Chapter 20
MAGMATIC CONTRIBUTIONS TO LOW.SULFIDATION EPITHERMAL DEPOSITS Stuart F. Simmons Geothermal Institute, The University of Auckland, Private Bag 92019,Auckland, I,{ewZealand
INrRonucrron Low-sulfidation epithermaldepositsform at <300'C and depths<2 km within meteoric-waterdominated hydrothermal systems that are genetically linked to magmatism at convergent plate boundaries.Despite this strong association with intrusionscausingconvection,the occurrence of magmatic fluids in the low-sulfidation epithermal environment is difficult to detect becauseof scant diagnosticevidence.There are threereasonsfor this: 1 . M a g m a t i c s i g n a t u r e sm a y b e m a s k e db y water-rock interaction along the flowpath separatinga magma from a suprajacentlowsulfidationenvironment. 2. Magmatic contributionslikely wax and wane with time, occurring in some cases in sharp pulses. Consequently,their preservationmay be concealed within complex inter-mineral banding and/or intra-mineral zonation, at centimeterto micrometerscale. 3. Waning hydrothermalactivity may dilute or erasemagmaticsignatures. Despite these geologic processes,the results of a few detailed studies of low-sulfidation epithermal deposits indicate that magmatic contributionsdo exist and are possiblyimportant to ore formation. The salinity, and gas and isotope compositions of fluids in active geothermalsystemsof low-sulfidationcharacter substantiate this pointof view. In this paper, I review the evidence for magmatic contributionsin the formation of six world-class Iow-sulfidation epithermal deposits. drawingon the fluid-inclusionand isotopicstudies
publishedover the last twenty years.Along with being well-studied, the geologic settings and metal contents of the deposits are diverse, spanningthe spectrumof low-sulfidationdeposit types. The focus is on big deposits because, assuming that a cause and effect relationship exists, the potential for finding evidence of magmaticcontributionsand their possiblelink to ore genesis is greatesthere. Accordingly, the well-studiedbut small epithermaldeposits,such as Creede. USA. are not discussed. The framework for interpreting the data from these deposits,however,is basedon studiesof active geothermal systems, where the heirarchy of physical and chemical processes that affect thermal fluids can be directly assessedin a moderngeologiccontext. Therefore,I first review the evidenceof magmaticcontributionsin active geothermalsystems,referringspecificallyto data from the Taupo Volcanic Zone. In discussing low-sulfidation systems, I will emphasize the importanceof time scalesin which both long- and short-termmagmatic contributionsare indicated and will discusstheir relevanceto ore-formation. MAGMATICCon poNENTSIN ACTIVE GnorrrnRIrAL SYSTEMS Epithermaldepositsform in the shallow parts of hydrothermal systems where temperature, pressure,and chemical gradientsfavor efficient precipitation of metalsthroughboiling or mixing, from a deeply derived fluid (e.g, Heald et al. 1987; White & Hedenquist1990). Studies of active geothermal systems (e.g., Giggenbach 1981, 1987, 1988; Henley & Ellis 1983; Hedenquist1986)indicatethat the compositionof this deepfluid is shapedby its circulationhistory:
455
S.F.Simmons
l) incorporationof magmaticvolatilesinto large inputs of meteoric waters at deep levels; 2) water-rock interactionat intermediatelevels;and 3) boiling (stearn and gas loss) and mixing (typicallydilution)at shallowlevels(Fig. l). The aqueousand gaseouscomponentsthat precipitate as mineralswithin epithermaldeposits,or that are trappedas inclusionsin minerals,can thereforebe influencedby one, or a combinationof three By consideringonly those rnajorsetsof processes. which are non-reactive or components conservativeto the gas and liquid phases,the masking effects of shallow and intermediate processes can be deterrnined. Table I lists the componentsfrom which magmatic signaturesin epithermalenvironments can be interpreted,basedon studiesof volcanic gases, geothermal fluids, and hydrothermal minerals. These signaturesinvolve a ratio of components (N2-Ar-He) or isotopes (helium, oxygen,hydrogen).Only for chloride(or salinity) can a magmatic signaturebe inferred from its concentration alone. The most diagnostic signaturesare indicated by helium isotopesand N2-Ar-He,both of which occur in small to trace in amounts,thoughanalysisof thesecomponents not widely applied. minerals has been epithermal In contrast,interpretationof the more abundant and commonly determinedchloride (reportedas of eq. wt. % NaCl from ice melting temperatures fluid inclusions),and oxygen and hydrogen isotopes,provide permissiveevidencethrough construction of internally consistentarguments that establishbaselineandlorend-membercompositionsto interpretdata trends. Theseconcepts are elaborated below by considering these components and their magmatic signaturesin geothermalfluids from the TaupoVolcanicZone.
C\ -Nore fotming environment
@.c
E
II I
A, ,: . : . t , ,i:,t ,
: nagru :
Figure l. Schematic diagram showing the for low-sulfidation environment epithermal mineralizationwithin a magma-related hydrothermalsystem relief (after Henley& Ellis with low topographic 1986). 1983;Hedenquist
(Table 2; Hedenquist1986; Giggenbach1995). Thesedata were correctedfor effects due to steam loss and dilution and therefore represent the compositionof the deep Iiquid prior to boiling. Host rocks comprise a sequenceof Quaternary volcanicsof predominantlyrhyolitic composition that unconformably overlie weakly rnetamorphosed graywacke basementof Mesozoic age (Cole 1990);the relativelyuniform composition of the stratigraphyhere helps in assessingthe contributions of rock-leachableconstiuents to geothermalwaters.Direct evidenceof intrusions is rare (Browne et al. 1992),but their existenceis inferred from the high heat flow associatedwith geothermal activity, and the distribution and occurrenceof volcanicevents.Fluid samplesfrom active andesiticvolcanoesat White Island and Ngaruhoe,at the north and southendsof the TVZ, Geothermul Systemsin the Taupo Volcanic Zone respectively(Fig. 2), provide constraintson the (TVZ) end-membercompositionof magmaticgasesin 'lYZ contains about twenty equispaced the region. The geothermalsystemsassociatedwith calc-alkaline The TYZ geothermal systems also contain magmatism(Fig. 2). About half of thesesystems geochemicalenvironmentsthat are analogousto havebeendrilledto depthsof 500 to 2800 m; the thosededucedto be responsiblefor forming lowdeepfluids dischargedfrom thesewells havebeen suffidation epithermal deposits (e.g., Henley analyzed for rnajor aqueous and gaseous 1985). This perspective is substantiatedby 6180 and 6D components,plus helium-isotopes, occurrenceand zonationDatternsof hvdrothermal
456
L ow-suI/idati on Epi thermaI Deposi ts
Table L Componentsindicating magmatic signaturesin low-sulfidationepithermal deposits magmatic components signature 'r Ili Ra>I r"/H"
evidence diagnostic
source fluid inclusions
temporal resolution poor to moderatc- rcquiresbulk anall'sis of rnaterialpotcntialll'containingrnutiplc generations of fluid inclusions
N, Ar-Hc
IlciAr ..i N.,/lc .."10
6'to
8 to l0
"/o.,
diagnostic('l) chcckfbr manllehclium isotope signature
fluidinclusions
pcrmissive- check
quarlz.calcite(requires good-rcquires a lew tensof milligrarns estimatcof cquilbration matcrial temperaturc,e 9., 7-r): fluid inclusions
with 61,)
6l)
-20 to -80 "/".,
fluid inclusions
poor 1ornodcratc- requircsbulk lnalvsis of materialpo{cntialiycontainingrnutiplc g c n c r r t i o nrs' ! -l l u i d i n c l u s i o n s
> 1 0 . 0 0 0m g / k g permissivconly 1br fluid inclusions rclativclyyoung crust lackingsonnatc brint-sor cr aDoritcs
excellent- individual1'luidinclusionscan be measured
pcrmissivefor data in trendsassesscd conccrtrvith 6r8o valucs
C l lc o n c c n tration
good to moderate- requiresanallsis o1' matcrialpotentiallyconlainingmutiple generations o1'f'luidinclLrsions: individual inclusionssampledbv thermal decrepitation
alteration.the stateof hydrotherrnalfluid-mineral "ore-grade" concenequilibria, occLlrrenceof trations of precious and base metals in precipitates,and relatively high precious-metal fluxes (e.g.,Brown 1986;Brow'ne1969;Browne & Ellis 1910 Browne & Lovering 19'73; Hedenquist & Browne 1989; Hedenquist&. Henley l985at Henley 1985; Krupp & Seward 1987: Simmons et al. 1992; Simmons &. Christenson1994;Weissberg1969).The compositions of deep geothermalwaters are dominated of chloride(up to 2500 mg/kg) by concentrations and dissolved carbon dioxide (up to 20,000 mg/kg), occurrenceof reduced aqueous sulfur pH (e.g.,Ellis species(t{2S,HS-) and near-neutral 1919),with their equilibriumrnineralassemblage comprising quartz,orthoclase("adularia"), illite. chlorite, pyrite and calcite. + epidote and wairakite (Giggenbach 1988). These deeply derived waters also transport relatively large quantities of precious and base metals, which precipitatein hot springs,fractures,and vugs, and in wellhead piping. Of these, the spectacular sulfldescales,containingup to 6 wt. 0%Au and 30
wt. o Ag, on wellhead back-pressureplates at Kawerau, Broadlands-Ohaaki, and Rotokawa demonstrate: 1) that high concentrations of aqueousAu and Ag (-l to 2 pg Au/kg; 8 pg Ag/kg) in chloride waters are possible due to bisulfidecomplexing(Seward 1973 ). and 2) that boiling from about 260 to 180 'C is an efficient mechanismto precipitate Au and Ag (Brown 1e 8 6 ) . Metal fluxes in these three easternsystems (i.e., Kawerau, Broadlands-Ohaaki, and Rotokawa;seeFig. 2) appearhigher than those in the centraland westernsystemsdue to the higher gas (CO2 and H2S) concentrations(Henley & Hedenquist1986). The mass flow for the preexploitation state of Broadlands-Ohaaki is estimatedat 100 kg/sec;thus, about 6500 years are requiredto flux I million oz Au (0.03 x 10" kg) thro"ughthe upflow zone,which coversabout 10 km'(Brown 1986). Shorter periods for equivalentAu flux are inferredfor both Kawerau and Rotokawa because both aqueous sulfide concentrations and massflows are higher (Henley & Hedenquist1986; Krupp & Seward 1987).
S.F. Simmons
White lsland6 of Plenty
^1(;Tarawera
,*Lffin.""n"
-{A"Waiman Mangaki
^"fwuiotuo, lod c""ooro" Ngatamariki Wairakeij
BROADLANDS/ OHAAKI
G E O T H E R M AALR E A S bY drilling Delineated D e l i n e a t ebdY g e o p h Y s i c s Otherthermalareas VOLCANOES
Geothermalsystemsremainactivefor severaltens of thousandsto severalhundredsof thousandsof years(e.g.,Browne 1979;Henley& Ellis 1983), indicating that periods of metal deposition at epithermaldepthscan be relativelyshort-livedand still accountfor very large, high-gradeorebodies of gold (Henley 1985),basedon the concentration and silver directly measuredin the high-gasTYZ systems.
