Phanerozoic Evolution of Bolivia and Adja Adjace cent nt Regi Region onss Thierry Sempere Convenio YPFB–Orstom Santa Cruz, Bolivia
Abstract
T
he geologic evolution of Bolivia and the central Andean system during the past 500 m.y. was largely controlled by the geodynamics of the South American margin of western Gondwana. The Phanerozoic strata were deposited in mainly marine environments until the Early Triassic, after which continental environments predominated. However, there were six restricted marine transgressions in the Late Cretaceous–Danian and one in the late Miocene. The Late Cambrian–Early Ordovician margin was initially a passive margin. It became an active one during a Middle Ordovician compressional episode and was controlled by large-scale transtensional or transpressional conditions from the Late Ordovician to the Triassic. The Late Ordovician–Mississippian evolution was characterized by vigorous subsidence of the marine foreland, which was filled with thick, shallowing-upward sequences showing northeastward onlaps. Ashgill and latest Devonian–Mississippian glaciomarine and fluctuating sea level processes are recorded in the succession. Shallow marine carbonates, marls, and sandstones, as well as some evaporites and eolianites, were deposited during Pennsylvanian–Early Triassic time. After Middle Triassic rifting was aborted, the Bolivian basin behaved in a cratonic way until it was caught up in the Andean system due to the onset of transtension along the margin in the Late Jurassic. It became part of the Andean foreland domain in early Senonian time. Andean thrust deformation propagated into Bolivia from the west in the late Oligocene and progressed eastward through Neogene time. Organic-rich units correlate with Paleozoic highstand deposits and younger transgressions. Generation, migration, and trapping of hydrocarbons depended mainly on Cenozoic sedimentary burial and tectonic loading and hence on propagation of Andean deformation.
Resumen
L
a evolución geológica de Bolivia y de los Andes centrales durante los últimos 500 Ma fue controlada en gran parte por la geodinámica del margen sudamericano del Gondwana occidental. Los estratos fanerozoicos se depositaron en ambientes principalmente marinos hasta el Triásico inferior, y predominantemente continentales posteriormente. Sin embargo, seis transgresiones marinas restringidas ocurrieron en el Cretácico superior-Daniano, and una en el Mioceno superior. El margen Cámbrico superior-Ordovícico inferior era inicialmente un margen pasivo. Se volvió activo durante un episodio compresivo en el Ordovícico medio, y estuvo controlado por condiciones transtensionales o transpresionales de gran escala del Ordovícico superior al Triásico. Una fuerte subsidencia del antepaís marino, rellenado por espesas secuencias de somerización que presentan un traslape hacia el noreste, caracterizó el período Ordovícico superior-Mississippiano. Procesos glaciomarinos (Ashgilliano y Devónico terminal-Mississippiano) y fluctuaciones eustáticas se registran en la sucesión. Carbonatos, margas, y areniscas, de agua somera, y algunas evaporitas y eolianitas, se depositaron durante el intervalo Pennsylvaniano-Triásico inferior. Posteriormente al aborto del rifting del Triásico medio, la cuenca boliviana se comportó de manera cratónica, hasta su captura en el sistema andino por la iniciación en el Jurásico superior de condiciones transtensivas a lo largo del margen. Se volvió parte del antepaís andino en el Senoniano inferior. En el Oligoceno superior, la deformación andina empezó a propagarse en Bolivia desde el oeste por medio de cabalgamientos, y progresó hacia el este durante el Neogeno. Las unidades ricas en materia orgánica corresponden a depósitos paleozoicos de alto nivel marino y a transgresiones más jóvenes. La generación, la migración y el entrampamiento de los hidrocarburos han dependido principalmente del soterramiento sedimentario y de la carga tectónica cenozoicos, es decir de la propagación de la deformación andina.
Sempere, T., 1995, Phanerozoic evolution of Bolivia and adjacent regions, in A. J. Tankard, R. Suárez S., and H. J. Welsink, Petroleum basins of South America: AAPG Memoir 62, p. 207–23 207–230. 0.
207
208
Sempere Figure 1—General structure of the area of interest. Study area and Bolivia-Peru basin are shown in insert. Symbols: dotted lines, drainage divides; finely dashed lines, national boundaries; hatchured, western Andes (Sempere et al., 1989, 1990b, 1991); crystalline pattern, Khenayani– Turuchipa paleostructural corridor (CPKT) in the northwest, and northeastern boundary of Chaco Boreal high in the southeast. Structural elements and domains (most are Spanish): CALP, main Altiplanic thrust; CANP, main Andean thrust; CFP, main frontal thrust; FA, Apurimac fault; FLIA, intra-Andean boundary fault (conjectural); SFK, Khenayani fault system; CB, Chapare buttress; CI, Cuzco indenter; CK, Calama knot; MI, Mizque indenter; SB, Susques buttress; VH, Vilcamba hinge; FPCH, Huarina fold and thrust belt; SAB, sub-Andean belt; SBS, Santa Bárbara system; TTB, TarijaTeoponte belt; UCU, UllomaCoipasa-Uyuni domain. Cities: An, Antofagasta; Ar, Arequipa; As, Asunción; Cb, Cochabamba; Ch, Charagua; Cr, Corumbá; P, Potosí; S, Salta; SC, Santa Cruz; T, Tarija.
