearing turbidites of the Eastern Alps: Sedimentology, v. 34, p. 1055·1077. Hesse, A., 1990a, Early diagenetic pore water/sediment interactions: Modern offshore basins, in Mcllreath, LA. and Morrow, D.W., eds., Diagenesis: Geological Association of Canada, Geoscience Canada Reprint Series 4, p. 277-316. Hesse, A., 1990b,Origin of chert: Diagenesis of biogenic siliceous sediments, in Mcllreath, LA. and Morrow, D.W., eds., Diagenesis: Geological Association of Canada, Geoscience Canada Reprint Series 4, p. 227-251.
Hinman, N.W., 1987, Organic and inorganic chemical controls on the rates of silica diagenesis: A comparison of a natural system with experimental results, unpublished Ph.D. thesis, University of California at San Diego, 381 p. Jacka,A.D., 1974, Replacementoffossilsby length-slow chaldecony and associated dolomitization: Journal of Sedimentary Petrology, v. 44, p. 421-427. Kastner, M., 1980,Length-slow chalcedony: the end olthe newtestament: EOS,v. 61, p.399. Keene, J.B., 1983,Chalcedonic quartz and occurrence of quartzine (length-slow chalcedony) in pelagic sediments: Sedimentology, v. 30, p. 449-454. Knauth, L.P., 1979, A model for the origin of chert in limestone: Geology, v. 7, p.274-277. Knauth, L.P. and Epstein, S., 1976, Hydrogen and oxygen isotope ratios in nodular and bedded cherts : Geoch imica et Cosmochimica Acta, v. 40, p. 1095-1108. Land, L.S., 1979, Chert-chalk diagenesis: The Miocene island slope of north Jamaica:Journal of Sedimentary Petrology, v. 49, p. 223-232. Lohmann , K.G. and Meyers, W.J., 1977, Microdolomite inclusions in cloudy prismatic calcites : A proposed criterion for former high magnesium calcites: Journal of Sedimentary Petrology, v. 47, p. 1078-1088. Lovering, T.G. and Patten, L.E., 1962,The effect of CO2 at low temperature and pressure on solutions supersaturated with silica in the presence of limestone and dolomite: Geochimica et Cosmochimica Acta, v. 26, p. 253-284. McBride, E.F.,1988,Silicification of carbonate pebbles in a fluvial conglomerate by groundwater: Journal of Sedimentary Petrology, v. 58, p. 862-867. McBride, E.F. and Folk, A.L., 1977, The Caballos Novaculite revisited: Part II: Chert and shale members and synthesis: Journal of Sedimentary Petrology, v. 47, p. 1261-1286. Meyers, W.J., 1977, Chertification in the Mississippian Lake Valley Formation , Sacramento Mountains, New Mexico: Sedimentology, v. 24, p. 75-105. Meyers, W.J. and James, A.T., 1978, Stable isotopes of cherts and carbonate cements in the Lake Valley Formation (Mississippian), Sacramento Mts., New Mexico: Sedimentology, v. 25, p. 105-124. Michel-Levy, A.andMunier-Chalmas,C.P.E., 1892 , Memo ire sur diverse formes attectees par Ie reseau elementaire du Quartz: Bulletin de la Societe Francalse de Mineralogie, v. 15, p. 159-190.
C ( ( ( ( ( ( ( ( (
C ( (
C C C C C C C C
C C
C (
C (
C C C (
C C
C C C C
C C
r r
273
Origin of Inorganic and Replacement Cherts
r r
r
o o o o o o o o
o o o
o o o o o
-.
Milliken , K.L., 1979, The silicified evaporite syndrome - two aspects of silicification history offormer evaporite nodules from southern Kentucky and northern Tennessee: Journal of Sedimentary Petrol ogy, v. 79, p, 245-256. Namy, J.N., 1974, Early diagenetic chert in the Marble Falls Group (Pennsylvanian) of central Texas: Journal of Sedimentary Petrology, v. 44, p. 1262-1268. Richter, D.E., 1972, Authigenic quartz preserving skeletal material: Sedimentology, v. 19, p. 211-218. Schmitt, J.C. and Boyd,D.W., 1981, Patternsof silicification in Permian pelecypods and brachipods from Wyoming: Journal of Sedimentary Petrology, v. 51, p. 1297-1308. Simonson , B.M. , 1987, Early silica cementation and subsequent diagenesis in arenites from four Early Proterozoic Iron Formations of North America: Journal of Sed imentary Petrology, v. 57, p.494·511, Swett, K., 1965, Dolomitization, silicification, and calcitization patterns in Cambro-Ordovician oolites from Northwest Scotland : Journal of Sedimentary Petrology, v. 35, p. 928-938. Wilson, R.C.L., 1966, Silica diagenesis in Upper Jurassic limestones of southern England: Journal of Sedimentary Petrology, v. 36, p. 1036-1049. Wood, G.V. and Armstrong, A.K., 1975, Diagenesis and stratigraphy of the Lisburne Group limestones of the Sadlerochit Mountains and adjacent areas , northeastern Alaska : United States Geological Survey, Professional Paper 857,41 p.
Chert Replaced Elfaporites Boulter, C.A. and Glover, J.E., 1986, Chert with relict hopper moulds from Rocklea Dome , Pilbara Craton, Western Australia: An Archean halite-bearing evaporite : Geology, v. 14, p. 128-131. Chowns, T.M. and Elkins , J.E., 1974, The origin of quartz geodes and cauliflower cherts through the silicification of anhydrite nodules: Journal of Sedimentary Petrology, v. 44, p. 885-903. Folk, R.L. and Pittman , J .S., 1971, Lengthslow chalcedony: a new testament for vanished evaporites: Journal of Sedimentary Petrology, v. 41, p. 1045-1058. Friedman , G.M. and Shukla , V., 1980, Significance of authigenic quartz euhedra after sulfates: Example from the Lockport Formation (middle Silurian) of New York: Journal of Sedimentary Petrology, v. 50, p. 1299-1304. Geeslin. J .H. and Chafetz. H.S ., 1982, Ordovician Aleman ribbon cherts: An example of silicification prior to carbonate lithification: Journal of Sedimentary Petrology, v. 52, p. 1283-1293.
Milliken , K.L., 1979, The silicified evaporite syndrome - two aspects of silicification history of former evaporite nodules from southern Kentucky and northern Tennessee: Journal of Sedimentary Petrology, v. 79. p. 245-256. Siedlecka, A., 1972, Length-slow chalcedony and relicts of sulphates - evidences of evaporitic environments in the Upper Carboniferous and Permian beds of Bear Island, Svalbard: Journal of Sedimentary Petrology, v. 42, p. 812-816. Siedlecka, A., 1976, Silicified Precambrian evaporite nodules from northern Norway: a preliminary report: Sedimentary Geology, v. 16, p. 161-175.
Magadi-type Cherts Eugster, H.P.,1967, Hydrous sodium silicates from Lake Magadi, Kenya: Precursors of bedded chert: Science, v. 157, p. 11n Eugster. H.P.,1969,Inorganic bedded cherts from the Magadi area, Kenya: Contributions to Mineralogy and Petrology, v. 22, p. 1-31, Eugster,H.P.andSurdam, A.C., 1973, Depositional environment of the GreenRiver Formation: Geological Society of America, Bulletin. v. 84. p. 1115-1120. Gude, A.J. and Sheppard, R.A., 1972, Silhydrite, 3Si0 2 • H20, a new mineral from Trinity County, California: American Mineralogist, v. 57, p. 1053-1065. Hay, R.L., 1968, Chert and its sodium silicate precursors in sod ium-carbonate lakes of East Africa: Contributions to Mineralogy and Petrology, v.17, p.255-274. Jones, B.F., Rettig, S.L. and Eugster, H.P., 1967, Silica in alkaline brines: Science, v. 158, p. 1310-1314. Maglione, G.• 1980. An example of Recent continental evaporitic sedimentation; The Chadian Basin, in Busson, G., ed., Evaporite Deposits: Illustration and Interpretation of Some Environmental Sequences: Edition Technip, Paris, p. 5-9. Parnell, J., 1986, Devonian Magad i-type cherts in the Orcadian Basin, Scotland: Journal of Sedimentary Petrology, v. 56, p.495-500. Peterson, M.N.A. and Von der Borch, C.C ., 1965, Chert: modern inorganic deposition in a carbonate precipitating locality: Science, v. 149, p. 1501-1503. Rooney, T.P.. Jones, B.F.and Neal,J:r., 1969, Magadiite from Alkali Lake,Oregon:American Mineralogist, v. 54, p. 1034-1043. Sheppard, R.A. and Gude, A.J., 1986, Magadi-type chert - A distinctive diagenetic variety from lacustrine deposits. in Mumpton, FA, ed., Stud ies in Diagenesis: United States Geological Survey, Bulletin , Report B1578, p. 335-345 .
Surdam, R.C., Eugster, H.P. and Mariner, R.H ., 1972, Magadi-type chert in Jurassic and Eocene to Pleistocene rocks, Wyoming: Geological Society of America, Bulletin , v. 83, p. 2261-2266. White , A.H. and Youngs, B.C., 1980. Cambr ian alkali playa-lacustrine sequence in the northeastern Officer bas in , South Australia: Journal of Sedimentary Petrology, v. 50, p. 1279-1286.
Inorganically Precipitated Lagoonal and Lacustrine Cherts Peterson, M.N.A. and Von der Borch, C.C., 1965, Chert: modern inorganic deposition in a carbonate precipitating locality: Science , v. 149, p. 1501-1503. Renaut, R.W. and Owen, A.B., 1988, Opaline cherts associated with sublacustrine hydrothermal springs at Lake Bogor ia, Kenya Rift valley: Geology, v. 16, p.699-702. Renaut, R.W., Tiercelin, J.J. and Owen, R.B .• 1986, Mineral prec ipitation and diagenesis in the sediments of the Lake Bogoria basin , Kenya Rift Valley, in Frostick, L.E., Renaut, R.W., Reid,l. and Tiercelin, J.J., eds. , Sedimentation in the African Rifts: Geological Society of London , Special Publication 25 , p.159-175. Wheeler, W.H. and Textoris, D.A., 1978,Triassic limestone and chert of playa origin in North Carolina: Journal of Sedimentary Petrology, v. 48, p. 765-776.
Pedogenic Cherts: Silcretes Bruckner, W.D., 1966, Origin of silcretes of central Australia: Nature, v. 209, p.496-497. Exon , N.F.• Langford-Smith, T. and McDougall , I., 1970, The age of geomorphic correlations of deep weathering profiles, silcrete and salt in the Roma-Amby region, Queensland: Journal of the Geological Society of Australia, v. 17, p. 21-30. Frankel, J.J., 1952,Silcrete near Albertinia, Cape Province: South African Journal of Science. v. 48. p. 173-182. Frankel, J.J. and Kent, L.E., 1937, Grahamstown surface quartzites (silcretes): Geological Society of South Africa, Transactions, v. 40, p. 1-42. Khalaf, F.I., 1988, Petrography and diagenesis of silcrete from Kuwait , Arabian Gulf: Journal of Sedimentary Petrology, v. 58, p. 1014-1022. Hutton, J.T., Twidale, C.R., Milnes, A.A. and Rosser, H., 1972,Composition and genesis of silcretes and silcrete skins from the Beda Valley. Southern Arcoona Plateau, South Australia: Journal of the Geological Society of Australia, v. 19, p.31-39.
c
274
Diagenesis
Geoscience Canada Reprint Series 4
c (
Langford-Smith, T., 1978, ed., Silcrete in Australia: Armidale, University of New England, p. 304. Langford-Smith, T. and Drury, G.H., 1965, Distribution, character and attitude of the duricrust in the northwest of New South Wales and the adjacent areas of Queensland: American Journal of Science , v. 263, p. 170-190. Meyer, R. and Pena dos Reis, R.B., 1985, Paleosols and alunite silcretes in continental Cenozoic of western Portugal : Journal of Sedimentary Petrology, v. 55, p.76-85. Mountain , E.D., 1952, The origin of silcrete: South African Journal of Science, v. 48, p.201-204 . RUbin, D.M. and Friedman, G.M., 1981, Origin of chert grains and a halite-silcrete bed in the Cambrian and Ordov ician Whitehall Formation of Eastern New York State: Journal of Sedimentary Petrology, v. 5, p. 69-72. Smale, D., 1973, Silcretes and associated silica diagenesis in southern Africa and Australia: Journal of Sedimentary Petrology, v. 43, p. 1077-1089. Summerfield, M.A., 1983, Petrography and diagenesis of silcrete from the Kalahari Basin and Cape coastal zone, Southern Africa : Journal of Sedimentary Petrology, v. 53, p. 895-909. Swineford, A. and Franks, P.C., 1959, Opal in the Ogallala Formation in Kansas, in Ireland, H.A., ed., Silica in Sediments: Society of Economic Paleontologists and Mineralogists, Special Publication No. 7, p. 111-120. Thiry, M. and Millot , G., 1987, Mineralogical forms of silica and their sequence of formation in silcretes: Journal of Sedimentary Petrology, v. 57, p. 343-352. Watts , S.H. , 1978, A petrographic study of silcrete from inland Australia: Journal of Sedimentary Petrology, v. 48, p.987-994. Woolnough, W.G., 1927, Presidential address: Journal and Proceedings of the Royal Society of New South Wales, v. 61, p.1-53. Hydrothermal-volcanogenic Cherts
Anderson, R.N., Hobart, M.A. and Langseth, M.G., 1979, Geothermal convection through oceanic crust and sediments of the Indian Ocean: Science, v. 204, p. 828-832 . Barrett, T.J., 1979, Origin of bedded cherts overlying ophiolitic rocks in the Italian North Apennines, and implications of the ophiolite-pelagic sediment sequences for seafloor processes, unpublished D. Phil. thesis, Oxford University, v. 1, 288 p. Bertine, K.K. and Keene, M.J., 1975, Submarine barite-opal rocks of hydrothermal origin: Science , v. 188, p. 150·152.
Bishoff , J.L., 1969, Red Sea geothermal brine deposits: their mineralogy, chemistry, and genesis , in Degens, E.T. and Ross, D.A., eds., Hot Brines and Recent Heavy Metal Deposits in the Red Sea: Springer, New York, p. 368-40t Constantinou, G. and Govett, G.J.S., 1972, Genesis of sulphide deposits, ochre and umbre of Cyprus: Institute of Mining and Metallurgy, Transactions, v. 81, p. B34B46. Davis, E.F., 1918, The radiolarian cherts of the Franciscan Group: University of Californ ia, Publication Bulletin , Department of Geology, v. 11, p. 235-432. Fehn, U., 1986, The evolution of low-temperature convection cells near spreading centers: A mechanism for the formation of the Galapagos mounds and similar manganese deposits: Economic Geology, v. 81, p. 1396-1407. Fournier, R.O. and Rowe, J.J., 1966, Estimation of underground temperature from the silica content of water from hot springs and wet-stream wells : American Journal of Science , v. 264, p. 685-697. Hesse, R., 1~90b, Origin of chert: Diagenesis of biogenic siliceous sediments, in Mcllreath, LA. and Morrow, D.W., eds., Diagenesis: Geological Association of Canada, Geoscience Canada Reprint Series 4, p. 227-251. Hutchinson, R.W., 1982, Syn-depositional hydrothermal processes and Precambrian sulphide deposits, in Hutchinson, R.W., Spence, C.D. and Franklin, J.M., eds., Precambrian Sulphide Deposits: Geological Associat ion of Canada, Special Paper 25, p. 761-791. Ivanov,S.N., 1971, Deposits of pyrite ores of eugeosynclinal regions, in Proceedings ofthe IMA-JAGOD meetings '70, IAGOD volume: Society of Mining Geologists of Japan,Tokyo,p.193497. Keith, T.E.C., White, D.E. and Beeson, M.H., 1978,Hydrothermal alteration and self-sealing in Y-7 and Y-8 drill holes in northern part of Upper Geyser Basin, Yellowstone National Park, Wyoming: United States Geological Survey, Professional Paper 1054-A, 26 p. Lalou, C., Brichet, E., Ku,T.L.and Jahanno, C., 1977, Radiochem ical, scanning electron microscope (SEM) and X-ray dispersive energy (EDAX) studies of a FAMOUS hydrothermal deposit : Marine Geology, v. 24, p. 245·258. Meyer, C. and Hemley, J.J., 1967, Wall rock alteration, in Barnes, H.L. , ed., Geochemistry of Hydrothermal Ore Deposits: Holt, Rinehart and Winston , New York, p. 166-235. Rimstidt, J.D. and Barnes, H.I., 1980, The kinetics of silica-water reactions: Geochimica et Cosmochimica Acta, v. 41, p. 1683-1699.
Rimstidt, J.D. and Cole, D.R., 1983, Geothermal mineralization, I: The mechanism of formation of the Beowawe, Nevada, siliceous sinter deposit: American Journal of Science, v. 283, p. 861-875. Robertson, A.H.F., 1976, Origins of ochres and umbres: evidencefrom Skouriotissa, Troodos Massif, Cyprus: Institute of MiningandMetallurgy, Transactions,v. 85, p. B245-B25t Robertson, A.H.F. and Hudson, J.D., 1973, Cyprus umbres: chemical precipitates onaTethyan oceanridge:EarthandPlanetary Science Letters, v. 18,p. 93-10t Russell, M.J., 1975, Lithogeochemical environment of the Tynagh base-metal deposit, Ireland, and its bearing on ore deposition: Institute of Mining and Metallurgy, Transactions,v. 84, p. B128B133. Taylor, G.R., 1974, Volcanogenic mineralization in the islands of the Florida Group, B.S.I.P.: Institute of Mining and Metallurgy, Transactions, v. 83, p. B120B130. Troop, D.G., 1984, The petrology and geochemistry of Ordovician banded iron format ions and associated rocks at the Flat Landing Brook massive sulfide deposit, northern New Brunswick, M.Sc. thesis, University of Toronto, 178 p. Precambrian Cherts
Cloud, p., 1973, Paleoecological significance of banded iron-formation: Economic Geology, v. 68, p. 1135-1143. Drever, J.T., 1974, Geochemical model for the origin of Precambrian banded iron formations: Geological Society of America, Bulletin, v. 85, p. 1099-1106. Eugster, H.P., 1969, Inorganic bedded cherts from the Magadi area, Kenya: Contributions to Mineralogy and Petrology, v. 22, p. 1-31. Eugster, H.P. and Chou, I-M., 1973, The depositional environments of Precambrian banded iron formations: Economic Geology, v. 68, p. 1144-1168. Garrels, R.M., 1987, A model for the deposition of the microbanded Precambrian iron formations: American Journal of Science, v. 287, p. 81-106. Holland, H.D., 1973,The oceans: A possible source of iron in iron formations : Economic Geology, v. 68, p. 1169-1172. Holland, H.D. and Malinin, S.D., 1979,The solubility and occurrence of non-ore minerals, in Barnes, H.L., ed., Geochemistry of Hydrothermal Ore Deposits, Second Edition : Wiley Interscience, New York, p. 461·508.
c
c c c c c c c c c c c c c c c c c c c c c c c C E C C C C
c e c c c c c c c c
c
r r
o o c o c c o o o o o
o o o
o o o o o o o o o o
o o ;
Origin of Inorganic and Replacement Cherts
Hughes, C.J., 1976,Volcanogenic cherts in the Late Precambrian Conception Group, Avalon Peninsula, Newfoundland: Canadian Journal of Earth Sciences, v. 13, p. 512-519. Laberge, G.L., 1973, Possible biological origin of Precambrian iron formations : Economic Geology, v. 68, p. 1098-1109. Pettijohn, F.J., 1975, Sedimentary Rocks, Third Edition: Harper & Row, New York, 628 p. Ross, G.M. and Chiarenzelli , J.R., 1985, Paleoclimatic significance of widespread Proterozoic silcretes in the Bear and Churchill Provinces of the northwestern Canadian Shield: Journal of Sedimentary Petrology, v.55, p. 196-204. Simonson, B.M., 1985, Sedimentology of chert in the Early Proterozoic Wishart Formation , Quebec-Newfoundland, Canada : Sedimentology, v. 32, p. 2340. Simonson, B.M., 1987, Early silica cementat ion and subsequent diagenesis in arenites from four Early Proterozoic Iron Formations of North America: Journal of Sedimentary Petrology, v. 57, p.494-511. Walker,J.C.C ., 1984,Suboxic diagenesis in banded iron formations : Nature, v. 309, p.340-342.
Accepted, as revised, 1989
275
276
Diagenesis
Geosc ience Canada Reprint Series 4
c c c c c c c c c c c c c c c c c c c c
c c c
c c c
c
Backscattered-electron SEM photomicrograph of a pyritic aggregate from the Cretaceous Harmon Member in the Peace River Formation of Alberta. Early framboidal pyrite, the bright globular areas, coalesced into a cauliform body which was , in turn, overgrown by an euhedrally terminated rim (indicated by arrow). This progression of morphologies indicates that early, rampant nucleation ofpyrite under open-system conditions was followed by a reduction in the supply of sulphate that led to slower crystal growth and the development of a euhedrally terminatedrim . The bright zone along the margin is an edge effect. Barscale indicates 20jJm. Photograph courtesy of J. Bloch (University of Calgary).
c c c
c c c c e c c c C l
c c
c c
Early Diagenetic Pore Water/Sediment Interaction
277
C.'· (
o o ( o c )
(
Early Diagenetic Pore WaterI Sediment Interaction: Modern Offshore Basins
o o o
Department of Ge%gical Sciences McGill University 3450 University Street Montreal, Quebec H3A 2A 7
o o o
o o
o o o
o o )
o o o r
r:
Reinhard Hesse
INTRODUCTION Rock-water interaction, which receives so much attention by geothermal engineers, igneous and metamorphic petrologists, is of tantamount interest to sedimentary petrologists studying diagenesis. Water is the principal transport medium in diagenesis acting as the carrier of dissolved substances from sites of dissolution to sites of precipitat ion in the sediment column and facilitating chemical exchange between sedimentary basins and the hydrosphere. Where water is absent because mineral cements or non-aqueous fluids such as liquid or gaseous hydrocarbons fill the pores, diagenetic reactions are inhibited or slowed down drastically. For most diagenet ic reactions the presence of water in the sediment pores is thus a prerequisite. Pore-water analyses are, therefore, of great significance in the study of diagenesis, particularly early diagenesis for which water is as volumetrically important as the solid phases. Pore waters are much more sensitive to changes in diagenetic environments than solids and therefore are the preferred diagnostic objects for the study of early diagenesis. There are three aspects of diagenesis to which pore-water studies contribute most: (1) The detection of diagenetic reactions actively occurring in the sediment at the time of sampling (drilling) by tracing the products of these reactions
released to, or the reactants consumed from, the pore water through observed concentration changes in vert ical or lateral profiles. (2) Distinction between the principal modes of solute transport, i.e,. between advection (convection) and diffusion, and determination of their flux rates. (3) Assessment of mineral-solution equilibria. As it is a widely held view that diagenetic processes in actively subsiding sedimentary basins do not lead to thermodynamic equilibrium between the potential reaction partners, porewater chemistry can provide clues to this question by providing the data necessary for calculation of the saturation state with respect to the minerals of interest. Such calculations have been carried out for common authigenic minerals in ancient basins on land (Merino, 1975; Nesbitt, 1980, 1985)and in a few cases also for modern offshore basins. The general plan for this review is to follow the early diagenetic pore-water evolution in modern offshore basins tracing the changes that occur with increasing burial from the sources to the sinks of the dissolved species. The three principal processes causing these changes are reactions, diffusion and advection whose combination produces nine different types of porewater profiles. These will be discussed in order of increasing complexity. Relatively simple, more or less straight-line pore-water profiles are encountered in low sedimentation-rate environments of the pelagic realm and can be understood in terms of diffusive and convective transport of solutes. These are discussed first together with some mineralization reactions that are characteristic for the early diagenetic history of these environments. The more complex profiles observed in organic-matter rich, high sedimentation-rate basins of the continental margins (with their reaction-dominated trends) follow. In these basins, extensive mineralization reactions serve as sinks for the dissolved substances. They are treated in a separate section. Ion exchange reactions and adsorption are additional sinks that may help to understand the behaviour of some ionic species in these basins. The definition of the boundary between the stages of early and intermediate diagenesis is deferred to the end of the article, because it is based on the deliberations of this review.
TERMINOLOGY The diagenetic evolution of pore water starts with "connate water" which is the water trapped between the sediment particles during deposition. Initially this connate water may considerably exceed the volume of the solids. Freshly deposlted clayey-silty sediments may contain as much as 90% (by volume) water, and sands up to 40% or 50%. However,with burial to depths of only a few metres the water content is very rapidly reduced in volume. There has been considerable confusion in the literature concerning the term "connate water". Due to reactions with the solids, pore waters usually undergo sign ificant compos it ional changes in the subsurface. Buried interstitial waters commonly are strikingly different in composition and total amount of dissolved species from their connate-water precursors in marine or terrestrial environments. The term connate water, in the sense advocated by Case (1955), however, should be used only for pore waters still displaying the compositional and isotopic signatures inherited from their parental source in the ocean or lake water bodies. Generally this applies only to the shallowest levels of burial, often not more than a few centimetres or millimetres below the sedimentlwater interface over which distance the pore water is still in diffusional contact with the overlying water body. Only in exceptional situations, such as certain types of pelagic sediments, may this zone of diffusional or convectional exchange extend downward a few hundred metres in the subsurface . The definition of "connate water" by Kharaka et a/. (1985) as "water that was deposited with sediments or other rocks in the basin and which has been out of contact with the atmosphere since its deposition" is similar to White 's (1965) definition . This definition appears too broad, because it does not take into account compositional changes that may take place immediately after deposition and which may become so profound as to obscure the origin and source of the water. To call such pore waters "connate" would not be in line with the very meaning of the word and intent ofthe earlyauthors (e.g., Case, 1955) who introduced the term. Changes in pore-water chemistry at depth are often coupled with increasing concentrations of dissolvedsolids. Maxi-
278
imum salinities are encountered in the deep parts of sedimentary basins, particulary in the vicinity of salt domes or deeply buried evaporites where values of 200-300 g. L-1 of dissolved solids are not exceptional (an up to eight-fold increase compared to normal seawater salinity). Salinities as high as 60%0 have been reported, but these occur in ore fluids of porphyry copper deposits (Roedder, 1977) not related to burial diagenesis . Highly saline waters are called brines. According to Carpenter (1978), brines have salinities in excess of 10%0, which would limit application of the term to a relatively small sample of ali deep subsurface waters. Following Kharaka et al. (1985a),salinities above seawater salinity of 3.5%0 are sufficient for a pore water (or any other hypersaline body of water) to qualify as a brine. In the petroleum industry it is common practice to call pore waters encountered in oil-bearing sedimentary formations "formation waters", a term which strictly applies to the "water present in the rocks immediately before drilling", whereas the general term" interstitial water" is applicable to all types of pore waters from all depths and different kinds of sedimentary basins: modern offshore basins on the continental margins and ocean floors, modern lakes or river basins and ancient basins on land, irrespective of age or sampling procedure. Waters with salinities lower than those of brines, i.e., between 3.5 and 1%0, are simply called "saline waters", those with salinities between 1.0 and 0.1%0 "brackish", and those with still lower concentrations of dissolved salts , "fresh waters". Another important term is " meteoric water " originally defined as "water that was recently involved in atmospheric circulation" (White , 1957, p. 1661). This is "water derived from rain, snow, water courses , and other bodies of surface water that percolates in rocks and displaces their interstitial water" (Kharaka and Carothers, 1988). The time of the last contact with the atmosphere has been dropped intentionally from the definition by these authors, because water with the characteristics of meteoric waters may be very old. A modifier such as " Miocene" or"Pleistocene meteoric water" may be used, however, to denote its age, i.e., the time of its removal from contact with the atmosphere if known.
Diagenesis
SOURCES OF PORE-WATER DATA Information concerning the chemical evolution of pore waters during burial diagenesis in modern offshore basins comes from three principal sources: (1) studies of piston and gravity cores; (2) the interstitial water-sampling program and related sediment studies of the Deep Sea Drilling Project (DSDP) and its successor program, the Ocean Drilling Program (ODP); and, (3) offshore oil and gas wells. The first is restricted to the upper 10to 20 m of the sediment column in modern subaqueous environments, which is the maximum depth of penetration by conventional coring devices such as piston and gravity corers. Since the classical study by Emery and Rittenberg (1952) on California Borderland basin sediments, this type of analysis has been applied to surface sediments all over the world's oceans and of many lakes. It covers the very earliest stages of diagenesis for which Berner (1980) provides an excellent theoretical treatment and review. In the past two decades, a second important source of information has been the Deep Sea Drilling Project. Although perhaps not as widely noticed as some other of its spectacular results, the interstitial water program has been one of the outstanding successes of this project. It provided a consistent data body of thousands of pore-water analyses covering the depth range from a few metres to about 1500 m subsurface. As an outcome of this program and the piston core studies mentioned before, we now have a fairly good understanding of the variations in porewater chemistry that occur in the upper 1000 or 1500 m of the sediment column in different open-ocean as well as deepwater continental margin environments. Between 1968 and 1983 more than 1000bore holes were drilled by the DSDPat 624drill sites and for more than half of these pore-water data are available. Results have been summarized during various phases of the project (Manheim and Sayles, 1974; Sayles and Manheim, 1975; Gieskes, 1975, 1981, 1983) and this article includes an update of those earlier reviews. Innumerable holes drilled for oil and gas or other resources in sedimentary basins on the continental shelves and slopes, from which formation waters have been sampled, provide the third
Geoscience Canada Reprint Series 4
source of information . There is not much overlap with the results of the DSDP, which concentrated on the ocean basins and deeper portions ofthe continental margins and stayed away from potential hydrocarbon provinces on the upper margins and shelves on which offshore oil drilling has focussed. Penetration reached in exploration wells for oil and gas generally exceeds that of the drill holes of the DSDP and, therefore , pore-water compositions from deeper, more advanced levels of diagenesis may be obtained, which are the subject of a forthcoming article. Extraction of the pore water from unconsolidated to moderately consolidated sediments by squeezing with stainless steel squeezers (Manheim and Sayles, 1974) may lead to systematic errors due to temperature increases between the ocean floor and the laboratory. Increases of 1 to 50% in chloride and dissolved silica , for a temperature difference of 20°C (Bischoff et al., 1970; Fanning and Pilson, 1971) have been observed. For potassium , enrichments of up to 13% have been found, whereas magnesium and calcium can show depletions of up to 2.5 and 5%, respectively. These shifts can be avoided by standardized extraction procedures under regulated temperatures, including centrifugation for less consolidated samples.
c c c c c c c c c c c
c c c c
c c c c c c c c c c c c C G C C C C
c
e
0.1
I
10
100
Sed rmenrat ron Rate (g/cm 7/10 Ja)
Figure 1 Organic carbon accumulation rate as a function of sedimentation rate. (From Heath et aI., 1977).
c c c c
c c c
c c c c. o o ( (
c o c
(;
o o o o o o o o o o o o o o (
o
Early Diagenetic Pore Water/Sediment Interaction
DIAGENETIC REGIMES OF PORE-FLUID EVOLUTION IN MODERN OFFSHORE BASINS Diagenetic environments of subsurface pore-water evolution in modern offshore basins can be divided into two main regimes: I. Low to intermediate sedimentationrate basins (or environments) mostly associated with oxic and suboxic diagenesis and convection- and diffusion-controlled pore-water profiles, and II. High sedimentation-rate basins characterized by anoxic diagenesis and reaction-controlled pore-water profiles. Despite the great variation in the chemical composition of pore-water analyses from offshore basins all over the world, this broad classification in terms of two main environments or regimes does justice to most of the data and will help not to miss the wood for trees. Situations not covered by this simple two-end-memberscheme are included in a third category: III. Special environments. The reason for the dominant effect of sed imentation rate on pore-water chemistry, particularly during early diagenesis in marine basins, is its close positive correlation with organic matter content of the sediments and its tendency to counteract the role of diffusion . Organic matter is the least stable and most reactive sediment constituent. It is metabolized rapidly by bacteria during early diagenesis and the metabolites including the basic nutrients carbon, nitrogen, phosphorus, sulphur, and hydrogen are released to the pore water. These become involved in the first diagenetic mineralization reactions and trigger other reactions. Organic matter concen-
o o
Table 1
(I
G l C
II
III
tration, therefore, determines the early diagenetic reactivity of the sediment, that is, rates and types of reactions. The correlation between sedimentation rate (expressed in g' cm-2 per 1000 years (103 a» and organic carbon accumulation rate (expressed in the same units) has been shown to be linear on a logarithmic scale (Figure 1) for open marine pelagic and hemipelagic environments (Heath et el., 1977; MOiler and Suess, 1979). This is basically the result of increased preservation of organic matter in more rapidly deposited sediments. The faster the sediment is buried, the more rapidly its organic matter is removed from contact with oxidizing bottom and pore waters, in which bacteria are most efficient in decomposing organic matter. Rapid sedimentation, therefore, is a prerequisite for organic matter to survive the first steps of deposition and burial and to become available for subsequent diagenetic reactions . Variations in primary production rate in the surface waters appear to be of secondary importance only, as demonstrated by those authors. Changes in productivity, to some extent, parallel changes in sedimentation rate thus enforcing the positive correlation. It is not clear whether this correlation also applies to terrestrial organic matter because of its greater resistance to bacterial decomposition in the oxidation zone (Waples , 1983). Positive correlation between sedimentation rate and organic carbon content of the sediments is not universal, however, and does not apply for stagnant or semistagnant basins with oxygen-deficient bottom waters. The same applies to areas with extremely high sedimenta-
Major types of interstitial water profiles
Low to Intermediate sedimentation rate basins 1. Convection-controlled profiles - no chemical gradients (Figure 6. DSDPsite 495) 2. Diffusion-controlled profiles with linear correlation between changes in Ca and Mg concentration (Figure 7) 3. Reaction-controlled profiles in suboxic environments (Figures 4, 6, site 495) High sedimentation-rate basins 4. Reaction-controlled profiles in anoxic environments with no chloride gradients (Figure 6, upper part of DSDP site 570) 5. Profiles with fresh water influx (Figure 14) 6. Gas-hydrate affected profiles with downward decreasing chlorinity and increasing 0'80 values (Figures 6, 17) Special environments 7. (Thrust-)faulted sect ions with discontinuous chemical trends indicating lateral pore-water flows (Figure 18) 8. Evaporite-dissolution affected profiles (Figure 19) 9. Hydrothermal activity and intrusion of igneous dykes and sills (Figure 20)
279
tion rates. In stagnant or euxinic basins, bottom waters act as a trap for organic matter protecting it from oxidation. Such basins are rare , however, in modern oceans . Excessive input of terrigenous materials has a diluting effect. thus also producing a negative correlation between sedimentation rate and organic carbon content (Gautier et al. , 1984; Johnson Ibach, 1982; Stein, 1986). The effect of sedimentation rate on the diagenetic evolut ion of pore waters is two-fold. Apart from the effects caused by coupling with organic matter concentration, sedimentation rate also controls pore-water evolution through its effects on diffusion rates. Where sedimentation rates exceed diffusion rates, diffusive transport cannot keep pace with burial. Chemical concentration changes produced by diagenetic reactions will be preserved much better during burial and for much longer periods of time. An analogous relationship exists between sedimentation rate and advection rate. The linkage of sedimentation rate to the three pr incipal diagenetic processes that control pore-water chemistry, i.e., (1) chemical reactions in the sediment/pore-water system triggered by organic matter decomposition, (2) transport of solutes by diffusion, and (3) transport by advection (convection), explains the overriding effect ofthis single parameter on early diagenetic porewater evolution. Where sedimentation rates are low, the reactivity of the sediment during early diagenesis is low due to reduced organic matter concentrations. and diffusion (or advection) may efficiently dissipate existing concentration gradients. The result are simple. often straight-line vertical concentrat ion profiles. Where sedimentation rates are high, the chemical reactivity of the sediment is high due to generally increased organic matter preservation. Consequently. early diagenetic reactions are intense. In these environments, the efficiency of diffusion (and advection) in dissipating gradients established by the reactions is low. Complex pore-water profiles result. Interaction of the three general processes. i.e., reactions. diffusion, and advection , in the two or three main types of environments of pore-water evolution gives rise to the nine different types of vertical pore-water profiles (Table 1 and Figure 2) mentioned in the
280
Diagenesis
CICa t~· :':no'"
'J .:( :
~;i
;~ :~
2+Mg2+
It :' 5 1 4'.1:;,
Geoscience Canada Reprint Series 4
80:NH3 PO:- Alk , ~rn!.,~: i'~'rf :J.Jo.~~.