Location map of Figure 2. geothermal systems TVZ described in the text (from Hedenquist1990).
isotopic ratios are compared in the form R/Ra. The MORB R/Ra value of 8-9 is determinedfrom gas analysesof mid-oceanridge basaltsand axial vent hot springs(Craig & Lupton 1981). Deep undepletedmantlemay have much higher R/Ra of 15-30(Lupton 1983). The continentalcrust R/Ra value of <0.1 resultsfrom the decay of uranium and thorium (Gerlinget al. 19'11;Morrison & Pine
1es5).
Volcanic and geotherrnal gases from the have helium-isotopecompositions circum-Pacific Helium Isotopes The large differences in helium-isotope that range from I to 8 times Ra, indicatingthat they contain a significantcomponentof mantlecompositionsfor atmospheric,upper-mantle,and derived helium (e.g., GiggenbacheI al. 1994;' useful He a crustal or radiogenicsources,makes Hilton et al. 1993; Sano et al. 1985, 1981; tracer in volcanic and geothermalgases(Fig. 3). 3He/He Torgersen& Jenkins1982;Torgersenel al. 1982). ratio (Ra) equals1.39+ The atmospheric Mantle helium is thoughtto rise throughthe crust 0.01 (Clarkeet al. 1969;Mamyrin et al. 1969)and is usedas a referencestandardto which all sample with buoyantmelts,exsolvingwith other gasesat
458
Low-suUidation Epithermctl Deposits
Atmosphere Rlta=1
Continental crust R/Fa<0.1
from the Figure 3. Helium isotopiccompositions upper sources, includingatmosphere, mainterrestrial crust. mantle.andcontinental
relativelyshallowdepths(<10 km) wheremagmas crystallizein subvolcanicchambers. Solid rocks 'H. to account for its relacontain insufficient tively high concentrationsin geothermalfluids due to leaching.Therefore,mantle helium must indicatea direct magmaticinput and is considered here to be the single most diagnosticindicatorof magmatic contributionsto geothermalfluids and low-sulfidationenvironments. In the TVZ, geothermalgas samplesrange from 3.5 to 8 R/Ra and vary independentlyof COz, Cl, or other component concentrations (Giggenbach1989; see Table 2). This likely resultsbecauseHe is a noble gas,and becausethe processes affecting He-isotope ratios are decoupledfrom those affectingother components (Giggenbach et al. 1994; Sano et al. 1981;' Torgersenet al. 1982).
Table 2. Chemical characteristicsof TVZ geothermal fluids (Hedenquist1986; Giggenbach1995) l. Broadlands Ohaaki 2. Kawerau
970
4s50
8l 0
3160
3. Rotokawa
850
4 38 0
4. Wairakei
1700
I 820
6 .l
-4.5
-40
-3.75 -3I 7.1
-a.J
u.2
-6
-43 -45
-+J
2400
t275
7.0
-6 5
6. Waimangu
5tl5
l875
7.5
-4 75 -38
7. Waiotapu
1250
t620
7.0
-2.5
-40
2500
2000
4.0
-3 s
-10
>10,000
1720
6.6
+7.l
11
5. Mokai
8. I'okaanu Waihi 9. Whitc Island
Nit r og en-He I i um-Arg on The relative concentrations of N. Ar. and He sampled from active volcanoesassociatedwith mantle hot-spots,crustal rifts, and convergent plate boundariesplot in two distinct groups(Fig. 4) that define mixing trends involving basaltic, andesitic, and meteoric water end-members (Giggenbach1992a).The "andesitic"end-member is characterizedby N2/He values that range from 1700 to 5000. By contrast,the "basaltic" endmember is characterized N2/He values that range from l0 to 220, thoughthe relativeenrichmentin helium may alternativelysuggesta crustal source (Giggenbach1986). Values of He/Ar for these two sourceregionsare closeto the mantleratio of about 3. Air-saturated meteoric groundwater (N2/Ar value about 50) forms the common endmemberto both trendsand indicatesthe influence of the atmosphere (N2/Ar 82) on gas compositions. Distinct enrichmentsin CO2/He valuesfor volcanicgasesalso exist for "andesitic" versus"basaltic"gases,implying that both N2 and CO2 are derived from the same source, which Giggenbach (1992a) speculates is subducted sediment. The N2-Ar-He signatureof geotlrermalfluids from high-temperaturesystemsin Japan (Chiba 19911'Ueda et al. 1991; Yoshida l99l), New Zealand(Giggenbach1992a),and the Philippines (Giggenbach & Poreda 1993) overlaps with volcanic gas data and indicatesthe existenceof deep inputs from either basaltic or andesitic sourceregions in these systems. As nitrogen is rnostly non-reactivein geothermalfluids, as are argon and helium, the relative concentrationsof these three gases complement helium-isotope ratios as a potential tracer of deep magmatic contributions. For example,geothermalsystems locatedalongthe easternmargin of the TYZltave N2-Ar-He compositionsthat plot between the andesitic and meteoric end-members,whereas westernTVZ geothermalsystemshave N2-Ar-He compositionsthat plot between a basaltic and meteoric end-member(Fig. 4). These results indicatethe incorporationof magmaticgasesinto TVZ geothermal fluids. In addition, these signaturescorrelatewith high and low total-gas for the "andesitic" and "basaltic" concentrations systems, respectively, and indicate that the
4s9
S.F. Simmons
And Bas D Geother mal O Volcanic /
Zsn\ \ x
w
/"c.
a
\
\
?-.\
{Qi13j"'ir -7-
of Figure 4. Relativeconcentrations N, Ar, and He in gases fiom andesitic(And) and basaltic (Bas) volcanoes around the world and 'fVZ, all geothermalgasesfiom the i s o t o p i c r a t i o s of having helium >4 (Giggenbach 1992a). R/Ra
a
50 ASW r.'- 20
l O O OH e
2OOAr
magmaticgascomponents,when present,resultin a much higher gas concentrationthan that of meteoric or crustal source (Giggenbach19924' Giggenbachet al 1994). Oxygen and Hydrogen IsotoPes Oxygen and hydrogen are the principal components of geotherrnal waters, and their isotopic compositionsplot close to that of the localmeteoriccomposition(Figs.5a,b),indicating this is the primary sourceof water (Craig 1963). The commonly observedpositive enrichmentsin of 6180and 6D resultfrom one,or a combination, affectingthe composition the threemain processes fluids(Fig. I ). of geothermal of At shallowlevels,the isotopecompositions the to the deep fluid can increaseupon ascent surf'ace through steam loss associated with modes boiling. Both continuousand single-step assessed, of steam loss have been quantitatively assuming adiabatic cooling, and the results indicate that single-stepseparationproducesthe maximum isotope enrichments(Truesdell e/ a/. 1977;Giggenbach& Stewart1982). For example, single-stepsteam loss for a liquid which boils "C producespositive shifts of from 260 to 100 "/oo "/oo 6D (Fig. 5c). 6180 and 10 about 1.5 of compositions the from Subtractingtheseeff-ects surface waters sarnpledfrorir wells and springs
indicates that deeper level processestnust be consideredto explaintlre isotopecompositionsof the pre-boiledparentwaters. Intermediate-levelprocessesinvolve isotope exchange through water-rock interaction aI elevatedtemperaturesthat depletesthe 6l80 and 6D compositionsof fresh rocks and enrich the 6180 and 6D compositionsof geothermalwaters (Craig 1963; Taylor 19731'Blattner 1993). As fresh rocks contain large amountsof oxygen (up to 46 wt. o/o) and relatively small amounts of the 6r80 compositionof hydrogen(< 0.2 wr.o/o), either the rock or meteoric water is rnore easily shiftedthan the 6D compositionas controlledby the bulk water-rock ratio and temperaturedependent mineral-water fractionation factors. These isotopic shifts have been modelled quantitatively,assumingsirnple closed and open systems(e.g., Ohmoto & Rye 1914, Field & -faylor 1986),and more Fifarek 1985; Criss & complex dyrramicallyevolving infiltratiorrfronts (e.g., Blattner & Lassey 1989; Blattner 1993). a simple (1993)has also constructed Giggenbach on hydrothermal model based isotope-exchange alteration as represented by the following reaction: + ll,O + COr -+ K-mica + 2 quartz + K-f'eldspar Ca-f'eldspar -r calcire (1)
Low-sulfidation Epithermal Deposits 6D%"
O local groundwaters O geothermaldischarges I volcaniccondensates -15 6t"o %'u
6D%" 0
aoitino pc
. SMOW
c
,*€ -20
.oG
White lsland \->
rr'i'x\$g2.'P
-40
-ou
-80 -6
-4
1. Broadlands Ohaaki
5. Mokai
2. Kawerau
6. Waimangu
3. Rotokawa
7. Waiotapu
4. Wairakei
8. Tokaanu-Waihi
-2 6180 %,o
Figure 5. The oxygen and hydrogenisotopic compositionsof geothermaland volcanic fluids: (A) geothermalsystemsassociatedwith andesiticvolcanism (from Giggenbach 1992b); (B) TVZ geothermal systemsin Table 2; (C) the oxygen and hydrogen isotopic enrichmentsin liquid due to adiabaticsinglesteamlossfrom 260 to 100 oC (Giggenbach& Stewart 1982). The water-rock interactioncurve in (B) assumesequilibrium isotopic o o/oo; exchangebetweenlocal water (6'tO : -7 looand6D : -45 and K-mica at 300 'C (see Giggenbach1993);dotsrepresent water-rockweightratios.