INTRODUCTION Bolivia is of particular interest because it possesses an almost complete Phanerozoic stratigraphy that has recorded the evolution of the nearby Pacific margin of western Gondwana. New data collected in the past 10 years, combined with the use of modern methods in stratigraphy and basin analysis, permit a better understanding of its regional Phanerozoic geologic history. Bolivia was located on the edge of cratonic Gondwana during most of the Phanerozoic. This geotectonic location explains the high subsidence rates reflected in the great thicknesses of many of the stratigraphic units and the comparatively small amount of magmatism and tectonism prior to Neogene time. The purpose of this paper is to summarize new and mostly unpublished stratigraphic and paleogeographic data concerning Bolivia, to outline the Phanerozoic geologic evolution of the central Andean area, and to suggest how these data could be useful to hydrocarbon exploration.
STRUCTURAL SETTING Three broad geomorphic provinces are recognized in Bolivia: the Andes mountains in the southwest, the Guaporé shield in the northeast, and between them, the Beni-Chaco plain, which is the present-day foreland basin basin (Figure (Figure 1). The Bolivi Bolivian an Andes are comparat comparative ively ly
young and only started to form in the late Oligocene (Sempere et al., 1990a). In the Bolivian orocline, late Oligocene–Recent deformation propagated toward the east and northeast through what had been the eastern part of the Andean depositional area. Consequently, the Bolivian Andes provide many excellent exposures of Phanerozoic strata, of better quality and easier access than their counterparts in the sub-Andean rain forests of central and northern Peru. Andean Bolivia is an ideal place to study the stratigraphy and paleogeography of the margin of western Gondwana. During most of Late Ordovician–Paleogene time, Bolivia was located close to the southern tip of a trough that ran parallel to the Pacific margin of South America and that generally deepened toward the north or northwest along its axis. This trough, called the BoliviaPeru basin in this paper, was confined to the south by the Sierras Pampeanas area in northwestern Argentina during this time interval. An important consequence of this geometry is that, from Late Ordovician to early Paleocene time, marine transgressions flooded into Bolivia from the northwest. Between about 450 and 145 Ma, most of Bolivia consisted of a large sedimentary basin that was subjected to only local compressional deformation. The Silurian– Devonian to Middle Jurassic stratigraphy of Bolivia closely resembles that of the widespread Chaco-Paraná basin basin of of southe southern rn Brazi Brazil, l, east eastern ern Para Paragua guay, y, nort northwe hweste stern rn Uruguay, and northeastern Argentina (Sempere, 1990), suggesting that the Bolivian strata were deposited in a
Phanerozoic Evolution of Bolivia and Adjacent Regions region that was relatively stable and cratonic in nature (Oller and Sempere, 1990). The present-day Beni-Chaco basin, into which Andean deformation is still propagating, is the undeformed remnant of the Bolivia-Peru basin. The extensional processes that culminated in the opening of the South Atlantic from the latest Jurassic onward dissected and diminished the old depocenter, the western part of which became involved in the Pacific margin geotectonic system (Sempere, 1994). Little sedimentation but considerable pedogenesis occurred in the sub-Andean-Beni-Chaco basin between Late Jurassic and early Oligocene time. The productive sub-Andean-BeniChaco basin thus records two superimposed histories of evolution spanning the early Paleozoic–Jurassic (pericratonic Gondwana) and late Oligocene–Recent (Andean) intervals. Andean thrust deformation propagated into Bolivia after 27 Ma by taking advantage of several décollement levels in Paleozoic shales of the Bolivia-Peru basin (Sempere et al., 1989, 1990a). A close control on the geometry of Andean deformation was imposed by the geometry of the Paleozoic basin and its sedimentary fill (Sempere et al., 1989, 1991; Hérail et al., 1990). Propagation of thrust deformation and sedimentary and tectonic bur ying of Paleozoic organic-r ich uni ts have als o controlled hydrocarbon migration and distribution. Present-day geometries of Phanerozoic paleogeographies suggest that Andean deformation was considerable in the Bolivian orocline, involving a higher shortening in its northwest-trending segment and transcurrent motions (Figure 2). The southern half of Andean Bolivia is crossed by a major northeast-striking structural element known as the Khenayani-Turuchipa paleostructural corridor (CPKT) (Figure 1), which controlled Phanerozoic sedimentation and deformation in southwestern Bolivia (Sempere et al., 1991). Paleogeographic data suggest that the northeastern edge of the Chaco Boreal high was the continuation of the CPKT before Andean deformation (Figure 1). It is believed that offset of this paleogeographic marker by Andean deformation offers an eloquent illustration of the dextral tectonic displacement that occurred in the southern part of the orocline (~500 km cumulative). Large amounts of transcurrent displacements in this region are attributed to rotation of the Vilcabamba hinge zone in southern Peru at the northern end of the orocline (VH in Figure 1) (Sempere et al., 1989).