18
£»
0
n""r" r r r""f1' ~I
~I
r,
sr\.~lrV\·;
~I
;~:n
\V
urlJV ® ~[JnrlJrr @ :: If?rlTvvrr ® :f71?I vPvf\ @ ::F rllJr@ 5~O Q' I I I ~4~~ I CD ~.
C C C
c
C
~ c
c
~
~c
~ ~
~ ~c
:r::h~rlrl@ ~
:F" """fVlrl ® ~ Figure 2 Chemical characteristics of major types of interstitial water profiles (see Table 1).
L
(
c c· c o o (
C
o o o o o o o o o o o o o o o
Early Diagenetic Pore Water/Sediment Interaction
not the rule although they may cause significant modification in certain basins or parts thereof .
introduction. This list of nine is not meant to be exhaustive and is merely intended to provide guidance in recognizing general trends among the highly variable pore-water compositions encountered in the subsurface. It will serve as a frame of reference in the following discussion of the two or three end-member environments of porewater evolution . Before addressing these environments , two of the general processes, i.e., (1) early diagenetic organic matter oxidation as the key mechanism in early diagenetic reactions and (2) diffusion will be discussed in more detail in the next two sections. The role of the third mechanism, (3) advection, will be dealt with in connection with the relevant types of pore-water profiles (i.e., types 1, 8 and 9). Other geologic factors besides sedimentation rate such as the composition of the starting fluids, i.e., the connate waters (seawater versus fresh water) or the tectonic stage of basin evolution that would be equally important in a global survey, may be considered secondary for the purpose of this review article which is concerned solely with modern marine basins. For these the effects of diastrophism (uplift and deformation) and meteoric water influx are
EARLY DIAGENETIC ORGANIC-MATTER OXIDATION Organic matter represents highly reduced carbon compounds which are among the strongest reductants in freshly deposited sediments . The organic matter dispersed in the sediment provides an energy source for sediment-dwelling organisms, foremost bacteria, to maintain their metabolism through oxidation reactions. Bacterially mediated oxidation of organic matter becomes the first and most efficient mechanism to alter the chemical composition of pore waters during early diagenesis. The well-established organic matter oxidation reactions start immediately after deposition. According to the source of the oxidant (electron acceptor), six different zones or stages have been distinguished in the oxidative breakdown of organic matter during burial (Claypool and Kaplan, 1974; Curtis, 1978). These are (Figure3): (1) the oxidation zone, (2) the nitrate reduction zone, (3) the sulphate reduction zone, (4) the carbonate reduction zone,
V.ntilat.d Basin
Stagnant (.uxinic) Basin
~.!.!=.!~~--=I . .
o o o
o o o o
281
: i } ..:
l
. ,I
\
Oxic bottom wat.r
; .
'.
:
(
\
\:
l
;
I Oxidation Zan. (2) Nitrat. r.ductionZon.
Anoxic bottom water. edimentl ~~;j Water = Interface
(3) Sulfat. reduction Zone
c 6
(4) Carbonate reduction
C
(5) and fermentation Zones
@>
cOt. CH4
®
@)
and fermentation
(6) Thermocatalytic
decarboxylation
Figure 3 Stages of organic matter oxidation in anoxic sediments. (Modified from Claypool and Kaplan, 1974).
(5) the fermentation zone, and, (6) the thermocatalytic decarboxylation zone, which represent a burial-depth zonation. The first five zones require the presence of bacteria. At the lower boundary of zone (5), bacterial activity ceases and thermocatalytic reactions take over. This is a natural lower boundary for the zone of early diagenesis. (1) In the oxidation zone, freely dissolved oxygen is available from seawater trapped in the sediment or supplied by diffusion from the overlying bottom water. The pore water of the oxidation zone may, therefore, be called "connate" in the strictest sense of the word. Bacteria are not the only oxygen-consuming organisms active in this zone, which is the habitat of a diverse group of endobenthonic microand macro-organisms. Aerobic bacteria are the most efficient users of oxygen in this zone, however. The amount of dead animal or plant matter left undestroyed in the sediment after passage through the oxidation zone is a function of the residence time in this zone. In oxidizing organic matter and breaking it down into smaller molecules the bacteria gain energy for their metabolism. Because of their small size, in the micron range, they possess a large surface-area/mass ratio which favours the exchange of dissolved substances with the pore water through the cell walls. Microbial degradation of organic molecules leads to a loss of functional groups, particularly hydrocarbonchains and carboxyl-groups and the smaller molecules are partly converted to carbon dioxide according to an overall reaction which may be written as CH20 + O2
->
CO2 + H20
where the bulk composition of organic matter has been simplified to CH20. The efficient use of oxygen by aerobic respirators may lead to temporary or local oxygen depletion, particularly in regions of rapid sedimentation where replenishment from the overlying bottom water by diffusion may not keep pace with consumption. Obligatory aerobic organisms will then cease to exist, but facultative aerobeswhich can switch from an aerobic to an anaerobic mode of respiration may still be present. The oxidation zone is easily recognized by its brownish-yellowish sediment color, where it is present. In
Diagenesis
282
organic matter-rich sediments, it may be only a few centimetres or millimetres thick - in contrast to the carbon-free brown abyssal clay discussed below. In stagnant or euxinic basins the oxidation zone is absent. (2) The nitrate reduction zone commences where the concentration of dissolved oxygen drops below about 0.5 mL 02 ·(L H20)-' (Devol, 1978), which correspondsto the oxygen levelat which most benthonic macro-organisms disappear from the sediment (Rhoads and Morse, 1971). Dysaerobic bacteria, which live at dissolved oxygen levels between to and 0.1 mL O2 . (L H20)-', are characteristic for this environment. which may be called suboxic. The term "suboxic diagenesis" thus refers to the nitrate reduction zone. Concentration of nitrate tends to increase from the ambient bottom water value of 0.03-0.04 mM (mM = millimoles . L") to a maximum in the oxidation zone Table 2
where ammonia released from organic matter decomposition is oxidized (Froelich et a/., 1979) . From th is maximum, nitrate decreases downward to zero at the base of the nitrate reduction zone. The lower boundary of the nitrate reduction zone is characterized by the change from positive to negat ive electrochemical potentials and the appearance of anaerobic bacteria. When the electrochemical potential is lowered sufficiently, oxides and hydroxides of manganese, which at higher oxidation potentials have a very low solubility, will be reduced and go into solution as shown schematically in Figure 4 (Froelich at a/.• 1979)thus serving as an important source of oxidant in
Geoscience Canada Reprint Series 4
CONCENTRATION -+
~
~
!
Figure 4 (right) Schematic diagram for the depth distribution of dissolved 02' NO-3 Mn 2 + and Fe2 + in suboxic environments (pore-water profile type 3). (From Berner, 1980, after Froelich et at, 1979).
Stages of organic matter oxidation. (Modified from Froelich et al., 1979).
C : N : P = 106: 16 : 1 = Redfield ratio for primary organic matter. The organic matter involved in reactions (1) to (4) does not necessarily have this ratio . ~GO' of
metabolic processes coupled with oxidation reactions (in kcal- (mol)-' glucose)
(1) Oxidation by freely dissolved O2 (aerobic respiration) -763 Manganese reduction (CH 20)'06(NH3l'6(H3P04 ) + 236 Mn02 + 472 H+ ..... 236 Mn 2+ + 106 CO 2 + 8 N2 + H3P04 + 366 H 20
-748 (birnessite) -698 (pyrolusite)
(2) Nitrate reduction (CH 20)'06(NH 3),6(H 3P04 ) + 99.4 HN03 ..... 106 CO 2 + 55.2 N2 + H3P04 + 177.2H20
-724
(CH 20),06(NH3l ,6(H3PO..) + 84.4 HN03 ..... 106 CO 2 + 42.4 N2 + 16 NH 3 + H3PO.. + 148.4 H20
-658
Iron reduction (CH 20),06(NH3),6(H3P04 ) + 212 Fe 203 (or 424 FeOOH) + 848 H+ ..... 424 Fe 2+ + 106 CO 2 + 16 NH3 + H3PO.. + 530 H 20 (or 742 H20)
-337 (hemat ite) -318 (limonitic goethite)
(3) Sulphate reduction (CH 20)'06(NH3l'6(H3PO..) + 53 (504 ) 2 - ..... 106 CO 2 + 16 NH 3 + 535 2 - + H3PO.. + 106 H20
-91
(4) Carbonate reduction -84
c c c c c c c c c c c c c c c c c c c c
c c c
c c c c c c c c c c e c c c c c c c c
c
c c c o o r o CO
o o o o o
o o o o o o o
o o o
o o
o
Early Diagenet ic Pore Water/Sediment Interaction
addition to nitrate of the pore water. The exact Eh-Ievel at which manganese reduction occurs depends on the different oxides and hydroxides involved having different Gibbs free energies (Table 2). Iron reduction will occur after manganese reduction at somewhat greater depth, often overlapping with the sulphate reduction zone when the redox potential has been lowered further. Both manganese and iron reduction are hydrogen-ion consuming reactions, as the equations in Table 2 show, increasing the pH which may be a prerequisite for the precipitation of early diagenetic carbonates in concretions (see later section). (3) Suboxic conditions of the nitrate reduction zone are followed by the truly anoxic conditions in the underlying sulphate reduction zone. Although the sulphate-reducing bacterium Desulfovibrio desulfuricans is the dominant bacterial species, as only few species can tolerate the toxic effects of the hydrogen sulphide produced in this zone, it is not the only one. The sulphate reducers oxidize relatively small organic molecules such as lactic acid and fourcarbon dicarboxylic acids. These are produced by fermenting bacteria whose symbiosis with the sulphate-reducing bacteria is required by the process . A simplified equation for this rather complex process is 2CH20 + SO~--+S2- + 2H+ + 2HC03. As soluble ferrous iron usually becomes available through reduction of ferric oxyhydroxides and oxides in the sulphate reduction zone and the lower part of the nitrate reduction zone of terrigenous sediments, dissolved sulphide concentrations remain relatively low (less than 5 mM), because of the instantaneous precipitation of metastable iron-monosulphides which later transform to pyrite. In iron-poor marine carbonate muds the process will take a different course of events. Sulphate reduct ion is diagnostic for the pore waters of marine sediments as compared to freshwater sediments, which are often low in sulphate and lack the sulphate reduction step in the early diagenetic evolution of their pore waters. In marine sediments the sulphate reduction zone is considerably thicker than the nitrate reduction zone because dissolved sulphate is about three orders of magnitude more abun-
dant (27 mM) in oceanic bottom waters than nitrate. However, in organic-matter rich sediments the combined thickness of the nitrate and sulphate reduction zones is often less than a few metres and in certain regions may be as small as 10or 20 cm (Reeburgh, 1983). (4) In each of the three oxidation steps discussed so far one of the main products of bacterial organic matter decomposition has been carbonic acid and its dissociated species bicarbonate HC030r carbonate CO~-. Below the sulphate reduction zone the carbonate itself becomes one of the main oxidants for further bacterial oxidation of organic matter. In the carbonate reduction zone bacter ial carbonate reduction leads for the first time in the burial history of organic-matter rich sediments - to the production of methane. The process appears to require a reducing intermediate (e.g., hydrogen) also derived from bacterial organic matter degradation (Berner, 1980). For example, the oxidation of ethanol will yield acetic acid and hydrogen according to the reaction CH3CH20H + H20-+CH 3COOH + 2H2· The hydrogen may be used by true methanogenic bacteria such as Methanobacterium thermoautotrophicum to reduce CO2 produced in this and the previous zones (according to a reaction of the general type CH20 + H20 -+CO2 + 4H which is enzyme catalyzed and produces atomic H) by the redox reaction 8H + CO2 -+ CH4 + 2H20 . The process of methane formation thus appears as a disproportionation of organic matter into carbon dioxide and methane (Goldhaber and Kaplan, 1974) and it is not clear at present whether a direct transformation of organic compounds into methane by bacteria is possible. Although methanogenic bacteria exist over a wide temperature range (0°-75°C, Zeikus and Wolfe, 1972), the optimum for individual species spans only a few degrees. For example, the thermophylic M. thermoautotrophicum has its optimal temperature range between 65°C and 70°C. Methane production and sulphate reduction are not mutually exclusive precesses (Claypool and Kvenvolden, 1983), but they seem to be fairly well separated, because CH4 levels in the sulphate reduction zone are low and
283
significant CH4 production seems to start only after the disappearance of more than 80%, if not complete depletion of the dissolved sulphate (Sansone and Martens, 1981). A possible reason is that methane producing bacteria may not be able to tolerate the levels of H2S or HS- existing in the sulphate reduction zone, although this seems questionable in view of the fact that methane bacteria have been found in the presence of unreacted HS- below the sulphate reduction zone. Alternative explanations are (i)that free hydrogenmay not be available for CO2 reduction in the presence of sulphate reducing bacteria or (ii) that the methane which may be produced in the sulphate reduction zone is immediately oxidized to CO2, This latter process would also inhibit the appearance of CH4 in the sulphate reduction zone due to upward diffusion . (5) The fifth oxidation mechanism is bacterial fermentation where the oxygen contained in organic compounds is transferred to oxidize organic matter, simultaneously yielding CO2 and CH4 • From what has been said above it is not clear whether this step is a separatestep or whether in fact bacterial carbonate reduction always requires fermentation processes as symbiotic reactions. (6) Bacterial activity decreases decisively at temperatures above 75°C and only few populations still exist at higher temperatures. Organic matter oxidation, however, continues to proceed at temperatures above 75°C, but under abiogenic conditions . Once the activation energy has been reached, the reactions will occur spontaneously. These thermocatalytic reactions in-: elude the decarboxylation of organic acids according to the general formula R-COOH -+ RH + CO2 where the organic acid group COOH is converted to CO2 and R- is the organic radical. This process occurs at more elevated temperatures (above 75°C) and deeper burial levels and is probably very important in generating aggressive acidic pore waters at depths which may be responsible for the dissolution of carbonates and feldspars and the generation of secondary porosity. These are additional arguments for choosing the boundary between zones (5) and (6) as the boundary between early (or shallow) and intermediate diagenesis as discussed more fully in the final section of this article.
Diagenesis
284
Table 3
Diffusion in Sediments. (After Berner, 1980).
Fick's First Law (1)
J. I
Geoscience Canada Reprint Series 4
Jj
ac.I = _D·_ az
OJ
I
= flux of species i = diffusion coefficient [cm 2 • S-1 ]
aC azj
Fick's Second Law
= verticaI concentratton gra dllent °
°
c c c c
Fick's Second Law in polar co-ordinates
3)
= _ aJ =
aCj
j
at
ar
a(r2 0 j aa~j} .....,1"....-~_ _ r2
or
Fick's First Law for charged species i
= mobility of i-th ion at infinite dilution
Vi
~~ = gradient of electrical potential E Zj
B" R T
Sediment diffusion coefficient Os (6)
8
o o, = 02
al 8
= charge of the i-th species = Faraday constant = gas constant
= temperature in degrees Kelvin
al = az?; 1 tortuosity °
= actual path length of ion in solution
= q,F,
Diffusion flux in sediments J s
Js
ac
(7)
ac
= -q,Os---az 1
aJ s
1 a(cP°s
at = --;az- = --;
az
ac
c c c c c c c c c c c
F
=
R . e -R,where
az} F
where q, is porosity F is formation factor
= cP-n
o
Re is electrical resistivity of sediment Ro is resistivity of pore water
,for n =2: D,
= cP20
Mean diffusion path length
c c c c
c c
c c c c c c C E C C C
C
c
e
General diagenetic equation v = velocity of advective flow Rk = k-th reaction
c (
c (
c (
c
c
c
c c r. r
o r o o c o o o o o o
o o
o o o
o o
o o o o o o
285
Early Diagenetic Pore Water/Sediment Interaction
The sequential order in which zones (1) to (6) follow one another, most of them apparently without significant overlap, is related to the difference in the Gibbs free energy associated with each of the oxidation reactions listed in Table 2 (based on glucose as organic substrate). Utilization of dissolved oxygen as an oxidant is evidently the most energy efficient process. For each later step in the sequence the energy yield is less suggesting that specific bacterial populations associated with the various reactions in the subsurface follow each other in the order of decreasing energy efficiency of their metabolic reactions. Details of the biochemical reactions involved in bacterial metabolism are still poorly understood. Probably all of them include enzymatic reactions. It is known, for example, that methanogens can only use acetic acid or acetate and shorter hydrocarbon molecules for conversion into methane (Mechalas, 1974). Because acetic acid levels in the pore waters of marine sediments are relatively low, other strains of bacteria must be present to produce the intermediate substances which are then being used in the methane generating process. Similarly, sulphate-reducing bacteria use relatively short-chained organic acids which must be provided by symbiotic bacteria breaking down larger organic molecules. Thus different interacting bacterial populations must be present at anyone time in the sulphate reduction and methane producing zones which are involved in a series of simultaneous and partially symbiotic reactions. With the conventional spectrum of pore-water analyses we only see the end products of these chain reactions.
o
c
Table 4
DIFFUSION AS A DIAGENETIC TRANSPORT MECHANISM Diffusion operates to dissipate concentration differences. Thus the driving forces for diffusion are chemical potential gradients and gradients of the electrical potential caused by charged species. The flux J j of a dissolved chemical species i due to molecular diffusion is directly proportional to the concentration gradient oCj • The factor of pro-
OZ
portionality is the diffusion coefficient OJ' If we restrict ourselves to vertical concentration gradients, which are dominant during early diagenesis, then the flux J j is given by the one-climensional form of Fick's first law of diffusion (equation (1), Table 3). Most dissolved species are electrically charged ions . Therefore electrical potential gradients have to be taken into account resulting in a more complex version of Fick's law (equation (4), Table 3). The unknown
species J j or its gradient oCj is the sum electrical potential gradient oE can be
OZ
eliminated from (4) by using the electrical neutrality condition which requires that the sum of the electrical charges associated with the fluxes of all species be balanced. In other words, E;ZjJj = 0 which leads to equations (4a) and (5) in Table 3. Variations in concentration with time
oCj which may result, e.g., from varia-
OZ
tion in the input rate of soluble substances during sedimentation, will cause vertical gradients of fluxes oCj which
oz
are obtained from Fick's second law. In the form given (equation (2), Table 3) it is based on the assumption of a con-
Diffusion coefficients and mean diffusion path length. Os Zm (in m) Na+ Ca2 + CI(50 4)2-
stant diffusion coefficient OJ which, however, is not applicable for many dissolved species in natural sediments undergoing compaction. For the common occurrence of concretion growth during diagenesis, which is a diffusion-controlled process, flux gradients are perpendicular to a sphere or ellipsoid, for which Fick's law is more conveniently expressed in spherical co-ordinates (equation (3), Table 3). Reservoirs that are in diffusive communication are connected by concentration profiles of the dissolved species which bridge the concentration difference and are straight lines, if the diffusion coefficients are constant, or curved ones, if the diffusion coefficients vary with time or depth of burial. Curved profiles, on the other hand, may also result from transport by advection or reactions within the sediment. In the general diagenetic equation (equation (9), Table 3) the total flux of a dissolved
7.4 4.4 10.2 5.0
154.2 117.8 179.0 125.8
Where,
D, is the bulk sediment diffusion coefficient (1(}-6 cm 2·s·'). (Data from Li and Gregory (1974) for brown abyssal clay with a porosity of 71% at T = 20-25°C, quoted in Berner, 1980). Zmis the mean diffusion path length calculated for t (time) = 1 Ma.
oz
of three terms, the diffusive flux, the advective flux and the flux resulting from ongoing reactions. To distinguish between the various contributions each situation has to be analysed individually. A depth dependency of the diffusion co-efficient is to be expected from its relation to porosity via tortuosity. Tortuosity 0 is a measure of the distortion of the diffusion pathway of solute species due to the presence of solid particles. Tortuosity is a function of porosity and the formation factor F, which is the resistivity ratio R.,IRo of a natural sediment (Re ) and its pore water (Ro) (see Table 3). The bulk (or sediment) diffusion coefficient Os, therefore, depends strongly on porosity and, for that matter, on depth. For more compacted sediments considerably smaller coefficients result. Apart from its effect on the sediment diffusion coefficient Ds ' porosity cf> has to be included in all of the equations (1) to (4) as a separate factor, because under diagenetic conditions diffusion occurs only in the pore water. This leads to equations of the form given in equation (7) (Table 3). Over what distances can diffusion affect transport of dissolved species in actively subsiding sedimentary sequences? It is possible to calculate the mean diffusion path length Zm (Gieskes, 1975), which is the distance over which
286
Diagenesis
diffusion will diss ipate concentration anomalies for a given period of time (equation (8), Table 3). For a period of one million years diffusion path lengths for young, unconsolidated sediments are characteristically between 100 and 200m (Table4), which exceed the thickness of most pelagic sediments accumulated in a million years. It is, therefore, entirely feasible that in pelagic sediments the sediment-water interface and the igneous rocks (layer 2) of the oceanic crust are still in diffusive communication, even after deposition of several hundred metres of sediment. This will always be the case, if the diffusive communication length Z, which is the diffusion coefficient Os divided by sedimentation rate CAl (Gieskes. 1975), exceeds sediment thickness.
eupelagic or true pelagic sediments. They represent the most slowly deposited sediments on earth including the brown abyssal clay, the prototype of a pelagic sediment, with sedimentation rates of less than 5 rn- Ma-'. Biogenic pelagic sediments are deposited somewhat faster, at rates ranging from less than 10 rn- Ma-' up to 100 m -Ma-' for most modern siliceous and calcareous oozes (e.g., Scholle et et., 1983, p. 640). Pore-water profiles for pelagic sediment sites of the DSDP display three principally different trends: those with no pronounced gradients at all (type 1); others with vertical gradients but linear correlation between major ions', especially calcium and magnesium (type 2); and a third group with gradients but no linear correlation between the major ions (type 3). The first group of sites occurs close to mid-ocean ridges and has a limited sediment thickness, generally less than 200 m, the second group at greater distances from the ridges and under a th icker sediment
I. LOW TO INTERMEDIATE SEDIMENTATION-RATE BASINS (Pore-water Profile Types 1 to 3) These have sedimentation rates of less than about 50 m- Ma-1 characteristic for Sal
pH
a
N. K
~
r-""1n
Sr
C"" I%.' 781620242832361418400 10IS0.0S0.2 r1i
iii
i
I
1~ t t
Co
...
Alk
r"'"""'T"'""
~
1""""'.1 101S203040!iO 02040
t'"T'TT"1r-T'l
Geoscience Canada Reprint Series 4
cover (Figure 5), the third group in more reactive pelagic sediments. The first group is convection-controlled, the second diffusion-controlled and displays 1\ 25
o
SITES I I1HOUI mDIUl
20
•
15
10
Figure 5 Frequency distribution of drill sites with and without concentration gradients for major pore water cations in pelagic (and hemipelagic) sediments as a function of sediment thickness on basaltic crust (based on DSDP legs 1-85, data for legs 1-53 compiled by McDuff, 1981). n number of drill sites. Si0 2
o,--,-,10
00.1
.,-,
SllES11TH ;UDlUT
o
1020 0.7
rTTT"T'1
j
1.0 ,
6"01%.1 (SMOW) 1.201.0
i i i 1'1
I \i ~f )J ~~ ~:l/!l!t;; W '\
200
3001 400 ~
\..
I
400~
,...., 10 14
711620242132311418400 10 0.05 ,.,1
!
J
i
,
or 100
~
I~!(,)
I: ~t
,
I
j
il~""""r-
r
~
I"TI
0 1020
M
010
rT"1i
02040
iii
0
I~I
I
,
,
,
iii
i
~,..--.r--"'T"
~1
Ii
1 ]
n
Iil0
0.1
,,......,....,
5 10 15 0 0.1 0.2
~
0 1020
i
o "un
0
01.0
~
I
o 1.0
c
c c c c
c E
0.3
~
\\
c c c c c c c c c c c c
C
3.0
~
,.-,
",.......,
"-
c c
0.4
20304OtiO 020 408080- 5 101600.10.20.305150.10.70.8 ,
495
I
0.1 ' 0.3 ' O:S ' 0:7 0.1 0.20.3 781620 24 28 32 36 1418400 10 0.05 0 10 20 3040 !i080 70 408080100 101520 020.3 SO.5 0.7 -r-r-1n t"""T-r-1 ni i i i i i'r--n~n n t'T"1i i i i · i i i I i i i i i ~
C r
c
\( r
"T'T'T"M
c c c
497
Ii '0
Figure 6 (Above and opposite page) Pore-water chemistry and Isotopic composition of DSDP drill sites 495 (pore-water profile, mixed type 1 & 2 & 3), 496 (type 6), 497 (type 6), 565 (type 6), 568 (type 6), and 570 (mixed type 4 & 6). Sites 495, 496, 497 (leg 67)(Harrlson et al., 1982), Sites 565, 568, 570 (leg 84)(Hesse et aI., 1985; carbon-isotope ratios from Claypool et aI., 1985) on Middle America trench slope and oceanic crust of the Cocos plate (site 495) off Guatemala. Sites 565 (continental slope off Nicoya Peninsula, Costa Rica) and 568: squares identify In situ water samples.
c c c c e c c c C l
C C
C
(
c c r co r o (
287
Early Diagenetic Pore Water/Sediment Interaction
pH
,~ 200 300
Sal.
cr
(g/kgl
(g/kgl
K+ Sr 2 + Ca 2 +
rmf$f l ~
Alk .
(mMI
SO/-
(meq/kg)
ty.....,..,,0:--~=-30 .-=-~. '~'
o ,
6 '8 0
(mMI ~MI ~/OO SMOW) 51015 0510'5200.40.60.81 .03.05.0".00.'0+3.0 I
1: \~~ :f~ "' 1\ 1
:. ~ 't ( 1\ > 4
I
6510
81216
o
o o
•
_ _I
o
,
0:20:4
o o o
o o
A1 3 +
H4Si04
510
30
o o o
o
0
o
i
•
-200 -180 0
.
••
200
E
s
15
D
.Q
300
:>
'"
t OO
200
,- 200
•
..:
-180 - 160
400
i
i
i
r
• ••
••
~.
300
I
-80
- 60
.(
.' . i
-40
-20
•
0
, ,. 40,
•
,\
,
•••
,
-60
i
i
."\, \
•
tCO? ImMI
, •• • • •• •• •
40
60
I
i
20
40
20
i
'. '-'
•• •
• •
\ - 40
0
~
\
-80
I.
0
•
•
•
• •
•
• -20 • •
i
,
0
, ,-40,
- 20
•• • • •
,
80 I
:00 i
585
••
-20 i
i_I
•• •
•• •
400
0
i
, , ,
•• • •• • •
.\
••••
100
~
~
-160
•• ••
Q
,,
tC0 ;, t 3 C to 001 2 -20 . 40 0
-20
• • • •
300
:§
i
I
i
200
o o
e
-80
No dala
o
o
,
-180 -160
100
o o o o
CH ~ 13 C (" '00 1 4 - 40 -60
CH 4 ~D fl/ool -200
60
80
tOO
i
•• • • • 588
••
•
•
• 0
20
i
, 60 •• • • • 40
I
80 i
,
tOO
i .'
•
• •
•
•
•
570
c 288
Diagenesis
Geoscience Canada Reprint Series 4
(
C oxic to suboxic diagenesis. The third group is characterized by suboxic (to anoxic) diagenesis and is react iondominated, but formerly may have been convection-dominated. Convective Pore-water Circulation on the Flanks of Mid-ocean Ridges (Type 1 Pore-water Profiles). The first type of behaviour, i.e., profiles with no gradients for most ions, is illustrated by DSDP site 495 (Figure 6). McDuff (1981) has argued that the straight vertical profiles of seawater composition are probably caused by convection of seawater through the sediment column. This is in line with geothermal studies (Anderson et et., 1977, 1979)which suggest that heat loss on the flanks of mid-ocean ridges is by convective rather than conductive transport and that the heat-transporting fluids must percolate through the sediments (layer 1) of the oceanic crust. Postulated convective flow cells have a diameter (or wavelength) of 5 to 10 km and several of them may occur side-byside on the ridge flanks thus extending from the ridge crest outward to basement ages of up to 80 Ma. The cell structure requires that regions of discharge from the crust alternate with regions of recharge, but it is not clear how the former manifest themselves in the pore-water chemistry except at the ridge crest where fluids are issued in the "black and white smokers" of the hydrothermal vent system (Edmond et a/., 1979,1982; VonDamm et a/., 1985a, 1987). Active hydrothermal systems that have been drilled at various sites by the DSDP are characterized by ongoing reactions in the sediments (Von Damm et al., 1985b)or by intrusions of igneous dykes and sills producing complicated pore-water profiles that are far from straight lines (pore-water profile type 8). Discharge associated with simple, convection-controlled vert ical profiles may have been encountered in one area at 20 0 S on the East Pacific Rise (Bendereta/., 1986),but only the uppermost 2 m of the sediment were sampled. For discharge areas, which are probably much narrower and more focussed than the broad recharge areas, non-linear profiles would be expectedwith greater depth of penetration. Once the sediment cover on the oceanic crust exceeds a certain thickness, it becomes a seal for convection due to porosity reduction. When con-
vection is no longer operative , diffusion sets in to bridge the concentration differences between seawater at the top and hydrothermally altered fluids at the base of the sedimentary section, thus smoothing out concentration gradients. In discharge areas, porosity-reducing processes, such as recrystallization and cementation , may be accelerated due to elevated temperatures sealing off convection early, i.e., even under a relatively thin sediment cover. Thus discharge-related straight-line profiles would not be expected. An alternative explanation for the straight-line profiles is by diffusion from seawater reservoirs located both above and below the sediment section (McDuff, 1984). This is based on the discovery of cold seawater circulating in the highly permeable uppermost basaltic oceanic crust under a thin cover of young pelagic sediments which form a relatively impermeable lid (Langseth et a/.• 1984). Site 495 on the Cocos plate off Guatemala is located about 20 km oceanward from the Middle America Trench and bottomed in basalt of Early Miocene oceanic crust. At this site convection may have ceased not too long ago, but the previous convection-controlled profiles are still preserved due to a lack or sluggishness of reactions in layers 1 and 2. Its straight-line porewater profiles with near seawater concentrations for most ions (Harrison et a/., 1982)are the more remarkable, because major lithological changes occur in the sediment column penetrated by the drill - from Pleistocene to Upper Miocene olive-grey hemipelagic mud (upper 170 m) through Middle Miocene brown abyssal clay (10 m) to Middle to Lower Miocene white, pale brownish and greenish calcareous ooze and chalk (120m)and pale brown to purplish manganiferous chalk and limestone (30 m) overlying the basalt. As Figure 6 shows, none of the major ions seems to be affected in its vertical trend by any of the lithological boundaries (at 170 m, 180 rn, 399 m and 428 m depth), although these are relatively sharp. The only exceptions are silica and strontium and, in part, calcium, which display appreciable variations although these are also not related to any of the major compositional boundaries between the lithologic units. (These variations are exaggerated in Figure 6 due
to the different scales used for Sr2 + and Si02 compared to most other ions.) Si02 , Sr2 + and Ca2 + are species involved in active diagenetic reactions taking place in the sediment column at the time of drilling. Strontium is released by carbonate recrystallization. Its maximum near 250 m sub-bottom depth occurs approximately at the level where the transition from calcareous ooze to chalk is observed in the cores. Silica concentration profiles are known to be highly variable with depth and dependent on lithology; site 495 is an excellent example. Increases in dissolved silica in the upper few 100 m of biogenic pelagic sediments are related to the dissolution of opal-A of the tests of siliceous organisms , mostly diatoms and radiolaria, while marked decreases in silica which occur at greater depths are related to the reprecipitation as opal-CT. In site 495, the marked decrease in dissolved silica near 325 m sub-bottom coincides with the first porcellanite nodules observed in this hole, conf irming the relationship with opal-CT precipitation. The diagenesis of siliceous oozes and associated mechanisms of chert formation are discussed in a separate article (Hesse,this volume, p. 227-251.) Thus the two dissolved species of site 495 which display pronounced maxima in their concentration profiles, Sr2 + and Si02 , are clearly involved in ongoing reactions. The convective flow system appears to have ceased to operate not too long ago at this site because it . failed to obliterate the effects of these reactions, and the reaction products liberated are redistributed by diffusion (as in the case of Sr 2 +) or consumed by the precipitation of solid phases (as in the case of Si02 ) . Site 495 thus, in fact, is a mixed site showing the effects of former convection, of diffusion and of ongoing reactions. Its sediment thickness exceeds 400 m, which is higher than for most convection-controlled sites (Figure 5), but this is notsurprising in view of the mixed nature of this site. Diffusion-controlled Pore-water Trends (Type 2 Pore-water Profiles). The brown abyssal clay is a good example for diffusion-controlled pore-water profiles. It is virtually free of organic carbon. Any organic matter that may be deposited is removed by oxidation in the oxidation zone. A few tenths of a percent or less of highly refractory
C C C C C C C
C C C C C
C C C ( (
C
C C C C C C C E C C C
C
c e c l
C (
C C
C C
c
c
c c c o c o o c
c o o o o o o o o o o o o o o o
o o
Early Diagenetic Pore Water/Sediment Interaction
organic matter may remain, according to analyses of many drill holes in the Pacific (Initial Reports of the DSDP). In the absence of reactive organic carbon, which is the main reducing agent in marine sediments, iron and manganese remain in their oxidized, higher valence states in the sediment which therefore retains its brown colour during burial. During later stages of burial the sediment may gradually change to red when the brown iron hydroxides are converted to hematite. Although freely dissolved oxygen may disappear completely from these pore waters at deeper burial levels, they do not normally become reducing enough to change the Fe3 + to Fe2 + and thus leave the red colour unchanged. Various other types of pelagic sediments besides the brown abyssal clay display diffusion-controlled pore-water profiles. These are characterized by linear correlation between the concentration changes for calcium and magnesium as well as for oxygen isotope ratios. This phenomenon, particularly the inverse relationship between calcium and magnesium (Figure 7) had attracted considerable interest in the past
8 C
r.