Giggenbachpointedout that pure water on its own is relatively inert and unlikely to break structural bonds and drive the hydrothermal reactionsrequired for isotopic exchange,hence the relevant inclusion of CO2 as the main acid species,promoting the hydrolysis of feldspars ( s e eC i g g e n b a c h1 9 8 1 ,1 9 8 4 ,1 9 8 8 ) . T h u s ,o n e
mole of CO2 is consumedin convertingCa- and K-feldsparsto K-mica. Accordingly, minimum water-rockratiosinvolvedwith isotopicexchange will range between 0.15 and l, given that H2O1CO2rnole ratios in volcanic fluids are > 5 and in geothermalfluids >30. Using thesevalues in calculations (see Giggenbach1993) indicates
S.F.Simmons oloo m a x i m u m p o s i t i v ee n r i c h m e n t os f a b o u t 9 u/nn 6'nO and 6.5 6D due to water-rockinteraction ( F i g .s b ) . Water-rock interactionhas long beenthought adequate to explain the isotopic enrichments observedin geothermalfluids. However, recent scrutiny of the isotopic compositionsof geothermal waters associatedwith arc volcanism in the circum-Pacificindicatesa common input of magmaticwater to geothermalsystems(Fig. 5a). For these geothermalwaters, both 6180 and 6D values are higher in comparisonwith those in local meteoricwater.The straightlinesconnecting Iocal meteoricand geothermalwaters increasein slope with increasein latitude. The lines form trajectoriesthat point to a common end-member "/oo o/oo compositionof about 9 6'tO and -20 6D, close to the composition of high-temperature dischargesfrom nearby andesiticvolcanoes(Fig. 5a). This so called"andesitic"water(Giggenbach o/on 1992b)has a 6D value about 45 + l 5 heavier than that which was previously estimated for magmatic waters on the basis of analysesof amphiboles arrd micas from eroded plutons (Sheppardet al. 1969 Taylor 1986). The lighter 6D signature results from isotope fractionation betweenwater in a melt and water vapor exsolved during crystallization(Taylor 1986), and the actualcompositionof water in an undegassed melt is likely intermediatein composition(Taylor 1992). The trendsin Figure 5a thus indicatethat rnagmaticwater contributesto some geothermal waters, in proportionsup to 20 %, since isotope enrichmentsdue to water-rock interactionalone cannotexplainthe changein slopesof meteoricgeothermal water trajectoriesas a function of latitude(Giggenbach1992b,I 993). 'fhe stable-isotope compositionsof parentwaters in TYZ geothermalsystems(Fig. 5b) illustrate that resolution of the relative contributionsof intermediate(water-rock interaction)and deeplevel (magmatic and meteoric water mixing) effects is difficult for many cases becausethe strengthsof their respectivesignals significantly overlap. Hence, stable-isotopecompositions provide permissive,but not diagnostic,evidence of a rnagmaticcontribution to deep geothermal waters.
462
Chloride Chlorideconcentrations for geothermalwaters in systems associated with arc magmatism, isolated frorn seawater recharge and lacking evaporitesat depth or a sourceof connatebrines, typically range from about 500-2500 ppm for rhyolite-relatedsystems(e.g., Table 2) and 6000 to 12,000 ppm for andesite-relatedsystems (Hedenquist & Henley 1985b). As is the casefor stable isotopes,the chloride concentrationsof geothermal waters can be related to one, or a combination,of the threemain processes affecting the compositionof geothermalfluids (Fig. I ). Because the meteoric recharge for deep convectivesystemsstafts as almost pure water, the concentrationsof chloride measured in geothermal waters must be derived through processesother than boiling. The magnitudeof boiling effects, however, can be calculated to determinethe parentcomposition(see Henley er al. 1984). For example,a rising liquid that boils adiabaticallyfrom 300 to 100 oC producesabout 40 Yo steamby weight, and concentrates chloride in the liquid by a factorof about 1.7; in this case, 1700mg/kg chloride in a water dischargedat the surface indicates its deep pre-boiled parent contains1000mg/kg chloride. Suchenrichments are likely to be closeto the maximum,given that the temperatureof first boiling in most geothermal systemsrarelyexceedsatlout300 'C. Experimentalresults indicate up to several hundredmg/kg aqueouschloride can be derived through leaching of crustal rocks by pure hot water (Ellis & Mahon 1964. 1967. Fig. 6). Acceptinga minimum bulk water - rock weight ratio of 0.1 for geothermalsystems,maximum chlorideconcentrations of up to severalthousand mg/kg are possibleand overlapwith the rangeof compositions observed in active geothermal systems. Alternatively, magmatic contributions can account for chloride in geothermalfluids, as is indicatedby the concentrationsof up to a few weight percent in high-temperaturevolcanic emissions(e.9.,Hedenquist& Lowenstern1994) and high-salinityfluid inclusions(>35 wt. % NaCl) associatedwith igneous intrusions (e.g., Roedder 1984). The processesrelating to tlre
Low-sulfidation Enithermal Deposits
0.075 400
Ei o O.O5Oo.:g
ct) CD
E o.o,,H z
tr 200 o :
100
-
9E 991 o
g =H EEEEE c - o
r :
!
= c
!
o x (!
3 o
of aqueouschloride Figure 6. Concentrations from differentrocktypesfor a teached experimentally periodof 14 daysat >500 oC,usingpurewaterand water-rockratios* I (Ellis& Mahon1964,1967). transfer of magmatic chloride into the overlying geothermal convective cell are not wellunderstoodbut are likely to be more complexthan can be assumed from simple mixing (e.9., Fournier 1987),and this makes magmatic inputs difficult to quantify. Deep pre-boiled TVZ geothermal waters containup to about2500 mg/kg chloride,whereas the magmatic water from White Islandscontains >10,000mg/kg (Table 2). Even so, the sourceof chloride in thesegeothermalwaters,whetherit be derived from water-rock interaction or direct magmatic input, cannot be distinguishedon the basis of the chloride concentrationalone. The range of chloride concentrations in parent geothermal waters feeding systems across the TYZ (Table 2), which varies by a factor of 5 despitethe passageof the watersthrough similar countryrocks,suggestsa sourceotherthan simple rock leaching. The variation in B/Cl ratios (Giggenbach1995)acrosstheTYZ, and chlorineisotopestudies(Hedenquistet al. 1990), instead point to a deep igneousorigin for chloride.In the absenceof these parameters,only from extreme enrichmentsin chloride(e.g.,>10,000mg/kg) can a magmaticsourcebe inferred. IN EPITHERMAL MAGMATICCONPONNNTS ORn Dnpostts Table 3
the salient geologic charac-
teristics of six well-studied,world-class,lowsulfidationepithermaldepositsfor which evidence of magmaticcontributionsexists. Thesedeposits have been mined for precious and base-metals (Fig. 7), which occur as open-spacefillings in veins and stockwork structuresor within pore spacesamong sedimentaryclasts. The geologic sequencesassociatedwith these deposits are diverse,but the rocksthat host orebodiesare made up of volcanics (Antamok-Acupan, Comstock, Emperor,Tayoltita) and/or underlying and older basement units containing sedimentary and lithologies(Fresnillo,Hishikari). metasedimentary None of the deposits have rock types in their stratigraphythat are likely to containevaporitesor connatebrines. The main geologicfeaturecommon to all of thesedepositsis their close spatial and temporal relationshipto regional magmatism associated with convergentmovementsalong a plate margin. The compositionof nearby igneousrocks may itt fact havea geneticinfluenceon the metal contents of ores (Sillitoe this volume), as is postulatedfbr Emperor, where gold-silver-tellurideores are related to alkalic mafic igneous compositions (Richardsthis volume). lgneousintrusionsoccur at all of the deposits except Hishikari, which formedat shallowestdepth.At Antamok-Acupan, Au (x100)
A=Antamok-Acupan C=Comslock E=Emperor F=Fresnillo H=Hishikari N=NZTvZ T=Tayoltita
Cu+Pb+Zn
Ag (x100)
Figure 7. Temary plot of relative metal contentsof ores producedfrom low-sulfidationdepositsin Table 3. The metal ratios of TVZ geothermal precipitates are similar to the metal ratios of ores from Comstock and Tayoltita.
463
S.F. Simmons
Table 3. Geologiccharacteristicsof low-sulfidationepithermal deposits Approx Age Au Principal areal Orebodies Country Deposit Ma rocks metals extent veins, stock volc-intru0.5 5 (?) Au-Ag AntamokAcupan, Philippines
Associated igneousrock
disseminations; p r e - e x i s i n gh i g h S & porph min
diatreme-sed intermediate (meta)
v e l n s ;p r e existing high-S & porph min veins, stockwork; pre-existing high-S & porph min
volc-intrumetasedmetavolc volc-intr (sed)
intermediatel'elsic
Rct'ercnces omsti e/ al. 1990 Cookc & Bloom 1 9 9 0 ;M i t c h e l l & l , e a c hl 9 9 l ; Sawkins et al. 1979 Vikrc e/ a/. 1988; Vikre 1989
Comstock l . o d e ,U S A
0.28
6.56
Ag-Au
9 km2
Empcror, I'rj i
0 . II
0.04
Au-AgTe
6 km2
Fresnillo, Mexico
0.015
10.0
Ag-PbZn
l0 km2
veins, stockwork mantos,chimneys
-29
sed-volcintru
intermediatcfelsic
Ilishikari, Japan
0 25
0.l 5
Au-Ag
I km2
vetns
-l
scd-volc
intermcdiale
Ahmad et al. 1987; Andorson & Baton 1 9 9 0 ;E a t o n& S e r t e r f i e l d1 9 9 3 ; Kwak 1990, Setterficldet al 1992 G s m m e l fe t a l . 1 9 8 8 . I.angetal. 1988; Macdonald et a/. I 986; RuvalcabaRuiz & Thompson 1 9 8 8 ;S i m m o n s I991 lzawa et al. 1990
Tayoltita, Mexico
03
15.0
Ag-Au
4 km2
veins
volc-intru
intermediate
Smith el al. 1982
malic (alkalic)
rAbbreviations: volc:volcanic, high-S & porph min=high-sulfidationand porphyry-stylemineralization,intru:intrusive, sed:scdimentary, meta:metamorphic.
inclusion data and hydrothermalminerals, with Comstock,and Emperor,prolongedperiodsof "adularia"-illite in hydrothermal activity produced spatially related the occurrenceof sulfidesand every deposit, indicate conditions of mineraliporphyryand high-sulfidationepithermalstylesof zation of 300 to 150 oC from fluids of lowpreceded emplacement of the mineralizationthat sulfidation character, having reduced aqueous Cooke& low-sulfidationores(Comstiet al. 19901. Bloom 1990; Mitchell & Leach 1991; Setterfield sulfur speciesand near-neutralpH, similar to the deep fluid compositions of TVZ geothermal et al. 1992; Sillitoe & Gappe 1984; Yikre et al. 1988: Vikre 1989). Their occurrence,however, systems.These similaritiesprovide the basis for interpretations below. does not influencethe discussion.below. on lowsulfidationores. Chloride Characteristicsof the fluids associatedwith Of all the componentsthat could be used to mineralization are listed in Table 4. Fine are the basis indicatea magmaticsource,chloride, or salinity, mineralparageneses centimeter-scale for fluid-inclusion and stable-isotopestudies, is the easiestto measureby observingice melting except for Comstock, for which the samples temperaturesin fluid inclusions.Given careful investigated were obtained from universities, fluid-inclusionpetrographycombined with these museums,and private collections(Vikre 1989). measurements,salinity variations provide the clearestsign of changein fluid compositionwith Hydrogen-isotope analyses are available for time. The biggest drawback is the uncertainty Antamok-Acupan, Comstock, Emperor, and regarding the salinity, as the ice melting Fresnillo,but helium-isotopesand N2-He-Ar gas (7.) between0 to - 1.5'C may alsobe Fluidtemperature to Fresnillo. analysesare limited mostly
464
Low-sulfidation Epi Iher maI D ep osi Is
Table 4. Characteristicsof mineralizingfluids in low-sulfidationepithermal deposits Deposit
'I'"Cl
Salinity wto/o
l{ydrothcrmal mineralsr
'Hei
6r8oqtz"/oo 618c)"rooloo6Dn,o
l-7
Comstock Lodc.[JSA
250-300
Bmperor.