STRATIGRAPHY In Andean Bolivia, the basement consists of Upper Proterozoic or Lower Cambrian rocks that crop out in a few places. Deformed epimetamorphic rocks are known in the Tarija area, along the Chapare road northeast of Cochabamba, and at Cerro Chilla south of Lake Titicaca. These rocks show similarities to the Puncoviscana Formation of northwestern Argentina. The Yanahuanca Formation of the central Peruvian Andes (Dalmayrac et
209
al., 1980) and the basal part of the Marcona Formation of southern coastal Peru (Faro and San Juan members) (Shackleton et al., 1979) apparently have similar lithologies and deformation styles. Deformed rocks of Neoproterozoic–Eocambrian age in eastern Bolivia define a west-northwest striking wrench fault zone along the southern edge of the Guaporé shield. This fault zone splays in central Bolivia and becomes an east-northeast trending fold and thrust belt (Zubieta-Rossetti et al., 1993). This large-scale structure is Early–Middle Cambrian in age; the Chapare buttress (Hérail et al., 1990; Sempere et al., 1990a) is related to it. The Phanerozoic chronostratigraphy of Bolivia is not well constrained and requires more biostratigraphic, geochronologic, and sequence stratigraphic work. For instance, the Devonian and Carboniferous biostratigraphy suffers from many discrepancies among authors and among fossil groups (e.g., see Racheboeuf et al., 1993, for the Devonian). The origin of some of these discrepancies may partially lie in continental-scale diachroneity of fossil groups. Most Late Cambrian–Early Triassic strata are of marine origin, with many dark shale units recording sea level highs. Middle Triassic–Recent strata were mainly deposited in continental environments. The Bolivian Phanerozoic stratigraphy can be divided into eight main supersequences (with approximate durations given in parentheses) (modified from Sempere, 1990): 1. Tacsara supersequence: Late Cambrian–middle Caradoc (80 m.y. or more) 2. Chuquisaca supersequence: late Caradoc–middle Famennian (85 m.y.) 3. Villamontes supersequence: late Famennian– Mississippian (40 m.y.) 4. Cuevo supersequence: Pennsylvanian–Early Triassic (85 m.y.) 5. Serere supersequence: Middle Triassic–Middle Jurassic (95 m.y.) 6. Puca A+B supersequence: latest Jurassic–Turonian (56 m.y.) 7. Puca C supersequence: Senonian–middle Paleocene (31 m.y.) 8. Corocoro supersequence: latest Paleocene–Recent (57 m.y. or more) Definition of these eight stratigraphic intervals is based solely on the sedimentary record and not on tectonic or magmatic information. The boundaries between these supersequences are generally sharp. In the Chaco basin, the boundary between the Villamontes and Cuevo supersequences is not as sharp as it is in northwestern Bolivia. It is difficult to separate the Cuevo and Serere intervals in western Bolivia and southern Peru because of magmatism and deformation during the middle Permian–Early Jurassic. Grouping of the Villamontes, Cuevo, and Serere supersequences, which coincide with the formation and subsequent dispersal of Pangea, would thus be possible.