Sulphate (mM)
Ca and Mg are essentially diffusioncontrolled. Considerable variations in the numerical value of the diffusion coefficients result from changes in porosity, temperature and tortuosity with depth and these can be measured directly (porosity, drill hole temperature) or indirectly (tortuosity) using electrical resistivity determinations. By setting the velocity of advective flow v and the contribution from reactions R in equation (9) of Table 3 equal to zero, that is, assuming that the two elements Caand Mg behave conservatively and do not experience lossesto, or additions from, the sediment column, McDuff and Gieskes (1976) were able to match calculated profiles with the measured ones. Thus, only a diffusive flux seems to be important. Successful modelling suggests that such an interpretation is acceptable; it does not prove that it is the only possible one. What arethe reactions in the basaltof the oceanic crust that lead to the release of Ca and the consumption of Mg underlying the linear correlation between the changes (ACa-AMg)of these two elements? These are generally hydrolysis reactions of the alumino-
Calc ium and magnesium (mY)
2 4 ,...,--r-~~.• ",~T"'>
80 I
I 200
IT
400
ill
600
TIl"
E
-
Stront rum (mM)
s:
Potassium (mM)
Q
02
QI
o
I
l'
and led to different hypotheses. Sayles and Manheim (1975) thought that the diagenetic formation of high-magnesium calcite was responsible for the ACa-AMg correlation, a conclusion which, if substantiated, would have been in conflict with the common experience of carbonate petrologists that high-Mg calcite transforms diagenetically to low-Mg calcite, but not vice versa. Because the starting material in deep-sea sediments is low-magnesium calcite, a stable phase would have been replaced by a less stable phase during progressive diagenesis, in violation of Ostwald's rule. Donnelly and Merrill (1977) related the magnesium decrease with depth to silica diagenesis. Lerman (1975, 1977) and Lerman and Lietzke (1977) considered the effects of diffusion but neglected to include variable diffusion coefficients which forced them to invoke reactions in the sediment to explain the gradients. In their reevaluation of the problem, McDuff and Gieskes (1976) and McDuff (1978, 1981) included depth-variable diffusion co-efficients, with the result that they were able to argue successfully that these profiles with coupled changes for
Alkalinity i meq l-I)
o,..---
289
200 400
04
•
~
eI•
~
f-
600 L..f-
o
06
.-+-..:,.,I
(·l
\---<,
!•
t.••
------
e:::. ...J
••
Figure 7 DSDP site 528 on the Walvis Ridge, South Atlantic, with linear correlation between downward Ca2 + increase and Mg2+ decrease (pore-water profile, type 2). Dissolved Mg2 + drops rapidly to zero below the first basalt, at the same level Ca2 + increases rapidly. Dissolved Sr2+ shows typical trend for pelagic carbonates with a maximum around 100 m, resulting from carbonate dissolution. Lithology: I: calcareous ooze, II: nanna-fossil ooze and chalk, III: chalk with interbedded turbidites, IV: interbedded sedimentary rocks and basalt. (From Gieskes, 1983).
c 290
Diagenesis
Geoscience Canada Reprint Series 4
(
C silicates of the primary basaltic rocks. They produce cations, hydroxyl ions and silicic acid as soluble species. Mg2+ produced in these reactions is taken up entirely by the basalt in secondary minerals such as brucite or saponite. The hydroxyl ions produced are largely removed from solution by precipitation with excess Mg2+ , i.e. , the Mg2+ supplied by diffusion. In this way basalt alteration provides a sink for Mg2+ and also a source for Ca2+. According to McDuff (1981), Mg2+ uptake may exceed Ca2+ release at low hydrolysis rates. Because the reactions affecting the basaltic rocks may also occur in the lower parts of the sediment column, if these are rich in volcanic material, dissolved Mg2+ may be reduced to zero above the sediment/basalt boundary. In this case, the continued increase in Ca2+ downward, and the hydroxyl production associated with it, are compensated by a flux of Na + supplied from seawater by diffusion into the oceanic crust, which is reflected in a distinct depth gradient for Na + . The strongest argument in favour of diffusion-controlled gradients for Ca2+
and Mg2+ that are linearly correlated is their correlation with changes in oxygen isotope ratios (Figure 8). Often these display a distinct downward decrease of 0180 resulting in moderately negative values at a few 100m subsurface depth. As pointed out by Lawrence et a/. (1975) and Gieskes and Lawrence (1981), the only quantitatively important fractionation mechanism in pelagic sediments that could bring about a preferential removal of heavy oxygen from the pore water is the formation of phyllosilicates in alteration reactions of basaltic minerals and glasses, either in the igneous part (layer 2) of the crust or in volcaniclastic sediments (layer 1). It has been suggested, however, that slight downward decreases in the oxygen isotope ratio observed in pore waters of Miocene pelagic carbonates in the South Atlantic (hole 525B, leg 74 of the DSDP on the eastern flank of Walvis Ridge) represent a fossil seawater trend reflecting the withdrawal of isotopically light water from the ocean and storage in continental ice caps (Bath and Shackleton, 1984). An ice cap started to form on Antarctica in mid-
Miocene time. Hydrogen isotope ratios of these pore waters also decrease downward at a rate of about eight times the 0180-decrease (Figure 9a) corresponding to the isotope fractionation effect associated with evaporation of seawater. The absolute values of the 0180 shift are small, however, amounting to not more than 0.5%0 over 250 m depth (Figure 9b).ln their detailed study Bath and Shackleton compared the pore-water data with the 0180 values for benthonic foraminifera from the same hole (Shackleton et a/., 1984)which displaya downward decrease of 1.4%00180 over the same depth interval. The narrower range forthe pore water is attributed to diffusive dissipat ion. This raises the difficulty that the diffusion coefficient required to match the observed profile has a value of only 10-7 cm2. s-\ whereas diffusion coefficients commonly found for the types of sediments in question are in the range of 10-5 to 10"6. With these higher coefficients the gradient would have been smoothed out at least 30% more. A large reservoir of paleo-oceanwater with low oD and 01 8 0 values might exist, however, in the
C C C C C
C C C C C C C C
C
c <:
c c
c c c c c c C E
C C
C C
c
e (
c
c
Figure 8 Negative correlation between dissolved Ca concentration and 6180 of DSDP pore-water profiles. (Modified after Gieskes, 1983; data from Lawrence and Gieskes, 1981).
c c
c c c
( ( (,
c c o ( o ( c
(a)
+
+
+ +
,
(
l
40
80
E
o
+2.6 -0.4
+2.4 -0.6
+2.0 -1.0
+2.2 -0.8
EE
Elk EE
(b)
I I I I
..
831
I I
Co
0
+
EE
s: Q)
+2.8 -0 .2
+3.0
%0 POB +3.2 %0 SMOW +0 .2
o
l
~/
-4
o o o o o o
6
+
/
CO
o o o o o o o o o o o o o o
291
Early Diagenetic Pore Water/Sediment Interaction
120
831I
..E
,0=
I I
0 0
.c 160 .iJ :l en
I
200
I
240
I
-7
em
2 ·1 5
I
I I
83 83
I
D
= 10
·6
I I
em I I I I I
280
10
I
2.1--.1 5
EE
EE EE
Figure 9 (a) Oxygen and hydrogen isotope data for DSDP hole 525B from Bath and Shackleton (1984). The error bars are derived from the topmost 12 samples (small crosses), which all plot close to 0 on both co-ordinates. An error ellipse (1 standard deviation) has been drawn around the mean of these values and the diameters of this ellipse have been used for the errorbars of the remaining samples assuming that the variability among the top samples is a reliable estimate of the analytical uncertainty. The dashed line has a slope of 8, indicating that the isotopic trend reflects evaporation of seawater. (b) 3-point running average data (open circles) for pore water 0' 80 values and (squares) for benthonic foraminifera from the same hole. The less steep gradient for the pore-water data is attributed to diffusive dissipa tion. The two solid curves are calculated diffusion-controlled trends resulting from an assumed isotopic composition of mid-Miocene (14 Ma) ocean water of -0.9%0 018 0 and two different values for the average sediment diffusion coefficient.
basalt of layer 2 of the ocean crust and might allow higher gradients in sediments of layer 1 to persist, as suggested by Lawrence and Gieskes(1981), iftransport in layer 2 is still by advection and in layer 1 by diffusion . The 018 0 values show a linear correlation with Ca (Figure 8, interstitial water data for Ca in site 5258 from Gieskes eta/., 1984)and, alternatively, could be related to alteration reactions in the volcanic basement, whose products are distributed upward by diffusion. However, this does not explain the hydrogen data, because no hydrogen-isotopic fractionation is expected for this process. All known hydrogen-bearing low-temperature silicates are depleted in deuterium relative to seawater or interstitial waters (Savin, 1980). oD values of interstitial waters are, therefore, more reliable indicators for the isotopic composit ion of the original connate water than 0'80 values. Withdrawal of fresh, isotopically light water from the world ocean during glacial times is also the basis for the oxygen isotope stratigraphy of the PlioPleistocene (e.g., Shackleton and Cita, 1979). Evidence for an associated increase in seawater chloride has been presented (McDuff, 1985). Reaction-controlled Pore-water Profiles in Suboxic (to Anoxic) Pelagic to Hemipelagic Environments (Type 3 Pore-water Profiles). These have somewhat higher organic matter contents than the previous group (type 2) but still low enough values so that diagenesis remains oxic to suboxic (Figure 4). In other words, the oxidant demand is such that the free oxygen reservoir will be depleted but not the nitrate reservoir. Many of the DSDP drill sites grouped as type 2 or type 1 sites may in fact be mixed sites and belong to group 3, because the upper few metres of the sediment column, which are often left unsampled, may show the effects of suboxic diagenesis. Deeperdown in the holes, however, these effects may have been obliterated by diffusion. Piston-core studies with closely spaced samples in the upper few metres of the sediment are the appropriate approach to this environment. MOilerand Mangini (1980) estimate that sedimentation rates of less than 40 m- Ma" are required for suboxic diagenesis . For higher rates, increased organic matter content will eventually exhaust the nitrate reservoir
292
and establish anoxic conditions of the sulphate reduction zone. "Mixed" site 495 apparently belongs to this latter category. A close look at Figure 6 reveals that some of the dissolved species, especially ammonia, phosphate and alkalinity show slight enrichments whereas sulphate shows a slight decrease near the top of the sediment column. This reveals that the typical organic-matter decompositionreactions discussed before do in fact occur at present in this site, albeit apparently rather sluggishly. This is what is expected, because the hemipelagic muds in the upper 100 m of the column still contain between 1 and 2% organic carbon (Figure 10). They will have undergone at least the first two stages, but also part of the third stage of bacterial organic matter oxidation, although not much of this can be seen because most of the existing gradients were probably dissipated initially by convective flow and later by diffusion. In certain regions alternating zones occur of oxic and suboxic diagenesis on the one hand and anoxic diagenesis on the other, which show up in vertical profiles, e.g., as alternating brown and green layers and corresponding variations in pore-water chemistry. They are caused by temporal fluctuations in organic mattersupply, which,for example, may be associated with the influx of rapidly deposited, organic-matter rich, terrigenous muddy turbidites into an environment of pelagic sedimentation (Colley et a/., 1984). If the turbidites are rich enough in organic matter, their chemical signature may not be obliterated by diffusion although an oxidation front may move down (or up) from the overlying (or underlying) pelagic sediment and turn the upper (lower) part of the green or grey turbidite brown (Wilson et a/. , 1985). Recrystallization of Biogenic Pelagic carbonates. Biogenic pelagic carbonates provide another example of diffusion- and reaction-controlled pore-water profiles in low to intermediate sedimentation-rate areas, which unveil the role of solid sediment constituents other than organic matter in early diagenesis. It may be surprising that the recrystallization of pelagic carbonates which is reflected by adistinctSr2 + anomalyata 100toafew 100 m subsurface depth (Figure 11 a) does not seem to have a significant
Diagenesis
Geoscience Canada Reprint Series 4
effect on Ca 2 + concentration gradients. Recrystallization of deep-sea pelagic carbonates in the subsurface occurs because the small and delicate tests of foraminifera and coccoliths, which are the only major carbonate constituents of pelagic sediments, have a large reactive surface area easily subjected to pressure solution. Their Sr content is three to five times higher than that of the inorganic calcite reprecipitated from solution (Baker et al., 1982; Elderfield et a/., 1982). Thus a distinct Sr signal is observed, whereas the calcium (and magnesium) concentration is not much affected, particularly not in low sedimentation-rate sites with low alkalinity levels. In these sites about the same amount of calcite that goes into solution appears to be immediately reprecipitated, often as overgrowths on single-crystal skeletal elements of the Coccolithophorida, especially discoasters (Figure 12).In site 495, a small Ca2 + maximum is associated with the Sr2 + maximum (Figure 6), possibly indicating that in the absence of a diffusion supported Ca-flux from the basaltic basement a signal stemming from the recrystallization reaction may still be preserved . The maximum of dissolved Sr2 + near the calcareous ooze/chalk boundary
appears to be associated with a maximum in the recrystallization rate. From this maximum, dissolved Sr2 + is transported by diffusion both upward toward thesedimentlwater interface and downward toward a sink deeper in the sediment column or the basaltic oceanic crust. The ooze/chalk boundary may be viewed as a diagenetic front which moves upward through the sediment column as burial proceeds (Gieskes et a/. , 1986). Below the boundary, in the chalk, recrystallization is slowed down considerably. According to Stout (1985) the dissolved Sr2 + profiles with the gradual downward increase toward the maximum can be successfully modelled assuming diffusive transport and a constant production rate of Sr2 + for the reaction term in the general diagenetic equation (equation 9 in Table 3). As shown by Baker et a/. (1982)this would still allow for variable recrystallization rates, because the inorganically precipitated calcite consumes much less Sr2 + than is released by the dissolving biogenic calcite. The lower sink toward which dissolved Sr2 + diffuses downward, is provided by volcanogenic sediment deeper in the column or basaltic basement which behave similar to midocean ridge basalt during low-temperature alteration. The rate of carbonate Corg(%1
0
0
1
2
3
4
2 I
50
]
150
~
§ 200
j .i>
ci! 250
fl •
300
350
J
I1 i
~
I
>
400
496
I I I
i
lj
~II •.,.
...
£
t
4
~
t
100
3
Cf-\
-----t-.
~
(
I
.---1 ..-1 ~ l
497
5 i
o
1
2
3
~
o i
-=:ttl
1
2
3
~! \J 1i ~I
.>:
4
0
1
2
i
i
i
.>
3
I I
~l J
I I
•
I
•
?! •
494
I
499
495
Figure 10 Organic carbon distribution in DSDP drill sites on the middle (496, 497) and lowerslope (494) ofthe Middle America Trench,in the trench (499) and on the adjacentoceanic crust of the Cocos plate (495) off Guatemala. (From Harrison et at, 1982).
c c c c c c c c c c c c c c c c c c c c c c c c c c C E C C C C
c e c c c c c c c c
c (
Early Diagenetic Pore Water/Sediment Interaction
293
(
( (
G ('
o ( c ( c o o o o o () o o o
recrystallization in pelagic sediments depends also on sedimentation rate (Gieskes and Johnston, 1984), which controls the production of CO~- or HC03' in the sediment, as discussed in the section on "early diagenetic organic matter oxidation". The Sr2 + maximum occurs thus at greatly variable subsurface depths. Direct evidence for the diffusive transport of Sr2 + in pelagic carbonates is provided by strontium isotope ratios. Sr-isotope curves for the pore waters of pelagic carbonates break down into two parts (Figure 11 b). In the upper part above the Sr-maximum 87Sr/86Sr ratios are generally lower than the contemporaneous seawater curve of Burke at al. (1982) and Palmer and Elderfield (1985). Below the maximum they tend to be higher than or equal to paleo-seawater ratios for coeval sediments. This requires both upward and downward diffusion of dissolved Sr2 + from the site of maximum recrystallization (i.e., the dissolved Sr-maximum), where the isotopic composition is closest to that of contemporaneous seawater. Because seawater has become increasingly
(a)
o o o
o o C l
e C
0.8
1.0
rn.o.
1.2
o
40
.709
200 300 400
E
~
-
Cf) CD
:i: 500 a.
.708
CI)
<,
CP
0600
0
~
,..Cf)
288
CI)
700
.707 800
1000
c
20
100
900
"
paleo-seawater curve (Figure 11b). The dissolved 87Sr/ 86Sr ratios are too low to be due to upward diffusion, partlcularlyas insufficient carbonate is available for recrystallization. Here, alteration of volcanic sediment particles with low Sr-isotope ratios (of about 0.704) must be invoked as the dominant source of the anomaly (Gieskes et al., 1986). Alteration of volcanogenic sediment as a source of a light Sr-isotope signal is often accompanied by magnesium losses from the pore water in excess to those attributable to diffusive fluxes. OtherEarly Diagenetic Mineralization Reactions in Pelagic Sediments. Zeolites provide additional, well-documented examples of early diagenetic mineralization in slowly deposited pelagic sediments, particularly volcanogenic pelagic sediments (Kastner and Stonecipher, 1978). The crystallization of zeolites from volcanogenicmaterial proceeds through the formation of palagonite, a yellow-brownish alteration product of volcanic glass, which appears to be a common precursor of zeolites. The downward decrease of K + in the pore waters of many pelagic drill
(b)
Sr 2+ (mM) 0.6
o o
o o
heavier in strontium isotopes during the last 100 Ma toward the present, the Srmaximum is a source of relatively light Sr isotopes for the section above it and of relatively heavy isotopes for the section below it. DSDP site 289 on the Ontong-Java Plateau in the Southwest Pacific differs from other carbonate sites in thatthe Srisotopic composition of the pore water below the ooze/chalk boundary is very close to the paleo-seawater curve (Figure 11 b). Inthis site, maximum dissolved Sr-concentrations are reached at about 400 m sub-bottom and below this depth no significant decrease is observed (Figure 11 a) suggesting that the downward diffusive flux in this thick chalklimestone section is very small or nilin line with the isotopic data which also suggest that in the chalk section of this hole isotopic equilibrium exists between the pore waters and their host sediments . In carbonate-poor sediments of sites 541 and 543 near the toe of the Lesser Antilles island-arc slope, Sr-isotopic ratios of the pore waters in the upper part of the holes fall much below the
.706
• = 525
0 =528 0=543 o = 289 ... = 315 ~ = 357 • = 541
Figure 11 (8) Typical dissolved Sr2 + concentration profiles in pelagic sediments, Pacific drill sites, DSDP. (Modified from Baker at aI., 1982). (b) Sr-isotope ratios versus sediment age in pore waters from various DSDP drill sites and Sr-isotope curve of seawater (stippled) for the last 80 Ma . (From Gieskes et at, 1986).
294
Diagenesis
Geoscience Canada Reprint Series 4
( ( (
sites may be related to the formation of the main potassium-bearing zeolite, phill ipsite (K,Ca)AI 3Si sO'6 . 6H 20. Other common zeolites formed during early diagenesis in pelagic sediments such as clinoptilolite and analcite leave a less characteristic imprint on porewater composition . As Kastner and Stonecipher (1978) have shown, phillipsite is most abundant in the youngest sediments at shallowest subsurface levels and decreases in abundance with age and burial depth, whereas clinoptilolite shows the opposite behaviour. The transformation of zeolites into K-feldspar with progressive burial has been documented repeatedly and is another sink for K + , as are adsorption and ion exchange with clays. Widespread neoformation of clay minerals does not seem to occur in pelagic environments at shallow burial levels (Kastner, 1981). " Rev ersed weathering" (MacKenzie and Garrels, 1966), which would have involved clay mineral reconstitution in the oceans through the uptake of silica and cations by amorphous aluminosilicates, has not been found to occur on a large scale. Small amounts of very fine-grained smectite may form, however, in early diagenetic environments (Chamley and Millot, 1972; Johnson, 1976; Moberlyet a/., 1968; Hein et a/., 1979a) and the «5'80 and Mg2+ decreases mentioned earlier most likely involve newly formed clay minerals besides zeolites . DeLange and Rispens (1986) related sharp decreases in dissolved silicon and iron below mud-turbidites of the Nares Abyssal Plain to the precipitation of small quantities «0.1 wt.%) of an amorphous Fe-Si mineral, possibly an iron-smectite precursor. Decreases of dissolved aluminium at shallow subsurface depths may also require the formation of small amounts of authigenic clay minerals (Stoffyn-Egli, 1982) and this mechanism may also operate at deeper levels in the subsurface (e.g., highsedimentation rate DSDP sites 565 and 568, Hesse et al., 1985; Figure 6). The formation of AI-rich di-octahedral chlorites was suggested by Mackin and Aller (1984) for shelf sediments of the East China Sea where other aluminium removal mechanisms such as complexation with dissolved organic matter or adsorption on amorphous silica surfaces are less likely to occur, whereas in the above DSDP sites these and the
formation of smectite (Helm, 1985) might be sufficient to account for the dissolved aluminium decrease. These examples show that in a given drill site usually several mechanisms contribute to the make-up of the porewater chemistry. In low sedimentationrate environments such as the pelagic realm , convection- and diffusion-dominated pore-water profiles prevail, but reactions may significantly modifyvertical trends as shown for Sr 2+, K + and Si02. Thus every drill hole and every element has to be investigated specifically. In the following section, reactiondominated pore-water trends will be analysed which are characteristic for high sedimentation-rate environments. Some of these can also be found, however, in low-rate environments as examples of the foregoing discussion have shown. II. HIGH SEDIMENTATION·RATE BASINS These are characteristic of the continental margins, where sedimentation rates up to 500 rn- Ma-' and more, i.e., rates two orders of magnitude higher than for the brown abyssal clay, are no exception. High sedimentation rates at the continental margins result from the increased input of terrigenous material which tends to be rich in terrestrial organic matter. Rapid burial enhances preservation of this organic material. An
additional source is marine organic matter which is supplied at increased rates in areas of high biogenic surface productivity associated with regions of upwelling of nutrient-rich deeper waters. Because zones of upwelling are concentrated around the margins of the continents, continental slope and outer shelf sediments may contain signif icant proportions of marine organ ic matter which is highly reactive and has a high potential for liquid hydrocarbon generation. In high sedimentation-rate basins, the role of diffusion in smoothing out concentration anomalies is suppressed, because sedimentation rates exceed diffusion rates. In addition, anomalies produced by ongoing reactions will be much more pronounced than in low-rate basins because of the elevated concentration of reactive organic matter. Initial organic carbon contents of these sediments often exceed 2 or 3% (Figure 10). Pore-water Trends in Continental Margin Environments: Major Element and Isotope Chemistry in Anoxic Sediments (Type 4 Pore-water Profiles). The sequence of organic matter oxidation reactions discussed in a previous section is best reflected in the vertical trends of pore-water chemistry in rapidly deposited sediments of the continental margins. These sediments pass
Figure 12 Calcite overgrowth with well-developed crystal faces on discoaster. Distal ends of originalrays of discoaster markedby arrows. DSDP site 199, Caroline AbyssalPlain, 6090 m water depth, Early Pliocene carbonate turbidite, 59 m sub-bottom depth.
(
C C ( ( ( (
C C C C C C C (
c c c c c c
c c
c c c c c c c c c c c c c c c c
c r
c
c
(' I....
o o c
o o
c
c o o o
o
o
o n
~
o r:
r
o
Early Diagenetic Pore Water/Sediment Interaction
very rapidly through the oxidation and nitrate reduction zones during burial and essentially experience anoxic diagenesis. The main chemical species released to the pore water from the microbial breakdown of organic matter and the concomitant reduction of oxidants are the nutrients EC02 (including the species CO 2 , H 2C0 3, HC03" and CO~-), phosphate, ammonia and sulphide. The main oxidants consumed in the process are °2(aq), N03" and SO~ and also CO~ - . CO2 production occurs at all stages from (1)to (6), consumption from stage (4) downward. A distinction between the CO2 released and the CO 2 consumed at different stages is possible due to the strong isotopic fractionation effects associated with some of the reactions, particularly methane generation beginning in stage (4). In order to understand fully the chemical trends, reactions other than organicmatter decomposition have also to be considered. These include ion-exchange reactions, adsorption, and the precipitation of (early) diagenetic minerai phases. These, and reactions coupled with phase changes (i.e., gashydrate formation). which introduce further fractionation effects and diversification of the chemical trends, will be discussed in the following sections. Drill holes of the DSDP in the East Pacific off Guatemala will again be used as examples, because they are representative for high-sedimentation rate continental margin environments and display some trends but also some special features with unusual clarity. The landward slope of the Middle America trench off Guatemala was drilled during two legs of the DSDP (legs 67 and 84) in two parallel transects close to each other thus providing a unique opportunity to check the reproducibility of the results of interstitial water analyses in nearby drill holes. A third transect of holes was drilled in the same general area on the slope off Mexico (DSDP leg 66). Drill sites 496. 497 and 568 which are located on the mid-slope in water depths between 2000 and 2400 mare characterized by extreme maxima for carbonate alkalinity, ammonia and dissolved phosphate in the upper parts between 50 and 200 m sub-bottom ' (Figure 6). These maxima occur in the carbonate reduction and fermentation zones, i.e., below the sulphate reduction zone which is about 5 m thick in this
region as sulphate is absent in samples taken below that depth. Carbonate alkalinity rises to more than 120 milliequivalents (meq)- kg" between 23 and 45 m sub-bottom in hole 496 - a record value only exceeded during recent drilling on the Peru continental margin (ODP leg 112Scientific Party, personal communication). PO~- reaches a maximum of 0.4 mM and NHt of 21 mM in the same hole. In comparison with the alkalinity maximum at 35 m sub-bottom depth, the other two maxima are displaced downward to 56 and 175 m, respectively. The fact that the carbonate alkalinity maximum occurs within the carbonate reduction zone (and not at its upper boundary) shows that carbonate (or bicarbonate) production continues in this zone due to various fermentation reactions, initially at a faster rate than consumption. Eventually, however, consumption by methane generation (and precipitation of carbonate concretions, as discussed later) become dominant. Direct insight into the interplay of the various processes may be gained from carbon-isotope analyses. During the first three steps of organic-matter decomposition (i.e., in the oxidation, nitrate reduction and sulphate reduction zones) negligible isotopic fractionation occurs. The CO2 (or HC03" , CO~ -) released has about the same isotopic composition as the parent organic material, i.e., a O'3C of about - 25%0 (relative to the PDB standard) for marine organic matter. The pore water becom ing enriched with CO2 from this source in the upper three zones will gradually approach a O'3C value of - 20%0 to - 25%0 (Figure 13, especially site 174A). In the carbonate reduction and fermentation zones, however, the disproportionation of organic matter into CH 4 and CO 2 is accompanied by a strong isotope fractionation effect. The CH 4 generated has a C-isotopic composition which is about 70% lighter than the carbon of the parent material and may thus attain o'3C-values as negative as - 90%0 or - 100%0. This is a kinetic effect (Rosenfeld and Silverman, 1959) by which the CO 2 with light carbon is preferentially reduced to CH 4 • The residual CO 2 consequently becomes enriched in O'3C reaching positive o-values (Figure 13)as high as + 15%0 to + 25%0 (e.g., Curtis et al., 1972). As the dissolved carbonate becomes progres-
295
sively heavier, so does the methane produced from it at deeper levels. Through this feed-back mechanism (a Rayleigh-distillation process) the O'3Ccurves for both methane and dissolved carbonate show parallel trends (about 70 o-units apart) with depth. These are often characterized by a slight decrease in O'3C at greater depths due to the release of relatively light carbon by the continuing breakdown of organic matter in fermentation reactions . The parallelism of the carbon-isotope curves for methane and CO2 is convincing evidence that carbonate reduction is actually involved in biogenic methane generation. More recently (Claypool et al., 1985)O'3Cvalues as high as + 36%0 to + 38%0 have been measured for CO2 co-existing with methane as heavy as - 41 %0 in deeper parts of DSDP holes 568 and 570 (Figure 6). Methane of biogenic origin is usually lighter than - 55%0, except where it is generated from acetate, the fractionation factor of which is considerably lower (about 50%0). Jenden and Kaplan (1986) estimated that about 20% of the CH 4 in the uppermost metres of the Guatemalan trench slope sediment could be acetate-derived and that this fraction decreases to 0% at the bottom of the drill holes at 400 m. o'3C-values heavier than - 45%0 are generally considered characteristic for thermogenic gas (Schoell, 1983) with the exception of acetate-derived CH4 • However, there is no evidence for upward migration of thermocatalytic methane on the Guatemalan trench slope . In, orlrnrnedlatelybetow, the sulphate reduction zone, O'3C values for CO2 • more negative than - 28%0 have been observed locally, e.g., in DSDP hole 533 on the Blake Bahama Outer Ridge (-31.4%0, Claypool and Threlkeld, 1983), which require a source more depleted in heavy carbon than average organic matter. This could be due to the predominance of isotopically lighter terrestrial organic matter (Sackett and Thompson , 1963) or the anaerobic oxidation of methane diffusing upward into the sulphate reduction zone. In ancient rocks, a record of these processes may be preserved in carbonate concretions which incorporate carbonate of given isotopic composition from the pore water (with a small liquid to solid fractionation) in their concentric growth shells (see below).
c Diagenesis
296
Before following further the fate of the products of organic matter decomposition in mineral izat ion processes, the behav iour of some other major elements will be dealt with. Of the two main dissolved constituents of seawater, chloride and sodium , the former does not pa rticipate in reactions, because of its large ionic radius which prevents it from incorporation into the crystal lattices of common authigenic or hydrothermal mineral phases. As a nonreactive species , chloride often does not display vertical concentrat ion gradients, at least not at shallow bur ial depths. If, on the other hand, chloride gradients are observed, one of the following processes may be inferred : (a) dilution by mixing with fresh water released from gas hydrates or by mixing with meteoric waters; (b) salinity increasedue to halite dissolution or loss of water in hydration reactions, e.g., in basalt of the oceanic crust; or (c) slight changes in seawater chlorin ity during glacial-interglacial cycles. The Effects of Gas Hydrates on Pore-water Chemistry: Salt and Isotope Fractionation (Type 6 Pore-water Profiles). An interesting freshwate r source of a different nature is the occurrence of gas hydrates which will also leave a distinct isotopic imprint on the pore water. A significant downward chlorinity decrease coupled with a 01 8 0 increase (up to 3.3%0) has been observed in gashydrate bearing sections of DSDP drill holes (Harrison et al. , 1982; Jenden and Gieskes, 1983; Gieskes et al. , 1985 ; Hesse at al. , 1985) and interpreted as release of fresh water from the thawing of hydrates, either at the base of hydrate zones or in the sampling process (Hesse and Harrison, 1981). Gas hydrates are ice-like substances which form in organic-matter rich sediments with strong bacterial methane production. The crystal structure of gas hydrates (clathrates) is similar to that of zeolites containing large cavities (cages) in which gas molecules, such as CH4 , CO2 , H2S , N2 but also higher hydrocarbons up to the size of isobutane, may be accommodated. These gas molecules are not chemically bound in the structure, but are held in the cages by van der Waals forces and hydrogen bonds, which make the struc ture more stable than that of normal ice. The format ion of gas hydrates from
SCl3 0 100
(,,_
Geoscience Canada Reprint Series 4
v. POB)
-90 -80 -70 -60 - 50 -40 -
-20 -10
.--
<,
~
-a. . . . .
-,
]200
+10 +20
0 -,
\
\
:I:
I I
I
t- 300
0..
ijJ
UJ
0400
I
d1
J
~
I
t t
gsoo
t-
I
I
I
gsoo
10
,
r1J I
lD
::::>
en
I
SCl3 (%0 v. POB) - 40
-60
-80
+20
0
-20
~~
E
i
\
0\
'a.'Q lD
50
t-
0..
\ 0 \
~ \
UJ
0
~
01
I
e,
80
gloo
I
t-
I
, riP rtJ
'8 8'
0 lD lD
::::>
en 150
> I
4> I
b SC I3 (%. v. POB) -so
-80
- 20
it- 200
0
.e--
G....
~
100
E
(!I}
0»
~4oo
\
[J
~soo
0
J
/ /
cp
clIIJ
0 gjsoo
0
/
t
0
+20
'~,
..
~3oo 0
o !COz
008
SCI3 (%0 v. POB) - 80 100 ]200 :I:
ti:
300
UJ
0400 ~
g500
<,
-60
-40
-20
0
+ 20
( <,
<,
Q ......
,
'&b I
12I
I
c c c ( c c c c c <: c c c c c c c c E C C
SITE 174A I!l CH4
700
c c c c c
C
\
I
::::>
en
- 40
c c c
6-
..... <,
I
<;)
/
0
/
6
I
tgsoo !II
::::>
en
Figure 13 /l' 3C depth trends in CH. and CO2 in four DSDP drill sites (102: Blake-Bahama Outer Ridge; 147: Cariaco Trench; 174A: Astoria Fan; 180: Aleutian trench floor). (From Claypool and Kaplan, 1974).
C C
c e c c c c c c c
r:
Early Diagenetic Pore Water/Sediment Interaction
o 8
water and methane involves a negative molar volume change. Prerequisites for the formation and stability of methane hydrates are high gas concentrations,
297
elevated pressures and relatively low temperatures. These conditions exist in terrigenous continental slope and rise environments, particularly in the lower
~
o o o
1000
~
o o o o o
-'E
10 0
GAS
~
w a:
5
::>
r:
(f) (f)
o
a: a..
+ WATER
w
7
o o
10
[
o o o
1 2
Pure methane + J . 511 Na Cl Pure methane + pure water 3 91' methane + 7' C02 4 98' methane + 2' C02 5 90.4' methane + 9.6' ethane 6 56.4' IIICthane + 43.6' ethane 7 Pure ethane
8
Pure propane
2~---'----'--~-~-~---'----'--~-""""
o
10
20
30
40
TEMPERATURE (DC) Figure 14 Stability field of hydrates with various gas compositions. (From Hesse and Harrison , 1981, after Tucholke et al. , 1977). Boxesindicate estimated poTconditions for DSDP sites 497 (closely spaced diagonal lines) and 498A.
sw
:=J
ME BASEOf' GAS H'fDIUlTE ( . S R)
•
r
-; ~
Figure 15 Multichannel seismic reflection profile of gas hydrate zone with bottom simulating reflector (BSR) on Blake-Bahama Outer Ridge. (From Shipley et al ., 1979). Vertical scale: Two-way seismic travel time in seconds.
latitudes where a prolific supply of terrestrial organic matter to the oceans may be added to abundant marine organic matter below zones of upwelling. In water depths between 500 and 5000 m, hydrostatic pressures are sufficient for gas hydrates to be stable up to 1O o-30 o e (Figure 14). Gas-hydrate zones can thus attain a thickness of up to 1 km (or more), depending on the availability of methane-generating organic matter in outer, deeper continental margin environments. Because the ground is frozen or partially frozen in hydrate zones, they appear as relatively transparent zones on seismic profiles , often bound at the bottom by a strong seismic reflector, the so-called "bottom simulating reflector" (BSR) (Figure 15)which is named for its tendency to occur at a constant distance below the ocean-bottom reflecting the pressure dependence of hydrate stability. The BSR is caused by the presence of free gas at the base of the hydrate zone, where the hydrates thaw and reo lease quantities of gas well in excess of methane solubility in the pore water. An amplitude reversal associated with the seismic signal from the BSR attests to the presence of free gas, which is sealed in by the gas-tight cap of hydrateimpregnated sediments above. Like normal ice, gas hydrates do not incorporate dissolved salts, but concentrate heavy isotopes of oxygen and hydrogen in their structure thus causing salt and isotope fractionation. As a consequence, the remaining pore fluid not involved in hydrate formation will be enriched in dissolved salts and light isotopes. During burial and compaction when partial separation of the pore fluids and the solid sediment constituents takes place, - the solids being buried, the liquids moving up relative to the solids, - gas hydrates carry down fresh water in solid, ice-like form to greater burial depth. When the hydrates thaw at the base of the hydrate zone or in the sampling process, they release this fresh water, which is also isotopically heavy. Mixing with the remaining pore water, which due to compaction is much reduced in volume compared to the uncompacted sediment near the sea floor where the hydrates first form, produces the observed trends of down-hole chlorinity decrease and 018 0 increase (Figure 6, sites 496, 497, 565, 568, 570).
298
Positive <') 1 8 0 values are a reliable indicator for gas hydrates as fresh water source, because fresh water from a meteoric source would produce negative <') 1 8 0 values. However, positive <') 1 8 0 values are not invariably associated with a chlorinity decrease related to hydrate decomposition, as shown by the pore-water profile for site 565 drilled in 3 km water depth on the continental slope off Costa Rica (Figure 6). Here, a more or less continuous downward chlorinity decrease is accompanied by a zone of negative 018 0 values at subbottom depths between 95 and 170 rn, with a minimum o-value of - 1.26%0, (Hesse at al., 1985). Note that the elevated chlorinity values between 235 and 255 m are probably due to contamination by seawater, because Mg2+ and SO~- are also anomalously high. If the negative 01 8 0 values were the result of meteoric water influx, one would not expect the chlorinity to be lowered significantly, because pore waters in aquifers at 3.5 km subsurface depth tend to have chlorinities considerably higher than fresh water. Alternatively, if the lowered chlorinity in those isotopically light zones is due to hydrate decomposition as suggested by Hesse et al. (1985), a different isotopic fractionation mechanism producing isotopically light water has to be superimposed on the hydrate fractionation mechanism. This would be the alteration of volcanic glass to zeolites and smectites discussed before. Hesse and co-workers (1985) discussed the chlorinity versus depth tre-nds for sites 496 and 568 and concluded that, if the low-chloride waters in these sites were produced by downward diffusion of CI- toward (or upward advection of fresh water from) a pool of low-chlorinity water below the hydrate zone, a trend indicated by the dash-dot line in Figure 16 would be expected reflecting downward decreasing diffusion coefficients. The trends actually observed suggest that diffusion and advection in these high sedimentationrate sites are insignificant and that the gradual chloride profiles result from a rather uniform distribution of hydrates with gradually increasing concentrations downward. In other drill sites, e.g., site 570 on the Guatemalan trench slope, sites 490 to 492 on the Mexican trench slope (Gieskes et al., 1985) or site 533 on the Blake-Bahama Outer
Geoscience Canada Reprint Series 4
Diagenesis
Ridge (Jenden and Gieskes, 1983) chlorinities fluctuate and minima probably coincide with local concentrations of gas-hydrates as massive layers or nodules in preferred horizons. These considerations militate against the possibility of upward migration by diffusion or advection to the hydrate zone of isotopically heavier waters that form at great depths in sedimentary basins due to isotopic exchange reactions between the pore fluids and solids at more elevated temperatures (Clayton et al., 1966; L. Land, personal communication, 1985). If the isotopically heavy pore waters recovered from hydrate zones on the continental slope were derived by advection or diffusion of heavier flu ids from below, rather than by hydrate fractionation, why then do we not see the same isotopic signal in hydrate-free environments of similar tectonic settings? More importantly, why then do the hydrogen isotopes show the same trends as the oxygen isotopes in hydrate zones (e.g., in site 533 of the DSDP, Jenden and Gieskes, 1983), whereas in deep formation waters they do not, i.e., the deuterium enrichment characteristically lags behind the enrichment in 5180? This again is evidence that advection or diffusion of deeper pore waters in modern con-
tinental margin environments does not occur on a broad scale as a uniform advection or diffusion front. Attempts at sampling the pore fluids from hydrate zones in situ and retrieving the expected high-salinity (high chlorinity) and isotopically light waters that have been rejected from the crystal lattices of hydrates have only been partially successful. In situ water samples taken during leg 84 of the DSDP at sites 565 and 568 showed major element concentrations similar to the analyses of pore waters squeezed onboard the drill ship except for the deepest samples in hole 568 which, however, were probably contaminated by seawater (Hesse et al., 1985; see Figures 6, 16). Oxygen and hydrogen isotopic analyses, on the other hand, showed the expected depletions in the heavy isotopes (Hesse, in press), but the number of samples was too limited for the results to be conclusive, demonstrating the need for further sampling. A modified sampling procedure with highpressure gas displacement of pore fluids might shed light on this unsolved problem in the future. Gas-hydrate zones provide an opportunity to evaluate the amount of organic carbon converted into methane during early diagenesis, because most of the
c,-
(g/kg)
20
10
0 0
•+ 568
+
568 in situ
0
Advect ion-diffusion (no pore-water flow)
E s:
0.. 200 Ql "0
E
Q'
.9
.~/'o I,"
:J
C/)
I I I
i
,0, I
400
I I I
i i
,0'+ ' e ,P
'0 .0 .0
,
/0
r+ e
I I
0
e
496
, »e ~~'
Ie) i /
e
/
/ -
; ;+
+
/ ;' ;' ;'
Figure 16 Chlorinity versus depth profiles for DSDP sites 496 and 568. (From Hesse etal., 1985).
c c c c c c c c c c c c c c c c c c c c c c c c c c C E
C C C C
c e c c c c c c c c
c
c c c o o o
Early Diagenetic Pore Water/Sediment Interaction
r:
o
(
o o o o o o
o
o o o
o
o o o o o o o o o o (
)
e
c r
methane generated is probably still trapped and little has escaped to the ocean floor, in contrast to regions without gas hydrate development and their sealing effects. From the assumptions IistedinTable5afigureofabout 1.0wt.% of carbon is calculated that is lost to the hydrate zone, or 25 to 30% of the organic carbon initially embedded in the sediment. This is somewhat less than the loss of organic carbon seen in holes 496 and 497, where the downward deTableS
299
crease between the sea floor and 400 m sub-bottom is about 3%, or 1.5% Co r g averaged overthe entire column (Figure 10). Thus, most of this decrease can be explained by bacterial organic matter decomposition and little is due to depositional factors . These calculations also show that, in sediments with initial Co rg concentrations above 3 or 4%, there is more than enough organic matter available for the formation of gas hydrates. The figure of at least 50% dilution of
Amount of organic carbon converted to methane which is trapped in gas-hydrate zones.