160-300 4 . 5 - 7
IUl
Frcsnillo, Mexico
200-300
l -l 5
I Iishikari, Japan
I 50-250
He (R/Ra)
- 7 . 8t o L 9
q t z - c a l c - s u l l ' - a d - 1 .I t o 8 . 9 il-chl-mont
qtz-calc-sulf'-ad il-chl-ro-mont
1 6 . 9t o 20.5
qtz-calc-sulf'-adil-chl-mont
l3.9to t7.4
qtz-calc-sulf'-ad 7.3to 9.8 il-chl-mont
-69to -
O ' N e i l & S i l b e r m a n1 9 7 4 ; 1 ' a y l o r1 9 7 3 ;V i k r e e t a / . l 9 8 t ' i ; Vikre 1989
r33 6.5to 12.9
- 1 0t o 58
3.9to'7.7 -30to74
- 5 . 7t o - 0 . 1
Rcf'crcnces e t a l . 1 9 ' 7 9S . i m m o n s1 9 8 6
82
il-chl-ep-mag
n cupan, Philippincs
o/oo
Ahmad et al. 1981, Anderson & Eaton 1990; Eaton & Setterlield 1 9 9 3 ;K w a k 1 9 9 0 ;S e t t e r l l e l de t al 1992
1t o 2
Macdonald et al. 1986, R u v a l c a b a - R u i& z 'l'hompson 1 9 8 8 ;S i m m o n se t a l . 1 9 8 8 . S i m m o n s ,1 9 9 1 lzawaet al. 1990 Matsuhisa& Aoki 1994
- 5 . 8t o I l C h u r c h i l l 1 9 8 0 ;C l a r k e& ' l ' i t l e y qtz-calc-sull'-ad- 3 l to 8.0 1988;Conrad et al. 1992 Smith ab et al 1982 il-chl-aot-eD adularia.calc_ca|cite.ch|_ch|orite.ep:cpidote.i|_i||ite. mag=magnetitc.mont- montmorillonite, q1z:quartz,ro-roscoelite, sulFsulfides, zeol=zeolite l ayoltita, Mexico
250-280
0-13
derived from dissolved gases, mainly COz (Hedenquist & Henleyl9S5b).Thus,the only data solutions can be confidently saline which for o interpretedare those indicating >2.5 eq. wt. NaCl. Furthermore, because of the vagaries associated with fluid-inclusion trapping. ice melting studiesrequirea minimum of five to ten from a few spatiallydistributedbut measurements parageneticallyconstrainedsamplesto be certain of the rangeof fluid comPositions. The salinity data for the six deposits are plotted in Figure 8 and show a rangefrom <1 to 15 eq. wt. o NaCl. The data for Hishikari and Tayoltitaindicatethat the mineralizingfluids were dilute waters,which is typical of the majority of gold-silver epithermal deposits (Hedenquist & Henley 1985b);note that for Tayoltitaa few fluid inclusionscontaining>4 wt. % NaCl exist, and Conrad et al. (1992) suggestthat theserepresent inputs of magmatic fluids. In contrast, m ineralizing fluids for Emperor, Fresnillo, Comstock, and Antamok-Acupan were saline (>2.5 eq. wt. yo NaCl), and the chloride in these systemsmay have a magmatic origin. Only for Fresnillo. however, can saline fluids be
geneticallylinked to mineralization,with Ag-PbZn transportbeingfavoredby chloridecomplexes. Figure9 showssomeof the resultsof a fluidwith the infilling of the inclusionstudy associated -1 m wide Santo Nino vein in the Fresnillo District. Spatial and temporal variations in compositionsindicateepisodicinjectionsof brines (up to 12 eq. vrt. % NaCl) into fracturesotherwise fifled with relatively dilute fluids (-3 eq. wt. yo NaCl). Brines are closely associatedwith sulfide mineralization,occurring primarily in sphaleritehostedfluid-inclusions,whereasthe low-salinity fluid-inclusionsare hostedby barren quartz and calcite.This relationshipis found elsewherein the Fresnillo district, indicating that mineralization was coincidentwith the repeatedintroductiorrof saline liquids. The high salinity is interpretedto have been derived from a brine reservoir of magmatic origin, which was situatedbeneatha dilute geothermal convection cell (Simmons 1991). The rocks in the districtmostlyconsistof a highly deformedgraywacke-argillitesequence overlain by younger rhyolite volcanics,unlikely hosts for connatebrines or evaporites.Nor can simple rock leaching account for the slrarp 465
S.F. Simmons
I I cr
1n
I I
{ B
6(u
:ctl E 40,000E ()
o
z
20,000 contributionofaq. I t carbondioxidetoTm
E i =p g = l g i j
r
F
j
a
E1 l bi o t = g 3 FE EE E b l d
u
r
n
'
!
-
g
Figure 8. Range of salinitiesdeterminedfrom fluid(7.) for epithermal inclusionice melting measurements deposits(Tables3, 4). The shadedregionshowsthe 1,, range for which the freezing-point depressiondue to dissolved salts or carbon dioxide cannot be d i s t i n g u i s h e(dH e d e n q u i s&t H e n l e y1 9 8 5 b ) .
variationsand high salinitiesover time. Further evidence that the high salinities at Fresnillo indicate a magmatic contribution is based on genetic similaritiesfound in a number of other nearby Ag- base-metaldeposits which form a metallogenicbelt that transectsnorthern Mexico (Clark et al. 1982). Other than their contained metals, these deposits share similar ages of mineralization, closeassociationwith magmatism, and occurrencesof saline fluid-inclusions.some of which, basedon their enrichedstableisotopes, very high salinitiesexceeding35 eq. wt. o/oNaCl, and proximity to intrusions,are clearly magmatic in origin (e.9., Rye 1966: Sawkins 1964). Therefore,chloride in the Fresnillo brines seems to be not only of magmatic origin, but was necessary to the introductionof metal-transporting fluids. Orygen and Hydrogen Isotopes Stable-isotope studiesof epithermaldeposits are largely restricted to quartz, calcite, and
SantoNifroVein 425 LevelWest-Central StageI
Stagell
Stagelll
StagelV
O - 6
=z
Es
'i;
F }
r
(J
0 250"
I
E o o E
@ F
**sE;*g;**s o ll o
MEFigure 9. Sharpvariationsin fluid salinitiesover time during formationof the SantoNino vein, FresnilloDistrict; data compiled from observationsof the west-centralsector on the 425 level. Quartz and calcite volumetrically dominatethe vein-fillingsof stagesl, II, and III, for which silver sulfosaltsand base-metalsulfidesare the orebearingminerals;stageIV consistsonly of calciteand is barren.No dataareavailablefor sulfidesin stageIII. Brine pulses are associatedwith deposition of sulfides and sulfosalts in Stages I and II, whereas dilute fluids are with depositionof quartzand calcitegangue.SeeSimmonset al. (1988)and Simmons(1991)for further associated d e s c r i p t i oa n n dd i s c u s s i o n . 466
Low-sulfi dation Ep i t hermaI Deposi ts
Comstock
Tayoltita
Hishikari
htrdn
,r
Fresnillo
{
Figure 10. The shadedregions represent the rangeof quartzoxygen-isotope compositions versus equilibrium temperalure. estimated from fluid-inclusion data, for epithermal deposits (Tables 3, 4). The curvesrepresentcalculatedcompositionsof "/*) local meteoric(-6 to -18 and magmatic o/oo; water (8 in equilibrium with quartz as a function of temperature, based on fractionation factors determined bv Matsuhisaet al. (1919\.
Emperor
,y {
6t8o %n "adularia" gangue, as theseare the most common oxygen-bearingminerals.The 6l8O.in.,ulcompositions are converted to 6lsowatercompositions from mineral-water equilibrium fractionation e q u a t i o n s( e . 9 . , F r i e d m a n & O ' N e i l 1977; Matsuhisaet al. 1979) and estimatedequilibrium temperatures, the latter typically determinedfrom fluid-inclusion homogenizationdata. The minimum range of homogenizationtemperaturesfor any one generationof primary fluid-inclusions from thesedepositsis about20 "9, corresponding "/no to an uncertaintyof about* 0.5 in calculating equilibriumwater compositions.Note that oxygen isotopes can also be measured directly on inclusion fluids (e.g., Yigk et al. 1994),though such results are not available for the deposits discussedhere.
compositionat the time of mineralization.The composition of local meteoric water may be inferredfrom trends in combined6r80- 6D data; in very young deposits,modern meteoric water compositionscan be used. Oxygen-isotopecompositionsof quartz from the six epithermal deposits are plotted versus temperaturein Figure 10. Curvesrepresentingthe isotopiccompositionsof magmatic(6t80 : 8 o/oo) and local meteoric water in equilibrium with quartz are shown for comparison. The 6180 values of fluids from these depositseither plot between meteoric and magmatic compositions (Antamok-Acupan, Comstock, Hishikari, Tayoltita) or overlap with the magmatic composition(Emperor,Fresnillo),with maximum positive enrichmentsrelative to local meteoric Hydrogen-isotope compositionsare measured water that range from 7 o/oo6180 at Hishikari to directly on inclusionfluids releasedundervacuum l8-20o/no 5l80 at Comstock, Emperor, and by crushing; thus, ore-relatedsulfide minerals, Fresnillo.Acceptinga value of about9oloo6180a, along with quartz and calcite, can be analyzed. the maximum possiblefor enrichmentto meteoric The uncertaintiesin bulk analysisare unknown water due to water-rock interaction(as previously becauseseveral fluid-inclusion generationsmay described),suggeststhat the waters at Comstock, be present,but averageresultssuggesta rangeof Emperor,and Fresnillo had minimum magmatic "/on up 20 are possible.The 6D dataare difficult to water inputsof 50o%,and the watersat Hishikari, interpretwithout knowledgeof the meteoricwater Tayoltita, and Antamok-Acupan had no magmatic
467
F S.f-. Simmons
input but instead were derived from shifted rneteoricwater. Alternatively,assurningthat there were no positive enrichmentsdue to meteoric water-'rock interaction.the minimum magmatic inputsinto depositwatersrangedfrorn about 100% at Ilishikari to 90yo at Emperor and Fresnillo. -fhese two models bracketthe range of possible 'Ihe resultsfrom Hishikari,which interpretations.