210
Sempere
Phanerozoic Evolution of Bolivia and Adjacent Regions (facing page) Figure 2—Paleozoic paleogeography for various time intervals. Geographic reference marks and Neogene tectonostratigraphic boundaries as in Figure 1. Hachures show approximate extent of basin. Arrows are mean paleocurrents; arrow tips are location of data. (A) Late Cambrian–Early Ordovician paleogeography. Depositional facies are mainly turbidites around the Puna aulacogen and distal to proximal shelf sediments in other outcrop areas. 1, Mainly mafic eruptive rocks; 2, high gravity anomaly of Götze et al. (1988). Paleocurrents outside Bolivia after Niemeyer et al. (1985), Moya (1988), and Bahlburg (1990). (B) Llandeilo–middle Caradoc paleogeography. All known depositional facies are shallow marine. 1, Caradoc mafic volcanics (Avila, 1992); 2, acid plutonism, possibly of magmatic arc; 3, known area affected by Middle Ordovician deformation; 4, structural strikes. (C) Ashgill– Llandovery paleogeography. Facies are mainly resediments. Dotted line, eastern limit of Tokochi Formation; tight hachures, Cancañiri Formation is >250 m thick; loose hachures, Cancañiri is <100 m thick; 1, Iron-rich rocks; 2, probable ongoing magmatism. Note influence of CPKT on paleogeography. The Caradoc mafic volcanics of shown in (B) may have been erupted at beginning of this interval. Data from Paraguay after Dyck (1992). (D) Wenlock–middle Famennian paleogeography. Loose hachures, mainly shallow marine shale and sandstone (Bolivia-Peru basin); tight hachures, Devonian–Early Mississippian turbidites, highly deformed in northwest-trending recumbent folds during Mississippian (El Toco Formation, northern Chile); 1, Wenlock limestone; 2, Pridoli–Lochkovian paleocurrents; 3, latest Eifelian–early Givetian paleocurrents; 4, Frasnian–middle Famennian paleocurrents; 5, little deformed shallow marine Middle Devonian sandstones associated with Arequipa massif; 6, interstratified tuffs; 7, San Nicolás batholith; 8, erosional boundary of Devonian in Bolivia. Successive edges of basin: A-L, Ashgill–Llandovery; W, Wenlock; L, earliest Lochkovian; E, Emsian. Mafic sills, dikes, and flows occur in Devonian strata of the Peruvian and northwestern Bolivian sub-Andean belt. Data partially from Mégard (1978), Dalmayrac et al. (1980), Breitkreuz (1986), ENAP (1988, personal communication), Mukasa and Henry (1990), and YPFB (proprietary data). (E) Mississippian paleogeography. 1, continental to shallow marine facies; 2, slope and basinal deposits, mainly resediments; 3, most distal facies; 4, Late Devonian or Mississippian folding; 5, structural strikes; U, uplifted areas (CBU, Chaco Boreal high; MU, Madidi uplift; GCU, Guaporé craton uplift). Although they varied with time, all Mississippian mean paleocurrents are integrated here, including data from Helwig (1972) and Breitkreuz (1986). Note reorganization of paleocurrents and arches compared with previous period. Hercynian deformation is indicated only where it is reliably identified; it was confused with Late Permian–Early Triassic or Late Triassic deformation in some areas. Paleogeography in eastern Peru is tentative; it probably included Mississippian uplifts, but deposits in this area were locally eroded prior to the Cretaceous (Mégard, 1978). (F) Pennsylvanian–Early Permian paleogeography. Wavy line, erosional boundary of basin in subAndean-Beni-Chaco region; 1, shallow marine carbonates, Bolivia-Peru basin; 2, fluvioeolian sandstones; 3, limestone intercalations in Cangapi Formation; 4, marine deposits (carbonates) of northern Chile and southern coastal Peru (after Breitkreuz, 1991); 5, marine or fluvial paleocurrents; 6, progradation of eolianites; 7, axis of Bolivia-Peru basin; 8, magmatism; 9, interstratified volcanics; 10, Late Pennsylvanian intraarc lake of Breitkreuz (1991).
211
In this paper, the terms supersequence, megasequence, and sequence, in order of decreasing thickness, are used for stratigraphic units bounded by sharp contacts.
LATE CAMBRIAN–MIDDLE CARADOC Strata of Late Cambrian–middle Caradoc age are exposed in the Cordillera Oriental, southern Altiplano, and some parts of the sub-Andean belt of Bolivia (Suárez-Soruco, 1976). They are unknown in the UllomaCoipasa-Uyuni domain (Figure 1). The oldest outcropping rocks in the north are of late Proterozoic or Early Cambrian age and are overlain by Upper Cretaceous strata, whereas in central and southern Bolivia, the Ashgill age Cancañiri Formation crops out as the basal décollement of the local thrusts. Avila (1992) has reviewed the coeval igneous rocks of Bolivia. The base of this supersequence is exposed in the Tarija area and along the Cochabamba-Chapare road. In most of the outcropping area, the basal décollement of the Andean thrusts is located in Lower–Middle Ordovician shales and no older strata are exposed. These successions are generally thick and monotonous. In the Tarija area, the sequence begins with shallow marine clastics, which grade upward to thick, open marine graptolitic shales with subordinate turbidites and slumps. Paleocurrent data and facies distributions from southern