Assumptions Thickness of gas-hydrate zone Average porosity over this thickness Total water content of 400 m thick sediment column Chlorinity reduction from top to bottom of hydrate zone (e.g., from 19% to 9.5% in sites 496 and 497) Dilution of pore water by fresh water released from hydrates at base of hydrate zone at top of hydrate zone Average dilution over 400 m thick sediment column Average bulk density of 400 m thick sediment column
400 m 500/0 20 L H2 0 - cm·2 50%
500J0 00J0 25%
1.5 g . cm-3
Calculations Total fresh water released from hydrates over 400 m Number of moles H2 0 released Number of moles CH4 corresponding to 277 moles H2 0 in methane hydrate of compos ition CH4 • 6H2 0 Weight of carbon corresponding to 45 moles CH4 Total weight of 400-m-thick-column' cm-2 Percent of organic carbon converted to methane still contained in hydrate zone
5L
277 45 540 g 60,000 9 0.9
the pore water by fresh water released from hydrate decomposition at the base of a 400 m thick gas hydrate zone (or during sampl ing from within this zone) offers an interesting possibility to evaluate the efficiency of the isotopic fractionation mechanism . Experimental determinations have confirmed that the isotopic fractionation factor for a clathrate-watersystem(a = 1.027, Davidson et a/., 1983) is close to that of the icewater system (a = 1.03, O'Neil, 1968). With this fractionation factor 018 0 values at the base of a hydrate zone, where at least 50% of the pore water is derived from hydrate melting, should be about 1.4%0, compared to values of up to 3.3%0 actually observed. Because advection or diffusion of isotopically heavy waters from greater depths in the active margin environments considered here is not very likely as discussed earlier, a multiple fractionation mechanism has been invoked (Hesseeta/., 1985),by which the base of the hydrate zone migrates vertically up and down with time , e.g., as a result of heat-flow variations. Heavy water released by melting hydrates could again be used when the hydrate base moves downward to form new, heavier hydrates. Besides hydrate decomposition, there are other dehydration reactions causing fresh water release manifesting itself as salinity and chlorinity decrease at depth,
km FROM SHORELI NE
FERNANDINAJACKSONVILLE 0 TEST WELLS
100 I
50 I
-20t.. _
150
I
200
250
300
I
I
I
350
I
400 I
450
I
4
3
10
20
0
10
20
-
30
9
2[0 30_. MIOCEN[ O\.IGOCEIlt
=f"\ OLIGO(
I:
UOPiil EOC[NE
Jsm[
tp
20
POST~ I
NLEOCENE
o
0
U"'u .... 1IODl.[ "0:[1<[ lOWER
O\.IGOCENE LOWEREOCENE
100 400
Ol..lGQC(H[ .
_racENE 4
200
"«[H[
L~NE
6
DePTH m ft
3
r~~i lOffEA
100
200
EOCEN[
100
XJQ.J.. ,ooo
....... r-
'--
Figure 17 Fresh and brackish water in drill holes on the Florida continental shelf indicating subsurface discharge within carbonate aquifers up to 120km from the coast (pore-water profile, type 5). (From Manheim and Sayles, 1974).
300
Diagenesis
particularly in zones of abnormally high fluid pressure . Because these are generally associated with more advanced stages of diagenesis (middle diagenesis) at greater burial depth, they will not be discussed and we shall turn our attention to other processes operating at shallow burial levels during early diagenesis. Profiles With Meteoric Water Influx (Type 6 Pore-water Profiles). The influx of meteoric waters in subsurface aquifers at considerable depth and distance from shore was demonstrated in some test drill holes for the Deep Sea Drilling Project on the continental shelf off Florida (Manheim, 1967). In permeable Eocene to Cretaceous limestones, fresh water was encountered up to 120 km from shore (Figure 17). Fresh water buried during Pleistocene low sea-level stands has been reported from the Black (Manheim and Chan, 1974; Manheim and Schug, 1978)and Baltic Seas (Suess, 1976), when these inland seas became lakes. (Thrust-) faulted Sections With Active Lateral Pore-water Flow (Type 7 Pore-water Profiles). Sediments of subduction zone complexes beneath modern trench slopes undergo active tectonic deformation leading to thrust faulting, early penetra-
tive fracturing , development of "scaly clays", rehealing of the fractures by early diagenetic cements and a generally high degree of compaction. Associated with these processes is intense dewatering that should cause largescale fluid expulsion from these imbricated wedges (Carson et al., 1974; Carson, 1977; Bray and Karig, 1985). Theoretically, dewatering might be a diffuse, trench-slope wide process or, alternatively, a process of concentrated flow along structurally controlled pathways. In the pore-water profiles hitherto studied, there is no indication whatsoever for a diffuse, uniform process. Intergranular permeability in the finegrained sediments of imbricated tectonic wedges is very low and does not permit dewatering. Water escape by diffusion or advection from these wedge sediments is several orders of magnitude slower than the subduction process. On the other hand, direct or indirect evidence for fluid expulsion byfocussed lateral flow along highly fractured fault and shear zones has been reported from at least four active margins, i.e., Barbados (Westbrook and Smith, 1983; ODP Leg 110 Scientific Party, 1987), Oregon (Kulm et al., 1986; see also Ritger at al., 1987 referred to in the next
SoN «1"
(a)
Geoscience Canada Reprint Series 4
section) , Banda (Breen et aI., 1986)and Japan-Nankai (Laubier et et., 1986; Ohta and Laubier, 1987). ODP Leg 110 succeeded, for the first time in the history of deep-sea drilling , to penetrate the master thrust between the upper and lower tectonic plates in an active margin. Site 671 at the toe of the Lesser Antilles imbricated wedge off Barbados penetrated the sediments accreted to the upper (Caribbean) plate, the main decollement zone and the sediments of the incoming lower (Atlantic) plate. A distinct geochemical anomaly in the pore fluids is associated with the decollement which calls for active fluid flow.The methane concentration, which is close to zero in the upper plate sediments, rises sharply (up to 0.5 mM) in the decollement zone and the closest fault zone immediately above it, whereas chlorinity shows a sharp drop (Figure 18a).A permeable Eocene sand horizon below the decollement also shows elevated methane levels, whereas faults in the imbricated wedge only a few tens of metres above the decollement zone do not display the methane anomaly, but still carry low-chloride water (site 674, Figure 18b). As the detailed chemical analyses of these pore-fluids are in progress, it is premature at the time of revision of this review (May 1989) to
(b)
0
100 Site 674
o.. 200 '
s.,:I> -
-
-
-
j
JOO
s:
f
c c
c c c c c c c c c c c c (
] c:
c c c c c
(=
I
lQ{· ;
c c c c c c c c c
C C C
4 00
4Y.' ~ -
600
I
700~-===~---,.----,-----~
soo
,·00 l40 C
!>40
c r.·c....ct"
580 0
1""\~ 1
200 Mr':"a'l,,'
400 ,~M)
44:;
480
!i20
560
0 100
10
20
600 I C'
Figure 18 (8) (left) Decollement zone between upper (Caribbean) plate and underthrusting lower (Atlantic) plate at the toe of the Lesser Antilles island-arc slope with methane and chloride anomalies (site 671). Note thatthe methane anomaly is assoc iated only with the decollement zone and a fault immediately adjacent to it as well as with an Eocene sand horizon below the decollement, but not with faults higher in the accretionary wedge (neither in hole 671 nor in hole 674). (b) (right) Low-chloride waters, on the otherhand, are alsofoundin faultzones at greater distance from thedecollement(e .g., site 674, drilledupslope 17 km west of site 671). (From ODP Leg 110Scientific Party, 1987).
c e c (
C l. l
C C C
r r:
Early Diagenetic Pore Water/Sediment Interaction
301
r
c ( o r:
o r o o o o o o o o o o r
speculate on the origin of the methanebearing low-chlorinity waters, except that they have to come from a deeperseated source that is tapped by the decollement zone probably several tens of kilometres away under more interior parts of the wedge . Evaporite Dissolution at Depth (Type 8 Pore-water Profiles). Halite dissolution in the vicinity of saltdomes and evaporite layers is the main, although not the only source for high salinity NaCI and (Ca, Na:z)CI2 brines. These typically occur at great depths in sedimentary basins penetrated by deeper wells (e.g., Hanor, 1987). However, relatively shallow increases in chlorinity have been encountered at a number of drill sites of the DSDP in regions known to be underlain byevaporites, e.g., the Mediterranean Sea (McDuff et aI., 1978;Sayles et aI., 1972), Red Sea (Manheim et al., 1974), Gulf of Mexico (Manheim at al. , 1969), and Atlantic continental margin provinces offthe Guyanas (Waterman et a/., 1972), Namibia (Sotelo and Gieskes, 1978) and Morocco (Couture et et., 1978;
Gieskes et a/., 1980). In some of these the increase in chloride concentration is not accompanied by a corresponding sodium increase, e.g., at site 374 in the Balearic Basin of the Mediterranean Sea (McDuff et a/., 1978) indicating dissolution of other complex chlorides (Figure 19). At this site, the rare magnesium-rich mineral lueneburgite (M93(P04)2B20(OH)4 . 6H20) was also found (Muller and Fabricius, 1978). Another mechanism that has been invoked for the origin of high chlorinity brines at great depths in sedimentary basins in which evaporites are absent is membrane filtration by shales (e.g., Graf, 1982). As the discussion on membrane filtration has focussed on the deeper parts of sedimentary basins, the somewhat controversial issue (e.g., Manheim and Horn, 1968; Dickey, 1969) will not be taken up here . Hydrothermal Activity and Intrusion of Igneous Dykes and Sills (Type 9 Pore-water Profiles). Hydrothermally influenced drill sites mentioned earlier in conjunction with convection-controlled pore-water pro-
Alkal in ity (meq lol)
o o
°
o o o
2
Chlor ide (mM)
4 I
°
~.
1 1
,, \ \ \
2000
4000
I
I
,
.~. ~
n '--"
..
."'.'!!.,
f-
E 1000 2000 0°;..--..----:.::;:.::.-....---=.::;:.::....... I I I I
200
400
o I
I
I
2
3
I
I
,
\
I 1
\
• -~
I
• 200f-~
o
I I
I
I
r
Strontium (mM)
Cctcrurn (mM)
MagneSium (mu)
files may also show downward increases in chlorinity as a result of water removal in hydration reactions (Figure 20). For example, site 477 in the Guaymas Basin, Gulf of California, is located on a high heat-flow anomaly associated with a local spreading axis, for which a geothermal gradient of 88°C ·100m-1 was estimated for the sediment column leading to temperatures of 200°C at the bottom of the hole at 300 m sub-bottom depth. Distinct downward increases in Li +. K+ • and Rb + in the pore waters of this site are attributed to hydrothermal alteration of the sediments, which become depleted in alkali metals as these are released to the pore water (Gieskes et el., 1982b). The observed Ca 2+ increase and Mg2 + decrease are similar to diffusion-controlled sites, although sedimentation rates are very high (greater than 2000 rn- Ma-1 ) . These changes probably result from the release of Ca2 + and uptake of Mg2 + by the hydrothermally altered volcanics of the basaltic crust and the volcaniclastic sediment particles. The Sr2+ maximum at 140m sub-bottom depth may indicate removal of Sr2+ deeper in the hole by basalt-sea water interaction at low rockl water ratios, as observed elsewhere (Menzies and Seyfried, 1979), and this may also be responsible for the downward decreasing 87Sr/86Sr ratios. The basaltic sill at 58 to 105.5 m subbottom was apparently cool at the time of drilling. Its age of probably more than 20,000 years indicates that the thickness of the sediment cover at the time of the intrusion was probably less than 25 m at this high sedimentation-rate site. The sill has a distinct effect on the dissolved Mg2+ and ammonia con centrations, which are lowered, and the dissolved SUlphateconcentration which is raised in the pore waters of the sediment immediately above and below.
"----..~ .
•
--._'"
<,
iJ.
~---'
Figure 19 Interstitial water profiles for DSDP site 374 in the Balearic Basin of the western Mediterranean Sea indicating dissolution of evaporite minerals at 380 m sub-bottom depth (pore-waterprofile, type 8). (From Gieskes, 1983, after McDuffet aI., 1978). Lithology: I, marls; II, nannofossil ooze; III, dolomitic marls; IV, gypsum, anhydrite; V, halite.
EARLY DIAGENETIC MINERALIZATION REACTIONS IN HIGH SEDIMENTATION-RATE BASINS Processes and chemical reactions of pore-water evolution discussed so far have centered on organ ic matter decomposition reactions and the anions and neutral species produced. In order to account more fully for the observed vertical trends, consumption of the dissolved metabolites by precipitation of
Diagenesis
302
Geoscience Canada Reprint Series 4
c c
c (
authigenic minerals will be discussed next, which will also direct attention to some of the dissolved cations. The first mineralization reactions occurring in freshly deposited sediments include the formation of iron sulphides, rarely manganese sulphide, and various carbonates which form the main authigenic constituents of concretions. They are a familiar feature of organic matterrich sediments and have been studied both in modern anoxic environments and their ancient equivalents, i.e., black shales, although the number of studies directed at the solid authigenic phases, particularly from modern offshore basins, is remarkably limited compared to the wealth of pore-water data available. In part , this reflects the difficulties of detecting some of the authigenic solid phases. Pyrite and carbonates are noticeable exceptions . Other examples of early diagenetic precipitates include the zeolites and certain clay minerals such as smectites, which are characteristicforvolcanogenic and slowlydeposited pelagic sediments discussed in a previous section. Early Diagenetic Sulphide Precipitation. In the presence of soluble ferrous iron the sulphide produced in the sulphate reduction zone will immediately precipi-
tate as metastable iron monosulphides such as mackinawite (FeSO•9 0 to FeSO•9 6 with up to 10% Fe substituted for by Ni2+ and C02+) , greigite (Fe3S 4 , a cubic mineral probably with the structure of an inverse spinel) and amorphous FeS.These intermediate phases are kinetically favoured at higher supersaturation over the direct precipitation of pyrite, which has a much lower solubility product (2.4 x 10- 28 as compared to 2.8 x 10- 18 for mackinawite ; Berner, 1967; Goldhaber and Kaplan, 1974). Pyrite originating by transformation from these precursor phases displays a characteristic framboidal (raspberry-like) structure composed of tiny euhedral crystals of uniform size. This particular structure does not require the involvement of spherical micro-organisms, organic globules, a gel stage or gaseous vacuoles as sometimes assumed in the older literature, because it has been produced inorganically in experiments by reacting iron monosulphides with elemental sulphur to pyrite (Berner, 1969; Farrand, 1970). Euhedral pyrite, which often overgrows the early framboidal pyrite , appears to form somewhat later and at lower saturation levels. The process may also be pH-controlled, because at low pH the solutions may be undersatue Sulfate
Alkal i nit Y (meq l-I)
o
40 -.,..
. ..
,
.
..
80 0 10 20
200
0
1--r"T"'T...,-iir'-T-rT..:r-r:...r-rT-n"l
I
J
. . . .. . ... oe .. "' .
I~ ,
Calcium and mcqnesrurn (mill)
(mMl
I~
'" '.
rated with respect to mackinawite but supersaturated with respect to pyrite and precipitate pyrite, whereas at higher pH (above 6.5) they may be supersaturated with respect to both minerals and favour precipitation of the more soluble phase mackinawite (Goldhaber and Kaplan, 1974). These relationships are corroborated by the distribution of sulphur isotopesin pyrite-bearing carbonate concretions. A large isotopic fractionat ion effect is related to bacterial sulphate reduction. The sulphide produced is about 50% lighter than seawater sulphate (Goldhaber and Kaplan, 1980) which at present has a 534S value of + 20%0 relative to the COT (Canyon Diablo Troilite) standard. Open-system conditions asin stagnant basins where sulphate reduction starts above the sediment/water interface, should supply light sulphide with 534S values in the range of - 20%0 to - 30%0 to the sediment. Closed-system conditions, as during rapid sulphate reduction in organic-matter rich sediments on the other hand, should involve a Raleigh distillation process leading to increasingly heavier sulphides as the light sulphur is preferentially withdrawn in the early stages. The average isotopic composition, however, would be equal to seawater sulphate. The sulphur-iso-
.
,
.
..: ,. '.... -
.. .... ...... .
~
.. '
r
":
c
I'
l·-e
..............
I
(i~
?
AmmOnia 5 10 E Stront ium (mill)
01 100
200
03
02
~ ....
. :;:::..
Lithium (f£'IIl)
I
:r~::
-e-e
I
.. . . . ~
c ...
0704
o
0708 I
'I
I
..."
..........
.... .
"
1',.
~...
Potassium (mill)
Rub id ium (f£'IIl)
20
40
I
I
c (
( (
C C C C C C C C C C C C C C C C C C C C C C C C l
e
..
....
c
~
?~
(
o
300 Figure 20 Hydrothermally influenced pore-water profiles of DSDP site 477, Guaymas Basin, Gulf of California (pore-water profile, type 9; Gieskes, 1983, after Gieskes et aI., 1982b). Open circles: site 477A.
c
c c c
c c
c Early Diagenetic Pore Water/Sediment Interaction
c
'!
S''cCQCO.' ,
-I'~
.
o r:
'0
o o o l.."
'0
u
'"
Pyrite 20
o o
~:f\.!. ~j
o
Figure 21 Centre-to-rim mineralogical and isotopic variations in Liassic (Lower Jurassic) carbonate concretions from northern England. (From Coleman and Raiswell, 1981).Centre of concretion: UA at sample 5; US at sample 6.
n
'--'
I Z:S .. 5 . 7' tKo,t I Z S" 5 17 1'1011"", UA S'llillN"" US .......
CJ
topic composition of early diagenetic pyrite may then provide information on the environment, in which anoxic, organic-matter rich sediments were deposited. As Gautier (1985) found in Cretaceous shales from the Western Interior Seaway, samples with high organic matter concentrations (4-10%) display a narrow range of very light isotopic sulphur compositions in pyrite (034S = - 35%0 to - 25%0) indicating open-system sulphate reduction, in large part probably in the water column above the sediment under stagnant conditions. Samples with less than 3% organic carbon, on the other hand, show a wide range of 034S values (from - 35%0 to + 18%0) suggesting closedsystem conditions or a rapid change from initially open to closed system conditions probably under oxidizing bottom waters. A similar grouping of sulphur isotopic analyses into two main environments was reported by Schwarcz and Burnie (1973). In Lower Jurassic carbonate concretions from Yorkshire, England, Raiswell (1982) found the lightest 034S values (-23%0 to -26%0, Figure 21) associated with dispersed framboidal pyrite occurring both in the host sediment and in the concretions. In this case, the framboidal pyrite is interpreted as a phase that formed prior to concretion
CONCENTRATION (mM!L)
o
10
303
50
DIAGENETIC ZONES O2
DIAGENETIC PRODUCTS IN GAMMON SH
-.,,, :
growth under open-system conditions, not above the sediment/water interface but close to it in the sediment (Figure 22). Under these conditions isotopically light sulphur may be replenished from seawater sulphate and the alkalinity generated in the process of sulphate reduction may still be lost to the ocean bottom by diffusion, which would explain the delay of carbonate precipitation. Molecular diffusion and bioturbational irrigation are also required to explain the excess of solid sulphide found in some organic-matter rich sediments over the sulphide concentration that could be derived from the sulphate initially present in the pore water pius the sulphur contained in the organic matter (Goldhaber and Kaplan, 1980). Monosulphidic precursors of the framboids are favoured by high supersaturation associated with the initial stages of bacterial sulphate reduction, when reactive organic matter concentration is highest resulting in maximum reduction rates. In non-euxinic terrigenous sediments, iron availability from highly reactive ferric hydroxides is generally not a rate-limiting factor for sulphide precipitation at this stage (Berner, 1984; Raiswell and Berner, 1985). Euhedral pyrite, which in part nucleated on the framboidal pyrite, becomes progressively heavier from the center (034S = 13
8 -110
C .... vsPDB
,.0
FeS2·:~
o 8
sol'
In ~
Z
AEDUCEO
:::>
>a: < a: iii ~
:t
c.co 3
...
...r ...o
a' J
-3Z~
'H.
GENEAATION
e,
4
C'·%2-
F.C0 <§> 3 .'3,. -1O~
~ ,'3C
•
.'S"'.
@ .13c •. &~
- 20
-60
• Figure 22 Relationship between early diagenetic mineralization reactions in carbonate concretions and depth trend of dissolved chemical species and carbon isotopes in pore waters from which concretions precipitated. (From Gautier and Claypool, 1984). Note that the alkalinity maximum often does not coincide with the base of the sulphate zone , but occurs deeper in the methane generating zone (cf. Fig. 6).
304
- 24%0 to -14.5%0) toward the concretionmargins(c534S = -5.5%oto -2.5%0) reflecting the establishment of closedsystem conditions. Because no new light sulphate is supplied from seawater by diffusion, preferred utilization of light sulphur early in the reduction process makes the remaining sulphate reservoir progressively heavier. Isotopically heavy, large euhedral pyrite crystals which line the rims of these concretions show displacive and sectorial growth forming hopper-shapes and sometimes cone-in-cone structure (Carstens, 1986). The rare occurrence of authigenic manganese sulphide (alabandite) in modern anoxic sediments of the Baltic Sea (Baron and Debyser, 1957; Suess, 1979) has been ascribed to high dissolved Mn2 + concentrations that are typical for anoxic bottom water in local depressions of this restricted basin. Thus alabandite precipitates rather than the usual iron sulphides. Authigenic Carbonates: Calcite and Siderite (Figure 22). A prerequisite for authigenic carbonate precipitation in organic matterrich sediments is a relatively high pH which is buffered by reactions other than those of the carbonate system. If pH is controlled by the carbonate system, increased PC0 2 due to organic matter oxidation will raisethe bicarbonate rather than carbonate activity and lower pH. This appears to be the case in oxic environments, in contrast to anoxic environments. pH buffering by hydrogen sulphide and other weak acids in the sulphate reduction zone as well as hydrogen ion uptake by the reduction of manganese oxides and hydroxides in the nitrate reduction zone (and iron oxides and hydroxides in the sulphate reduction zone) will raise the pH sufficiently to cause supersaturation with respect to carbonates (Suess, 1979). This involves reactions of the type
Diagenesis
which increasesthe pH such that supersaturation with respect to carbonates and precipitation result. (The pH measured in DSDP water samples squeezed onboard the ship or in other deep subsurface samples is generally different from the in situ value due to lossesof dissolved gases (C0 2 , H 2S) and changes in temperature and pressure affecting the stability of hydrogen-bearing aqueous species (Kharaka et al., 1980). With the computer program SOLMNEQ II (Kharaka et al., 1985b) the necessary corrections for in situ conditions can be calculated). Under these circumstances calcite, dolomite, siderite, rhodochrosite and more complex iron-manganese carbonates may form in organic-matter oxidation zones (2) to (5). Fe-carbonates cannot form in the sulphate reduction zone in the presence of dissolved sulphide which competes for the dissolved iron. Likewise, experiments by Baker and Kastner (1981) suggest that dolomite formation is inhibited in the presence of substantial dissolved sulphate. Authigenic calcite, on the other hand, is among the earliest diagenetic minerals and may start to precipitate in the sulphate reduction zone together with or shortly after the first monosulphides. The most conspicuous result of early diagenetic carbonate precipitation is the formation of concretions. However, authigenic carbonates may also occur as dispersed crystals in the sediment, form overgrowths on detrital cores, or coalesce to produce laterally extensive beds. In deep-water black-shale sequences of the Cambro-Ordovician Cap-des-Rosiers Formation in Quebec, diagenetic envelopes have been observed by the author on calcilutitic turbidites and around starved ripples of clastic carbonate. These are similar to the "basal micritic limestone" below Devonian carbonate turbidites from the Rhenish Schiefergebirge described by Ma20a + 3H20 + 2e- -+ 2Me2+ + 60H-, Eder (1970). Later diagenetic fabrics and 4MeOOH + H20 + CH2 0 ..... also include replacements, particularly 4Me2+ + CO2 + 80Hby dolomite crystals. where, Me2 + represents Mn2 + or Fe2 + In concretions or laterally extensive ions. In the carbonate reduction and diagenetic beds, the carbonates charfermentation zones, it is probably the _ acteristically are pore-filling cements hydrogen consumption in the bacterial whose decreasing concentration from formation of methane in the presence centre (up to 85 or 90% by volume) to of metal ions according to a reaction rim (as little as 25%) reflect the porosity Me2 + + 2HCO a- + 8H+ +8e- ..... at the time of precipitation. For porosities estimated in this way the term CH 4 + MeCOa + 3H zO
Geoscience Canada Reprint Series 4
"minus-cement porosity" is usedin the literature . High minus-cement porosities in the range of 70 to 90% in the inner shells of many concretions attest to the common early diagenetic initiation of concretion growth. It is important that normally the growth of carbonate concretions is not displacive. Particles of the host sediment are not pushed aside by the growing carbonate crystals, in contrast to gypsum or anhydrite nodules precipitated from hypersaline brines. Grain-size of the carbonates is usually micritic. An exception are spherulitic carbonate concretions and slowly growing conein-cone cements which cause some mechanical displacement of the host sediment. In this case the carbonate content may not be used for porosity estimates. The samewarning appliesto concretions showing signs of recrystallization. Cementation in the early stages of concretion growth, on the other hand, is often not complete as shown by septarian concretions. Their cracks, which characteristically are widest in the centre and become narrower toward the concretion margins, indicate post-cementation compaction. They are shrinkage cracks showingthat cementation initially left some pore space unfilled. Near the termination of concretion growth , cementation and compaction in the outer shells reach about the same degree as in the surrounding host sediments thus forming a solid carapace around the concretion, which allowed the centre to respond to further dewatering by shrinkage. However,the total volume of septaria in concretions is generally small (less than 3%, Gautier, 1985) and so is the pore volume initially not filled by cement. Supporting evidence for the early diagenetic initiation of growth of commoncarbonate concretions comesfrom numerous observations besides carbonate concentration, including the undisturbed nature of original bedding features , the degree of preservation of the original shape of fecal pellets and fossils , clay fabric analyses but most notably isotopic analyses. Oertel and Curtis (1972) have shown that in the centre of some Carboniferous carbonate concretions from England clay particle orientation is essentially random as it would be expected for a flocculated clay suspension with 80 or 90% water content in
c c
c
c (
c
c c c
c
c c c c
c
c c c c
c c c c c c c c c c c c c C <..
e c
c c (
c (
( (
Early Diagenetic Pore Water/Sediment Interaction
305
( (
(
C' (
( (
c o o o ( o o o o o (
o (
o
o o r
r:
r-.
r:
which carbonate precipitation occurred. Toward the concretion margins a distinct trend emerges of preferred orientation of the basal planes of clay particles parallel to bedding. Preferred orientation is best developed in the host sediment. This reflects the increasing effect of burial compaction with time as the concretion continues to grow. Chemical and isotopic compositions of many carbonate concretions in black shale sequences reveal centre-to-rim variations that can be explained in terms of the organic-matter oxidation processes and related pore-water trends discussed before (Curtis et a/., 1972; Irwin et a/., 1977). In the case of a concretion starting to form in the sulphate reduction zone and continuing to grow in the carbonate reduction and fermentation zones, an early calcitic nucleus will be surrounded by a sideritic rim. Examples of this kind have been described by Gautier and Claypool (1984) and Gautier (1985) from the Upper Cretaceous Gammon Shale of the northern Great Plains of the United States. In these concretions a calcitic core with a 013C value of about -22%0 and a carbonate concentration of 85% iron-free calcite is mantled by an isotopically heavier (013C increasing outward from -10%0 to -4%0) sideritic rim with 70-80% carbonate. Oxygen isotopes are similar to those of calcite precipitated from modern seawater at temperatures above 15°C, i.e., 0180 values are between - 2%0 and - 3%0 and get progressively lighter outward in the siderite rim (to about - 7%0). Because the calcite is associated with abundant framboidal pyrite, there can be little doubt that the iron-free calcite formed in the sulphate reduction zone. More specifically, it probably formed in the lower part of the sulphate reduction zone, because the 013C of - 22%0 shows the maximum effect of organic-matter oxidation and the carbonate crystals in contact with the pyrite framboids radiate outward from the latter indicating that these formed first. Uptake of iron in the carbonate phase during siderite precipitation, on the other hand, shows that the concretion continued to grow in the methane generation zones (4) and (5) where dissolved iron becomes available and the carbonate becomes isotopically distinctly heavier because of the withdrawal of light carbon in methane. In the sulphate reduction zone,
ferrous iron concentrations in the presence of dissolved sulphide (at concentrations typically on the order of 10- 3.5 moles-Lr t) are 10- 16 moles L-1 (for pH = 7.5 and Eh = -0.245 mV), far too low for iron-carbonates to form. Siderite formation requires dissolved F&+ concentrations of at least 10-7 moles· L -1 (Curtis, 1967). Numerous examples have now been documented showing that concretions may start to grow close to the sediment/ water interface. In the case of the Gammon Shale, concretion growth was presumably complete a few tens of metres below the sea floor, perhaps as little as 10m (Gautier, 1985). In other examples there is evidence that siderite formation may have occurred at considerably greater burial depth, e.g., 800 to 1200 m sub-bottom on the East Coast continental rise off the United States (DSDP leg 93, Botz and von Rad, 1987). Interpretation of 0-, C- and S-isotopic results from the same concretion sometimes seems to lead to conflicting conclusions as in the case of a Lower Jurassic concretion described by Coleman and Raiswell (1981). 018 0 values as low as -9%0 to -10%0 require burial temperatures of 60°C or depths of 1to 2 km, which are incompatible with a shallow burial origin suggested by carbon and sulphur isotopes and other evidence. Alternative isotopic fractionation mechanisms, such as alteration of volcanic glass to clay, would shift the oxygen isotopic composition of the water, from which the carbonate precipitated, in the rightdirection, toward negativeovalues. However, it is difficult to accomplish the magnitudeoftheshift required ( -6%oto - 7%0) in the short depth interval available. If the shift is not caused by meteoric water influx in the subsurface, as suggested for late diagenetic calcite filling septarian cracks in concretions of the Jurassic Oxford Clay in England (Hudson, 1978),then additional mechanisms are required for which fractionation by carbonate precipitation itself (Irwin et a/., 1977), but also gas-hydrate formation are possibilities. As noted by Curtis and Coleman (1986), advection of connate meteoric waters from freshwater deposits into overlying marine sediments in transgressive sedimentary sequences as in the lower Westphalian coal measures of Yorkshire may also explain 180 depletion in early diagenetic calcite concretions (so-called
"coal balls"). Downward diffusion of marine pore-water and mixing with the pore waters of underlying brackish or freshwater sediments may have some effects upon concretion precipitation. Authigenic Dolomite ("deep-sea dolomite"). Dolomite and high-magnesium calcite are other common authigenic carbonates in anoxic sediments besides calcite and siderite. The burial diagenetic origin of dolomite in deep water, organic matter-rich terrigenous muds and biogenic siliceous oozes, i.e., of the socalled "deep-sea dolomites" is now an accepted fact since a number of studies were performed on DSDP material (i.e., leg 63, Pisciotto and Mahoney, 1981; leg 64, Kelts and McKenzie, 1982; leg 66, Wada et a/., 1981; leg 76, Matsumoto, 1983). Identical occurrences of bedded or nodular dolomite are exposed on land, e.g., in the Ordovician Cloridorme Formation in Quebec (Islam, 1981) or the Miocene Monterey Formation of California, for which an early diagenetic origin was already recognized by Bramlette (1946). Diatoms may be well preserved in the concretions as opal-A when the surrounding siliceous sediments have reached the opal-CT or quartz stage. These dolomites display a wide range in 013C values from -30%0 to + 30%0 which has been interpreted as evidence for their formation in the sulphate reduction, carbonate reduction and fermentation zones (Arthur at a/., 1983, Gautier, 1985, Hennessy and Knauth , 1985).Baker and Burns (1985), for example invoke dolomite precipitation in the SUlphate reduction zone and supply of magnesium by diffusion from overlying seawater in the case of dolomite-rich deep-sea sediments (containing more than 2 wt.% dolomite). Dolomite formation in sulphate-bearing environments is controversial, however (see above), and is not required to explain light carbon-isotopic compositions. As the findings of Ritger at a/. (1987)and earlier results of Russell et a/. (1967)from the Oregon and Washington continental slope show, dolomite, highmagnesium calcite and aragonite with extremely light carbon (013C ranging from - 32%0 to - 66%0) form as concretions, crusts and chimneys on the sea floor where they have been sampled by piston cores and from submersibles. They are apparently related to active sediment dewatering in the accretionary
306
Diagenesis
Geoscience Canada Reprint Series 4
c c
c prism above the subduction zone of the Oregon-Washington margin leading to upward methane migration concentrated along localized, structurally controlled pathways. The bicarbonate derived from oxidation of the methane to CO2 raises alkalinity leading to carbonate precipitation. Dolomite concretions and beds of the Monterey Formation have received special attention (Garrison et al., 1984), because of the interest in the Monterey as a major source for hydrocarbons in California. A detailed isotopic study by Kushnir and Kastner (1984)showed that nodular dolomite concretions span a wide range in c5 13 C values indicating growth over a considerable burial depth interval and contrasting remarkably with bedded dolomite with a very narrow range in isotopic values. Compton and Siever (1986) who analysed the problem of an adequate magnesium source concluded that diffusion from seawater was adequate for these dolomites if precipitation had occurred. In drill sites of the DSDP with authigenic carbonates, Ca and Mg depth profiles are clearly reaction-controlled as shown by sites 496,497, 565 and 568 to 570 (Figure 6). Ca decreases from the seawater value are rapid and concentrations remain low over most of the depths penetrated with only slight increases at the deepest levels. This contrasts markedly with diffusion-controlled sites and their pronounced downward Ca-increases. Mg, after increasing initially, is reduced considerably more rapidly than in diffusioncontrolled sites. The initial increase is probably related to ion exchange reactions with clays, which, however, may also contribute to the later decrease, in cases where this is not accounted for by carbonate precipitation. In these sites, carbonate consumption by precipitation of solid carbonates is also required in order to model successfully the EC02 and c5 13 C (of CO2 and of CH 4 ) depth profiles (Claypool et al., 1985). Mg-depletion without attendant carbonate alkalinity reduction in anoxic sediments of the Santa Barbara Basin has been attributed to uptake by clay minerals with high cation-exchange capacity (Sholkovitz, 1973). The cation-exchange capacity is increased when iron-oxide coatings are removed from clay minerals in reducing environments (Carroll, 1958).