6D%.
x Emperor indicatechangesin the 6l80 water composition 1 Frssnillo over time. are discussedfurtherbelow. O Antamok-Acupan . Comst@k Matsuhisa& Aoki (1994) used the banded Ryosen of the filling vein quartz-adularia-sLrlfide 6"0 %. 5 vein (average vein grade of 280 g Au/t) compoFlishikarito assesschangesin 6'oo'u,,,.. Figure ll. fhe oxygen and hydrogen isotopic of epithermalfluids for epithermal sitionswith time. Highestore gradeswithin the compositions of localmeteoric (Tables3, 4). Thepositions vein occur in vcry fine-grained sulfide-rich deposits Emperor are based and Antamok-Acupan to the for water parallel and lie near Ginguro bandsthat fbr Comstock whereas compositions, present-day on are rvallrock contacts. Sulfide-poor bands fiom stable-isotope Fresnillothev are estimated conrposedof fine-grainedquartz and ofthoclase and trends. physically. (adularia)that are difficult to separate versus By ingeniouslycomparing6lsOqurnz-aduraria Silberman1974;Sau'kinset al. 1979;Simmonse/ analyzed material tlf the ratio the quartz"-adularia (determined frorn XRD and Al analyses), ,z/.1988;Vikrc 1989). Boiling cannotaccountfor calculated Matsuhisa& Aoki (1994) defined two distinct these trends as indicated by the (Fig. 5c). lineartrends.which in turn were usedto interpret enrichmentsdue to adiabaticsteamloss Here again the Fresnillo and Emperor data plot pure encl-membercompositionsftrr 6'8Our"r,,and closeto the field of magmaticwater, though their associatedwith high-grade(>1000 g E'tO"dul",iu deuteriumvaluesare difficult to distinguishfrom Auit), and low-grade(<20 g Au/t) stagesof vein compositionsof local Ineteoricwaters.A filling; late barren quartz was also analyzed' the 6D much strongerindicationof a magmatic compoOxygen-isotopeequilibriurn temperatureswere nent in ore fluids is suggestedby' the Comstock quartz-adularia pairs, and calculated for data.basedon two different lines of evidence:l) homogenizationtemperatureswere measuredfor the increasein the positive slope of trajectories barren quartz. The results indicate 6'tO*u,., connectingmeteoricand hydrothermalwatersas a with for fluids associated of -0.1o/o., compositions function of increasing 6D, accepting that the Au-Ag mineralization,in contrast^to 618O*u,., source of meteoric water remained constant compositionsof -3.6 and -4.3to -5.7"/uofor fluids (Vikre 1989), the periodof mineralization during associatedwith later low-gradeand barrenstages, with associated one sample and 2) the positionof respectively. Thus a rninimum enrichment of "ore-stage" bonanzasilver ore from the 1200-footlevel ofthe about 3.5"/oo6'tO is indicated for water "gangue-stage" waters, Con Virginia mine, which has a n/oo u/,,,, waters relative to -69 6D 6'*o and composition of 1.9 sLrggestingthat mineralizing fluids contained (O'Neil & Silberman 19'14). Accordingly, the inputsof 30oh-or more.magmaticwater. proportion of magmatic input into Comstock Hydrogen-and oxygen-isotopiccompositions fluids rangesfrom about30 to 75 %o' of waters associatedwith four of the epithermal The wide range of 6D values indicated for deposits(Table 4) are plotted in Figure 11, and both Emperor and Cornstock data sets deserve show broad patternsfurther suggestingthat the further comment in the light of their potential mineralizingfluids were mixturesof meteoricand relation to inputs of magmatic waters. Taylor magrnaticwater (Ahmad el al. 1987; O'Neil &
468
Low-sulfidat ion Ep it hermaI Dep osi ts
$
I I I
i
(1986, 1992)has shownthat the 6D composition of water vapor in equilibrium with a hydrous magma decreases as the fraction of water remaining in the melt decreases,and that the resulting range of 5D water-vaporcompositions dependson the mode by which a hydrousmagma degases.Thus, the 6D compositionsof water and vapor evolvedthroughcontinousopen-system degassingrangefrom -25 to <-125 closed-system o/oo, n/uo respectively,as the fraction and -25 to -60 from 1 to in the rneltdecreases of water remaining 0.1 (Taylor 1986, 1992). The 5D values of Emperor fluids range from - l0 to -58o/oo, overlappingwith magmatic vapor compositions derived from either mode of degassing.The 5D values for Comstockfluids^aremuch lighter and range from -69 to -133"1oo,suggesting that magmaticvaporswere mostly derivedfrom opensystem degassing. The relatively wide range of 6D compositions in these two deposits may processes. thereforerelateto magmadegassing Helium Isotopes Helium isotopes were measured for vein materials obtained from subsurface mine workings (>100 m depth) at Fresnillo and Antamok-Acupan (Table 4). Inclusion fluids hosted by quartz, calcite, and sulfides were releasedunder vacuum by thermal decrepitation or crushing10- to 20-g monomineralicsamples; their isotope compositionswere measuredon a Nier-type double focussing mass spectrometer. '7 at AntamokThe R/Ra values, between 6 to Acupan and between I and 2 at Fresnillo (Fig. 12), indicate a component of mantle He, presumablytransportedto shallow crustal levels by ascendingmagmas(Simmons1986;Simmons et al. 1988). The Fresnilloresultsseemlow for arc-related fluids, and they may result from three different lHe processes:1) preferentialoutwarddiffusionof "He; over 2) in-.situradiogenicaccumulationof -He *H"; or 3) radiogenic accumulation of associatedwith long magma residencetimes and crustal contamination.Of these, the last seems most likely, given the relativelythick continental crust through which Fresnillo magmasmigrated (see Simmonset al. 1988). The helium-isotope
O a
A Fresnillo O Baguio
(g
E CE
tAtA a t m o . p h " r ihc " l i u r l
6 He concentration (x 10 ) cc STP/g Figure 12. Helium-isotopecompositions(R/Ra) versus heliumamountfor inclusion fluidscontained in quartz,calcite,andsulfidesfrom Fresnillo,andquartz (Simmons1986;Simmonset from Antamok-Acupan a/.1988).
ratios of geothermalemanationsassociatedwith arc volcanismand similarly thick crustrangefrom 1.30to 2.16 R/Ra in the SouthernVolcanicZone. andl.44to 6.47 R/Ra in CentralVolcanic Zone of the Andes, thus supporting this interpretation (Hilton et al. 1993). For Fresnillo, the results further indicate relatively uniform compositious throughtime irrespectiveof mineral host or fluidinclusion salinities, confirming that processes governinghelium input are decoupledfrom those affecting other fluid components (including metals), consistent with observationsof TVZ geothermal fluids and data from other (Simmonset al. 1987). hydrothermaldeposits Note that investigatorswishing to pursueHeisotope analysesof epithermal materials should ensurethat their samplesare shielded from the effects of cosmic radiation,especiallyat higher elevations of about 1500 m asl or more, as 'H" "un be produced by nuclear cosmogenic reactions involving spallation or neutron by'Li (e.g ,Kurz 1986). Sucheffects absorption are interpretedfor surfacevein-quartz samplesat 2200 m asl from the Fresnillodistrict,which have anomalousvaluesexceeding100 R/Ra (Simmons 1986; Simmons et al. 1986). The observed exponential decreasein R/Ra with depth (from
r S.F.Simmons
I l6 R/Ra at the surfaceto 65 R/Ra at l.l m depth) along with calculations of the cosmic-ray attenuationlength in rocks,however,indicatethat these radiation effects are unlikely to penetrate depths greater than about l0 m at Fresnillo (Simmons1986;Simmonsel al. 1986).
Fresnillo
2000 N, /He
N2-Ar-He Ratios Problemsassociatedwith the analysisof gas speciesin f'luid inclusionsare relatedto lossof H2 and H2S (through diffusion and post-extraction reaction;see Graney & Kesler this volume) and these artefactsalter the redox state calculatedfor er publisheddata(e.g.,Roedder1984;Hedenquist al. 1992). However, such problems should not of N, Ar and He, thoughthere affect measurement are few available measurementson epithermal materials.Norman & Musgrave (1994) reported data from three epithermaldeposits,includingthe SantoNiflo vein, Fresnillo(Fig. l3). Gaseswere for measuredby a quadrupolemass spectrometer decrepithermal vacuum by under fluids released tation or by crushingof 0.1 to 5 g of inclusionbearing material.The two smaller depositsfrom New Mexico (not\shown) both have gas trends "basaltic" signature.The indicating a possible Fresnillo data (Benton 1991)form a broad linear pattern that roughly overlaps with the mixing envelope having andesitic and meteoric endmembers. These results are consistent with salinity, and stable- and helium-isotope data, which support the existence of magmatic contributions in the Fresnillo fluids. Unfortunately,data from coexistingfluid inclusionsand stable isotopes are unavailable to assess covariationsor furtherconstrainthis interpretation'
1000
100 Air
looo He
2oo Ar
of nitrogen,argon Figure13.Relativeconcentrations and heliumin gasesfrom Fresnilloinclusionfluids 1994). (Benton1991,Norman & Musgrave
one component's origin can be checked for internal consistenciesby comparing it to other componentsthat behavein a similar manner(e.g., oxygen and hydrogenisotopescan be comparedto chloride,and helium isotopescan be comparedto COz, N2, and Ar). ln epithermal deposits, temporal constraints are relative as determined from the mineralogicrecord, with millimeter to centimeter-scale parageneses restricted to distancesof a few hundredmetersor less;hence, interpretation of spatial variations in the compositionsof paleo-fluidsat a fixed point in time acrossa depositis extremelydifficult. Data quality is also restrictedby the errors inherent to analyses and interpretationsof minerals and inclusionfluids. Thus, active geothermalsystems provide a scale of comparison for spatial and DISCUSSION temporal relations and a framework for interpretation not available from study of lowIn this paperI haveattemptedto documentthe deposits. sulfidation main geochemical evidence which indicates With these caveats,helium isotopesand N2magmatic contributions to low-sulfidation epiAr-He ratios can provide the most diagnostic thermal environments. The strength of this evidenceof magmaticcontributionsto epithermal evidenceis bolsteredby studiesof activesystems' deposits;the few availabledata indicatethat these where fluid compositions from geothermal componentsare promisingtracersof fluid origins systems in different stagesof evolution can be and deserve much further investigation. In comparedon a regionaland global scale.Besides contrast,enrichmentin both oxygen and hydrogen this temporal constraint,the capacityto analyze isotopes relative to meteoric water, and high of of chloride,can providepermissive concentrations all fluid componentsmeansthat interpretations