Bolivia, northwestern Argentina, and northern Chile define a north to north-northeast trending trough (Figure 2A) (Sempere, 1989, 1991a; Bahlburg, 1990). In the southern part of this trough, ophiolitic rocks (Allmendinger et al., 1983) and a large positive gravity anomaly suggesting the presence of a dense substrate (Götze et al., 1988) are documented. This early Paleozoic trough was apparently underlain by a substantially thinned crust. In southwestern Bolivia, the Upper Cambrian– Llanvirn sequence is affected by compressional deformation. In northernmost Argentina, beds of Ashgill and Llandovery age clearly postdate this deformation (Isaacson et al., 1976; Benedetto et al., 1992; M. C. Moya, 1992, personal communication). In the Tarija area, the Ashgill age Cancañiri Formation directly overlies Arenig–Llanvirn strata. This Ocloyic deformation increases from east to west, and a strong cleavage appears in southwestern Bolivia. It therefore documents a significant compressional event and uplift of Llandeilo–Caradoc age (Sempere, 1989, 1990, 1991a). However, this deformation is unknown north of 20° S lat in Bolivia. The orientations of cleavages (Figure 2B) suggest that the compressional fabric may continue below the present-day volcanic Cordillera Occidental, that is, along an inferred northeastern boundary of the Arequipa massif. Although no direct evidence supports this interpretation west of Lake Titicaca (Laubacher, 1978), the intensely deformed Llanvirn strata of central Peru (Mégard, 1978) may be a counterpart. North of 20° S lat in Bolivia, the exposed succession consists of a thick sequence of Llanvirn–Caradoc shal-
212
Sempere
Figure 3—Reconstruction of late Proterozoic supercontinent (Bond et al., 1984; Murphy and Nance, 1991). AF, Africa; APT, Appalachian-Patagonian terranes (Figure 6); B, Baltica; NA, North America; SA, South America; thick line, outline of cratonic Gondwana; dashed line, margin of displaced terranes (Ramos, 1988a); circle, paleoequator; central cross, late Proterozoic south pole.
lowing-upward sedimentary rocks, which is conformably capped either by the upper Caradoc Tokochi Formation (Sempere et al., 1991) or by the Ashgill Cancañiri Formation. Thus, deposition of thick shallow marine strata during Llandeilo–Caradoc time coincided with the final development of the Ocloyic deformation in southwestern Bolivia and northwestern Argentina (Figure 2B). This thick sequence of interbedded sandstone and siltstone is interpreted as the fill of a marine foreland basin related to this deformation (Sempere, 1989, 1991a). There is no consensus about the geodynamic system that was active during this time, but the data are best explained if eastern North America and western South America were united in the late Proterozoic (Figure 3) and only rifted apart in latest Proterozoic–Early Cambrian time (Bond et al., 1984; Murphy and Nance, 1991). The western margin of South America was thus a passive margin in Early Cambrian–Early Ordovician time. The north-northeast trending, partially ophiolitefloored trough in the Puna and southern Bolivian Altiplano is here interpreted as an aulacogen (the Puna aulacogen) which formed as a result of dextral-normal strain along the southern Iapetus passive margin of Gondwana. The beginning of closure and deformation of the Puna aulacogen, and thus the onset of active behavior along the nearby margin, occurred near the
Figure 4—Map depicting present-day relationships between known Ordovician deformation (black) and the related basin (hachured). Deformation in northwestern Argentina and southwestern Bolivia is due to closure of the Puna aulacogen (Figures 2A, B). Black areas in the northern Andes correspond to rocks deformed during this episode and again during a Permian event (after Dalmayrac et al., 1980). Note offshore data at 9°S lat (Bourgois et al., 1990) and probable deformation in coastal southern Peru (Shackleton et al., 1979). Map suggests current truncation at ~12°S lat of Ordovician structures by trench and higher Andean shortening in Bolivian orocline. eAd, edge of Andean deformation.
Arenig-Llanvirn boundary according to data documented by Bahlburg (1990). The Atico or Marcona events on the coast of southern Peru (Shackleton et al., 1979) probably correspond to this major deformation. A Caledonian (i.e., Taconic) deformation is recorded from northwestern Peru up to northern Colombia (Dalmayrac et al., 1980). These data suggest that a major tectonic crisis affected the western margin of Gondwana during the Middle Ordovician (Figure 4). However, interpretations of the early Paleozoic magmatic record of the central Andes are contradictory (e.g., Breitkreuz et al., 1989; Damm et al., 1990; Rapela et al., 1992).
Phanerozoic Evolution of Bolivia and Adjacent Regions
213
LATE CARADOC–MIDDLE FAMENNIAN The late Caradoc–middle Famennian interval is represented by two sets of units that are believed to span the late Caradoc–early Llandovery and late Llandovery– middle Famennian time intervals, respectively (Figure 5). The Devonian organic-rich shales are the most important hydrocarbon source rocks in Bolivia.