Complex Authigenic Carbonates. The occurrence of complex authigenic carbonates rich in Mn and Fe, which show transitions to end-member rhodochrosite and siderite, has only been described from a few DSDP drill sites (e.g., Hein et aI., 1979b; Okada, 1980; Wada et ei., 1981). The isotopic composition of similar concretions in Cretaceous black shales of the Western Alps (c5 13 Cvarying between -11%oand -6%0, c5 1 8() between -9.6%0 and -4.4%0; 'fasse and Hesse, 1984) suggests that they formed at somewhat greater burial depths in the methane generation zone. This requires mobilization of manganese at those depths, which is in line with secondary maxima in Mn-depth profiles between a 100 and 200 m sub-
bottom at various DSDP drill sites discussed by Gieskes (1981). Because these show a close correlation with maxima in the dissolved silica profiles, Gieskes (1981) suggested Mn-release from opal-A dissolution as a by-product of silica diagenesis. Ion Exchange Reactions and Adsorption. None of the processes discussed so far for rapidly deposited, organic-matter rich sediments explains the depths profiles for dissolved ammonia and phosphate. As the reader may recall from a previous section, dissolved ammonia often has a maximum below the alkalinity maximum (Figure 6), while the phosphate maximum occurs in an intermediate position. Build-up of both ammonia
c (
c
400
c c ( c c c c c c c c c c c c c
Eh
c
FeS2
1000
OXIDAT~.ON
'!!o.0
WATER
800
c c
c c c
mY 0
c c
-400 WATER UNSTABLE
, ",
II -800
ACID
o
pH 2
4
GEOTHERMAL FIELD CONNATE WATER & EOGENESIS
8
" ALKALINE 8
•
10
,~It.,. Jj~......
12
~ RAINWATER. EPIGENESIS
mm1 IiWI
IIESOGENESIS
Figure 23 pH/Eh fields of natural waters. (Adapted by II. Schmidt and the author from Fairbridge , 1967, based on data from Baas Becking et aI., 1960and Garrels, 1960). Stippled area: realistic stability field of water.
c c c c e (
l
C C C
c (
c (
Early Diagenetic Pore Water/Sediment Interaction
307
c (
( (
( (
C'
o o o o o o o o o o o c o o o o o o o o o (
E7
c c
and phosphate concentrations to their maxima in the methane generation zones underlines the importance of continuing fermentation processes (e.g., deamination of proteins). There are no known ammonia-bearing minerals in anoxic sediments except the highly unstable struvite (NH 4Mg (P0 4) . 6H 20), but its occurrence in modern marine sediments has yet to be demonstrated. Decrease of ammonia from its maximum in the methane generation zone can, in part, be explained by downward diffusion (Lerman, 1977). In gas hydrate-bearing sections, the decrease is partially due to dilution from hydrate water. However, the rapid drop generally seen in organic matterrich anoxic sediments requires additional uptake by ion-exchange reactions with clays, as the effects of diffusion in dissipating concentration maxima are minor for rapidly deposited sediments. Ammonium ions are incorporated into interlayer positions of clay minerals with high layer charges, particularly illites and vermiculites as shown by Muller (1977, and references therein) for oxidized pelagic sediments of the Central Pacific. Similar reactions are likely to occur in anoxic environments leading to the fixation of dissolved ammonium in crystal lattices which will carry it to great burial depth, down to the realm of metamorphism (ltihara and Honma, 1979). Phosphate diagenesis in rapidly deposited hemipelagic sediments of the continental margins is similar to that of Table 6
(
\....
c.
Molar concentration [02 ] [H 25]
Oxic
~
Anoxic Sulphidic
< 1Q-6
1Q-6
Diagnostic authigenic minerals and sediment characteristics
< 1Q-6
Hematite, goethite, Mn02 minerals, (no reactive organ ic matter)
~
Pyrite, marcasite, rhodochrosite, alabandite; organic matter. Light 034S, no bioturbation. Trend toward heavier 0345; bioturbated.
1Q-6
Stagnant Oxic bottom waters
Methanic
GEOCHEMICAL CLASSIRCATION OF EARLY DIAGENETIC ENVIRONMENTS Natural waters have been classified from a geochemical point of view in terms of their pH/Eh-fields (Figure 23). pH is the master variable controlling the solubility
Geochemical classification of early diagenetic environments. (Modified after Berner, 1981).
Environment
Nonsulphidic Post-oxic (=Suboxic)
ammonia (as the pore-water profiles suggest), but also different in that there are authigenic mineral phases incorporating phosphate, namely apatite (Cas(P04h(OH,F) or its cryptocrystalline varieties collophane (carbonate-fluorapatite) and dahllite (carbonate-hydroxylapatite). Vivianite (Fe3(P0 4b . 8H 20) is an iron-bearing phosphate. Apatite precipitation is favoured overvivianite in the presence of fine-grained calcium carbonate as nuclei. In the absence of such nuclei, apatite formation is hindered by Mg ions and vivianite may form instead. However, due to the lack of sufficient iron, supersaturation for vivianite is usually not reached (Berner, 1980). Earliest diagenetic precipitation of collophane in shelf muds in regions of upwelling off Peru is suggested by dissolved fluoride decreases in the pore waters to 35-45 x 10- 6 M (or about half the seawater concentration) within the uppermost 30-50 cm of the seamount (Froelich et a/., 1983). Solid phosphate minerals are often difficult to detect and a significant portion of the phosphate fixation that is required to interpret the downward decreases in the profiles of dissolved phosphate probably occurs by adsorption.
< 1Q-6
< 1Q-6 < 1Q-6
< 10-6
< 1Q-6
Glauconite and other Fe2+ - Fe3+ minerals; no sulphides . (Siderite, vivianite, rhodochrosite). Minor organic matter. Siderite, vivianite, rhodochrosite. Organic matter.
of many minerals under surface and subsurface conditions. Eh exerts a controlling influence on the solubility of ions which occur in more than one valence state, e.g., Feand Mn. Theclassificationis based on the classical work of Krumbein and Garrels (1952) and showsthe field for marine connate waters (or what is the same: marine eo(dia)genesis) to cover a pH range of 6 to 9.5 and Eh levels from + 600 to about 0 m\l. The diagenesis of marine sediments (beyond the connatewater or eogenetic stage) occurs under a pH of 6 to 9 and Eh levels ranging from mildly reducing to about - 300 m\l. This classification has been useful in characterizing the stability fields of detrital and authigenic minerals in sedimentary environments(interms of pH/Eh conditions). In practice it has not been too helpful, particularly in the study of modern unlithified sediments, as pointed out by Berner (1981), mainly for two reasons. First, the pH of most subaqueous sediments,marine and non-marine,variesrelatively little, i.e., between 6 and 8, and, in fact, 90% of Berner's measurements in modern marine sediments fell into the narrow pH-range of 6.5 to 7.5. Secondly, the Eh cannot be measured correctly because many ionic species do not easily react at a platinum or gold electrode to giveoff or accept electrons. Also,because of kinetic inhibition, unstable speciespersist into forbidden Eh-domains. The measured electrode potential, therefore, does not generally agree with the values calculated from thermodynamic data. Pt and Au eletrodes, so to speak, do not "see" ions that are not electro-active. Berner (1981), therefore, proposed a new, more practical geochemical classi-: fication of sedimentary environments in terms of the concentration levels of dissolved oxygen and total sulphide (H2S plus HS-, Table 6). This scheme has the advantage of wider applicability as the two parameters used involve species which are mutually exclusive at measurable concentration levels. That is, if sulphide and dissolved oxygen are brought together at concentrations greater than 1Q-6M,andatapHof6t08,theywitl reactin a matter of minutes or hours, to form sulphate and water. Furthermore, the presence of one or the other, even in trace amounts,has vital (or lethal)effectson the population of organisms living in the sediment and consequently on the bacterially mediated mineralization reactionsdiscussed before. Obligatory anaerobicbacteria
308
cannot tolerate the presence of dissolved oxygen, even in trace amounts - they die, and sulphide minerals will be oxidized. Vice versa, obligatory aerobic bacteria cannot tolerate H2S even in trace amounts and, in its presence, oxide minerals will be converted to sulphides. Dissolved oxygen and sulphide are, therefore, ideal candidates for parameters to be used in a geochemical classification of sedimentary, specifically early diagenetic environments. In the proposed classification scheme of Berner (1981) each category has its indicator minerals (Table6) which will, at least in part, be preserved if the sediment passes from one environment to another during burial. The diagnostic value of most ofthese indicator minerals is evident from the discussion in the section on "early diagenetic mineralization reactions" and needs no further comment. The dividing line between the two main environments "oxic" and "anoxic" is the presence (i.e., at concentrations greater than 1Q-6M) or absence (i.e., presence at concentrations less than 1Q-6M) of dissolved oxygen. The anoxic environment is subdivided further into "sulphidic" and "nonsulphidic" by the same principle, i.e., presence or absence of reduced sulphur compounds. In sulphidic environments alabandite is thermodynamically stable relative to rhodochrosite only at very high H2S concentrations. Rhodochrosite, which is the more common Mn minerai in this zone, however, is also characteristic for anoxic-nonsulphidic environments and, therefore, not diagnostic. Further distinction of the nonsulphidic anoxic environments into "post-oxic" and "methanic" is based on the presence or absence of pyrite. The postoxic (= "suboxic" sensu Froelich et af., 1979,used here) environment does not contain enough organic matter for sulphate reduction to take place. Because of the lack of H2S (and O~ Fe2+ can build up to saturation with glauconite. The methanic environment, however, may follow a sulphidic anoxic stage and, may therefore, contain pyrite, but this is not a necessary condition as freshwater methanic environments may be free of pyrite. Distinction between the two nonsulphidic anoxic environments, therefore, depends critically on the interpretation of types and concentrations of authigenic minerals found in
Diagenesis
paragenesis. Suboxic environments may, by definition, contain only small quantities of reduced authigenic FeMn-minerals. Berner's classification is independent of pH and salinity and should, therefore, apply to all sediments, although it may be difficult to be applied to certain low-iron and lowanganese sediments.
at about 75°C where biogenic decomposition reactions oforganic matter give way to thermocatalytic reactions, is fundamental enough to be used as the lower boundary of the "early diagenetic regime". Further burial and attendant temperature increase lead to thermal maturation of kerogen and to the release of carbonic acid and organic acids. Thesetrigger carbonate and feldspar dissolution, generate secondary porosity in sandstones, liberate potassium andaluminumandinitiatethe smectite to illite transformation inclayey sediments, which does not seem to start to any significant extent at temperatures less than about 75°C . A whole new set of important diagenetic reactions begins at this temperature which appears to be a threshold in terms of activation energies for thermocatalytic organic matter decomposition. The presence of this natural boundary shows up clearly on diagrams displaying the temperature variation of gas composition and yield in fine-grained, organic-matter rich sediments (Figure 24). It is
DEFINITION OF "EARLY" DIAGENESIS It seems that the term "early diagenesis" as generally used in the literature is rather loosely defined and, in the sense applied by the majority of authors, refers mostly to diagenetic processesthat can be studied in piston cores. Even Berner (1980)in his masterly treatment of the subject leaves the definition open-ended by stating that "early diagenesis ... refers to changes occurring during burial to a few hundred metres where elevated temperatures are not encountered...... In the writer's opinion the natural incision, which occurs in burial diagenesis
20
Geoscience Canada Reprint Series 4
Sapropelic Source
Humic Source
(68)
Diagenesis
50 (122)
,-. ~
°'o- ' °U
100 (212)
... eu au
-
Catagenesis
:s
Q.
f-<
150 (302)
(
200 (392)
Metamorphism
•
•
Relative Yield of Gas from Organic Matter in Fine-Grained Sediments Figure 24 Hunt, 1979).
c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c e
Yield andcomposition of natural gas from different types of organic matter. (From
c C (
C C C C
(
c c
c·
(,
c c c c c· c ( c o o o ( (
o o o
o o
(
o c o o o (
Early Diagenetic Pore Water/Sediment Interaction
therefore suggested that this temperature be used to differentiate early and intermediate diagenetic regimes. Organic geochemists, of course, use this temperature as the boundary for diagenesis versus catagenesis, the latter term being equivalent to intermediate plus advanced diagenesis in the petrologist's vocabulary. Early diagenetic burial depths range from about a 100 m for high heat-flow areas (such as the Salton Sea geothermal area) to 2 or 3 (or more) km for moderate to low geothermal gradient areas. Characteristic physical conditions of the "early diagenetic regime" defined in this way are listed in Table 7. Another important caesura in terms of burial temperatures which is tentatively proposed as the boundary between the intermediate and advanced stages of diagenesis occurs at about 140°C where the transition from random mixed-layer smectite-illite to ordered mixed-layer clay minerals takes place. As the transition from diagenesis to anchimetamorphism may be placed around 200°C, the realm of diagenesis is subdivided into three about equal temperature intervals, but it is beyond the scope of this paper to give a more elaborate justification for the proposed boundaries (and to outline their variability). SUMMARY AND OUTLOOK This review of pore-water studies from modern offfshore basins shows the remarkable progress that has been achieved over the last 15 years, largely reflecting the impact of the Deep Sea Drilling Project. As a result, it appears that, for these basins, the principal trends in the chemical evolution of pore waters with progressive burial down to about 1.5km subsurface depth are more or less established today. The observed trends provide important insights into the sources and sinks for the dissolved species and into ongoing diagenetic reactions in the subsurface.
E7 l
C C.
c.
Table 7 Temperature range:
Burial depths : Porosity of clayey sediments: Vitrinite reflectance under oil: Clay minerals:
There are, nevertheless, numerous open questions, which emerge from this review and need further studies . One of these is the mass-balance problem . Can the early diagenetic mineralization products such as carbonate concretions or pyrite nodules be accounted for by relatively short-distance diffusional transport within a limited sediment volume, that is, essentially by closed system conditions, or are open-system conditions required as suggested by Compton and Siever (1986)forthe early diagenetic dolomite in the Monterey Formation? The model of Wigley et al. (1978) for partially open-system diagenetic processes and its application by Claypool and Threlkeld (1983) are encouraging steps to tackle this problem. A second problem, which is indirectly related to the first, concerns the role of advection as a major transport process for solutes. This problem has major geodynamic implications for trench-slope areas and fore-arc basins of island-arcl trench systems where active dewatering of the sediment prism is known to occur, but the flow patterns are not known in detail. The first step in successfully tackling this problem was done by ODP leg 110, and similar attempts should be undertaken elsewhere. Another big unknown concerns the effects of lithologic alternations on porewater Chemistry, for example, of interstratified mud and sand layers, or interlayered hemipelagic sediments and turbidites, or any other short-distance alternations between lithologically different sediments. The works of Colley et al. (1984), Wilson et al. (1985) and Delange (1986) show great promise in this respect. In low-sedimentation rate areas, lithologic differences between interstratified sediments in a vertical column may have no effects on porewater chemistry whatsoever, as DSDP site 495 shows, because diffusion (or convection) may dissipate any chemical gradients or changes in the pore-water
Early diagenesis.
o to 75°C (Upper limit of 75°C is limit for biogenic methane generation) 0.1 to 2-3 km 90% (sea floor) to 30% Ro less than 0.5% detrital parageneses, no newly formed illite
309
profiles . The situation may be quite different, however, for sandlmud alternations with their porosity and permeability contrasts (as shown by DSDPsites 499 and 500 , Harrison et al. , 1982). Here, the problem is a technical one, that is, of obtaining uncontaminated water samples from the sand layers. These questions are of considerable interest for the petroleum industry. None of the studies described in this article is of purely academic interest. As has become obvious, organic matter plays the key role in early diagenetic pore-water/sediment interact ion. The early diagenetic changes that affect the dispersed organic matter in the sediment have important consequences for what happens later during burial , when the sediment enters the "oil window". Organic matter that has lost most of its oil-generating capacity before reaching the "oil window" because of intense early diagenetic modifications including biogenic methane production, obviously isof less interestfrom an economic point of view than that which passes through the early diagenetic stage with less severe alteration. In the years ahead, deeper penetration by ODP drillholes is required in orderto acquire the same level of knowledge for the intermediate stage of diagenesis that now exists for early diagenesis. This will be amajoreffortwhich awaits the application of higher technology to scientific ocean drilling . ACKNOWLEDGEMENTS This work and the author's research on gas hydrates, pore waters andcarbonate concretions was funded by the Natural Sciences and Engineering Research Council of Canada . J.M. Gieskes, B. Hitchon, L. Land, E. McBride, A. Mucci and another reviewer of the Journal provided helpful critical comments at various stages on earlier versions of the manuscript. A. Kosowski typed the manuscript and R. Yatesdid the photographic and drafting work. This helpand support is gratefully acknowledged.
310
REFERENCES General Thesecover the sections including Introduction , Terminology, Sources of PoreWater Data , Diagenetic Regimes of Pore-Fluid Evolution in Modern Offshore Basins, Geochemical Classification of Early Diagenetic Environments, Definition of Early Diagenesis, Summary and Outlook). Baas Becking, L.G.M ., Kaplan , LA. and Moore, D., 1960, Limits of the natural environments in terms of pH and oxidation-reduction potentials: Journal of Geology, v. 68, p. 243-284. Berner, A.A., 1980, Early Diagenesis. A theoretical approach: Princeton University Press, Princeton Series in Geochemistry, 241 p. Berner, R.A., 1981, A new geochemical classification of sedimentary environments: Journal of Sedimentary Petrology, v. 51, p.359-365. Bischoff, J.L. , Greer, A.E. and luistro, A.O., 1970, Composition of interst itial waters of marine sediments; temperature of squeezing effect : Science, v. 167, p. 1245-1246. Carpenter, A.B., 1978,Origin and chemical evolution of brines in sedimentary basins, in Johnson, K.S. and Russell, J.A., eds ., Geology of Industrial Minerals, 13th Annual Forum: Oklahoma Geological Survey, Circular 79, p. 60-77: Case, L.C., 1955, Origin and current usage of the term , " connate water ": American Association of Petroleum Geologists, Bulletin, v. 39, p. 1879-1882. Claypool, G.E. and Kaplan, LA. , 1974, The origin and distribution of methane in marine sediments , in Kaplan, LA. , ed ., Natural Gases in Marine Sediments: Marine Science, v. 3, p. 99-140. Claypool, G.E. and Kvenvolden , K.A., 1983, Methane and other hydrocarbon gases in marine sediment: Annual Reviews of the Earth and Planetary Sciences, v. 11, p.299-327. Claypool, G.E. and Threlkeld, C.N., 1983, Anoxic diagenesis and methane generation in sediments of the Blake Outer Ridge, DSDP Site 533, leg 76, in Sheridan, R.E., Gradstein, F.M., et al., Initial Reportsofthe Deep Sea Drilling Project, v. 76, p. 391-402: United States Government Printing Office , Washington, D.C. Colley, S., Thomson, J., Wilson, T.R.S. and Higgs, N.C., 1984, Post-depositional migration of elements during diagenesis in brown clay and turbidite sequences in the North East Atlantic: Geochimica et Cosmochimica Acta, v.48, p. 1223-1235.
Diagenesis
Curtis, C.D., 1978, Possible links between sandstone diagenesis and depth-related geochemical reactions occurr ing in enclosing mudstones: Geological Society (london), Journal, v. 135, p. 107-117. Delange, G.J., 1986, Early diagenetic reactions in interbedded pelagic and turbid itic sediments in the Nares Abyssal Plain (western North Atlantic) : Consequences for the compos ition of sediment and interstitial water: Geochimica et Cosmochimica Acta, v. 50, p. 2543-2561. Devol, A.H., 1978, Bacterial oxygen uptake kinetics as related to biological processes in oxygen-deficient zones of the oceans: Deep Sea Research, v. 25, p. 137-146. Emery, K.O. and Rittenberg, S.C., 1952, Early diagenesis of California bas in sediments in relation to orig in of oil: American Association of Petroleum Geologists, Bulletin, v. 36, p. 735-806. Fairbridge, R.W., 1967, Phases of diagenesis and authigenesis, in Larsen , G. and Chilingar, G.V., eds., Diagenesis in Sediments : Elsevier, Amsterdam, Developments in Sedimentology, v. 8, p. 21-89. Fanning, K.A. and Pilson , M.E.a., 1971, Interstitial silica and pH in marine sediments: some effects of sampling procedures : Science , v. 173, p. 1225-1231. Froelich, P.N., Klinkhammer, G.P., Bender, M.L., luedtke, N.A., Heath, G.A., Cullen, D., Dauphin, P., Hammond, D., Hartman, B. and Maynard, V.,1979,Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis: Geochimica et Cosmochimica Acta, v. 43, p. 1075·1090. Garrels, R.M., 1960, Mineral Equilibria at low Temperature and Pressure: Harper and Row, New York, 254 p. Gautier, D.L., Clayton, J.L. , Leventhal, J.S. and Reddin , N.J., 1984, Origin and source rock potential of the Sharon Springs Member of the Pierre Shale, Colorado and Kansas, in Woodward, J., Meissner, F.F. and Clayton, J.L., eds., Source rocks of the Rocky Mountains: Rocky Mountain Associat ion of Geologists, p. 369-385 . Gieskes , J.M., 1975, Chemistry of interstitial waters of marine sediments : Annual Reviews of the Earth and Planetary Sciences , v. 3, p. 433-453 . Gieskes , J.M., 1981, Deep sea drilling interstitial-water studies: Implications for chemical alteration of the oceanic crust , layers I and II, in Warme, J.E., Douglas, A.G. and Winterer, E.L., eds., The Deep Sea Drilling Project: A decade of progress: Society of Economic Paleontologists and Mineralogists, Special Publication , v. 32, p. 149-167. Gieskes,J.M., 1983, The chemistry of interstitial waters of deep-sea sediments: Interpretation of Deep Sea Drilling Data, in Riley, J.P.and Chester , R., eds., Chemical Oceanography, v. 8, p. 221-269.
Geoscience Canada Reprint Series 4
Goldhaber, M.B. and Kaplan, LA. , 1974, The sulphur cycle, in Goldberg, E.D., ed., The Sea, v. 5, Marine Chemistry: John Wiley and Sons, New York, p. 569-655. Harrison, W.E., Hesse, R. and Gieskes, J.M., 1982, Relationship between sedimentary facies and interstitial water chemistry in slope, trench and Cocos plate sites from the Mid-America Trench transect, active marg in off Guatemala, Leg 67, DSDP, in VonHuene, R., Aubouin, J., et al., Initial Reports of the Deep Sea Drilling Project , v. 67, p. 603-614: United States Government Printing Office , Washington, D.C. Heath, G.A., Moore, T.C. and Dauphin, J.P., 1977, Organic carbon in deep sea sediments, in Andersen, N.A. and Malahoff, A., eds., The Fate of Fossil Fuel CO2 in the Oceans: Plenum Press, New York, p.605-625. Hunt, J.M., 1979,Petroleum Geochemistry and Geology: Freeman, San Francisco, 617 p. Johnson Ibach , L.E., 1982, Relationship between sedimentation rate and total organic carbon content in ancient marine sediments: American Association of Petroleum Geologists, Bulletin, v. 66, p.170-188. Kharaka, Y.K. and Carothers , W.W., 1988, Geochemistry of oil field waters from North Slope, Alaska, in Grye, G., ed., Geology and Exploration of the National Petroleum Reserve in Alaska, 1974 to 1982: United States Geological Survey, Professional Paper 1399, p. 551-561. Kharaka, Y.K., Hull, A.W. and W.W. Carothers, 1985a, Water-rock interactions in sedimentary basins, in Gautier, D.L., Kharaka, Y.K. and Surdam, R.C., eds., Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralogists, Short Course, v. 17, p. 79-176. Krumbein , W.O. and Garrels , A.M., 1952, Origin and classification of chemical sediments in terms of pH and oxidationreduction potentials: Journal of Geology, v. 60, p. 1-33. Li, Y. and Gregory, S., 1974, Diffusion of ions in seawater and in deep sea sediments: Geochimica et Cosmochimica Acta, v. 38, p. 703-714. Manheim, F.T. and Horn, M.K. , 1968, Composition of deeper subsurface waters along the Atlantic continental margin: Southeastern Geology, v. 9, p. 215-236. Manheim, F.T. and Sayles, F.L., 1974, Composition and origin of interstitial waters of marine sediments, based on deep sea drill cores, in Goldberg, E.D., ed., The Sea, v. 5, Marine Chemistry: John Wiley and Sons, New York, p. 527-568. Mechalas, B.J., 1974, Pathways and environmental requirements for biogenic gas production in the ocean, in Kaplan, LA., ed., Natural Gases in Marine Sediments: Marine Science, v. 3, p. 11-26.
c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c
c c c c e (
l
C (
C C C C
,c.
Early Diagenetic Pore Water/Sediment Interaction
3 11
(
C' (
C
rr
o
o o
o o
a c
c
o o o o r
r (
r
C7 (
I....
'-..,
Merino , E., 1975, Diagenesis in Tertiary sandstones from Kettleman North Dome, California. II. Interstitial solutions: distribut ion of aqueous species at 100°C and chemical relation to the diagenetic mine raology: Geochimica et Cosmochimica Acta, v. 39, p. 1629-1645. MOiler, P.J .and Suess, E., 1979,Productiv ity, sedimentation rate, and sedimentary organic matter in the oceans - I. Organic carbon preservation: Deep-Sea Research, v. 26A, p. 1347-1362. Nesbitt, H.W., 1980,Characterization of mineral-formation water interactions in Carboniferous sandstones and shales of the Illinois Sedimentary Basin: American Journal of Science, v. 280, p, 607-630. Nesbitt, H.W., 1985,A chemical equilibrium model for the Illinois Bas in format ion waters : American Journal of Science, v. 285, p. 436-458. Reeburgh, W.S., 1983, Rates of biogeochemical processes in anoxic sediments : Annual Reviews of the Earth and Planetary Sciences, v. 11, p. 269-298. Rhoads, D.C. and Morse, J,W., 1971, Evolutionary and ecologic significance of oxygen-deficient marine basins: Lethaia, v. 4, p. 413-428. Roedder, E., 1977, Fluid inclusions as tools in mineral exploration: Economic Geology, v. 72, p. 503-525. Sansone, F.J.and Martens, C.S., 1981, Methane production from acetate and associated methane fluxes from anoxic coastal sediments: Science, v. 211, p. 707-709. Sayles , F.L.and Manheim, F.T., 1975,lnterstitial solutions and diagenesis in deeply buried marine sediments: Results from the Deep Sea Drilling Project: Geochim ica et Cosmochimica Acta, v. 39, p. 103-127. Stein, R., 1986, Organic carbon and sedimentation rate - further evidence for anoxic deep-water conditions in the Cenomanian/Turonian Atlantic Ocean : Marine Geology, v. 72, p. 199-209. Waples, D.W., 1983, A reappraisal of anoxia and organic richness, with emphasis on Cretaceous of North Atlantic: American Association of Petroleum Geologists, Bulletin , v. 67, p. 963-978. White , D.E., 1957, Magmatic, connate, and metamorphic waters: Geological Society of America , Bulletin, v. 68, p. 1659-1682. White , D.E.. 1965,Saline waters of sedimentary rocks, in Young, A. and Galley, J.A., eds., Fluids inSubsurface Environments: American Association of Petroleum Geologists, Memoir 4, p. 342-366. Wigley, T.M., Plummer, L.N. and Pearson, F.J., Jr., 1978, Mass transfer and carbon isotope evolution in natural water systems : Geochimica et Cosmochimica Acta , v. 42, p. 1117-1139.
Wilson, T.R.S ., Thomson , J., Colley, S., Hydes, D.J., Higgs, N.C. and Sorensen, J., 1985, Early organic diagenesis: The significance of progressive subsurface oxidation fronts in pelagic sediments: Geochimica et Cosmochimica Acta, v. 49, p. 811-822. Zeikus, J.G. and Wolfe, R.S., 1972, Methanobacterium thermoautotrophicum sp. n., an anaerobic, autotrophic, extreme thermophile: Journal of Bacteriology, v. 109, p. 707-713.
Low to Intermediate Sedimentation-rate Basins Anderson, R.N., Hobart, M.A. and Langseth, M.G., 1979, Geothermal convection through oceanic crust and sediments of the Indian Ocean: Sc ience , v.204, p.828-832. Anderson, R.N., Langseth, M.G. and Sclater, J.G., 1977, The mechanics of heat transfer through the floor of the Indian Ocean: Journal of Geophysical Research, v. 82, p.3391-3409. Baker, P.A., Gieskes, J.M. and Elderfield, H., 1982, Diagenesis of carbonate in deepsea sediments - Evidence from Sr/Ca ratios and interstitial dissolved Sr2 + data: Journal of Sedimentary Petrology, v. 52, p.71-82. Bath,A. andShackleton,N.,1984,Oxygenand hydrogen isotope studies in squeezed pore waters, Deep Sea Drilling Project, Leg 74, hole 525B: Evidence for mid-Miocene ocean isotopic change, in Moore, T.C., Jr., Rabinowitz, P.O., et a/., Initial Reports of the Deep Sea Drilling Project, v. 74, p. 697-699: United States Government Printing Office, Washington, D.C. Bender, M.L., Hudson, A., Graham, D.W., Barnes, R.O., Leinen , M., and Kahn, D., 1986, Diagenesis and convection reflected in pore water chemistry on the western flank of the East Pacific Rise, 20 degrees south: Earth and Planetary Science Letters, v. 76, p. 71-83. Burke, W.H., Denison, R.E., Hetherington, E.A., Koepnick, R.B., Nelson, H.F. and Otto, J.B. , 1982, Variation of seawater 87Sr/86Sr throughout Phanerozoic time: Geology, v. 10, p. 516-519. Chamley, H. and Millot, G., 1972, Neoformation de montmorillonite a partir de diatomees et de cendres dans Ie sediments marins de Santorin (Mediterranee orientale): Comptes Rendus, Academie des Sciences, Ser. 0, t. 272, p. 1132-1134. Colley, S., Thomson, J., Wilson, T.R.S. and Higgs, N.C., 1984, Post-depositional migration of elements during diagenesis in brown clay and turbidite sequences in the North East Atlantic: Geochimica et Cosmochimica Acta, v. 48, p. 1223-1235.
DeLange, G.J. and Rispens, F.B., 1986, Indication of a diagenetically induced precipitate of an Fe-Si mineral in sediment from the Nares Abyssal Plain, western North Atlantic : Marine Geology, v. 73, p. 85·97. Donnelly, T.W. and Merrill , L., 1977, The scavenging of magnesium and other chemical species by biogenic opal in deep-sea sediments: Chemical Geology, v. 19, p. 167-186. Edmond, J.M., Measures , C., McDuff, R.E., Chan, L.H., Collier, R., Grant, B., Gordon, L.1. and Corliss , J.B., 1979, Ridge crest hydrothermal activity and the balances ofthe major and minor elements in the oceans : The Galapagos data: Earth and Planetary Science Letters, v. 46, p. 1·18. Elderfield, H., Gieskes, J.M., Baker, P.A., Oldfield , R.K., Hawkesworth, C.J. and Miller, R., 1982, 87Sr/86Sr and 180 /160 ratios, interstitial water chemistry and diagenes is in deep-sea carbonate sediments of the Ontong Java Plateau: Geochimica et Cosmochimica Acta, v. 46, p. 2259-2268 . Fehn, U., 1986, The evolution of low-temperature convection cells near spreading centers: A mechanism for the formation of the Galapagos Mounds and similar manganese deposits: Economic Geology, v. 81, p, 1396-1407. Gieskes, J.M., Elderfield, H. and Palmer, M.R., 1986, Strontium and its isotopic composition in interstitial waters of marine carbonate sediments : Earth and Planetary Science Letters, v. 77,p. 229235. Gieskes, J.M. and Johnston , K., 1984,Interstitial water studies, Leg 81, in Roberts, D.G., Schnitker, D., et a/., Initial Reports of the Deep Sea Drilling Project, v. 81, p. 829-836: United States Government Printing Office, Washington, D.C. Gieskes,J.M .,Johnston , K., Boehm, M. and Nohara, M., 1984, Interstitial water studies, Deep Sea Drilling Project, Leg 75, in Hay, W.W., Sibuet, J.-C., et a/., Initial Reports ofthe Deep Sea Drilling Project, v. 75, p. 959-963: United States Government Printing Office, Washington, D.C. Gieskes, J.M . and Lawrence, J.R., 1981, Alteration of volcanic matter in deep sea sediments : evidence from the chemical compos ition of interstitial waters from deep sea drilling cores: Geochimica et CosmochimicaActa, v. 45, p. 1687-1703. Hein , J.R., Yeh, H.W. and Alexander, E., 1979a, Origin of iron-rich montmorillonite from manganese nodule belt of the north equatorial Pacific : Clays and Clay Minerals, v. 27, p. 185-194.
( 312
Helm, A., 1985, Mineralogy and diagenesis of slope sediments offshore Guatemala and Costa Aica, Deep Sea Drilling Project leg 84, in Von Huene, R., Aubouin, J., et a/., Intitial Aeports of the Deep Sea Drilling Project , v.85, p.571-594 : United States Government Printing Office, Washington, D.C. Johnson, T.C., 1976, Biogenic opal preservation in pelagic sediments of a small area in eastern tropical Pacific: Geological Society of America, Bulletin, v. 87, p. 1273-1282. Kastner, M., 1981, Authigenic silicates in deep-sea sediments: formation and diagenesis, in Emiliani, C., ed.,The Sea, v. 7, The Oceanic Lithosphere: John Wiley and Sons, New York, p. 915-980. Kastner, M. and Stonecipher, SA, 1978, Zeolites in pelagic sediments of the Atlantic, Pacific and Indian Oceans, in Sand, L.B. and Mumpton, FA, eds., Natural Zeolites: Occurrence, properties, use: Pergamon Press, New York, p. 199-220. Langseth, M.G., Hyndman, A.D., Becker, K., Hickman, S. and Salisbury, M.H., 1984, The hydrogeologic regime of isolated sediment ponds in mid-ocean ridges , in Hyndman, R.D ., Salisbury, M.H . et el., Initial Aeports of the Deep Sea DrillingProject, v. 78B, p. 825-837: United States Government Printing Office , Washington, D.C. Lawrence,J.R. and Gieskes, J.M., 1981,Constraints on water transport and alteration in the oceanic crust from the isotopic composition of pore water: Journal of Geophysical Research, v. 86 , p. 7924-7934. Lawrence, J .R ., Gieskes, J.M. and Broecker, W.S., 1975, Oxygen isotope and cation composition of DSDP pore waters and the alteration of Layer II basalts: Earth and Planetary Science Letters, v. 27, p. 1-10. Lerman , A., 1975, Maintenance of steady state in oceanic sediments: American Journal of Science, v. 275, p. 609-635. Lerman, A., 1977, Migrational processes and chemical reactions in interstitial waters, in Goldberg, E.D., McCave, LN., O'Brien, J.J. and Steele, J.H. , eds., The Sea, v. 6, Marine Modeling : John Wiley and Sons, New York, p. 695-738. Lerman, A. and Lietzke, T.A., 1977, Fluxes in a growing sediment layer: American Journal of Science, v. 277, p. 25-37. Li, Y. and Gregory, S., 1974, Diffusion of ions in seawater and in deep sea sediments : Geochimica et Cosmochimica Acta, v. 38, p. 703-714. MacKenzie , F.T. and Garrels , R.M., 1966, Silica-bicarbonate balance in the ocean and early diagenesis: Journal of Sedimentary Petrology, v. 36, p. 1075-1084.