410
Low-suffidationEpithermalDeposits evidence of magmatic contributions(Table 1). Currently,Fresnillo is the only depositfor which all of thesetechniqueshave been applied,though there is a gap in the continuity of samples investigatedfrom this deposit, and the isotope studiesare reconnaissance in scale. Although there is evidence of magmatic contributions,it shouldbe clearthat the dominant source of water entering most low-sulfidation epithermalenvironmentsis meteoric,but this is not the issue here. Instead, a number of researchers have summarily discounted the possibility of magmatic contributionsin magmarelated ore-forming hl,drothermal systems, arguingthat water-rock interactionis sufficientto explain enrichmentsboth for oxygen and hydrogen isotopes,and the origins of othercomponents, includingmetals(e.g.,Taylor 1973 Campbeller al. 1984; Seal & Rye 1992),notwithstandingthe fact that a few percentmagmatic water could also accountfor the same isotopic enrichments(e.g., O'Neil & Silberman 1974 Sawkinset al. 1979). The problem then relates to the framework of interpretation, and I believe this involves appreciation: I ) of the nature of magmatic components, 2) that magmatic components potentiallycontributeto ore formation,and 3) that magmatic contributionscan reach shallow epithermalenvironments. The nature of magmatic componentsis best understood from examination of degassing volcanoes and study of porphyry ore deposits (".9., Hedenquist & Lowenstern 1994). These componentsare mostly volatile and includewater, carbondioxide, chlorine (as HCI), sulfur (as SO2 and H2S), and base and preciousmetals, all of which are observedin low-sulfidationepithermal environments,with chlorine and sulfur being important for metal transport. The signatures which record the appearance of magmatic componentsin the epithermal environment are mostly restricted to those in Table l. Other potential tracers, such as the isotopic compositionsof carbon,sulfur, and lead,are commonly ambiguousdue to effectsrelatingto redox stateor crustalcontamination,and are difficult to interpret (seeHedenquist& Lowenstern1994). That magmaticfluids can reachand influence low-sulfidation epithermal environments is
probably better documented than mosr geo_ scientists realize, with a much clearer magmatic connection existing for high-sulfidation epithermal environments(Arribas this volume). At one extremeare the eruptionsof magma through geothermal systems, which in recent history includethe 1886eruptionof Mt. Tarawerain New Zealand (Simmons el a/. 1993), the 19?6-1977 eruption at Krafla in Iceland (with the first recordeddischargeof magma from a geothermal well: Larsenet al. 1979),andthe l99l eruptionof Mt. Pinatubo,Philippines,formerly a geothermal prospectof the PhilipineNationalOil Company (with two pre-eruptionexplorationwells: Delfin et al. 1992). At the other extreme is the evidencefor steady influx of magmatic components(helium, nitrogen, chlorine, and water) into geothermal systemsand epithermaldeposits.Theseextremes also representend-memberson a time-scaleof influence,one nearly instantaneous, from hoursto days, and the other continuous,over hundredsto tens of thousandsof years. At time-scales in betweenare the pulses of fluid that reflect the magmaticinputs inferredfor Fresnillo,Hishikari, and Comstock. These sharp changes in fluid compositionshave not been recognizedin active systems,though wells have only been monitored for a maximum of about35 years(e.g.,Wairakei). Fluid pulsesare also known from the mineralogic record of some active geothermalsystems; for example,618O.ut.it" valuesat Kawerauindicatethe former field-wide presenceof a carbon dioxide and 6r8o-enrichedthermal fluid (up to 5 "/un comparedto currentvaluesof -3.75 ''ln,,)of likely magmaticorigin (Christenson1989). Thus both transient and persistent influxes of maglnatic contributionsare possible. Only at Fresnillo, for which high-salinity brines of rnagmaticorigin are interpreted,can a causeand effect relationshipbetween magmatic inputsand mineralizationbe considered.For other deposits,such as Hishikari and Comstock. the availability of metal-transporting ligands in mineralizingfluids cannotbe assessed by current analyticaltechniquesand, therefore,any genetic link betweenmagmaticinputs and mineralization is inferred only by spatial associationbetween precious-metaloccurrenceand isotopicallv(6180
S.F. Simmons
and 6D) enriched gangue minerals. Even in the 'lVZ. where relatively high concentrations of "gassy" precious metals are being fluxed in magmatic containing lluids geothermal contributions, the cause and effect relation is ambiguous as the source of aqueous sulfur, assuming it accounts for the aqueous Au and Ag, cannot be traced. The ultimate source of metals in low-sulfidation epithermal deposits thus remains poorly understood. ACKNOWLEDGMENTS I thank J. W. Hedenquist, J. Margolis, R. Sherlock, and J. F. H. Thompson fbr their perceptive comments and criticisms of an earlier version of this paper, and thank Louise Cotterall, who drafted some of the Figures. REFERENCES AHMAD, M., SOLOMON, M. & WALSHE, J. L. (1987): Mineralogicaland geochemicalstudiesof the Emperor gold telluride deposit, Fiji' Econ Geol.82.345-370. A N D E R S O N .W . B . & E A T O N , P . C ' ( 1 9 9 0 ) : G o l d mineralizationat the Emperor mine, Vatukoula, Fiji. .t. Geochem.Explor. 36, 267-296. BENTON, L.B. (199l): Compositionand Source of the Hydrothermal Fluids of the Santo Nino Vein, Fresnillo, Mexico, as DetLrminedfrom'-S'loS', Stable Isotope and Gas Analysis M.S' thesis, New Mexico Tech. Soccoro,New Mexico, USA' "Andesiticwater":a phantom BLATTNER, P. (1993): of isotopic evolution of water-silicatesystems' E a r t h P l a n e t .S c i .L e t t . 1 2 05 l l - 5 1 8 . BLATTNER, P. & LASSEY, K. R' (1989): Stable isotopeexchangefronts,Damk6hlernumbers,and fluid to rock ratios.Chem.Geol. 78 381-392' B R O W N , K . L . ( 1 9 8 6 ) :G o l d d e p o s i t i o nf r o m g e o t h e r mal dischargesin New Zealand.Econ. Geol Sl 9'79-983. B R O W N E , P . R . L .( 1 9 6 9 ) : S u l f i d em i n e r a l i s a t i oinn a Broadlandsgeothermaldrill hole' Taupo Volcanic Zone, New Zealand. Econ. Geol- 64 156-159'
4'/2
BROWNE, P.R.L. (1979): Minimum age of the Kawerau geothermal field, North Island, New Zealand.Volcanol.Geotherm.Res.6 213-215. B R O W N E ,P . R . L .& E L L I S , A . J . ( 1 9 7 0 ) :T h e O h a a k i Broadlands geothermal area, New Zealand. mineralogyand relatedgeochemistry.Am. .1. Sci. 26997-131. B R O W N E ,P . R . L . ,G R A H A M , I . J . ,P A R K E R , R . J .& WOOD, C.P. (1992): Subsurfaceandesitelavas and plutonic rocks in the Rotokawa and Ngatamarikigeothermalsystems,Taupo Volcanic Zone, New Zealand. Volcanol. Geolherm. Research51 199-217. B R O W N E , P . R . L . & L O V E R I N G ,J . F . ( 1 9 7 3 ) :C o m position of sphalerites from the Broadlands geothermalfield and their significanceto sphalerite geothermometryand geobarometry. Econ. Geol. 68381-387. , . (1984): C A M P B E L L , A . , R Y E , D . & P E T E R S E NU of the San study isotope A hydrogenand oxygen for the role of Implications CristobalMine, Peru: genesis of for the the water to rock ratio 7 9 , l 8 l 8 1 8 3 2 . G e o l . . con. w o l f r a m i t ed e p o s i t sE C H I B A , H . ( 1 9 9 1 ) : A t t a i n m e n to f s o l u t i o na n d g a s equilibrium in Japanese geothermal systems. G e o c h e mJ.. 2 5 3 3 5 - 3 5 5 . C H R I S T E N S O N ,B . W . ( 1 9 S 9 ) : F l u i d i n c l u s i o na n d stable isotope studies in the Kawerau hydrothermalsystem,New Zealand: Evidencefor past magma-fluid interactionin the active system. /n Water-Rock InteractionWRI-6 (D.L Miles, ed.). Balkema,Rotterdam,155-158. CHURCHILL, R. K. (1980): Meteoric WaterLeaching and Ore Genesis ut the Tuyoltita Silver-Gold Mine, Durango, Mexico. Ph.D' thesis, Univ. Minnesota,Minneapolis,Minnesota. C L A R K , K . F . , F O S T E R ,C . T . & D A M O N , P . E . ,1982): Cenozoic mineral deposits and subduction-relatedmagmatic arcs in Mexico' Geol. Soc.Am. Bull. 93,533-544. C L A R K E . M . & T I T L E Y , S . R . ( 1 9 8 8 ) :H y d r o t h e r m a l evolution in the formation of silver-gold veins, Tayoltitamine, San Dimas district,Mexico. Econ. Geol. 83, 1830-I 840.
Low-sulf dation Epi therma l Depos its
C L A R K E , W . 8 . , B E G , M . A . & C R A I G ,H . ( 1 9 6 9 ) : tH. Excess in the sea: evidence fbr terrestrial primordialhelium. Eurth Planet. Sci.Lett.6,213220.
ELLIS, A.J. & MAHON, W.A.J. (1964):Natural hydrothermal systems and experimental hotrvater/rock interactions. Geochim. Cosmochim. Acta 28, 1323-1357.
C O L E , J . W . ( 1 9 9 0 ) : S t r u c t u r acl o n t r o la n d o r i g i n o f volcanismin the Taupo Volcanic Zone, New Zealand.Vol can ol o,st,52 445-455.
ELLIS. A.J. & MAHON, W. A. J. (1967):Natural hydrothermal systems and erperirncntal hotwater/rockinteractions(ll\. Geoch i m. (-ot ntoch rm. A c t a3 1 . 5| 9 - 5 3 8 .