Late Caradoc–Early Llandovery The first stratigraphic set includes the Tokochi, Cancañiri, and Llallagua formations, all of which reach their maximum thickness in the Ulloma-Coipasa-Uyuni domain. They thin out rapidly to the southeast and east (Gagnier et al., in press). The Tokochi Formation (Sempere et al., 1991) consists of black shales up to 200 m thick that overlie shallow marine Caradoc strata. It has yielded only an open marine fauna, including cephalopods and Caradoc pseudoplanktonic brachiopods (L. Branisa, 1986, personal communication). The Tokochi is thought to mark a sea level high (maximum flooding surface) and to have been deposited in a relatively deep, anoxic environment. Because the thickness of these black shales increases toward the west, they should be present at depth in the central Altiplano. The sharp or gradational contact between the Tokochi and Cancañiri formations indicates a drop in sea level. The latter consists mainly of greenish gray, generally unstratified, resedimented sandy mudstones and diamictites (mass flows) and minor sandstones (basinal turbidites and local shelf clastics). The Cancañiri Formation is of Ashgill age (Sempere, 1990; Benedetto et al., 1992; Toro et al., 1992), but its top is possibly early Llandovery. Thickness varies up to 1.5 km or more. The millimeter to meter size subrounded to angular clasts consist of a variety of sedimentary, igneous, and metamorphic rock types (Rodrigo et al., 1977). Some quartz and igneous rock clasts are clearly faceted and striated, indicating that they were derived from a glaciated area. Large sandstone blocks, up to 1000 m 3 in volume, are more abundant near the Khenayani-Turuchipa paleostructural corridor, suggesting that structural activity during deposition of the Cancañiri produced fault scarps and block sliding (Sempere et al., 1991). The Cancañiri Formation is interpreted as a lowstand deposit related to regional extension and associated high rates of resedimentation while a large part of Gondwana was glaciated. The Llallagua Formation overlies the Cancañiri Formation with a sharp contact. It consists of turbidite beds, one to several meters thick, and subordinate siltstone and mudstone; it can reach thicknesses of 1.5 km or more. These facies are attributed to lowstand deposition following the Cancañiri maximum lowstand. The Llallagua Formation grades transitionally into the Uncía (Kirusillas) Formation, forming a thinning- and fining-upward succession as a result of sea level rise. The Llallagua is attributed to the Llandovery because of its stratigraphic position. Its upper part is equated with the basal Kirusillas (Uncía) Formation, which occurs more to
Figure 5—Stratigraphic section for the late Caradoc– middle Famennian interval. Lithologic symbols are common use.
the east and has yielded a late Llandovery graptolite (Branisa et al., 1972). The thickness of the Tokochi-Cancañiri-Llallagua sequence increases dramatically in the eastern and southern Ulloma-Coipasa-Uyuni domain (Figure 2C), and might have reached 3.5 km in depth before Andeanage thrusting used the Cancañiri Formation as the main décollement level there. These values and the age constraints indicate maximum known decompacted sedimentation rates of about 250 m/m.y. for this time span, which is very high. The predominance of sediment gravity flows in the Cancañiri and Llallagua formations strongly suggests synsedimentary tectonic activity. It is possible that the Tokochi anoxic facies represent a starved basin stage in this evolution. Paleocurrents (Figure 2C) indicate sediment transport toward the west and northwest, that is, in the direction of thickening and toward the interior of the Ulloma-Coipasa-Uyuni domain. These characteristics suggest that deposition of the late Caradoc-Llandovery strata occurred in a faultcontrolled basin, the principal normal faults being reactivated and inverted during the Andean orogeny to form the Khenayani fault and Main Altiplanic thrust systems (Figure 2C) (Sempere et al., 1991). Resedimented facies appear to have evolved from mass flows that reworked glaciomarine deposits (Cancañiri Formation) to proximal
214
Sempere
(lower Llallagua Formation) and distal turbidites (upper Llallagua Formation), suggesting an overall sea level rise. This transition of the Llallagua Formation to the Uncía (Kirusillas) Formation also indicates waning of extensional activity.