Diagenesis
Mackin, J.E. and Aller, R.C., 1984, Dissolved AI in sediments of the East China Sea: implications for auth igenic mineral formation : Geochimica et Cosmochimica Acta , v. 48, p. 281-297. McDuff, R.E. , 1978, Conservative behaviour of calcium and magnesium in interstitial waters of marine sed iments: Identification and interpretation, unpublished Ph.D. thesis, Sc ripps Institution of Oceanography, University of California, San Diego, 183 p. McDuff, A.E., 1981, Major cation gradients in DSDP interstitial waters : the role of diffusive exchange between sea water and upper oceanic crust: Geoch imica et CosmochimicaActa, v. 45, p. 1705-1713. McDuff, A.E., 1984, The chemistry of interstitial waters from the upper ocean crust , site 395, Deep Sea Drilling Project Leg 78-B, in Hyndman , R.D., Salisbury, M.H. et et., Initial Reports of the Deep Sea Drilling Project, v. 78B, p. 795-799: United States Government Printing Office, Washington, D.C. McDuff, R.E., 1985, The chemistry of interstitial waters, Deep Sea Drilling Project Leg 86, in Heath, R.G., Burckle , L.H. et a/., Initial Aeports of the Deep Sea Drilling Project, v. 86, p. 675-687: United States Government Printing Office, Washington, D.C. McDuff, A.E. and Gieskes , J.M. , 1976, Calcium and magnesium profiles in DSDP interstitial waters: diffusion or reaction? : Earth and Planetary Science Letters , v. 33, p. 1-10. Moberly, A., Jr., Kimura, H.S. and McCoy, FW. , Jr., 1968, Authigenic marine phyllosilicates near Hawaii: Geological Society of America, BUlletin , v. 79 , p. 1449-1460. Muller, P.J . and Mangini, A., 1980, Organic carbon decomposition rates in sediments of the Pacific manganese nodule belt dated by 230Th and 231Pa: Earth and Planetary Science Letters, v. 51, p.94-114. Palmer, M.A. and Elderfield, H., 1985, Sr isotope composition of sea water over the past 75 Myr: Nature, v.314, p.526-528. Savin, S.M., 1980, Oxygen and hydrogen effects in low-temperature mineralwater interactions, in Fritz, P. and Fontes, J.C., eds., Handbook of Environmental Isotope Geochemistry, v. 1: Elsevier, Amsterdam, p. 283-327. Sayles, F.L. and Manheim, F.T., 1975, Interstitial solutions and diagenesis in deeply buried marine sediments : Results from the Deep Sea Drilling Project: Geochimica et Cosmochimica Acta, v. 39, p. 103-127.
Geoscience Canada Reprint Series 4
Scholle, PA , Arthur, MA and Ekdale, AA , 1983, Pelagic environment, in Scholle, P.A., Bebout, D.G. and Moore, C.H., eds., Carbonate Deposit ional Environments: American Association of Petroleum Geologists , Memoir33, p. 619-691. Shackleton , N.J., Hall, M.A. and Boersma, A., 1984, Oxygen and carbon isotope data from Leg 74 foraminifera, in Moore, T.C., Jr., Rabinowitz, P.O., eta/., Initial Aeports of the Deep Sea Drilling Project, v. 74, p. 599-612: United States Government Pr inting Office, Wash ington, D.C. Stoffyn-Egli, P., 1982, Dissolved aluminium in interstitial waters of recent sediments from the North Atlantic Ocean : Geochimica et Cosmochimica Acta, v. 46, p. 1345-1352. Stout, P.M., 1985, Chemical diagenesis of pelagic biogenic sediments from the equatorial Pacific, unpublished Ph.D. thesis, Scripps Institution of Oceanography, University of California, San Diego, 220 p. Von Damm, K.L. and Bischoff, J.L., 1987, Chemistry of hydrothermal solutions from the southern Juan de Fuca Aidge: Journal of Geophysical Research, v. 92, p. 11334-11346. Von Damm, K.L., Edmond, J.M., Grant, B., Measures , C.L , Walden, B. and Weiss, A.F., 1985a, Chemistry of submarine hydrothermal solutions at 21°N , East Pacific Rise: Geochimica et Cosmochimica Acta, v. 49, p. 2197-2220. Von Damm, K.L., Edmond, J.M., Measures, C.L and Grant, B., 1985b, Chemistry of submarine hydrothermal solutions at Guaymas Basin, Gulf of California: Geochimica et Cosmochimica Acta, v. 49, p. 2221-2237. Wilson , T.A.S., Thomson, J., Colley, S., Hydes, D.J., Higgs, N.C. and Sorensen, J., 1985, Early organic diagenesis : The significance of progressive subsurface oxidation fronts in pelagic sediments: Geochimica et Cosmochimica Acta, v. 49, p. 811-822. High Sedimentation-rate Basins
Berner, R.A., 1980, Early diagenes is. A theoretical approach: Princeton Series in Geochemistry: Princeton University Press, 241 p. Claypool , G.E. and Kaplan , I.A. , 1974, The origin and distribution of methane in marine sediments, in Kaplan, I.A., ed., Natural Gases in Marine Sediments: Marine Science , v. 3, p. 99-140. Claypool , G.E. and Kvenvolden, K.A., 1983, Methane and other hydrocarbon gases in marine sediment: Annual Reviews of the Earth and Planetary Sciences, v. 11, p.299-327.
C
C C C C C C C C C C C C C C
C C
c
c
c
c c
c
c c
c c
c
c c
c c
e c (
c (
C C
C
r Early Diagenetic Pore Water/Sediment Interaction
r
r:
r:
[
C r
r
Claypool , G.E. and Threlkeld, C.N., 1983, Anoxic diagenesis and methane generation in sediments of the Blake Outer Ridge , DSDP Site 533, Leg 76, in Sheridan, R.E., Gradstein, EM ., et a/., Initial Reports of the Deep Sea Drill ing Project, v. 76, p. 391-402: United States Government Printing Office , Washington , D.C. Claypool , G.E., Threlkeld, C.N., Mankiewicz, P.N., Arthur, M.A. and Anderson , T.E , 1985, Isotopic composition of interstitial fluids and origin of methane in slope sediment of the Middle America Trench, Deep Sea Drilling Project Leg 84, in Von Huene, R., Aubouin , J., et a/., Initial Reports of the Deep Sea Drilling Project, v. 84 , p. 683-691: United States Government Printing Office, Washington , D.C. Clayton, R.N. , Friedman, I., Graf, D.L., Mayeda, T.K., Meents, W.F. and Shimp, N.E, 1966, The origin of saline formation waters: 1. Isotopic composition: Journal of Geophysical Research, v. 71, p. 38693882. Curtis, C.D., Petrowski, C. and Oertel, G., 1972, Stable carbon isotope ratios within carbonate concretions: a clue to place and time of formation: Nature, v. 235 , p.98-100. Davidson , D.W., Leaist, D.G. and Hesse, R., 1983, Oxygen-18 enrichment in the water of a clathrate hydrate: Geochimica et Cosmochimica Acta, v. 47, p. 2293-2295. Froelich , P.N., Klinkhammer, G.P., Bender, M.L. , Luedtke, N.A., Heath, G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B. and Maynard, v., 1979,Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis: Geochimica et Cosmochimica Acta, v. 43, p. 1075-1090. Gieskes,J.M., Elderfield , H., Lawrence,J.R., Johnson, J., Meyers, B. and Campbell , A., 1982a, Geochemistry of interstitial waters and sediments, Leg 64, Gulf of California, in Curray, J.R., Moore, D.G., et a/.,Initial Reports ofthe Deep Sea Drilling Project, v. 64, part 2, p. 675-694: United StatesGovernmentPrinting Office, Washington, D.C. Gieskes, J.M., Johnston, K. and Boehm, M., 1985, Interstitial water studies, Leg 66, in Huene, R., Aubouin, J., et a/., Initial Reports of the Deep Sea Drilling Project, v. 84, p. 961-967: United States Government Printing Office , Washington, D.C. Goldhaber, M.B. and Kaplan, LA., 1974, The sulphur cycle, in Goldberg, E.D ., ed., The Sea, v. 5, Marine Chemistry: John Wiley and Sons, New York, p. 569-655 . Goldhaber, M.B. and Kaplan, LA., 1980, Mechanisms of sulfur incorporation and isotope fractionation during early diagenesis in sediments of the Gulf of California: Marine Chemistry, v. 9, p. 95-143.
Harrison , W.E., Hesse, R. and Gieskes, J.M ., 1982,Relationship between sedimentary facies and interstitial water chemistry in slope, trench and Cocos plate sites from the Mid-America Trench transect , active margin off Guatemala . Leg 67, DSDP, in Von Huene, R., Aubouin, J., et a/., Initial Reports of the Deep Sea Drilling Project, v. 67, p. 603-614: United States Government Printing Office, Washington, D.C. Hesse, R., 1988, Pore-water anomalies in gas-hydrate bearing sediments of the deeper cont inental margins: Journal of Inclusion Phenomena, in press. Hesse, R. and Harrison , W.E., 1981, Gashydrates (clathrates) causing pore-water freshening and oxygen isotope fractionation in deep-water sedimentary sections of the terrigenous continental margins: Earth and Planetary Science Letters, v. 55, p. 453-462. Hesse, R., Lebel , J. and Gieskes, J.M ., 1985, Interstitial water chemistry of gas hydrate bearing sections on the Middle-America Trench Slope Deep Sea Drilling Project, Leg 84, in Von Huene, R., Aubouin, J., et a/., Initial Reports of the Deep Sea Drilling Project, v. 84 , p. 727·737: United States Government Pr inting Office, Washington , D.C. Jenden, P.O.and Gieskes, J.M., 1983,Chemical and isotopic composition of interstitial water from Deep Sea Drilling Project sites 533 and 534 , in Sheridan , R.E., Gradstein, F.M., et a/., Initial Reports of the Deep Sea Drilling Project, v. 76, p.453-461: United States Government Printing Office, Washington, D.C. Jenden, P.O. and Kaplan, LA. , 1986, Comparison of microbial gases from the Middle America Trench and Scripps Submarine Canyon: Implications for the origin of natural gas: Applied Geochemistry, v. 1, p.631·646. Manheim, F.T., 1967, Evidence for submarine discharge of water on the Atlantic continental slope of the southern United States, and suggestions for further search: New York Academy of Sciences, Transactions, Series II, v. 29, p. 839-853. Manheim, F:T. and Chan, K.M., 1974, Interstitial waters of Black Sea sediments : New data and review , in Degens, E.T. and Ross, D.A., eds., The Black Sea - geology, chemistry and biology: American Association of Petroleum Geologists, Memoir 20, p. 155-180. Manheim , F.T. and SChug, D.M., 1978, Interstitial waters of Black Sea cores, in Ross, D.A., Neprochnov, Y.P., et el., Initial Reports of the Deep Sea Drilling Project, v.42, part II, p.637-651: United States Government Printing Office, Washington, D.C.
313
O'Neil , J.R., 1968, Hydrogen and oxygen isotope fract ionation between ice and water : Journal of Physical Chemistry, v. 72, p. 3683-3684 . Rosenfeld, W.O. and Silverman , S.A., 1959, Carbon isotope fractionat ion in bacterial production of methane: Science, v. 130, p. 1658-1659. Sackett, W.M. and Thompson, R.R., 1963, Isotopic organic carbon compositions of recent continental shelf derived clastic sediments of the Eastern Gulf Coast, Gulf of Mexico: American Association of Petroleum Geologists, Bulletin, v. 47, p.525-531. Schoell, M., 1983, Genetic character ization of natural gases: American Association of Petroleum Geologists, Bulletin, v.67, p. 2225-2238. Shipley, T.H., Houston, M.H., Buftler, A.T., Shaub, EJ ., McMillenik, J., Ladd, J.W. and Worzel, J.L., 1979,Seismic evidence for widespread possible gas hydrate horizons on continental slopes and rises: American Association of Petroleum Geologists, Bulletin, v. 63, p. 2204-2213. Suess, E., 1976, Porenlosunqen mariner Sedimente . Ihre chemische Zusammensetzung als Ausdruck friihdiagenetischer Vorgange, Habiliationsschrift, Universitat Kiel, 193 p. Tucholke, B., Bryan, G.M. and Ewing, J.I., 1977, Gas-hydrate horizons detected in seismic profiler data from the western North Atlantic: American Association of Petroleum Geologists, Bulletin , v. 61, p.698-707. Lateral Pore-water Flow Along Fault Zones
Bray, C.J. and Karig, D.E., 1986, Porosity of sediments in accretionary prisms and some implications for dewatering processes: Journal of Geophysical Research, v. 90, p. 768-778. Breen, N.A., Silver, E.A. and Hussong, D.M., 1986, Structural styles of an accretionary wedge south ofthe island of Sumba,lndonesia revealed by Seamarc II side-scan sonar: Geological Society of America, Bullet in, v. 97, p. 1250-1261. Carson, B., 1977, Tectonically induced deformation of deep sea sediments off Washington and northern Oregon : mechanical consolidation: Marine Geology, v. 24, p.289·307. Carson, B., Jennwei, Y. and Myers, P.B., 1974, Initial deep-sea sediment deformation at the base of the Washington continental slope: Geology, v. 2, p. 561-564.
( 314
Kulm, L.D., Suess, E., Moore, J.C., Carson, B., Lewis, B.T., Ritger, S.D., Kadko, D.C., Thornburg, T.M., Embley, R.W., Rugh, W.D., Massoth, G.J., Langseth, M.G., Cochrane, G.R. and Scamman, R.L., 1986,Oregon subduction zone: venting, fauna and carbonates: Science, v. 231, p.561-566. Laubier, L., Ohta, S. and Sibuet, M., 1986, Decouverte de cornmunautes animates profondes durant la campagne francojaponaise KAIKO des plonqses dans les fosses de subduction autour du Japon: Comptes Rendues de L'Academie des Sciences Francals, Paris, Series 3, v. 303, p. 25-29. ODP Leg 110 Scientific Party, 1987, Expulsion offluids from depth along a subduction-zone decollement horizon: Nature, v. 326, p. 785-788. Ohta, S. and Laubier, L., 1987, Deep biological communities in the subduction zone of Japan : observations of bottom photographs taken during Nautile dives in the KAIKO project: Earth and Planetary Science Letters, v. 83, p 329-342. Westbrook, G.K. and Smith, M.J., 1983, Long decollernents and mud volcanoes: evidence from the Barbados Ridge complex for the role of high porewater pressures in the development of an accretionary complex: Geology, v. 11, p.279-283.
Evaporite Dissolution and Hydrothermal Activity Couture, A., Miller, A.S. and Gieskes, J.M., 1978, Interstitial water and mineralogical studies, Leg 41, in Lancelot, V., Seibold, E. et el., Initial Reports of the Deep Sea Drilling Project, v. 41, p. 907914:United States Government Printing Office, WaShington, D.C. Dickey, P.A., 1969, Increasing concentration of subsurface brines with depth: Chemical Geology, v. 4, p. 361-370. Gieskes, J.M., 1983,The chemistry of interstitial waters of deep-sea sediments: Interpretation of Deep Sea Drilling Data, in Riley,J.P. and Chester,A., eds .: Chemical Oceanography, v. 8, p. 221-269. Gieskes, J.M., Graham, D. and Ellis, A., 1980, Interstitial-water studies, Deep Sea Drilling Project sites 415and 416, in Lancelot, V., Winterer, E.L. et si., Initial Reports ofthe Deep Sea Drilling Project, v. 50, p. 691-693: United States Government Printing Office, Washington , D.C. Gieskes , J.M ., Kastner, M., Einsele, G., Kelts, K. and Niemitz, J., 1982b, Hydrothermal activity in the Guaymas Basin, Gulf of California: A synthesis, in Curray, J.R., Moore, D.G. eta/., Initial Reports of the Deep Sea Drilling Project, v. 64, p. 1159-1167: United States Government Printing Office, WaShington, D.C.
Diagenesis
Graf, D.L., 1982, Chemical osmosis, reverse chemical osmosis, and the origin of subsurface brines: Geochimica et Cosmochimica Acta, v.46, p. 14311448. Hanor, J.S., 1987, Kilometre-scale thermohaline overturn of pore waters in the Louisiana Gulf Coast: Nature, v. 327, p.501-503. Manheim, ET. and Horn, M.K., 1968, Composition of deeper subsurface waters along the Atlantic continental margin: Southeastern Geology, v. 9, p. 215-236. Manheim, ET., Sayles, EL., and Friedman, I., 1969, Interstitial water studies on small core samples, Deep Sea Drilling Project, Leg 1,in Ewing, M., Worzel, J.L. et a/., Initial Reports of the Deep Sea Drilling Project, v. 1, p. 403-407: United States Government Printing Office, Washington, D.C. Manheim, F.T., Waterman, L.S., Woo, C.C. and Sayles, F.L., 1974, Interstitial water studies on small core samples, leg 23 (Red Sea), in Whitmarsh, R.B., Ross, D.A. at a/., Initial Reports of the Deep Sea Drilling Project, v. 23, p. 955-967: United States Government Printing Office, Washington, D.C. McDuff, R.E., Gieskes, J.M. and Lawrence, J.R., 1978,Interstitial water studies, Leg 42A, in HsO, K., Montadert, L. et aI., Initial Reports of the Deep Sea Drilling Project, v. 42, part 1, p. 561-568: United States Government Printing Office, Washington, D.C. Menzies, M. and Seyfried, W.E., 1979, Basalt - seawater interaction: Trace element and strontium isotope variations in experimentally altered glassy basalt: Earth and Planetary Science Letters, v. 44, p. 463-472. MOiler, J. and Fabricius, E, 1978, Luneburgite in Upper Miocene sediments of the eastern Mediterranean Sea, in HsO, K., Montadert, L. et a/., Initial Reports ofthe Deep Sea Drilling Project, v. 42, p. 661-664: United States Government Printing Office, Washington, D.C. Sayles, EL., Waterman, L.S. and Manheim , ET., 1972, Interstitial water studies on small core samples from the Mediterranean Sea, in Ryan, W.B.E, HsO, K.J. et a/., Initial Reports of the Deep Sea Drilling Project, v. 13, p, 801-808: United States Government Printing Office, Washington, D.C. Sotelo, V.and Gieskes , J.M., 1978, Interstitial water studies, Leg 40: Shipboard studies, in Bolli , H.M ., Ryan, W.B.E et a/., Initial Reports of the Deep Sea Drilling Project, v. 40, p. 549-554: United States Government Printing Office, Washington, D.C.
Geoscience Canada Reprint Series 4
Waterman , L.S., Sayles, EL. and Manheim, F.T., 1972, Interstitial water samples on small core samples, Leg 14, in Hayes, D.E., Pimm, A.C. et el., Initial Reports of the Deep Sea Drilling Project, v. 14, p. 753-762: United States Government Printing Office , Washington, D.C.
Early Diagenetic Mineralization Reactions in High Sedimentation-rate Basins Arthur, M.A., Anderson, T.E, Kaplan, LA., Veizer, J. and Land, L.S., 1983, Stable Isotopes in Sedimentary Geology: Society of Economic Paleontologists and Mineralogists, Short Course Lecture Notes, v. 10, p. 1-151. Baker, P.A. and Burns, S,J., 1985, Occurrence and formation of dolomite in organic-rich continental margin sediments : American Association of Petroleum Geologists, Bulletin, v. 69, p. 19171930. Baker, P.A. and Kastner, M., 1981, Constraints on the formation of sedimentary dolomite: Science, v. 213, p. 214-216. Baron, G. and Debyser, I., 1957, Sur la presence dans des vases organiques de la mer Baltique de sulfure manganeuxhexagonal: Comptes Rendues de l'Academie des Sciences Franyais, v. 245, p. 1148-1150. Berner, R.A., 1967, Thermodynamic stability of sedimentary iron sulfides: American Journal of Science, v. 265, p.773-785. Berner, A.A., 1969, The synthesis of tramboidal pyrite: Economic Geology, v. 64, p.383-384. Berner, A.A., 1980,Early Diagenesis. A theoretical approach: Princeton University Press, Princeton Series in Geochemistry, 241 p. Berner, A.A., 1984, Sedimentary pyrite formation: an update: Geochimica et Cosmochimica Acta , v. 48, p. 605-615. Beitz, A. and v. Rad, U., 1987, Authigenic FeMn carbonates in the Cretaceous and Tertiary continental rise sediments of DSDP Site 603 off the eastern U.S.A., in Van Hinte, J.E., Wise, S.W., Jr. er a/., Initial Reports of the Deep Sea Drilling Project, v.93, p. 1061-1077: United States Government Printing Office, Washington, D.C. Bramlette, M.N., 1946, The Monterey Formation of California and the origin of its siliceous rocks: United States Geological Survey, Professional Paper 212, 55 p. Carroll, D., 1958,The role of clay minerals in the transportation of iron: Geochimica et Cosmochimica Acta, v. 14, p. 1·27. Carstens, H., 1986, Displacive growth of authigenic pyrite: Journal of Sedimentary Petrology, v. 56, p. 252-257.
C
C C C C (
C
C C C
C C C C C
C C
c c c c c c c c
c
c
c c C l
(
c
c c (
c Early Diagenetic Pore Water/Sed iment Interaction
315
r> '-
c c
c) r>
l.
(
C
o o o
c
o
o o o o o o
o o o
o (;
b
Claypool , G.E., Threlkeld, C.N ., Mankiewicz, P.N., Arthur, M.A. and Anderson, T.F., 1985, Isotopic composition of interstitial fluids and origin of methane in slope sediment of the Middle America Trench, Deep Sea Drilling Project Leg 84 , in Von Huene, R., Aubouin, J. et el., Initial Reports of the Deep Sea Drilling Project , v. 84, p. 683-691: United States Government Printing Office, Washington, D.C. Coleman , M.L. and Raiswell , R., 1981, Carbon, oxygen and sulfur isotope variations in concretions from the Upper Lias of N.E. England: Geochimica et Cosmochimica Acta, v. 45, p. 329-340. Compton, J.S. and Siever, R., 1986, Diffusion and mass balance of Mg dur ing early dolomite formation, Monterey Formation : Geochim ica et Cosmochimica Acta, v. 50, p. 125-135. Curt is, C.D., 1967, Diagenetic iron minerals in some British Carboniferous sediments: Geochimica et Cosmochimica Acta , v. 31, p. 2109-2123. Curt is, C.D. and Coleman , M.L., 1986, Controls on the precipitation of early diagenetic calcite, dolomite and siderite concretions in complex depositional sequences, in Gautier, D.L., ed., Roles of Organic Matter in Sediment Diagenes is: Society of Econom ic Paleontologists and Mineralogists, Special Publication no. 38, p. 23-33. Curtis, C.D., Petrowski , C. and Oertel , G., 1972, Stable carbon isotope ratios within carbonate concretions: a clue to place and time of format ion: Nature, v. 235, p.98-100 . DeLange, G.J., 1986, Early diagenetic reactions in interbedded pelagic and turbiditic sediments in the Nares Abyssal Plain (western North Atlantic): Consequences for the composition of sediment and interstitial water: Geochi mica et Cosmochimica Acta, v. 50, p. 2543-2561, Eder, F.W., 1970,Genese riff-naher DetritusKalke bei Balve im Rheinischen Schiefergebirge (Garbecker Kalk): Verhandlungen der Geologischen Bundes-Anstalt, Vienna, Jg. 1970, p. 551-569. Farrand, M., 1970, Framboidal sulfides precipitated synthetically: Mineralium Deposita, v. 5, p. 237-247. Froelich, P.N., Kim, K.H. , Jahnke, R., Burnett, W.C., Soutar, A. and Deakin , M., 1983, Pore water fluoride in Peru continental margin sediments : Uptake from seawater: Geochimica et Cosmochimica Acta, v. 47, p. 1605-1612. Garrison, R.E., Kastner, M. and Zenger, D.H., 1984, eds., Dolomites of the Monterey Formation and organic-rich units: Society of Economic Paleontologists and Mineralogists, Pacific Section, Special Publication No. 41, 215 p.
Gautier, D.L., 1985, Interpretation of early diagenesis in ancient marine sediments, in Gautier, D.L., Kharaka, Y.K. and Surdam, R.C., Relationship of Organic Matter and Mineral Diagenesis: Society of Economic Paleontologists and Mineralog ists, Short Course Lecture Notes, v. 17, p. 6-78. Gautier, D.L. and Claypool, G.E., 1984, Interpretation of methanic diagenesis in ancient sediments by analogy with processes in modern diagenetic environments, in McDonald, D.A. and Surdam , R.C., eds., Clastic Diagenesis : American Association Petroleum Geologists , Memoir 37, p. 111-123. Gieskes, J.M., 1981, Deep sea drilling interstitial-water studies: Implications for chem ical alteration of the ocean crust , Layers I and II, in Warme, J.E., Douglas, R.G. and Winterer, E.L., eds., The Deep Sea Drilling Project: A decade of progress : Society of Economic Paleontologists and Mineralogists, Special Publication 32, p. 149-167. GoJdhaber, M.B., and Kaplan, LR., 1974, The sulfur cycle, in Goldberg, E.D., ed., The Sea, v. 5, Marine Chemistry: John Wiley and Sons, New York, p. 569-655 . Goldhaber, M.B., and Kaplan, LR., 1980, Mechanisms of sulfur incorporation and isotope fractionation during early diagenesis in sediments of the Gulf of California: Marine Chemistry, v.9, p.95-143. Hein, J.R., O'Neil, J.R. and Jones, M.G., 1979b, Origin of authigenic carbonates in sediment from the deep Bering Sea: Sedimentology, v. 27, p. 577-591. Hennessy, J. and Knauth, L.P., 1985, Isotopic variations in dolomite concretions from the Monterey Formation, California: Journal of Sedimentary Petrology, v. 55, p. 120-130. Hudson , J.D., 1978, Concretions, isotopes , and the diagenetic history of the Oxford Clay (Jurassic) of central England: Sedimentology, v. 25, p. 339-370. Irwin, H., Curtis, C.D. and Coleman, M., 1977, Isotopic evidence for source of diagenetic carbonates formed dur ing bur ial of organic-rich sediments: Nature, v. 269, p. 209-213. Islam, S., 1981, Thermal maturation patterns in Cambro-Ordovician flysch sediments of the Taconic Belt, Gaspe-Peninsula , unpublished M.Sc. Thesis , McGill University, 191 p. Itihara, Y. and Honma, H., 1979,Ammonium in biot ite from metamorphic and granitic rocks of Japan: Geochimica et Cosmochimica Acta, v. 43, p. 503-509.
Kelts, K. and McKenzie, J.A., 1982, Diagenetic dolomite formation in Quaternary anoxic diatomaceous muds of Deep Sea Drilling Project Leg 64, Gulf of California, in Curray, J.R., Moore, D.G. et et., Initial Reports of the Deep Sea Drilling Project, v. 64, part 2, p. 553-569: United States Government Printing Off ice , Washington, D.C. Kharaka, Y.K., Hull, R.W. and Carothers, W.W., 1985, Water-rock interactions in sedimentary basins , in Gautier, D.L., Kharaka, Y.K. and Surdam, R.C., Relationsh ip of organic matter and mineral diagenesis, Society of Economic Paleontologists and Mineralogists, Short Course Lecture Notes, v. 17, p. 79-176. Kharaka, Y.K., Hull, a.w and Specht, D.J., 1985b, SOLMNEQ II: Solution-mineral equilibr ium computat ions : Un ited States Geological Survey, Water Supply Paper, in review. Kharaka, Y.K., Lico, M.S. and Carothers, W.W., 1980, Predicted corrosion and scale-formation properties of geopressured geothermal waters from the Northern Gulf of Mexico Basin: Journal of Petroleum Technology, v. 32, p. 319324. Kushnir, J. and Kastner, M., 1984,Twoforms of dolomite occurrences in the Monterey Formation, California ; concretions and layers - a comparative mineralog ical, geochemical and isotopic study, in Garrison, R.E., Kastner, M. and Zenger, D.H., eds., Dolomites of the Monterey Formation and Organ ic Rich Units: Society of Economic Paleontologists and Mineralogists, Pacific Section, Special Publication no. 41, p. 171-183. Lerman, A., 1977, Migrational processes and chemical reactions in interstitial waters,ln Goldberg, E.D., McCaye, LN., O'Brien J.J. and Steele, J.H., eds., The Sea, v. 6: John Wiley and Sons, New York, p. 695-738. Matsumoto, R., 1983, Mineralogy and geochemistry of carbonate diagenesis of the Pliocene and Pleistocene hemipelagic mud on the Blake Outer Ridge, Site 533, Leg 76, In Sher idan, R.E., Gradstein, EM. et a/., Initial Reports of the Deep Sea Drilling Project, v. 76, p. 411-427: United States Government Printing Office, Washington, D.C. MOiler,P.J., 1977, C/N ratios in Pacific deepsea sediments : Effect of inorganic ammonium and organic nitrogen compounds sorbed by clays: Geochimica et Cosmochimica Acta, v. 41, p. 765-776. Oertel , G. and Curtis, C.D., 1972, Claystone-ironstone concretion preserving fabrics due to progressive compaction : Geological Society of America, Bulletin, v. 83, p. 2597-2606.
316
Okada, H., 1980, Pebbles and carbonate nodules from Deep Sea Drilling Project Leg 56 cores, in Honza, E. et aI., Initial Aeports of the Deep Sea Drilling Project, v. 56-57, p. 1089-1105: United States Government Printing Office, Washington, D.C. Pisciotto, K.A. and Mahoney,J.J ., 1981, Isotopic survey of diagenetic carbonates , Deep Sea Drilling Project Leg 63, in Yeats, A.S., Haq, B.U. et el., Initial Aeports of the Deep Sea Drilling Project, v. 63, p. 595-609: United States Government Printing Office, Washington, D.C. Aaiswell, A., 1982, Pyrite texture, isotopic composition and the availability of iron: American Journal of Science, v. 282, p. 1244-1263. Aaiswell, A. and Berner, A.A., 1985, Pyrite formation in euxinic and semi-euxinic sediments: Amer ican Journal of Science, v. 285, p. 710-724.
Diagenesis
Aitger, S., Carson, B. and Suess, E., 1987, Methane-derived authigen ic carbonates formed by subduction-induced pore water expulsion along the Oregonl Washington margin: Geological Society of America, Bulletin, v. 98, p. 147-156. Aussell, K.L., Deffeyes, K.S., Fowler, G.A. and Lloyd, A.M., 1967, Marine dolomite of unusual isotopic compos ition : Science, v. 155, p. 189-191. Schwarcz, H.P. and Burnie, S.W., 1973, Influence of sedimentary environments on sulfur isotope ratios in clastic rocks: a review: Mineralium Deposita, v. 8, p.264-277. Sholkovitz, E., 1973,lnterstitialwaterchemistry of the Santa Barbara Basin sediments: Geochimica et Cosmochimica Acta, v. 37, p. 2043-2073. Suess, E., 1979, Mineral phases formed in anoxic sediments by microbial decomposition of organic matter: Geochimica et Cosmochimica Acta, v. 43, p. 339352.
Geoscience Canada Reprint Series 4
Tasse, N. and Hesse, A., 1984, Origin and significance of complex authigenic carbonates in Cretaceous black shales of the Western Alps: Journal of Sedimentary Petrology, v. 54, p. 1012-1027. Wada, H., Niitsuma, N., Nagasawa, K. and Okada, H., 1981, Deep sea carbonate nodules from the Middle America trench area off Mexico, Deep Sea Drilling Project Leg 66, in Watkins, J.S., Moore, J.C. et a/., Initial Aeports of the Deep Sea Drilling Project, v. 66, p. 453-474: United States Government Printing Office, Washington, D.C.
Originally published in Geoscience Canada v. 13 Number 3 (September 1986) Revised 1989
c c c c c c c c c
c c
c c c c c c
c c
c c c c c c c c ~
c
c
c C
l
e ( (
l C ( (
( .
c
Fluid Geothermometry in Sedimentary Systems
317
c.
c
(
c (
C' C
o o ( o o c o o o o o o
o o (
( (
,~.
'-..
Fluid Geothermometry in Sedimentary Systems Ian Hutcheon Hugh Abercrombie David Gardner The Department of Geology and Geophysics The University of Calgary Calgary, Alberta SUMMARY Fluid geothermometers have demonstrated their utility in the geothermal industry, but have seen limited application to sediment diagenesis. If diagenesis is viewed as an active process, fluid geothermometry is a tool to observe active analogues to diagenetic processes. Understanding active processes should help to unravel diagenetic histories. Measurement of temperatures in well bores can be done by mechanical means, but chemical and isotopic geothermometers can be more useful in some circumstances. Fluid geothermometers are based on chemicalor isotopic equilibria among gaseous or aqueous species. Although gas geothermometers have not been utilized to estimate temperatures of produced fluids from oil and gas wells, they have some promise. Geothermometers based on the composition of aqueous fluids are used in geothermal applications and less so with fluids produced from oil and gas wells. The Na/K and chalcedony geothermometers give the most consistent results with fluids obtained from thermal recovery pilot sites, but other geothermometers hold promise. INTRODUCTION This paper discusses geothermometers based on the chemical or isotopic compositions of fluids, water or gas. These methods are useful to determine
the temperatures of fluids particularly if deep wells are available for sampling. The techniques discussed also are useful to monitor temperatures during thermal recovery of hydrocarbons. Although the " equilibrium" model for sedimentary systems is questioned by some workers, most fluid geothermometers (the Na/K fluid geothermometer, for example) are based on empirical correlations between observed flu id compositions and measured temperatures. The sil ica geothermometer is also useful and is based on the equilibrium solubility of silica species. We have tested these models in our own research on thermal pilots and find them to be effective. There are a multitude of methods that rely on the compositions or physical properties of minerals, for example isotopic compositions of minerals , fluid inclusion homogenization or mineral compositions (Walshe, 1986; Cathelineau and Nieva, 1985). Fluid inclusion analysis and methods that utilize minerai compositions are powerful, but the temperatures measured must be interpreted in terms of the diagenetic history ofthe rock. Forexample, fluid inclusions form at a certain stage of growth of a crystal and represent the conditions at that stage of diagenesis. Not only is it difficult to interpret when, in the burial history, the event in question occurred , it is difficult to see through subsequent processes of recrystallization or alteration. The same difficulties are encountered in methods that employ chemical or isotopic compositions of minerals. Direct measurements of temperature in modern diagenetic environments are difficult. Temperatures can be measured in oil wells by down-hole logging tools, but this is complicated because temperatures normally are measured during drilling and the injection of fluid during drilling upsets the thermal regime, complicating the interpretation oftemperature measurements. Producing wells are ideal candidates for temperature measurements, but rarely are temperature probes inserted in producing wells. Fluid geothermometers have been developed, largely as a tool in the geothermal industry, to determine the heat content, and thus the value, of geothermal resources. The pioneering papers of White (1965),Ellis and Mahon (1967) and Fournier and Truesdell(1973) present an empirically calibrated geo-
thermometer that relies on the Na/K ratio in the aqueous solution. Mariner and Kharaka (1988) have summarized the data for a number of aqueous geothermometers. Temperatures measured from fluid chemistry are useful in thermal recovery projects because down-hole measurements typically are not possible. Fluid geothermometers thus provide a means of monitoring how temperatures change during the production of a hot fluid from initially cold surrounding rocks. In examining the diagenesis of a sedimentary system subject to active burial, fluid geothermometry is a precursor to examination of the stability of waterswith respectto diagenetic minerals. Drill-stem test temperature measurementsare alsosubject to errors related to fluid circulation time. Formation water, rapidly retrieved from the rocks, should retain an accurate formation temperature. Paleo-temperatures during diagenesis are particularly difficult to estimate and methods such as fluid inclusions, fission track dating combined with burial history and vitrinite reflectance can be employed. These methods are not discussed in this paper. Mineral equilibria are intimately related to the fluid geothermometers mentioned above, butthe reader should be aware that these methods are useful only to determine the present condition of sedimentary systems in the subsurface and do not contain information regarding the past diagenetic history. The methods are still useful because examination of drill core samples from deeper wells indicates that diagenesis is an "active" process in these environments (801es, 1982).As such, an understanding of present-day conditions is useful to the petrologist studying diagenesis in the same way that modern analogues are useful to sedimentologists studying ancient depositional systems. Geothermal systems are recognized throughout the world as economic sources of energy and this has led to extensivesubsurface geothermal exploration in many countries (Ellis, 1979a). These present-day hydrothermal systems provide direct measurements of temperature and gas and water compositions. They also provide a good analogy for the rock-water interaction processes which accompany
( Diagenesis
318
diagenesis and thermal enhanced recovery. Thermal recovery, both steam injection and in-situ combustion, also provide an opportunity to obtain fluid samples and relate their composition to temperature. The reader is cautioned that the sampling, preservation and analysis of fluids is not a trivial procedure and unless the details of sampling are subject to careful control, the interpretation of analytical results can be equivocal. The techniques discussed here should be applied with caution to samples of uncertain derivation. Lico et at. (1982) provide documentation of methods for the sampling of oil field and geothermal waters. GAS GEOTHERMOMETERS Gas equilibrium in geothermal discharge has been used by many workers to estimate fluid temperatures at depth (Giggenbach, 1980; D'Amore and Panichi, 1980; Arn6rsson et a/., 1983a). Geothermometers based on gas composition either can be calculated from thermodynamic and experimental data or devised empirically. Typically, a number of assumptions are made which undoubtedly are not valid for all geothermal or thermal oil recovery systems. Obtaining representative water, steam and gas samples from geothermal systems is a difficult problem and well sampling and tubing has been standardized (Ellis and Mahon, 1977; Arn6rsson et a/., 1978). The more prevalent gases in hot springs and geothermal brines include CO 2, CH4 , H2S, S02 and H2. Smaller concentrations of NH 3 , N2 and O2 are also common and considered to be of geological interest. Investigations of heavy oil pilots show CO 2, CH 4 and H2S to be the most commonly produced gases. The relative abundance of CO 2, CH 4 and H 2S in gases, depends on temperature and can be used as a geothermometer in hydrothermal systems. These methods have not been applied to oil field systems, but, considering their successful application to geothermal systems, they warrant serious consideration. Arn6rsson et at. (1982), Giggenbach (1980), Arn6rsson et at. (1983a), and D'Amore and Panichi (1980) have formulated a number of simple equations which relate gas composition to temperature . The geothermometers have been calibrated using either
experimental or empirical methods. The gas reaction used most often for gas-reaction equilibria is based on the Fisher-Tropsch reaction (Nehring and D'Amore, 1984): (1) CO2 + 4H 2 = CH 4 + 2H20 Figure 1 shows the partial pressure of CO2 in geothermal waters from Iceland as a function of temperature (from Arn6rsson et a/., 1983). The obvious relationship between CO 2 concentration and temperature suggests the possible use of CO 2 as a geothermometer. The upper stability limit of calcite, approximately 300°C, is the upper limit of applicability. The equilibria between ammonia nitrogen and hydrogen (Ellis, 1979b) and the reaction of water with carbon to produce hydrogen and CO 2 (D'Amore and Panichi, 1980) have also been employed in geothermal systems. For reaction (1) the equilibrium constant (I<)
Geoscience Canada Reprint Series 4
can be calculated from the free energy change of the reaction (~G r) at a number of different temperatures and , assuming ideal mixing of ideal gases, is dependent on the gas species as follows: K
= PC02(PH)4/PCH4(PH20)2
I
I
-
....-..
en
•• •
..0
ca
•
10-1
en
~
ca
..0
o o
-
10-3
10-5
)
c c c c
c
c c c c
-
.-
-"-
a..
c c
c
-
~.