C O M S ' I I ,M . E . C . , V I L L O N E S ,R . I . J R . .D E J E S U S . C . V . , N A I ' I V I D A D , A . R . , R O L I , A N ,L . A . & D I . J R O Y ,A . C . ( 1 9 9 0 ) : M i n e r a l i z a t i o ni n t h e , h i l i p p i n e sf l:u i d K e l l y g o l d m i n e ,B a g u i od i s t r i c t P i n c l u s i o n a n d w a l l r o c k a l t e r a t i o ns t u d i e s . . , / . Geochem.Explor. 35 341-362. C O N R A D , M . E , , P E T E R S E N ,U . & O ' N E I L , J . R . (1992): Evolutionof an Au-Ag-producinghydrothermal system: The Tayoltita mine, Durango, M e x i c o . E c o n .G e o l .8 7 . 1 4 5 1 - 1 4 1 4 . C O O K E , D . R . & B L O O M , M . S . ( 1 9 9 0 ) :E p i t h e r m a l and subjaccntporphyry mineralization,Acupan, B a g u i od i s t r i c t ,P h i l i p p i n e sa: f l u i d i n c l u s i o na n d parageneticstudy. ./. Ger.tchent. Explor. 35, 297340. C R A I G , F l . ( 1 9 6 3 ) :T h e i s o t o p eg e o c h e m i s t royf r v a t e r and carbon in geothermal areas. /n Nuclear Geologyon GeothermalAreas(E. Toneiorgi,ed.). S p o l e t ol.t a l y , l 7 - 5 3 . 'He a n dm a n t l e C R A I G , I I . & L U P T O N ,J . E .( 1 9 8 1 ) : volatilesin the oceanand the oceancrust.In The S e a( C . E m i l i a n i ,e d . ) ;7 , 3 9 1 - 4 2 8 . '|AYLOR, H . P . ( 1 9 8 6 ) :M e t e o r i c C R I S S ,R . E . & lrydrothermalsystems.Reviewslllineral. l6 373424. D E L F I N , F . J . ,J I T . S , U S S M A N ,D . , R U A Y A , R . G . & of the REYES, A. G. (1992): Hazardassessment system:clues prior Pinatubovolcanic-geothermal to the June 15, 199I eruption. Geotherm.Res. CottncilTruns.16,519-521. , . N. (1993): The EATON, P. C. & SE'f'fERFIELDT relationship between epithermal and potphyry hydrothermalsystemswithin the Tavua Caldera, F i j i . E c o n .G e o l .8 8 , 1 0 5 - l - 1 0 8 3 . E L L I S , A . J . ( 1 9 7 9 ) :E x p l o r e dg e o t h e r m asl y s t e m s..I n Geochemistryof HydrothermalOre Deposits,2nd e d .( H . L . B a r n e se, d . ) ,6 3 2 - 6 8 3 .
IrlELD, C. W. & FIFAREK,R. H. (1985):Lrght stable-isotopesystematics in the eptthermal environment.ReviewsEcon.Geol. 2.99-l)8. : o n c e p t u anl r o d e l so f b r i n e F O U R N I E R ,R . O . q 1 9 8 7 )C evolutionin magmatichydrothermalsysten;:;. I-,1S Geol.Sun. Prof Paper 1350. 1487-1506 F R I E D M A N , I . & O ' N E I I - ,J . R . ( 1 9 7 7 ) :C o m p i l a t i o n of stable isotope fractionation lactors of geochemical interest. U.S. Geol. Surv. Pro/. [)aper 440-KK. G E M M E L L , J . B . , S I M M O N S ,S . F . & Z A N I ' O P , H . ( 1 9 8 8 ) : l ' h e S a n t o N i f l o s i l v e r - l e a d - z i nrcei n , Fresnillo district, Zacalecas,Mexico: l)afl l. Structure,vein stratigraphyand mineralogy.Ecor. Geol.83,Ls97-l6l8. GERLING, E. K.. MAMYRIN. B. A. & TOLSTIKHIN,L N. (1971): Helium isotope compositionin some rocks. Geochem.Inlernat. 8, 755-761. G I G G E N B A C F I ,W . F . ( 1 9 8 1 ) : G e o t h e r m a lm i n e r a i equilibria. Geochim. Cosmochim.Acta 15, 3934t0. G I G G E N B A C H ,W . F . ( 1 9 8 4 ) :M a s st r a n s l e ri n h i d r o thermal alteration systems - A conccptual approach.Geochim. Cosmochim..lcta 48, 2693-
27n. GIGGENBACH, W.F. (1986): The use of gas chemistry in delineating the origin of fluids dischargedover the Taupo Volcanic Zone.. [n ProceedingsSymposium V., Internat. VolcanofogicalCongress,Auckland,New Zealand, 47 -50. GIGGENBACH, W.F. (1987): Redox processes governing the chemistry of fumarolic gas dischargesfrom White Island, New Z.ealand. AppliedGeochem.2, 143-161.
S.F. Simmons
G I G G E N B A C H , W . F . ( 1 9 8 8 ) : G e o t h e r m a lsolute equilibria: Derivation of Na-K-Mg-Ca geoindicators. Geochim. Cosmochim.Acta 52, 2',7492 76 5 .
HEDENQUIST, J. W. (1986): Geothermalsystemsin the TaupoVolcanicZone:their characteristics and refation to volcanism and mineralisation. Bull. RoyalSoc.New Zealand23,134-168.
G I G G E N B A C H , W . F . ( 1 9 8 9 ) : P r o c e s s ecso n t r o l l i n g CO, and Cl contents of thermal dischargesfrom the Taupo-Rotorua volcanic-magmatic hydrothermal system, New Zealand. In \Nater-Rock Interaction WRI-6 (D.L Miles, ed.). Balkema, Rotterdam,259-262.
H E D E N Q U T S T J, . W . & B R O W N E , P . R . L . ( 1 9 8 9 ) : The evolutionof the Waiotapugeothermalsystem, New Zealand,basedon the chemicaland isotopic composition of its fluids, minerals and rocks. Geochim. Cosmochim. Acta 53. 2235-2251
cIGGENBACH, W.F. (1992a): The compositionof gases in geothermaland volcanic systemsas a function of tectonic setting. In Water-Rock InteractionWRI-7. Balkema.Rotterdam.873-878. GIGGENBACH, W.F. (1992b): Isotopic shifts in waters from geothermal and volcanic systems along convergent plate boundaries and their o r i g i n .E a r t h P l a n e t .S c i .L e t t . l l 3 , 4 9 5 - 5 1 0 . G I G G E N B A C H , W . F . ( 1 9 9 3 ) :R e p l y t o c o m m e n tb y "Andesitic water": a phantom of P. Blattner: isotopic evolution of water-silicate systems.Earth Planet.Sci.Lett. 120,519-522. GIGGENBACH, W.F. (1995): Variations in the chemical and isotopic composition of fluids dischargedfrom the Taupo Volcanic Zone, New Zealand, J. Volcanol. Geotherm. Research (accepted). GTGGENBACH,W.F. & POREDA, R.J. (1993): Helium isotopic and chemical composition of systemsin the gasesfrom volcanic-hydrothermal Philippines.Geothermics22, 369-380. GIGGENBACH, W.F., SANO, Y. & WAKITA, H. (1994): Isotopiccompositionof helium and COt and CHo contents in gases producedalong the New Zealandpart of a convergentplateboundary. Geochim.Cosmochim.Acta 57,3427 -3455.
H E D E N Q U I S T ,J . W . , G O F F , F . , P H T L L T P SF, . M . , ELMORE, D. & STEWART, M.K. (1990): Groundwaterdilution and residencetimes, and constraintson chloride source in the Mokai geothermalsystem,New Zealand,from chemical, '"Cl data. Geophys. stable isotope, tritium and Research95, 19365-1937 5. HEDENQUIST,J.w. & HENLEY, R.w. (1985a): Hydrothermal eruptions in the Waiotapu geothermalsystem,New Zealand: Their origin, brecciasand relationto preciousmetal associated deposition.Econ.Geol.80, 1640-I 668. : he H E D E N Q U I S T J, . w . & H E N L E Y , R . W . ( 1 9 8 5 b ) T effect of CO, on freezing point depression measurements of fluid inclusions:Evidencefrom active systems and application to epithermal studies.Econ.Geol.80, 1379-1406. H E D E N Q U I S T J, . W . ,R E Y E S ,A . , S I M M O N S ,S . F .& TAGUCHI, S. (1992): The thermal and geochemicalstructureof the epithermalenvironment: A frameworkfor interpretingfluid inclusiondata. Eur. .1.Mineral. 4, 989-1015. H E D E N Q U T S TJ,. W . & L O W E N S T E R N ,J . B . ( 1 9 9 a ) : The role of magmas in the formation of hydrothermalore deposits.Nature 370, 519-527. HENLEY, R. W. (1985): The geothermalframework for epithermalsystems.ReviewsEcon. Geol. 2, l1 A
GTGGENBACH, W.F. & STEWART, M.K. (1982): controllingthe isotopiccompositionof Processes steam and water dischargesfrom steamvents and steam-heated pools in geothermal areas. G e o t h e r m i clsl , 7 l - 8 0 .
H E N L E Y , R . W . & E L L I S , A . J . ( 1 9 8 3 ) :G e o t h e r m a l systems,ancient and modem. Earth Sci. Reviews
HEALD, P., FOLEY, N.K. & HAYBA, D.O. (1987): Comparative anatomy of volcanic hosted epithermaldeposits:acid-sulfateand adularia-sericite tvoes.Econ. Geol. 82 l-26.
HENLEY, R.W., TRUESDELL, A,H., BARTON, P.B., JR. & WHITNEY, J.A. (198a): Fluid* Mineral Equilibria in Hydrothermal Systems. ReviewsEcon.Geol. 1.267 o.
r 9 ,t - 5 0 .