Late Llandovery–Middle Famennian The second stratigraphic set includes several units that occur in the Bolivian Andes, sub-Andean belt, and Beni-Chaco plains, albeit with different names. Shallow marine, relatively warm water faunas were present as early as Llandovery (Isaacson et al., 1976) and early Wenlock time (Branisa et al., 1972; Merino, 1991; Gagnier et al., in press). Although the precise ages of the Devonian units are not firmly established, the chronostratigraphic data of Vavrdová et al. (1991) and Racheboeuf et al. (1993) are used here. Three overall thickening- and shallowingupward megasequences are recognized (Figure 5): late Llandovery–Lochkovian, Pragian–early Givetian, and late Givetian–middle Famennian. The first of these intervals is generally divided into two lower order sequences: late Llandovery–Pridoli and Lochkovian (Figure 5). Paleoenvironments include relatively shallow marine, distal to proximal shelf, deltaic, littoral, and locally even fluvial deposits (e.g., in the Lochkovian age Santa Rosa Formation). The lithologies and stratigraphic relationships are illustrated in Figure 5. Between Cochabamba and Santa Cruz, a thin limestone unit of early Wenlock age (Merino, 1991; Gagnier et al., in press) occurs in the basal part of the late Llandovery–Ludlow shales (Figure 2D), which postdates a Llandovery hiatus. Slumped sandstones and siltstones are common in the Pridoli portion of the series in several areas of Bolivia (e.g., Cocha bamba-Oruro road, Tarija, and Sierras Chiquitanas) and possibly indicate contemporaneous tectonic instability. Two prominent transgressions are recognized. The first is recorded by the lower Kirusillas Formation (Wenlock–early Ludlow) and is probably a continuation of the transgression that started in Llandovery time. The second is reflected by the Icla Formation (Pragian– Emsian). Two less important transgressions also occurred. The earliest Lochkovian transgression is recorded by a shale unit between the Tarabuco and Santa Rosa formations (Chululuyoj Formation) (Racheboeuf et al., 1993), whereas the late Givetian transgression is reflected by the Los Monos Formation (Sempere, 1990; Racheboeuf et al., 1993). However, the latter has been assigned an age as early as earliest Eifelian (Lobo-Boneta, 1975; McGregor, 1984; Pérez, 1990). These four transgressions were each followed by progradation of shallower facies (Figure 5). The Lochkovian Santa Rosa and the upper Eifelian–lower Givetian Huamampampa formations reflect important lowstand periods. Figure 2D shows paleocurrent data for the Pridoli– Lochkovian, upper Eifelian–lower Givetian, and Frasnian–middle Famennian deposits. Late Silurian– Middle Devonian paleocurrents do not sustain Isaacson’s (1975) assumption that coeval sediments were derived from the Arequipa massif. Changes in paleocurrent
Figure 6—Simplified Late Cambrian–Late Devonian evolution of the central Andean region. APT, AppalachianPatagonian terranes (see Figure 3) (Ramos, 1988a; Mpodozis and Kay, 1992); PPO, paleo-Pacific ocean.
patterns from Late Silurian to Late Devonian time suggest large-scale tectonic modifications in the architecture of the basin. The Late Devonian reorganization of paleocurrents toward the southeast may be related to the Hercynian deformation in Peru (Laubacher, 1978; Mégard, 1978). Onlap toward the northeast is conspicuous and possibly has exploration significance. Whereas there is no obvious stratigraphic gap in the Ordovician– Devonian succession in the Cordillera Oriental north of 20° S lat, the Ashgill diamictites, which disconformably overlie Caradoc sandstones, are disconformably overlain by upper Llandovery shales or by the lower Wenlock Sacta limestone in the mountain area between Cochabamba and Santa Cruz. The Sacta limestone overlies Eocambrian rocks in the nearby Andean foothills. Eocambrian rocks in the Sierras Chiquitanas are overlain by Upper Silurian sandstones (López et al., 1982). In contrast, in the northwestern sub-Andean belt, Upper Silurian shales and sandstones overlie deeply weathered Caradoc sandstones. In the Pando well in northern Bolivia, Lower Devonian sandstones rest on
Phanerozoic Evolution of Bolivia and Adjacent Regions
215
Figure 7—Simplified stratigraphic cross section of the Carboniferous and Permian showing chronology used in this paper. Localities (shown in Figure 2E): CT, Copacabana peninsula and Tiquina straits; RK, Río Kaka; CH, Charagua area; AP, Angosto del Pilcomayo. Black indicates carbonates; other lithologic symbols are common use. Information partially from YPFB proprietary data and Díaz-Martínez (1991). Rapid transition between Chuquisaca and Villamontes supersequences (RK and CH sections) is probably a characteristic of the basin “axis” (Figure 2E).
Precambrian gneisses. Onlap of this magnitude in the Beni-Chaco basin is expected to have formed potential stratigraphic traps. Maximum known sedimentation rates are high: they average about 70, 85, and 100 m/m.y. for the late Llandovery–Lochkovian, Pragian–early Givetian, and late Givetian–middle Famennian intervals, respectively. Subsidence was rapid and more intense to the southwest and increased through time. In contrast to the late Caradoc–Llandovery interval during which sedimentation rates were much higher (~250 m/m.y.) in the Ulloma-Coipasa-Uyuni domain, thickness variations are small and the principal units widespread. This broadscale regional subsidence is attributed to continuous tectonic loading of the craton margin transition by transpressional processes (Figure 6). Such a marine foreland basin interpretation is supported by the Late Devonian– Mississippian epeirogenic deformation of the BoliviaPeru basin. In summary, the overall stratigraphic succession reflects extensional subsidence of the trough that straddled the Ulloma-Coipasa-Uyuni domain in late Caradoc–early Llandovery time and transpressional foreland basin subsidence afterward. In northwestern Argentina and southern coastal Peru, numerous granitoids were intruded during the Late Ordovician–Early Devonian (Bahlburg, 1990; Mukasa and Henry, 1990; Rapela et al., 1992). They are believed to be the earliest evidence of subduction-related
magmatism in the Paleozoic. Mukasa and Henry (1990) believe that the San Nicolás batholith of southern Peru was probably emplaced in two short-lived magmatic events at about 425 and 394–388 Ma. These U-Pb dates indicate late Wenlock–early Ludlow and late Pragian– early Emsian ages, respectively, broadly corresponding to the Kirusillas-Uncía and Icla transgressions. In northern Chile, the first known onlap of the Ocloyic basement is of Early Devonian age and the upper part of the succession is of Early Carboniferous age (Breitkreuz, 1986). Intensely deformed Upper Devonian–Mississippian flysch occur west of this shallow marine belt and are thought to represent either fore-arc deposits deformed in accretionary wedges (Bell, 1987) or the fill of a turbiditic trough that was closed in Mississippian time and subsequently intruded by posttectonic granitoids (Breitkreuz et al., 1989).