I-
-
c c c c c c c c c c c c
c
•
"'-'"
C\J
(2)
In geothermal systems, temperatures are high and pressures are relatively low, a situation thatfavours the assumption of ideal gases mixing ideally. In real systems, the gas fugacities should be used and the mixing coefficients determined from van der Waals equation (Redlich-Kwong, for example) or a virial equation of state. Giggenbach (1980) plotted log K against 11T, with temperature (T) in kelvin, and derived the following equation: (3) log K = 10.76 - 93231T Because PH20 generally is not measured in a gas-producing well and most sedimentary and geothermal systems
(
c
c c
c c
.-
1-I
I
I
100
200
300
Temperature (OC) Figure 1 The partial pressure of CO2 in geothermal waters from Iceland. (After Arn6rsson at al., 1983).
e c
L
C (
C (. (
( (
319
(
(
( (
(
C ( (
o c o o o o o o o o o o o o o o o o o C)
l l
are a mixture of liquid water and vapour, the useful form of the equation is more complicated: logXC02 + 410gXH2 - logXCH 4
'" - 5.4398 - 13419/T + O.01959T
(4)
where Xi refers to the mole fraction of gaseous species in the vapour. Gas geothermometers are most accurate between 200° and 300°C, but empirical gas geothermometers may extend this range. Gas geothermometers have not been applied to hydrocarbon wells to our knowledge, but their use could be developed as a tool in exploration and production of gas and oil reservoirs. AQUEOUS GEOTHERMOMETRV Numerous chemical and isotopic geothermometers capable of remote determination of deep temperatures in thermal water systems have been proposed over the past forty years. Earliest work involved determination of the systematic variations with temperature in the fractionation of light stable isotopes between co-existing minerals, fluids, and gases (Urey, 1947; Urey et a/., 1951; Epstein et a/., 1951; Craig, 1953). These same methods have been used to estimate paleotemperatures in the ocean from fossil shells and diagenetic temperatures from cements. Investigations of lcelandicqeotharrnal waters by Bodvarsson (1960)and Bodvarsson and Palmason (1961) led these workers to conclude that the concentration of silica in thermal waters could be related to the
temperature of the underlying geothermal system (Ragnarsdottir and Walther 1983).White (1965) and Ellis and Maho~ (1967) published empirically derived curves relating the Na/K ratio in thermal waters to temperatures. These and other geothermometers have been used extensively to estimate temperatures at depth in natural thermal water systems, particularly in geothermal exploration and exploitation. Outside of these fields, the potential of geothermometry using the chemical and isotopic composition of fluids has received little attention. The abundance of chemical analyses of produced waters and gases in both conventional and unconventional petroleum recovery invites application of chemical geothermometers for remote temperature measurement. More particularly, the range of temperatures encountered in water or steam injection for heavy oil recovery techniques closely parallels the 25°-300°C temperature range observed for natural thermal and geothermal waters. The equilibrium concentration of chemical and isotopic species varies with changing temperature and, to a lesser extent, pressure. In theory, chemical and isotopic geothermometers can be calibrated using thermochemical data. In practise, existing thermochemical data are not adequate to allow accurate calibrations, necessitating empirical or experimental calibration of these geothermometers.
-1.0 -2.0 C\I
0 U5
n1
-3.0
C>
0
.....J
-4.0
(
.
o
100
200
300
TEMPERATURE (OC)
Figure 2 The solubility of amorphous silica, chalcedony and quartz as a function of temperature. '--
Silica. In natural water systems, silica may be present as amorphous silica (gelatinous silica, opal, sinter), chalcedony(a microcrystalline, fibrous variety of quartz), alpha-quartz, or cristobalite. Although metastable, amorphous silica and chalcedony commonly are present in thermal water systems at temperatures less than 180°C (Arn6rsson, 1975;Ellis, 1979b; Arn6rsson et et., 1983a). In thermal waters at temperatures above 180°C, alpha-quartz is the stable silica phase. The silica geothermometer is based upon the variation with temperature of ~he solubility of quartz or chalcedony In water. This is represented by the reaction: Si0 2 (qz,Ch) + 4H 20
-5.0
'-..
The basic requirements for geothermometers that involve an aqueous phase are (Fournier et st., 1974; Ellis, 1979b): 1. The elements, isotopes, or species involved are present within the waterrock system. 2. Chemical or isotopic equilibrium is attained in the time the fluid is present at the depth of interest. 3. Solution re-equilibration rates are slow enough that the equilibrium concentrations at depth remain essentially unaltered during migration of the fluid to the surface. Some additional considerations include ensuring that geothermometers involving equilibrium between fluid and minerai are applied to rocks that contain that mineral. Also, the gain or loss of fluid components through mixing or steam flashing must be considered. Reaction kinetics will dictate whether equilibrium distributions of chemical and isotopic species can be achieved under reservoir conditions and be maintained under surface conditions once the sample is collected. In addition many reactions chosen for use as geo~ thermometers, particularly isotopic and gas-gas reactions, show significant variations in reaction rate with temperature, pH and fluid composition.
= H4Si04 (aq)
(5)
where Si0 2 (qz ,c h) refers to quartz or chalcedony. Above a pH of 9.0, this geothermometer must be corrected for the ionization of H 4Si04 (aq) to H 3Si04(aq) (Ragnarsdottir and Walther, 1983). Figure 2 shows the solubilities of amorphous silica, chalcedony and
( 320
quartz as a function of temperature. Appropriate corrections are discussed by Arnorsson et al. (1983a). The chalcedony geothermometer has been experimentally calibrated by Fournier (1973, 1977). Arnorsson et al. (1983a) have empirically calibrated the chalcedony geothermometer by correlating measured drill-hole temperatures with calculated H4Si04(aq) molalities in geothermal fluids from Iceland. Agreement between the Fournier(1973, 1977) solubility curves and the empirical calibration of Arnorsson et a/. (1983a) is good in the range 0°-180°C. Numerous studies of quartz solubility as a function of temperature have been carried out, but results are not consistent. Ragnarsdottir and Walther (1983) combined measured quartz solubility with data from Dickson (1982), Crerar and Anderson (1971) and Fleming and Crerar (1982), to calibrate a pressuresensitive silica (quartz) geothermometer. This geothermometer is valid for solutions in equilibrium with quartz in the temperature range 180°-340°C and fluid pressures from water saturation to 500 bars. Application of the quartz geothermometer to heavy oil recovery is limited. Detrital quartz is the most abundant mineral in the majoroil sand deposits in Alberta and its dissolution during heavy oil recovery is well known (Boon, 1977a,b; Reed, 1980). Applying the quartz geothermometer requires that quartz dissolution under recovery conditions be an equilibrium process. This requirement is commonly assumed, either implicitly or explicitly (Hutcheon, 1984). Na/K and Na-K-ea. In contrast to the silica geothermometer, which is related to mineral solubility, the Na/K and Na-K-Ca geothermometers are dependent upon the temperature variation in the partitioning ofthese elements between the fluid and mineral phases (Ellis, 1979b). The Na/K ratio in natural thermal fluids is believed to be controlled by the exchange reaction (Ellis and Mahon, 1977): K + + Na-feldspar = K-feldspar + Na + (6)
requiring the presence of Na- and K-feldspar in the reservoir. At low temperatures, it would seem more likely that the Na/K ratio is controlled by reactions among clay minerals. This suggestion can be tested using data from Hutcheon et al. (1988). Figure 3 shows
Diagenesis
the Na/K ratio of fluids, produced from a heavy oil pilot site during steam injection and recovery, with respect to the exchange reaction (6) and a similar reaction between illite and Na-smectite. The analyses of produced fluids from this pilot fall closest to the phase boundary between albite and K-feldspar, suggesting that reactions among clay minerals may not control the Na/K geothermometer. Hutcheon et et. (1988) note that the Na content of produced fluids from this pilot is controlled by mixing of condensed steam and formation water and that the K concentration is in equilibrium with the K-feldspar-illite phase boundary. Clearly the Na/K geothermometer may be mediated by more than one mineral reaction. For waters at temperatures less than 180°C and containing high amounts of calcium, the Na-K geothermometer may give anomalous temperatures (Fournier and Truesdell, 1973; Ellis, 1979b). For this reason, Fournier and Truesdell (1973) proposed the empirical Na-K-Ca geothermometer, which accounts for the participation of Ca in aluminosilicate reactions. Arnorsson et al. (1983a) suggest that the Na-K-Ca geothermometer is less rigorous than the Na/K, because the former does not require equilibrium with a particular mineral assemblage and does not account for the possibility of Ca-Ioss through calcite precipitation upon flashing of water to steam as pressure drops. At low temperatures, there is little mutual solid solution of albite and
Geoscience Canada Reprint Series 4
K-feldspar, therefore the activities of Na- and K-feldspar can be assumed to be unity. Indilute waters, the equilibrium constant for reaction (6) is thus the anaIytical Na/K ratio, hence temperatures can be calculated. Above 250°C, at high sulphate levels, and in saline waters with more than 5000 rnq- L-1 chloride, activities and activity ratios must be calculated and then related to the equilibrium constant (Arnorsson, 1983a). Various computer programs, such as SOLMNEQ (Kharaka and Barnes, 1973) or SOLMNEQF (Aggarwal and Gunter, 1986) can be used for this calculation. The Na/K geothermometer hasfound wide use in geothermal studies at a temperature range between about 100° and 300°C (Fournier and Truesdell, 1973; Ellis and Mahon, 1977; Ellis, 1979b). Arnorsson et al. (1983a) have extended this range with two temperature functions, one from 25° to 250°C and the second from 250° to 350°C . Bojes and Wright (1986)have applied the Na/K geothermometer to thermal recovery, but the technique has not seen wide usage. Figure 4 shows estimated temperatures using the Na/K and silica geothermometers for a thermal recovery pilot after Hutcheon et a/. (1988). The correspondence of Na/K and chalcedony temperature estimates implies that the chalcedony geothermometer is effective in this temperature range (80°-180°C). The rocks from the steam pilot sites in the Cold Lake area typically are lithiclastic and contain volcanic fragments , chert and opaline sil-
5.0
---
4.0
C C
C C C
C C C C
C C C C C C C C C C C C
C C
C C
C C C
c c
+~
ctS
+ctS 3.0
c
ctS
2.0
<..
1.0
c
'-
Z
0> 0 --I
e <..
c
0 0
100
200
300
TEMPERATURE (DC) Figure 3
The Na/K ratio of waters from a thermal pilot as a function of temperature.
(
<.. (
c (
( (
Fluid Geothermometry in Sedimentary Systems
321
(
c c
c (
(
C' ( (
o o o o o o
o
o o o
c
o o o o o o o o o o C: b C
ica. Hutcheon at al. (1988) have also sampled produced fluids from a steam pilot dominated by quartz sandstones. Over the same temperature range, the silica concentration is apparently controlled by quartz saturation, rather than chalcedony saturation . One complication of the direct application of the Na/K geothermometer is the role played by the Na-smectite (beidellite) component of smectites that are known to form during heavy oil recovery (Perry and Gillott, 1979; Sedimentology Research Group, 1981). Weissburg and Wilson (1977) have discussed this problem as it applies to geothermal exploration . Mineral Saturation Index Geothermometer. The silica and Na/K geothermometers are based on control of the fluid phase composition by equilibrium with astable silica phase in the former case, and Naand K-feldspar in the latter. The mineral saturation index (MSI)geothermometer is similar in that it incorporates equilibrium between the fluid and all minerals that could conceivably be present (Reed and Spycher, 1983). The MSI geothermometer, like all geothermometers, requires accurately measured fluid compositions and pH values. The MSI geothermometer uses existing geochemical models such as SOLMNEQ(Kharakaand Barnes, 1973; Aggarwal and Gunter,1986)or SOLVEQ (Reed, 1982)to calculate ion activities at various temperatures. At each tem-
--~ --
180
I-
Q)
I
I
I
• Chalcedony
A.
A.A.
• •• •
-
• Wellhead
A.
"-
~ -.. rn 140
I-
tt
"Q)
a. E 100 Q) ~
0
-
•I ••
• •• • ••
A.
I
-
•
I
I
I
I
I
30
60
90
120
150
180
Days
. Figure 4
"
I
A. NalK
60 ,
I
220
perature, an ion activity product, Q, is calculated for each of the minerals that could possible be present. The calculated values of Q are then compared with the equilibrium constants, K, for each mineral and the function 10g(Q/K), commonly referred to as the mineral saturation index, is computed. Mineral saturation is achieved when Q equals K, hence 10g(Q/K) = O. Theoretically, by examining all possible equilibria, the MSt geothermometer should provide more accurate estimates of temperature than chemical geothermometers that relyon one chemical equilibrium. However, the MSJ geothermometer suffers the disadvantage of being totally dependent upon the thermodynamic data base selected, which, as discussed by Arn6rsson et al. (1983b), may be inadequate for chemical geothermometry. Geothermometers that are based upon particular fluidmineral equilibria and have been calibrated with measured temperatures, such as silica and Na/K geothermometers, offer the distinct advantage of being independent of errors in thermodynamic data. Other disadvantages of the MSJ geothermometer are the requirement for high-quality fluid analyses, including accurate AI determination, and the cost of multiple computer runs to calculate a single temperature . Possible Applications to Reservoir Diagenesis. Aqueous geothermometers provide a means of extending subsurface tem-
The temperatures of produced fluids from a thermal recovery pilot estimated using the Na/K (triangles) and silica geothermometers.
perature data. In areas of active exploration, particularly in basins with a relatively continuous history of burial, aqueous geothermometers could be used to extend the geographic and stratigraphic range of temperature information . It would be necessary to compare fluid temperatures estimated from aqueousgeothermometers to drillstem temperatures to verify that they gave reasonableestimates. If variations between drill-stem temperatures and fluid temperatures were systematic,the fluid temperatures could still be used to extend the temperature data base. In overpressured zones (regions in which fluid pressures are higher than hydrostatic pressure), temperature is observed to increase more rapidly with depth. Aqueous geothermometers might provide a method of delineating overpressured zones in areas of active exploration . It should be emphasized that the accuracy of temperature estimates is directly related to the quality of water samples. The procedures for obtaining useful water samples from geothermal and hydrocarbon wells are summarized in Lico et a/. (1982). CONCLUSIONS A number of geothermometers based on equilibria among gaseous species have been developed for usein geothermal systems and may potentially be useful in sedimentary systems. These geothermometers are dependent on the concentration of different gas species, including CH4 , CO2 , H2 , NH3 , N2 and H2S. High concentrations of dissolved salt in produced geothermal fluids and fluids from thermal and conventional oil recovery havebeen shown to effect these calculations to varying degrees . Although gas geothermometers have not been utilized to estimate temperatures of produced fluids from oil and gas wells, they have some promise. Geothermometers basedonthe composition of aqueous fluids have been used extensively in geothermal applications and less so with fluids produced from oil and gas wells. The Na/K and chalcedony geothermometers give the most consistent results with fluids obtained from thermal recovery pilot sites, but other geothermometers hold promise. Of the many versions of the Na/K geothermometer, the equations
322
developed byArn6rsson (1983) produce temperatures that best agree with other available temperature estimates. Aqueous geothermometers have demonstrated their utility in the geothermal industry, but have been neglected in applications to sediment diagenesis. Although there is some discussion, there is evidence that diagenesis, when viewed from the perspective of water-rock interaction, is an active process. Fluid geothermometry is then a tool to observe active analogues to diagenetic processes. Understanding active processes, should help to unravel diagenetic histories.
ACKNOWLEDGEMENTS The data on thermal recovery were obtained through the co-operation of Husky Oil Operations Limited under a research agreement with the Alberta Oil Sands TechnologyResearchAuthority. The assistance of Peter Putnam , Andrew Bau and Ken Applegate was invaluable.
REFERENCES Aggarwal, P.K. and Gunter, W.O., 1986, SOLMNEQF: A FORTRAN-77 version of SOLMNEQ: ARC/AOSTRA Joint agreement report 8687-4. Arn6rsson ,S., 1975,Application ofthe silica geothermometer in low temperature hydrothermal areas in Iceland: American Journal of Science, v. 275, p. 763-784 . Arnorsson, A., Gronvold, K. and Sigurdsson, S., 1978, Aquifer chemistry of four high·temperature geothermal systems in Iceland: Geochimica et Cosmochimica Acta, v. 42, p. 523-536. Arn6rsson, 5 ., Sigurdsson , S. and Svavarsson, H., 1982,The chemistry of geothermal waters in Iceland. I. Calculation of aqueous speciation from 0° to 370°C: Geochimica et Cosmochimica Acta , v. 46, p. 1513-1532. Arn6rsson, S., Gunnlaugsson, E. and Svavarsson, H., 1983,The chemistry of geothermal waters in Iceland. III. Chemical geothermometry in geothermal investigation: Geochimica et Cosmoch imica Acta, v. 47, p. 567-577. Bodvarsson, G., 1960, Exploration and exploitation of natural heat in Iceland: Bulletin Volcanologique, v. 23, p. 567578. Bodvarsson , G. and Palmason, G., 1961, Exploration of subsurface temperatures in Iceland : Proceedings U.N. Conference, New Sources of Energy G/24 , p. 82-90, Rome.
Diagenesis
Bojes, J.M. and Wright, G.B., 1986,Application of fluid analyses to the operation of an in situ combustion pilot: Annual Technical Meeting, Petroleum Society of CIM, paper 86-37-61, p. 315-349. Boles, J.A., 1982,Active albitization of Gulf Coast Tertiary: American Journal of Science, v. 282, p. 165-180. Boon, J.A., 1977a, Mass transfer of silica during steam injection: Proceedings 2nd International Symposium on WaterRock Interaction, Strasbourg, France, p.199-206. Boon, J.A., 1977b, Fluid-rock interactions during steam injection, in Redford, D.A. and Wlnestock, A.G ., eds., The Oil Sands of Canada - Venezuela 1977: CIMM, Toronto, p. 133-138. Cathelineau, M. and Nieva, D., 1985,A chlorite solid solution geothermometer, The Los Azufres (Mexico) geothermal system: Contributions to Mineralogy and Petrology, v. 91, p. 235-244. Crerar, D.A. and Anderson, G.M., 1971, Solubility and solvation reactions of quartz in dilute hydrothermal solutions: Chemical Geology, v. 8, p. 107-122. Craig, H., 1953,The geochemistry of stable carbon isotopes: Geochimica et Cosmochimlca Acta, v. 3, p. 53-92. D'Amore , F. and Panichi, C., 1980, Evaluation of deep temperatures of hydrothermal systems by a new gas geothermometer: Geochimica et CosmochimIca Acta, v. 44, p. 549-566. Dickson, F.W.,1982,Kinetics of quartz reaction with H2 0 at 200°C and 120 bars (abstr): EOS, v. 63, p. 1128. Ellis, A.J., 1957, Chemical equilibrium in magmatic gases: American Journal of Science, v. 255, p. 416-431, Ellis, A.J., 1979a, Explored geothermal systems , in Barnes, H.L., ed., Geochemistry of Hydrothermal Ore Deposits , Second Edition: John Wiley and Sons, New York, p. 632-683. Ellis , A.J., 1979b,Chemical geothermometry in geothermal systems: Chemical Geology, v. 25, p. 219-226. Ellis, A.J. and Mahon, W.A.J., 1967, Natural hydrothermal systems and experimental hot water-rock interaction: Geochimica et Cosmochimica Acta, v. 31, p.519-531, Ellis, A.J. and Mahon, W.A.J., 1977, Chemistry and Geothermal Systems : Academic Press, New York, 392 p. Epstein,S., Buchsbaum, R., Lowenstam , H.A. and Urey, H.C., 1951, Carbonatewater isotopic scale: Geological Society of America , Bulletin, v. 62, p. 417-425. Fleming, B.A. and Crerar, D.A., 1982,Silicic acid ionization and calculation of silica solubility at elevated temperature and pH: application to geothermal fluid processing and reinjection: Geothermics, v. 11, p. 15-29.
Geoscience Canada Reprint Series 4
Fournier, R.O., 1977, Chem ical geothermometers and mixing models for geothermal systems: Geothermics, v. 5, p.41-50. Fournier, A.O. and Truesdell,A.H., 1973,An empirical Na-K-Ca geothermometer for natural waters: Geochimica et Cosmochimica Acta, v. 37, p. 1255-1275. Fournier, A.O., White, D.E. and Truesdell, A.H., 1974, Geochemical indicators of subsurface temperatures - Part 1,basic assumptions: United States Geological Survey, Journal of Research, v. 2, p.259-262. Giggenbach, W.F., 1980, Geothermal gas equilibria: Geochimica et Cosmochimica Acta, v. 44, p. 2021-2032. Heald, E.F.,Naughton,J.J. and Barnes,l.L., 1963, The chemistry of volcanic gases: Journal of Geophysical Research, v. 68, p.539-557. Hutcheon, I.E., 1984, A review of artificial diagenesis during thermally enhanced recovery, in McDonald, D.A. and Surdam, R.C., eds ., Clastic Diagenesis: American Association of Petroleum Geologists, Memoir 37, p. 413·429. Hutcheon, I., Abercrombie, H., Shevalier, M. amd Nahnybidia, C., 1989, A comparison of formation waters in quartzrich and quartz-poor reservoirs during steam assisted recovery, in Meyer, A.F. and Wiggins , E.J., eds., Proceedings of the UNITAR/UNDP Fourth International Conference on Heavy Crude and Tar Sands, Edmonton, v. 2, p. 447-459. Kharaka, Y.K. and Barnes, I., 1973, Solution mineral equilibrium computations : United States Geological Survey, NTIS Document PB 215 899. Kharaka, Y.K. and Mariner, A.H ., 1989, Chemical geothermometers and their application to formation waters from sedimentary basins, in Naeser, N.D. and McCulloh, T.H.,ed., Thermal History of Sedimentary Basins: methods and casehistories: Springer-Verlag, NewYork, p.99-117. Lico, M.S., Kharaka, Y.K., Carothers , W.W. and Wright, VA , 1982, Methods of collection and analysis of geopressured geothermal and oil field waters: United States Geological Survey, Water Supply Paper 2194, 21 p. Nehring, N.L. and D'Amore, F.D., 1984,Gas chemistry and thermometry of the Cerro Prieto, Mexico, Geothermal Field: Geotherm ics, v. 13, p. 75-89. Perry, C. and Gillott, J.E., 1979, The formation and behaviour of montmorillonite during the use of wet forward combustion in the Alberta oil sand deposits: Bulletin of Canadian Petroleum Geology, v. 27, p. 314.
c c
c c ( c c c c c c c c c c c c c c c c c c c c c c ( c c c c C E C
c c c C l.
C
c r: \
323
Fluid Geothermometry in Sedimentary Systems
I
( ( (
C (,
C C (
o o ( c o o o
o o o
o o o o o o o o o o l l
Ragnarsdottir, K.V. and Walther, J.V., 1983, Pressure sensitive "silica geothermometer" determined from quartz solubility experiments at 250°C: Geoch imica et Cosmochimica Acta, v. 47, p. 941-946. Reed, M.G., 1980, Gravel pack and formation sandstone dissolution during steam injection: Journal of Petroleum Technology, June , 1980, p. 941·949. Reed, M., 1982, Calculation of multicomponent chemical equilibria and reaction processes in systems involving minerals, gases and an aqueous phase: Geochimica et Cosmochimica Acta, v. 46, p.513·528. Reed, M. and Spycher, N., 1983, Calculated pH at high temperature in hydrothermal waters with applications to geothermometry: 4th International Symposium on Water-Rock Interaction, Misasa, Japan , p. 401-408.
Sedimentology Research Group, 1981, The effects of in situ steam injection on Cold Lake oil sands: Bulletin of Canadian Petroleum Geology, v. 29, p. 447-478. Urey, H.C., 1947, The thermodynamic properties of isotopic substances : Journal of the Chemical Society, p. 562-581. Urey, H.C., Lowenstam, H.A., Epstein, S. and McKinney, C.R., 1951, Measurement of paleotemperatures and temperatures of upper Cretaceous of England, Denmark and the southeastern United States: Geological Society of America , Bulletin, v. 62, p. 399-416. Walshe, J.L. 1986,A six-component chlorite solid solution model and the conditions of chlorite formation in hydrothermal and geothermal systems: Economic Geology, v. 81, p. 681-703.
Weissburg , B.C. and Wilson , P.T., 1977, Montmorillonite and the Na/K geothermometer, in Ellis, A.J., ed., Geochemistry : New Zealand Department of Science and Industrial Research, BUlletin, v. 218, p. 31-34. White , D.E., 1965, Saline waters of sedimentary rocks, in Young, A. and Galley, J.E., eds., Fluids in Subsurface Environments -A Symposium: American Association of Petroleum Geologists, Memoir 4, p. 342-366.
Accepted
1988
( 324
Diagenesis
Geoscience Canada Reprint Series 4
C C
C C C C C (
C C C C C C C C
C
SEM photomicrograph of a K-feldspar overgrowth remaining after leaching of the core of the grain. Infiltration by low-salinity groundwater may have caused dissolution offeldspars. The sample is from the Cretaceous Sparky interval in the Aberfeldy heavy oil site, Lloydminster area, Saskatchewan. Photograph courtesy of I. Hutcheon and R. Lefebvre (University of Calgary).
c c c c c c c c c ~ c c c c C E C l l
C C <....
l (
325
Index
( (
(
c c c (
o o o
Abercrombie, H. Geothermometry in sedimentary systems .. . . . 317-323 Albandite Sulphide precipitation Manganese sulphide 304 Amorphous Inorganic gels Argillaceous sedimentary rocks (clay minerals) .. . 182 Silica 227 Aqueous geothermometry 319 Fluid geothermometry Mineral saturation index geothermometer 321 Na smectite (beidellite) 321 Na/K and Na/K-Ca . 320 Reservoir diagenesis . . . 321 Silica .. 319 Aragonite Crusts Coniatites 22 Strand line diagenesis . 22 Seas 28 Carbonate petrology . . Argillaceous Chert Lydite . . . . . 227 Sedimentary rocks Foscolos, A.E. 177-187 Pelitic sediments .. 178 Burial alteration Clays o 179,180 Mineral transportation in pelitic sediments . 178 Mixed layer system 180 Silicates 178 Clay minerals 182 Amorphous inorganic gels .. . Carbonates . 183 Chlorite 181 Definition Illite 180 Diagenesis, organic matter and cementation . 184 Kaolinite 181 Aromatic compounds Organic matter Burial diagenesis . 200,205,211 Authigenic Carbonates Calcite and siderite Early diagenetic pore water/sediment interact ion . 304 Cementation Siliciclastic sediments ... . 168 Dolomite Deep sea dolomite Early diagenetic pore water/sediment interaction . . 305 Autocementation 88,92 Liquid hydrocarbons ~ . . .. 0
0
(
0
•
•
•
•
•••
0
•
.
•
0
•
••
•
••••••
•
•••
••••••
•
••••••
•
•
•
0
0
0
.0.
•••••
••
••
•
•
••
•
•
•
•
••
•
•
•
•
•
•
•
0
•••••••••••••••••••
•
•
•
0
•
•
•
•
•
•
•
0
•••••
0
•
••
0
0
••
.
0
0
•••••
••••
•
••••
0
•
•
•
•
•
•
0
•••
•
0
•
•
0
•
•
•
•
•
•
0
0
••••••
0
•••
••
0
0
•
•
•••
•••
••
•
0
•
•
•
•
•
••
••
0
•
•
•
0
•
•
•
•
•
•
•
•
•
•
•
•
••
•
0
0
•••
0
0
•••••
0
•••
0
0
••
•
•
•
0
0
••
•
•
••••
0
•
•
0
•••••
••
•
•
00
0
•
•
••••
0
•
0
•••••
0
•
•
0
•
••
••••
0
0
•
••••••••••
0
•
0
0
••••••••••••
•••
••
0
•
•
•
0
•••
•
••
••
•••
•
•
••
•
0
0
•
•
0
•
•••••••••••••••
0
0
0
•
0
•••
0
•••
•
0
0
•
•
0
0
•
•
••••••
0
•••••••
••
0
0
0
0
•
•
•
••••••••••
•
0
0
0
•
0
0
0
•
0
••••••
••••
••
•
••••
••
0
•••••
•
••••
0
•
••
•
••
•••
•
•
•
•••
0
0
•
•
•
0
•
0
•
•••
•
•
Bacterial sulphate reduction Microbial metabolism 191,193 Barnes, M.A. 205-226 Organic diagenesis Organic matter 189-204 Barnes, WoC. Organic diagenesis 205-226 Organic matter 189-204 Basinal burial domain 99 Limestones - Burial diagenetic environment . . Beaches Strand line diagenesis 21 Beidellite Aqueous geothermometry Na smectite (beidellite) 321 Biogenic pelagic carbonates Recrystallization Pore water profiles . . 291,293 Biogenic siliceous sediments Burial diagenesis Stages 231 Chert 228 Depositional environments Formation of bedded chert 231 Ophiolite-associated chert 231 Shallow water 231 Deep-sea environments . . 241 Diagenetic grade Host rock lithology 242 Opal·CT deep-sea environment 241 Silica phases 231 Characteristics Terminology 227 Biological Alteration 21 Limestones - Sea-floor diagenesis Markers Geochemical fossils . 4 Bittern salts Potash deposits 155 Burial Cements Deeper burial Limestones - Burial diagenetic environment .. 89 Shallow burial marine Limestones - Burial diagenetic environment . ... 88 Compaction Model Dolomitization models ... 128,132,134,135 Diagenesis 2 Schematic figure Stages Biogenic siliceous sediments .. .. 231 Diagenetic models 99 Limestones Bustin, A.M. 205·226 Organic diagenesis . . Organic matter 189-204 0
•••••
0
0
•••••
0
••••
0
•
0
•
•
•
0
0
•
•••
•
•
••
•
•
•
••
0
•
•
••
0
•
0
•••
0
•••••
0
0
••
0
••••
0
•
•
•••••
0
•••
0
0
0
•••
•
••
••••
0
•
••
••••
•
•
•
0
•
•
•••
•••
0
•
••
•
•
0
••
0
0
••
•
•
0
•
0
•••••
0
•••
0
0
••
•
•
••
•
•
•
•
0
•••
•
•
••
•
••
••
••••••
••••
0
•
•
•
••
•
0
•
•
•
•
0
•
•
•
•
0
•••
0
••
•
•
•
0.
0
••••
•
0
0
•
0
•
•••
•
•
•
••••
•••••
••••
•
0
•••••
0
0
0
•
•
•
•
•
•••••
•••••
•
•
•
•
0
•
•
••
••••
•
0
0
0
••
0
0
0
•
0
•••
•••
•
••
0
0
•
0
•
0
•
••••
•
••••
0
••
•
•
••••••••••••••
0
0
•
0
••
•••
0
•••
0
0
0
•
•
•
•
••••••••••
•
•
0
•
0
•
0
•
•
(
C
326
C Calcareous Sinter (speleothems) Karst, subsurface 50 Tufa Karst, subsurface 50 Calcite Bladed-prismatic Limestones - Burial diagenesis 89 Seas Carbonate petrology 28 Calcium zeolite Chert - diagenetic silica transformations 246 Caliche Rhizoconcretions 44,45 (calcrete or duricrust) Definition 43 Facies Water-controlled, rock-air interface 43 Litholog ies Petrography 44 Nodules Glaebules 44 Carbon Isotope studies Geochemistry of kerogen '" 217 Preference index Definition 216 Carbonate Compensation depth Definition 14 Diagenesis Cathodoluminescence 10 Diagenesis Diagenetic environments 9 Dissolution Dissolution 14 Petrology Aragonite seas 28 Petrology Calcite seas 28 Caliche Lublinite 45,50 Microcodium 45 Precipitate Cement stratigraphy 13,16,17 Sediments Porosity ranges 96,97,98 Water-controlled precipitation 50 Carbonates and carbonate sandstones Chertification 253 Argillaceous sedimentary rocks (clay minerals) 183 Catagenesis Definition 1,1n ,195 Kerogen Organ ic matter - burial diagenesis 195 Cathodoluminescence Carbonate diagenesis .... . . . ... . . .. . .. .• ... .• . 10
Cement Authigenic cementation Siliciclast ic sediments Petrography Fibrous calcites Fibrous cements Spherulitic cements Sea-floor lithification Epitaxial cement Fibrous calcites Isotopic signature Microcrystalline calcite Scalenohedral cement Spherulitic calcite Stromatactis Stratigraphy Carbonate precipitate Cementat ion Definition Cements Evaporites Syndepositional features Cavity cements Chalcedony Silica Geothermometer Fluid geothermometry Changes with depth Porosity-depth curves Chemical Compaction Definition Composition Dolomite Signatures Isotopic composition of precipitates Structure Hydrocarbon generation Chert Biogenic siliceous sediments Diagenesis Physical Burial diagenesis Opal-A Physical characteristics Opal-CT Mineralogic phases Opal-CT Physical characteristics Definition Depositional environments Biogenic siliceous sediments Modern ocean floor Biogenic siliceous oozes Tectonics Diagenetic silica transformations Calcium zeolite
C C 168 26 26 26 . 26 26 26 26 26 26 26 13,16,17 52
C C C C (
C ( (
C C (
C 146 153 227,254,255 320 96,98
83 113 17,20 196
C C C C
C C C
C C C C C
247
232 236 234 227 228 228 230 246
C C C C
c
e (
c c C l.
C C
('. (
327
\.
c C
r. C C.
c r (
o o o (
o o o
o o o (
o o
c
o o
o r
o D , ~.