Low-sulJidati on Ep i t herm a I Deposi ts
HILTON, D.R., HAMMERSCHMIDT, K., TEUFEL, S . & F R I E D R I C H S E NH , . ( 1 9 9 3 ) :H e l i u mi s o t o p e characteristicsof Andean geothermal fluids and lavas. Earth Planet. Sci.Lett. 120.265-282. HTLTON, D.R & HEDENQUTST,J.W. (1936): Introductionto the geochemistryof active and fossil geothermalsystems.1r Monograph Series on Mineral Deposits26, GebriiderBorntraeger. ]ZAWA, E., URASHIMA, Y., IBARAKI, K., SUZUKI, R., YOKOYAMA, T., KAWASAKI, K., KOGA, A. & TAGUCHI, S. (1991): The Hishikari gold deposit: high grade epithermal veins in Quaternaryvolcanicsof southernKyushu, Japan..l.Geochem.Explor. 36, l-56. KRUPP, R. E. & SEWARD, T. M. (1987): The Rotokawa geothermalsystem,New Zealand: an active epithermal gold-depositingenvironment. Econ. Geol. 82, 1 109-1129. K U R Z , M . D . ( 1 9 8 6 ) : C o s m o g e n i ch e l i u m i n a terrestrialrock. Ilature 320, 435-439. KWAK, T. A. P. (1990):Geochemicaland temperature controlson ore mineralizationat the Emperorgold mine, Vatukoula, Fiji. .J. Geochem. Explor. 36, 297-338. LARSEN, G., GRONVOLD, K. & THORARINSSON, S. (1979): Volcaniceruptionthrougha geothermal bore hole at Ndmalall, Iceland.Nature 278,701710. LANG, 8., STEINITZ, G., SAWKINS, F. J. & S I M M O N S , S . F . ( 1 9 8 8 ) :K - A r a g e s t u d i e si n t h e Fresnillo silver district. Zacatecas.Mexico. Econ. Geol.83,1642-1646. LUPTON, J. E. (1983): Terrestrialinert gases:isorope tracerstudiesand cluesto primordialcomponents in the mantle.Ann. ReviewsEarth Sci. ll. 3714t4. M A C D O N A L D , A . J . ,K R E C Z M E R ,M . J . & K E S L E R . S.E. ( 1986): Vein, manto and chimney mineralization at the Fresnillo silver-lead-zinc m i n e ,M e x i c o .C a n .J , E a r t h S c i .2 3 , 1 6 0 3 - 1 6 1 4 . M A M Y R I N , B . A . , A N U F R I E V , G . S . ,K A M E N S K I I , I . L . & T O L S T I K H I N , I . N . ( 1 9 6 9 ) :D e t e r m i n a r i o n of the isotopic composition of atmospheric
helium. Geochem.Internat. 7 , 4gg-505. M A T S U H I S A , Y . & A O K I , M . ( 1 9 9 4 ) :T e m p e r a r u r e and oxygen isotopevariationsduring formation of the Hishikari epithermal gold-silver veins, southern Kyushu, Japan.Econ. Geol. 89, l60g1613. M A T S U H I S A ,Y . , G O L D S M I T H ,J . R .& C L A Y T O N . R.N. (1979): Oxygen isotopefractionationin the system quartz- albite - anorthite-water.Geochim. Cosmochim. Acta 43, 1131-l 140. M I T C H E L L , A . H . c . & L E A C H , T . M . ( 1 9 9 1 ) :E p t thermal Gold in the Philippines; lsland Arc Metallogenesis,Geothermal Systemsand Geology. AcademicPress,London,457p. M O R R I S O N ,P . & P I N E ,J . ( 1 9 5 5 ) :R a d i o g e n i o c rigin of the helium isotopes in rock. New york Acad. Sci.Annals 62,11-92. N O R M A N , D . l . & M U S G R A V E ,J . A . ( 1 9 9 4 ) :N , - A r He compositionsin fluid inclusions:Indicatorsof fluid source. Geochim. Cosmochim. Acta 58. lll9-ll3l. O H M O T O , H . & R Y E , R . O . ( 1 9 7 4 ) : H y d r o g e na n d oxygen isotopiccompositionsof fluid inclusions in the Kuroko deposits, Japan.Econ. Geol. 69, 947-953. O'NEIL, J.R. & SILBERMAN, M.L. (1974):Stable isotope relations in epithermalAu-Ag deposits. Econ. Geol. 69, 902-909. ROEDDER, E., editor (1984): FIuid Inctusions. ReviewsMineral. 12,644 p. R U V A L C A B A - R U I Z ,D . & T H O M P S O N ,T . ( I 9 8 8 ) : Ore deposits at the Fresnillo mine, Zacatecas, Mexico.Econ.Geol.83, 1583-1596. RYE, R. O. (1966):The carbon,hydrogenand oxygen isotope compositionof the hydrothermalfluids responsible for the lead-zinc deposits at Providencia,Zacatecas,Mexico. Econ. Geol. 61, 1399-1427. S A N O ,y . , N A K A M U R A , y . & W A K I T A , H . ( t 9 8 5 ) : Areal distributionof tH"/oH. ratios in the Tohoku District,northeastern Japan.Chem.Geol.52, l-8. SANO, Y., WAKITA, H. & GIGGENBACH. W.
r S.F. Simmons
( 1987): Island arc tectonics of New Zealand manifested in helium isotope ratios. Geochim Actu. 51, I 855-l 860' Cosmctchim. SAWKINS, F. J. (1964): Lead-zincore depositionin light of fluid inclusion studies, Providencia' Zacatecas.Econ. C eol. 59, 883-9I 9. S A W K I N S , F . J , O ' N E I L , J . R . & T H O M P S O N 'J . M . (1979):Fluid inclusionand geochemicalstudiesof v e i n g o l d d e p o s i t s ,B a g u i o d i s t r i c t ,P h i l i p p i n e s . Econ.Geol. 74. | 420-1434. S E A L . R . R . t l & R Y E , R . O . ( 1 9 9 2 ) : S t a b l ei s o t o p e study of water-rock interactionand ore formation, Bayhorsebaseand preciousmetal district,Idaho' E c o n G e o l .8 7. 2 1 1 - 2 8 1. & MUSSET, A.E. SETTERFIELD, T.N., and (1992): Magmatism R.D.J. OGLETHORPE, associated hydrothermal activity during the aoAr-3eAr dating evolution of the Tavua Caldera: events' hydrothermal and intrusive of volcanic, E c o n .G e o l . 8 7 .I 1 3 0 - l1 4 0 ' S E W A R D , T . M . ( 1 9 7 3 ) :T h i o c o m p l e x e so f g o l d a n d the transporlof gold in hydrothermalore solutions' (leochirn.Cosmctchim. Acta 31,319-399' S T I E P P A R DS, . M . ,N I E L S E N ,R . L . & T A Y L O R ' H . P . (1969): Hydrogen and oxygen isotope ratios in minerals fiom porphyry copper deposits.Ecor' G , ' t t l .6 4 . 5 l 5 - 5 4 2 . S I L L I T O E . R . H . & G A P P E , I . M . ( 1 9 8 4 ) :P h i l i p p i n e porphyry copper deposits:geologic setting and characteristics.Committeefor Co-ordination ctl ,loint Prospectingfor Mineral Restturcesin Asian Olfshore Arects(CCOP) Tech Pub. 14, Bangkok, 89 p. Naturc of SIMMONS, S.F. (1986): Physico-Chemical lVifio Santo lhe Minerulizing Solutions Jbr the Hydrogen, l/ein; Resultsfrom Fluid Inclusion, Oxygen and Helium Studies in the Fresnillo District.Zrtctttecas,Mexico. Ph.D. thesis Univ' Minnesota,Minneapolis,Minnesota. S I M M O N S , S . F . ( 1 9 9 1 ) : H y d r o l o g i ci m p l i c a t i o n so f alteration and fluid inclusion studies in the Fresnillo district, Mexico: Evidencefor a brine reservoirand a descendingwater table during the fbrmation of hydrothermalAg-Pb-Zn orebodies'
416
E c o n .G e o l .8 6 , 1 5 7 9 - 1 6 10 . P.R.L.B. &. S.F., BROWNE, SIMMONS, (1992): Active and extinct R.L. BRATHWAITE, New North lsland, the of systems hydrothermal l2l Series, Field Guide Geol. Econ. Zealand.Soc. p. s t M M o N S , S . F & C H R I S T E N S O N ,B . W . ( 1 9 9 4 ) : Origins of calcitein a boiling geothermalsystem. Am. J. Sci.294. 361-400. S I M M O N S ,S . F ,G E M M E L L , J . B . & S A W K I N S , F . J . ( - l 9 3 8 ) : T h e S a n t o N i n o s i l v e r - l e a d - z i n vc e i n , Fresnillo,Mexico: Part ll. Physicaland chemical natureof ore-formingsolutions.Ecttn. Geol. 83, 1619-1641. SIMMONS, S.F, KEYWOOD, M., SCOTT, I].J. & K E A M , R . F . ( 1 9 9 3 ) : l r r e v e r s i b l ec h a n g eo f t h e Rotomahana-Waimangu hydrothermal system (New Zealand) as a consequenceof a volcanic eruption.Geolog,,2l , 643-646. S I M M O N S ,S . F ,S A W K I N S , F . J .& S H L T J I ' T E RD, . J . (1987): Mantle-derivedhelium in two hydrothermal ore deposits,Peru.Nuture 329, 429-432. S I M M O N S , S . F . , S C H L U T T E R ,D . J . & N I E R , A . O . ( 1 9 8 6 ) :C o s m o g e n i a c n d m a n t l eh e l i u m i n t h e S t . Niflo vein, a silver-bearingquartz vein in the Fresnillo District, Zacatecas,Mexico. EOS, T'rans. Am. Geophys.Union 67, 1268. S M I T H , D . M . , J R . ,A L B I N S O N , I . . A . & S A W K I N S , F . j . ( 1 9 8 2 ) :G e o l o g i ca n d f l u i d i n c l u s i o t sr t u d i e s of the Tayoltitasilver-goldvein deposit,Durango, Mexico.Econ.Geol.71, 1120-1145T A Y L O R , B . E . ( 1 9 3 6 ) : M a g m a t i cv o l a t i l e s :i s o t o p i c variationsof C, H. and S. ReviewsMineral. 16, t85-225. TAYLOR, B.E (1992): Degassing of tlrO fiom rhyolite magma during eruption and shallow intrusion, and the isotopic composition of magmaticwater in hydrothermalsystems.Geol' Surv. .lapan RePort279, 190-194. '*O/'uO e v i d e n c ef o r n t e t e o r i c T A Y L O R , H . P .( 1 9 7 3 ) : hydrothermalalterationand ore depositionin the Tonopah,Comstock Lode and Goldfield mining districts,Nevada.Econ.Geol. 68,747-164.
Low-sulfidationEpithermal Deposits
T O R G E R S E N ,T . & J E N K I N S , W . J . ( 1 9 8 2 ) :H e l i u m isotopes in geothermal systems: Iceland, the Geysers, Raft River, and Steamboat Springs. Geochim. Cosmochim. Acta 46. 7 39-748. T O R G E R S E N ,T . , L U P T O N , J . E . ,S H E P P A R D D , .S. & G I G G E N B A C H , W . F . ( 1 9 8 2 ) :H e l i u m i s o t o p e variations in the thermal areas of New Zealand. Volcanol.Geotherm.Research12 283-298. T R U E S D E L L ,A . H . , N A T H E N S O N ,M . & R Y E , R . O . (1977): The effects of subsurfaceboiling and dilution on the isotopic compositions of Yellowstone thermal waters..J. Geoohvs.Research 82.3694-3704. UEDA, A., KUBOTA, Y., KATOH, H., HATAKEYAMA, K. & MATUSBAYA, O. ( l99l ): Geochemical characteristics of the Sumikawa geothermal system, northeastJapan. Geochem. J. 25, 223-244. V I K R E , P . G . ( 1 9 8 9 ) : F l u i d - m i n e r a lr e l a t i o n si n r h e ComstockLode. Econ.Geol. 84. 1574-1613.
VIKRE, P.G , McKEE, E.H. & SILBERMAN, M.L. (1988):Chronologyof Miocenehydrothermaland igneous events in the western Virginia Range, Washoe, Storey, and Lyon counties, Nevada. Econ. Geol. 83. 864-874. VITYK, M.O., KROUSE, H.R. & SKAKUN. H.R. (1994): Fluid evolution and mineral formation in the Beregovogold - basemetal deposit,Transcarpathia,Ukraine.Econ.Geol.89, S4'l-565. WEISSBERG, B.G. (1969): Gold-silver ore-grade precipitatesfrom New Zealand thermal waters. E c o n .G e o l . 6 4 , 9 5 - 1 0 8 . W H I T E , N . C . & H E D E N Q U I S T ,J . W . ( 1 9 9 0 ) :E p i t h e r mal environmentsand styles of mineralization: variations and their causes,and guidelines for exploration.J. Geochem.Explor. 36,445-474. YOSHIDA, Y. ( I 99 I ): Geochemistry of rhe Nigorikawa geothermal system, southwest Hokkaido, Japan.Geochem.J. 25,203-222.
417