Diagenet ic grade Opal-A to Opal-CT Rate-controlling factors 241,242 Temperature and time 241 Opal-CT Rate-controlling facto rs Precipitation . . . 244 Opa l-CT to quartz transformation 245 Silica stability fields 246 Transformat ional mechanisms 237 Flint Formation Geochemistry 263 Formation of bedded chert Biogenic siliceous sediments 231 Inorgan ic and replacement cherts Silica diagenesis 253 Origin Hesse, R. 227-251 Petrography Evaporites . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 265 Shallow water Biogenic siliceous sediments 231 Silica phases Solubility diagrams 233 -replacing evaporites Chertification 265 Chertification Carbonates and carbonate sandstones 253 Chert-replacing evaporites 265 Carbonates Quartz fabrics Replacement 257 Select ive replacement 258 Void filling 257 Silica fabric Equigranular 254 Fibrous 254,255 Silica sources 260 Silicification Tim ing 259 Chalk Paramoudra structure 263 Limestones Knauth 's Geochemical Model 261 Cherts Inorganic and replacement Geothermal silica 269 Hesse , R. 253-275 Hydrothermal and volcanogenic cherts Environments 268 Geothe rmal areas 269 Submarine exhalations 269 Lagoona l and lacustrine cherts 267,268 Magadi-type cherts . . . . . . . . . . . . 267 Pedogenic cherts 268 Precambrian cherts 270 * see also chertification 263-275 Silicification of wood 253 Volcanogenic cherts 268
Pelagic stratigraphy Siliceous sediments 229 Chloride comp lexes Metal comp lexes 3 Chlor ite Argillaceous sedimentary rocks (clay minerals) ... .. 181 Choquette , P.w. Limestones ................ . 9-111 - Burial diagenetic environment 75-111 - Meteoric diagenet ic environment 35-73 - Sea-floor diagenetic environment 13-34 Clay minerals Argillaceous sedimentary rocks Burial alteration 179,180 Climate Geologic time variation Limestones - Meteoric diagenesis 64,65 Compaction Model (burial) Dolomitization models 128,132,134 Complex authigenic carbonates Early diagenetic pore water/sediment interaction .. . 306 Coniatites Aragonite crusts 22 Connate water ·Definition 2n Pore water diagenesis 2n Continental margin environments Pore water profiles - type 4 294 Controls, processes and products Limestones - Burial diagenetic environment Deep burial 75,76 Convective pore-water circulation flanks Mid-ocean ridges Pore water profiles - type 1 288 Coorong Model (sabkha) Dolomitization models 127,131,133,134 Correlation chart Organic diagenetic indicators 206 Organic maturation indices 206 Corrosion Dissolution and precipitation Biogenic 39 Hydrostatic 39 Mixing 39 Simple 39 Cristabolite Silica 227 Criteria for bur ial alteration features Evaporites 147 Crystal structure Kinetic effects Dolomite 113,117 Crystalline framework fabrics Evaporites Syndeposit ional features 143 Crystallites Precipitated particles 18 Whitings 18
c (
328
C Darcy's Law Water movement 36 Deep burial diagenesis Diagenetic environments . . . . . . . . . . . . . . . . . . . . . . . 10 Deep phreatic zone Limestones - Meteoric setting 39 Deep-sea environments Biogenic siliceous sediments 241 Definition Argillaceous sedimentary rocks (clay minerals) 180 Caliche (calcrete or duricrust) 43 Carbon preference index 216 Carbonate compensation depth 14 1,1n,195 Catagenesis " Cementation 52 Chemical compaction 83 Chert 227 2n Connate water Diagenesis 1 Dolomitization reaction 115,116 Early diagenesis 308 Evaporites Primary depositional 141,142 Secondary diagenetic 141,142 Fluid geothermometry 317 Humin 193 Karst, surface 45 Limestones Burial diagenetic environment Pressure solution 83 Deep burial 75 Metagenesis 1,199 Neomorphism 10 Organic matter Eogenesis 189 Pelagic stratigraphy 229 Phreatic cements 53 Silica 227 Silicon, dissolved 227 Steinmann trinity 230 Telogenesis 199 Vadose cements 52 Vadose zone Zone of gravity percolation 36 Zone of infiltration 36 Deformation features Evaporites Burial alteration features 148 Detrital framework textures Evaporites Syndepositional features . . . . . . . . . . . . . . . . . . . . 143 Devonian Prairie Formation Potash evaporites 157 Diagenesis and porosity Siliciclastic sediments Burial alteration 166
Indicators Organic matter 218 Organic matter and cementation Argillaceous sedimentary rocks (clay minerals) ... 184 Physical Chert Biogenic siliceous sediments 247 Siliciclastic sediments Fluid flow 171 Geological controls . . . . . . . . . . . . . . . . . . . . . . . . 170 Heavy oil and tar sands 171 Hydrocarbon reservoirs 171 Diagenetic Classification Eogenesis 2 Mesogenesis 2 Telogenesis 2 Cycle Controls and products 5 Processes Porosity 3 Pressure solution 3 Environments Carbonate diagenesis 9 Deep-burial diagenesis . . . . . . . . . . . . . . . . . . . . . . 10 Meteoric diagenesis 9 Sea-floor diagenesis 9 Grade Host rock lithology Biogenic siliceous sediments 242 Opal-A to Opal-CT rate-controlling factors Chert - diagenetic silica transformations ... 241-242 Opal-A to Opal-GT, temperature and time Chert - diagenetic silica transformations ..... 241 Opal-CT - deep-sea environment Biogenic siliceous sediments 241 Opal-CT - mineralogic phases 241 Opal-CT rate-controlling factors, precipitation Chert - diagenetic silica transformations ..... 244 Opal-GT to quartz transformation Chert - diagenetic silica transformations 245 Models Organic matter 218 Pathways 10 Signature Fluid composition 9 Transport mechanism Diffusion Organic matter oxidation 285 Low to intermediate sedimentation rate Organic matter oxidation 286 Pore water profiles 286 Diatom ooze Siliceous ooze 227 Diffusion-controlled Gradient Pore water profiles - type 2 290
C C
C C C C C C C C
C C C
C C C C
C C C
C C C C C <; C C C
C C E l C C C C C l
C
( 329
(
(
( (
( (
(
Pore-water trends Pore water profiles - type 2 288 Disruption and destruction of sedimentary structures Evaporites Burial alteration features 148 Dissolution Carbonate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Fabric selective 20 Limestones - Burial diagenetic environment 93 and precipitation Corrosion Biogenic 39 Hydrostatic . 0 0 0 39 Mixing 0 0 0• . • . . • . 00. 00 0. 0.39 Simple 0 00 0" 0. 0.. 0" 0 0 39 Incongruent dissoution 00... . •... 00 42 Limestones - Meteoric setting 39 Dissolution-reprecipitation features Evaporites Syndepositional features 144 Mechanism Opal-CT - mineralogic phases . . 0. . . . •. . .. .• . 237 Dolomicrite 0 131,133 Dolostones, ancient Dolomite Chemical composition . 000.. 0... • 0........•• 0. 113 Crystal structure Kinetic effects . 0. . . . . .... 0.... 0.... 0.. 0113,117 Dolomitization Chemistry and precipitation . . . . . 0• 113-123,125-139 Dolostones 0 0... • . •. . . . . . . . .. 113-123,125-139 Morrow, DoW. 0" 0 0 113-123,125-139 Natural environment Stability 0. ... •. . . . •.... 0 00. . 115 Dolomitization 0. . • 0 0. • . . 125 Factors for process Ancient dolostones 00..• 0 0.. 0. •.... 0. . 130 Sucrosic or secondary replacement 131,132 Chemistry 0. .• •. . 0. 0• 0119 Sulphate ions Volume problem .. 00 0. . 0. 0.... •... 00. 0120 and precipitation Dolomite . . 0• . . . 0.. 0. 0... •.. 113-123,125-139 Models ... 0.... •. 00. •.... • .. . 00. 0.. 0. 0. 125-139 Burial compaction . 00. . 0....•... 128,132,134,135 Coorong 0..• 00" .. 0 0 00 127,131 Dorag 0. .. . • . 0. 113,261 Evaporite pumping . 0..• • 0. • 0. 0.. 0. 0. . . . • . . 113 Hydrothermal convection 129,132 Mixed-water or Dorag 0. 0. 128,131,132,134,135 Organogenic sea floor 0.. 0. 00. 0• . • . . . . 129 Recognition factors Ancient dolostones . 0.. 0. . . . . 0. . . . . . . . . . . 130 Reflux. 0 0 00 o' 125,130,132 Solution-Cannibalization 000. 0 0.•. 0. . 129,130 Reaction Definition 0. 0 0• . . 0 115,116 0
o o ( o o o o o o o
o o
(
r:
r:
•
•
•
•
0
••
•
0
•
0
•
Dolostones 0 0 113-123,125-139 Dolomite 0 Ancient Dolomicrite 0 0.. 0.. 131,133 Dolomitization . 0. • . . 0. 0•. 00•.. 0. . . • •. . . . • . 130 Megacrystalline white dolomite .. 0 132 Dorag Model Dolomitization models Dripstone Sinter . 0
00
113,261 0
0
0
0
0.. 50
Early diagenesis 0 0 00 00 308 Definition Organic matter oxidation Carbonate reduction zone . . 0. 0. . . . . 00... 281,283 Fermentation zone 00. 0... 0. . 000. . . . • . . . . . . 283 Nitrate reduction zone 0. . . . . . . . . • . . . . 0. . 282 ,283 Oxidation zone 000.. • 00.. 00 00 00 0281 Sulphate reduction zone 281,283 Thermocatalytic reaction zone '" 00 00 00283 Pore water/sediment interaction Authigenic carbonates Calcite and siderite . . . 000. . . • . . 0. • . . . . . 0304 Authigenic dolomite Deep sea dolomite 00......... •... 000... . 305 Complex authigenic carbonates 306 000. 0. . 0 2n-316 Hesse, R. 0...• 0 High sedimentation rate basins . . .. 294 Reactions .. 00... 0.... 00. 0. 00... • 0. 00• . 301 Sulphide precipitation . . .. . 0.. .. . 0 0302 Ion exchange reactions and adsorption . 00 0306 Rock-water interaction 9,2n Sulphide precipitation Framboidal pyrite . 0 0.... 0. . . . 0. . 302,303 Iron monosulphides 0 00. 0.. • . . . . 0... • . 302 Manganese sulphide 304 0
Electron paramagnetic resonance Geochemistry of kerogen
••
•
0
•••
•
•
•
•
0
•
•
•
0... • 0... •. 217
Eogenesis and catagenesis Porphyrins and isoprenoids . .... 0..... ... . . • 213 Carbon preference index 00. .. .•. . • • • . • . . • 216 Organic diagenesis Diagenetic classification 2 Organic matter Diagenesis 0 0.. 0 189,191,211 0
Evaporative pumping Sabkha Model
0
•
•
•
••
•
•
•
•
•
•
•
•
•
•
•
Evaporites Burial alteration Features Criteria Deformation features . 0. 00 0. . . . . . Disruption and destruction of sedimentary 0...• 0 structures Intersediment growth of nodules Polygonal mosaic textures 0. .. 00 0 0. 0
•
••
127
147 148 148 151 148
c (
330
C Textures Fluid inclus ions 154 Intrasediment growth of euhedra 149 of nodules 151 Secondary replacement Pseudomorph 151 Dehydration pair 151 Non-reaction pair 151,153 Chert Petrography 265 Definition 141 Primary deposition 142 Secondary diagenetic . . . . . . . . . . . . . . . . . . . . . . 142 Dissolution at depth 301 Pore water profiles - type 8 Lowenstein, T.K 141-163 Potash deposits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 141 Primary versus secondary features Pumping model Dolomitization models 113 Spencer, R.J 141-163 Texture Foam structure . .. .. . .. .... .• . . .. . .... . .. . 148 Polygonal mosaic 148 Syndepositional features Cements 146 Cavity cements 153 Crystalline framework fabrics 143 Detrital framework textures . . . . . . . . . . . . . . . . . . 143 Dissolution-reprecipitation features 144 Sedimentary structures 142 Exokarst Karst 45 Fabric selective dissolution Dissolution 20 Factors for process Dolomitization 125 Limestones - Burial diagenetic environment Deep burial diagenesis Extrinsic factors 77 Intrinsic factors 76 Pressure solution 87 Burial depth 87 Clay mineralogy 87 Early dolomitization 88 Liquid hydrocarbons 88 Metastable mineralogy 87 Pore water composition 87 Fibrous calcites Cement petrography 26 Fibrous cements Cement petrography 26 Flachkarren Karren 48 Flint Siliceous sediments 227
Flowstone Sinter Fluid composition Diagenetic signature Fluid geothermometry Aqueous geothermometry Mineral saturation index geothermometer Na smectite (beidellite) Na/K and Na/K-Ca Reservoir diagenesis Silica Chalcedony geothermometer Definition Geothermal systems Geothermometry in sedimentary systems Mineral equilibria Fluid inclusions Evaporites Burial textures Fluorescence microscopy Organic diagenesis Methods of quantifying Formation water Pore water diagenesis Terminology Foscolos, A.E. Argillaceous sedimentary rocks Fossil fuel resources Organic matter Fossil record Sea-floor lithification Frittenkarren Solution grooves Gardner, D. Geothermometry in sedimentary systems . .. .. Gas hydrates Pore water profiles - type 5 Isotope fractionation Salt fractionation Geochemical Aspects Limestones - Meteoric diagenesis Cathodoluminesence Minor elements Minor elements CL zones Minor elements strontium Stable carbonate (calcite) sediments Stable isotopes Classification of early diagenetic environments Geochemistry Criteria Ancient dolomitization recognition factors Fossils Biological markers Methods Organic diagenesis Methods of quantifying Biological markers
50 9 319 321 321 320 321 319 320 317 317 317-323 317
154
206
278 177-187 189-204 23 46 317-323
( (
C ( ( (
C
C C C
C C C C C C C C
C C C C
C 296 296
60 60 60,61 61 60 63 61 307 133 4
211
C
C C
C C
C C <...
e c l <...
l (
o -,
r:
331
(
c
c C' c (
c (
(
o o o o o o o o o o o o
o o o o
o o o o (
l; r
c c.
Carbon preference index 216 Extractable organic compounds 211,216 Light hydrocarbon analysis 216 Total extractable organic matter 216 Geochemistry Chert Flint Formation 263 Geochemical classification of early diagenetic environments . . . . . . 307 Kerogen 216 Pyrolysis 216 Carbon isotope studies 217 Electron paramagnetic resonance 217 Hydrogen index 217 Infrared spectroscopy 217 Oxygen index 217 Proximate analysis 218 Ultimate analysis 217 Silicification Cow Head Group 265 Geochronology Kerogen Carbon isotope studies 217 Knauth's Geochemical Model 261 Geothermal Silica Inorganic and replacement cherts 269 Systems Fluid geothermometry 317 Geothermometry in sedimentary systems Abercromb ie, H. . 317-323 Fluid geothermometry 317-323 Gardner, D 317-323 Hutcheon, I. 317-323 Ghyben-Herzberg princ iple Water table 37,39 Glaebules Caliche nodules 44 Gulf Coast fluid redistr ibution model Argillaceous sedimentary rocks 177 Hesse, R. Chert Origin 227-251 Inorganic and replacement cherts 253-275 Diagenesis of biogen ic siliceous sediments 227-251 Early diagenetic pore water/sediment interaction 277-316 High sedimentation rate reactions Basins Early diagenetic pore water/sediment interaction 294,301 Sulphide precipitation Early diagenetic pore water/sediment interaction . . . . . . . . 302 Humin Definition 193 Hutcheon , I. Geothermometry in sedimentary systems ... .. 317-323 Sandstone diagenesis . . . . . . . . . . . . . . . . . . . . . 165-176
Siliciclast ic rocks . . . . 165-176 Hydration shell Magnesium-bearing carbonate 117 Hydrocarbon generation Chemical structure 196 Hydrogen index Geochemistry of kerogen 217 Hydrothermal Activity and intrusion igneous dykes and sills Pore water profiles - type 9 301 and volcanogen ic cherts Inorganic and replacement cherts Environments 268 Geother mal areas 269 Submarine exhalations 269 Convection Model Dolomitization Models 129,132 Smokers Pore water profiles 288 Illite Definition Argillaceous sedimentary rocks (clay minerals) ' " 180 Incongruent dissoution Dissolution and precip itation . 42 Infrared spectroscopy Geochemistry of kerogen 217 Inside grains and skeletons Limestones Sea-floor cementation styles . . . . . . . . . . . . . . . . . . 18 Intersediment growth of nodules Evaporites Burial alterat ion features 151 Interstitial waters Pore water diagenesis Terminology 278 Intrasediment growth of euhedra Evaporites Burial textu res 149 of nodules Evaporites Burial textures 151 Ion exchange reactions and adsorption Early diagenet ic pore water/sediment interaction . .. 306 Isotopes Oxygen and carbon Ancient dolomitization recognition factors 135 Isotopic composition of precipitates Chemical signatures 17,20 Platform carbonates 17,20 James, N.P. Limestones 9-12 - Burial diagenetic environment 75-111 - Meteoric diagenetic environment 35-73 13-34 - Sea-floor diagenetic environment . . , Jasper Siliceous sediments 227
c (
332
Kamenitza Solution basins 46 Kaolinite Argillaceous sedimentary rocks (clay minerals) .. ... 181 Karren Flachkarren 48 Klutkarren 48 Rundkarren 48 Solution sculpture 46 Karst Exokarst . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Subsurface Calcareous sinter (speleothems) 50 Calcareous tufa 50 Lenticular zone processes Dissolution 50,51 Paleokarst 51 Speleothems 50 Syngenetic karst . . . . . . . . . 51 Vadose zone 48 Water table zone and phreatic zone 49 Processes Dissolution 49 Precipitation 49 Surface Definition 45 Facies Water-controlled rock-air interface 43 Landforms 48 Solution sculpture 46 Kerogen Colouration Organic diagenesis Methods of quantifying 207 Thermal alteration index . . . " 207 Geochemistry 216 Geochronology Carbon isotope studies . . . . . 217 Organic matter Catagenesis stage 195,196 Van Krevelen diagrams 4,217 Klutkarren Karren 48 Knauth's Geochemical Model Chertification Limestones 261 Geochronology 261 Lagoonal and lacustrine cherts Inorganic and replacement cherts 267,268 Lapies Solution sculpture 46 Lead-zinc deposits Megacrystalline white dolomite 132 Lenticular zone Meteoric setting 37 Processes Dissolution Subsurface Karst 50,51
Limestones Choquette, P.w. James, N.P Burial cements Origin Autocementation Hydrocarbons . . . . . . . Passive cementation Burial diagenesis Calcite Bladed-prismatic calcite . . . Models Origin of burial cements Passive cementation Autocementation Burial diagenetic env ironment Basinal burial domain Bur ial cements Deeper burial Shallow burial marine Choquette, P.w. Deep burial Controls , processes and products Definition Factors Extrinsic factors Intrinsic factors Mesogenetic realm Phreatic Thermobaric realm Dissolution James, N.P. Origin of burial cements Pressure solution Definition Factors Burial depth Clay mineralogy Early dolomitization Liquid hydrocarbons Metastable mineralogy Pore water composition Processes Chemical compaction Physical compaction Pressure solution Products Recognition of burial cements Isotopic composition Salinity Temperature Shelf/platform burial domain Meteoric diagenesis Choquette, P.W Climate Geologic time variation Geochemical aspects Cathodoluminesence
C C 9-111 9-111
93 93 92
89 99 92 92 99 89 88 75-111 75,76 75
n 76 75 75 75 93 75-111 92 83 87 87 88 88 87 87 83 79 . 83 76,79,99 90 91 91 90 100 35-73 64,65 60 60
c
C C C C C C C C
C C C
C C
c c c c c c c c c ~
c
c c c c e c c l l l. (
(\ (
333
c c ( ( .
C Ct ( .
( (
o o o o o o o o o o o o c. o o o o o (
l l 1'~.
r'
Minor elements 60,61 CL zones 61 Strontium 60 Stable carbonate (calcite) sediments 63 Stable isotopes 61 James, N.P. 35-73 Mineral-eontrolled cementation processes 52 Alteration of calcite 57 of magnesium 54 Cement stratigraphy 54 Diagenesis Freshwater-seawater mixing 59 Vadose versus phreatic settings 57 Phreatic 53 Vadose 52 Original mineralogy 35 Recognition of products 63 Setting Deep phreatic zone 39 Dissolution and precipitation .... .. . .. . .. . . • 39 Mixing zone . . . .• . . . . . . . . . . . . . .. . . . .. . . . 37 Phreatic 35 Vadose 35 Water-controlled 43 Sea-floor diagenesis Biological alteration 21 Cementation styles 18 18,19 Between particles-platform Inside grains and skeletons 18 Neomorphism 20 Shallow enclosed basin 19 Slope and deep basin 19 Choquette,P.W 13-34 James, N.P. 13-34 Lithification * see also Sea-floor lithification 23 Strandline diagenesis 21 Zonation Deep water 13 Latitude 14 Shallow 13 Vertical .... . ... • • •..... ... . . ........ . . • 13 Liquid hydrocarbons Autocementation 88,92 Lithification * see also Sea-floor lithification Limestones - Sea-floor diagenesis 23 Lowenstein, T.K. .Evaporites 141-163 Lublinite Carbonate petrology Caliche . ..... . . .• • .. . . . .. . . . . . ... .. .. .. 45,50 Lussatite Silica 234 Lutecite Silica ....•......... •................... 227,255
Lydite Argillaceous chert 227 Lysocline Vertical zonation 14 Magadi-type cherts Inorganic and replacement cherts 267 Magadilite Silica 227,266,267 Magnesium-bearing carbonate Hydration shell . . . . • . . . . . . . . . . . . . .. ... . .... . . 117 Mcllreath, I.A. Introduction 1-7 Megacrystalline white dolomite Dolostones, ancient 132 Lead-zinc deposits 132 Mesogenesis Diagenetic classification 2 Mesogenetic realm Limestones - Deep-burial diagenetic environment . . . 75 Metagenesis Definition ... . .. •.•. . ... ........... ..... ... 1,199 Metal complexes Chloride complexes 3 Metamorphism Diagenesis 1 Meteoric Diagenesis Diagenetic environments 9 Setting Lenticular zone ....... ... ............. •.... 37 Limestones Phreatic ..• .• .. . . .. .. . . . .. .. . . . . . . • ... . 35 Vadose 35 Water table 36 Water Influx Pore water Diagenesis Terminology 278 Profiles - type 6 300 Methods of quantifying Organic diagenesis Microbial metabolism Bacterial sulphate reduction Microcodium Carbonate petrology Caliche
205 191,193
45
Mineral-eontrolled cementation processes Limestones - Meteoric diagenesis Alteration of calcite 57 Alteration of magnesium-calcite .......... . . •.. 54 Cement stratigraphy 54 Diagenesis Freshwater-seawater .. . . ... . .... .. . . . .. •• 59 57 Vadose versus Phreatic Phreatic ............................. •.... 53 Vadose 52
(
C
334
( Mineral equilibria Fluid geothermometry Mineral transportation in pelitic sediments Argillaceous sedimentary rocks Burial alteration Mineralogy of precipitates Platform carbonates Mixed layer system Argillaceous sedimentary rocks
317
178 15,16
Buri~atteration 180 Mixed-water or Dorag Model 128,131,132,134,135 Dolomitization models Mixing zone Limestones - Meteoric setting 37 Models Diagenesis models Organic matter 218 125-139 Dolomitization models Solution-film model 88 Modern offshore basins Pore-fluid evolution Diagenesis regimes 279 Morrow, D.W. 113-123,125-139 Dolomite Introduction 1-7 Natural environment stability 115 Dolomite Neomorphism Definition 10 Limestones - Sea-floor diagenesis Cementation styles 20 Novaculite Siliceous sediments 227 Numerical models Graphical methods Organic matter diagenesis 219 Oceans Dissolved silica Silica budget 228 Oil window Organic matter Catagenesis stage . . . . . . . . . . . . . . . . . . . . . . . . . 197 Oligocene Rhine Graben Potash evaporites . . . . . . . . . . 156 Opal Silica 227,232,254 Opal-A Physical characteristics Chert Burial diagenesis 232 Opal-CT Mineralogic phases Chert Burial diagenesis 236 241 Diagenetic grade Dissolution-reprecipitation mechanism 237 Quartz transformation 237 Transformation mechanisms 237
Physical characteristics Chert Burial diagenesis
C 234
Ophiolite-associated chert Biogenic siliceous sediments
231
Ophiolite suite Steinmann tr inity
230
Organic Diagenesis Barnes, M.A. 205-226 Barnes , W.C 205-226 Bustin , A.M 205-226 Eogenes is and catagenesis Isoprenoids 213 Porphyrins 213 Carbon preference index 216 Methods of quantifying 205 Fluoresence microscopy 206 Geochemical methods 211 Biological markers Carbon preference index 216 211,216 Extractable organ ic compounds Light hydrocarbon analysis 216 Total extractable organic matter 216 Kerogen colouration 207 Vitrinite reflectance 205 Porphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Diagenetic indicators 206 Correlation chart Matter Barnes , M.A. 189-204 Barnes, W.C. . 189-204 Bustin, A.M. . 189-204 Burial diagenesis Aromatic compounds 200 ,205,211 Catagenesis 195 Metagenesis 199 Telogenesis 199 Diagenesis Catagenesis 195,196 Kerogen Oil window " 197 Eogenesis .. ...... .......•.... ... 189,191,211 Definition 189 Indicators . . .. .... .. .. ... .. ..... .. . •... 218 Models 218 Numerical models Graph ical methods 219 Paleogeothermal gradients . . . . . . . . . . . . . 219 191 Sources Terminology 190 Fossil fuel resources 189-204 Oxidation Diagenetic transport mechanism Diffusion 285 Low to intermediate sedimentation 286 Early diagenetic Carbonate reduction zone 281,283
C C C C C C
C C C
C C
C C
C
c c c c c c c c
C E
C C
C C
c e (
l
C l
C (. (
c c
335
(
c (
C.I C
c..
c (
(
()
o
o o o o (1 o o c o o o
o o r
o o o (1 (
e , ,-'
,'-.'
'--.' r-
Fermentat ion zone 283 Nitrate reduction zone 282,283 Oxidation zone 281 Sulphate reduct ion zone 281,283 Thermocatalyt ic reaction zone . . 283 Maturation indices Correlation chart 206 Organogenic Sea-Floor Model Dolomitization Models 129 Origin of bur ial cements Hydrocarbons 93 Limestones - Burial diagenetic environment 92 Passive cementation 92 Autocementation 93 Original mineralogy Limestones - Meteoric diagenetic environment . .. . 35 Ostwald processes Ostwald ripening 238,246 Quartz transformation 238,246 Oxygen index Geochemistry of kerogen . . . . 217 Paleogeothermal gradients Organ ic matter diagenes is " 219 Paleokarst Subsu rface karst 51 Paramoud ra structure Chertification of chalk 263 Pedogenic cherts Inorganic and replacement cherts 268 Silcrete . .. . . . . . • . .. . . . . . . . . . . . . . . . . . . .. . . . . 268 Pelagic Environments Pore water profiles High sedimentation rate 291,294 Stratigraphy Definition 229 Pelitic sediments Argillaceous sedimentary rocks 178 Permian Salado Formation Potash evaporites 157 Petrographic and textural criteria Anc ient dolomitization recognition factors 131 Petrography Caliche lithologies 44 of precipitates Aragonite Platform carbonates . . . . . . . . . . . . . . . . . . . . . . 17 Magnesium calcite Platform carbonates . . . . . . . . . . . . . . . . . . . . . . 16 Phreatic Cements Definition . . . . . . . . . . . . . . . . 53 Limestones - Deep-burial diagenetic environment 75 Phytokarst Solution sculpture 46
Platform carbonates Precipitates Isotopic compos ition Mineralogy Petrography of aragonite of magnesium calcite
17,20 15,16 17 16
Polygonal mosaic textu res Evaporites Burial alteration features
148
Porcelanite Siliceous sediments
227
Pore water Diagenesis Connate water Sources of data Terminology Format ion water Interstitial wate rs Meteoric waters Saline waters Profiles Biogenic pelagic carbonates Recrystallization Diagenetic transport mechan ism Low to intermediate sedimentat ion High sedimentation rate Pelagic environments Hydrothermal smokers - type 1 Convective pore-water circulation flanks Mid-ocean ridges -type 2 Diffusion-controlled Gradient Pore-water trends - type 3 Reaction-controlled, in subox ic pelagic -type 4 Continental margin environments -type 5 Gas hydrates Isotope fractionation Salt fract ionation -type 6 Meteoric wate r influx - type 7 Thrust faulted sections Active lateral pore-water flow -type 8 Evaporite dissolution at depth - type 9 Hydrothermal activity and intrusion Igneous dykes and sills Sedimentation • see also Early diagenetic pore water
2n 278
2n 278 278 278 278
291,293 286 291,294 288
288
290 288 291 294
296 296 300
300 301
301
c 336
Pore-fluid evolution Diagenesis regimes Modern offshore basins 279 Porosity Depth curves Changes with depth 96,98 Ranges Carbonate sediments 96,97,98 Secondary porosity Siliciclastic sediments 170 Porphyrins Organic diagenesis 213 Potash deposits Bittern salts 155 Evaporites 155 Precambrian cherts Inorganic and replacement cherts 270 Silcrete 270,271 Precipitated particles Crystallites 18 Pressure solution and quartz overgrowths Siliciclastic sediments 166 Stylolites 167 Primary versus secondary features Definition 141 Evaporites Processes Diagenetic cycle 2,3 Porosity 3 Pressure solution 3 Limestones - Burial diagenetic environment Chemical compaction 83 Physical compaction 79 Pressure solution 83 Products 76,79,99 Proximate analysis Geochemistry of kerogen 218 Pyrolysis Geochemistry of kerogen 216 Quartz Crystallinity index Silicification 253 Fabrics Replacement Chertification of carbonates 257 Selective replacement Chertification of carbonates 258 Void filling Chertification of carbonates 257 Transformation Opal-CT Mineralogic phases 237 Ostwald processes 238,246 Quartzine Silica 227,255 Radiolarian ooze Siliceous ooze 227 Reaction-controlled pore-water profiles in suboxic pelagic 291 Pore water profiles - type 3
Recognition Factors Ancient dolomitization Geochemical criteria 133 Oxygen and carbon isotopes 135 Petrographic and textural criteria 131 Stratigraphic and facies criteria 130 Trace elements 134 Ancient dolostones Dolomitization Models 130 of burial cements Limestones - Burial diagenetic environment 90 Isotopic composition . . . . . . . . . . . . . . . . 91 Salinity 91 Temperature 90 of products Limestones - Meteoric diagenetic environment . . . 63 Reflux Model Dolomitization models 125,130,132 Seepage Refluxion 125,131,133 Rhizoconcretions Caliche 44,45 Rillenkarren Solution flutes 46 Rimstone Sinter 50 Rinnenkarren Solution grooves 46 Rock-water interaction Early diagenetic pore water/sediment interaction •. 9,277 Rundkarren Karren 48 Sabkha Model (Coorong Model) Evaporative pumping 127,133,134 Saline waters Pore water diagenesis Terminology 278 Sandstone diagenesis Hutcheon, I. 165-176 Sea floor Diagenetic environments 9 Lithification Cement petrography Epitaxial cement 26 Fibrous calcites 26 Isotopic signature 26 Microcrystalline calcite 26 Scalenohedral cement 26 Spherulitic calcite 26 Stromatactis 26 Fossil record 23 Secondary Porosity Siliciclastic sediments 170 Replacement Pseudomorph dehydration pair Evaporites Burial textures 151
c c c c c c c c c c c c c
c (
c c c
c
c c c
c
c
c c c
c
c
c e l (
( (
C (
r
o c
337
(
c"
o
()
C ( 1 (
r
o o o o o o o o o o r:
o o o o o
o o r
'--"
~"
Pseudomorph non-reaction pair Evaporites 151,153 Burial textures Sedimentary structures Evaporites Syndepositional features 142 Seepage refluxion Reflux Model 125,131,133 Shallow enclosed basin Limestones - Sea-floor diagenetic environment Cementation styles 19 Shelf/platform burial domain Limestones - Burial diagenetic environment 100 Silcrete Pedogenic cherts 268 Precambrian cherts 270,271 Silica Amorphous silica 227 Chalcedony 227,254,255 Cristabolite 227 Definition 227 Lussatite 234 Lutecite 227,255 Magadilite 227,266,267 Opal 227,232,254 Quartzine 227,255 Tridymite 227 Silica budget Oceans Dissolved silica 228 Diagenesis Inorganic and replacement cherts 253 Fabric Chertification of carbonates Equigranular 254 Fibrous 254,255 Phases Characteristics Biogenic siliceous sediments 231 Sources Chertification of carbonates 260 of non-detrital silica Siliceous sediments 227 Stability fields Diagenetic silica transformations 246 Chert Silicates Argillaceous sedimentary rocks Burial alteration 178 Siliceous sediments Flint 227 Jasper 227 Mud 227 Novaculite 227 Ooze Biogenic Depositional environments Modern ocean floor 228
Diagenesis Hesse, R. . Diatom ooze Radiolarian ooze Sponge (spicule) ooze Porcelanite Silica sources of non-detrital silica Silicification of wood Inorganic and replacement cherts Quartz crystallinity index Timing Chertification of carbonates ..
227-251 227 227 227,253 227 227
253 253 259
Siliciclastic sediments Authigenic cementation Burial alteration Diagenesis and porosity Diagenesis Fluid flow Geological controls Heavy oil and tar sands Hydrocarbon reservoirs Pressure solution and quartz overgrowths Sandstones Secondary porosity Silicon, dissolved Definition
168 166 171 170 171 171 166 166,167 170 227
Sinter Dripstone Flowstone Rimstone
50 50 50
Slope and deep basin Limestones - Sea-floor diagenetic environment Cementation styles
19
Solubility diagrams Silica phases Chert
:
Solution Basins Kamenitza Surface Karst Flutes Rillenkarren Grooves Frittenkarren Rinnenkarren Sculpture Karren Lapies Phytokarst . . . . . Surface Karst Solution-cannibalization Model Dolomitization models
233
46 46 46 46 46
'"
46 46 46 46 129,130
Solution-film Model
88
Speleothems Subsurface Karst
50
( (
338
Spencer, A.J. 141-163 Evapor ites , Spherulitic cements Cement petrography 26 Sponge (spicule) ooze Siliceous ooze 227 Steinmann trinity Definition 230 Strandline diagenesis Aragonite crusts 22 Beaches 21 Limestones - Sea-tloor diagenesis 21 Tidal flats 22 Stratigraphic and facies criteria Ancient dolomitization recognition factors 130 Stratigraphy Pelagic Siliceous sediments Cherts 229 Stylolites Pressure solution and quartz overgrowths 167 Sucrosic or secondary replacement Dolomitization Ancient dolostones • . . . . . . .. . . . . . .. . . . . . . 131,132 Sulphate ions 119 Dolomitization chemistry Sulphide precipitation Early diagenetic pore water/sediment interaction Framboidal pyrite 302,303 Iron monosufphides 302 Manganese sulphlde Albandite 304 Syngenetic karst Subsurface karst 51 Tectonics Chert Depositional environments 230 Telogenesis 2,199 Definition Thermal alteration index Kerogen colouration 207 Thermobaric realm Limestones - Deep-burial diagenetic environment . .. 75 Thermocline Vertical zonation 13 Thrust-faulted sect ions Active lateral pore-water flow Pore water profiles - type 7 300 Tidal flats Strandline diagenesis 22 Trace elements Ancient dolomitization recognition factors 134 Transformation mechanisms Opal-GT Mineralogic phases 237 Chert Diagenetic silica transformations 237
Tridym ite Silica 227 Vadose Cements Definition , . . 52 , Zone of gravity percolation Definition 36 of infiltration Definition 36 Subsurface karst 48 Van Krevelen diag rams Kerogen 4,217 Vertical zonation Lysocline , 14 Thermocline 13 Vitrinite reflectance Organic diagenesis Methods of quantifying 205 Volcanogenic cherts Inorganic and replacement cherts 268 Water-eontrolled Precipitation Carbonate sediments 50 Rock-air interface Caliche facies 43 Surface karst facies 43 Water movement Darcy's Law 36 Water table 37,39 Ghyben-Herzberg principle Meteoric setting 36 Zone and phreatic zone Subsurface karst 49 Processes Dissolution Subsurface karst 49 Precipitation Subsurface karst 49 Whitings Crystallites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Zonation Limestones - Sea-floor diagenesis Deep water 13 Latitude 14 Shallow 13 Vertical 13 Zone I: zone of precipitation 15 Warm shallow water platform carbonates 15 Zone II: zone of dissolution/precipitation 15,20 Temperate water shelves, tropical carbonate slopes ' 20 Zone III: zone of active dissolution 15 Deep basinal environments 20 Zone IV: zone of no carbonate 15 Deep basinal environments 20
C C C C
C (
C C C C C C C
C C
C C C
C C C C C (
C
C C C C C C
c
e
(
c l
C C C (