LATE FAMENNIAN–MISSISSIPPIAN The Carboniferous succession of Bolivia is located mainly in the Beni-Chaco basin and adjacent subAndean belt, in the Tarija-Teoponte belt, and in the Huarina fold and thrust belt northeast and southeast of Lake Titicaca (Figure 1). A large portion of Bolivian oil and gas reserves occurs in Carboniferous strata. The chronostratigraphy of this supersequence is still being refined (Figure 7). There have been some recent
216
Sempere
significant biostratigraphic studies (Merino, 1987; SuárezRiglos et al., 1987; Rösler et al., 1989; Merino and Blanco, 1990; Vavrdová et al., 1991). In contrast with the underlying stratigraphy, “Carboniferous” facies are highly variable and inhibit lithologic correlation. The literature is replete with conflicting and chaotic stratigraphic schemes, although Helwig (1972) has addressed some of these issues. Furthermore, the “Hercynian” metamorphism and deformation of Bard et al. (1974) and Martinez (1980) is now known to be of Late Triassic age (Farrar et al., 1990). This deformation is not related to the Late Devonian or Early Mississippian deformation described in west-central Peru (Mégard, 1978) and west of Lake Titicaca (Laubacher, 1978), although there is evidence that some Late Devonian or Mississippian uplift and local folding did occur in Bolivia. The Late Mississippian U-Pb age obtained on the Amparaes syntectonic granite (Dalmayrac et al., 1980) is considered to be dubious because three concordant Late Triassic UPb ages have been obtained on similar plutons.
Northwestern Sub-Andean Belt and Adjacent Beni Basin In the northwestern sub-Andean belt and in adjacent Beni basin, the Carboniferous transitionally overlies the Upper Devonian and is capped by the major erosional unconformity at the base of the Jurassic Beu Formation. It comprises a lower succession characterized by black mudstones and common plant fossils and an upper succession of carbonates, eolianites, and subsurface evaporites. The lower succession (Retama Group, latest Devonian–Visean or early Serpukhovian) is equivalent to the Ambo Group of Peru. It contains conglomerates, thick mudstone slumps, some stratified sandstone slump blocks, and shallow marine black laminated shales, which grade upward into sandstones and coals. Dark diamictites occur in the upper part. The upper succession (Copacabana Formation, latest Mississippian or earliest Pennsylvanian to earliest Permian and locally Late Permian) consists of fossiliferous limestones and marls, white calcareous sandstones, thin green tuffaceous beds, dolomites, some anhydrite in the subsurface, some dark to mottled shales, and a few red siltstones, sandstones, and paleosols. It corresponds to the Tarma and Copacabana groups of Peru and is reviewed in the next section. The contact between these two successions is sharp and, on the basis of conodont biostratigraphy, would lie within the Serpukhovian or at the Mississippian–Pennsylvanian boundary (D. Merino, 1992, personal communication). This contact marks a significant paleoclimatic change in northwestern Bolivia, from cool-temperate to warm and arid conditions. The signature of this climatic change should facilitate regional correlation. Furthermore, the major global regressive-transgressive event of Serpukhovian to earliest Pennsylvanian age (Saunders and Ramsbottom, 1986; Veevers and Powell, 1987) is apparently recorded in facies near the Retama–Copaca bana contact and should be regionally widespread.
Huarina Fold and Thrust Belt Carboniferous strata in the Lake Titicaca area sharply overlie the Devonian. Díaz-Martínez (1991) have divided these strata into two units. First is the Ambo Group (latest Famennian–Visean or early Serpukhovian) (Rösler et al., 1989; Vavrdová et al., 1991), which includes a basal glaciomarine horizon and dark shales grading upward into cross-bedded sandstones with intercalated diamictite beds, coals, and a thin cap of reddish fluvial sandstones. The second unit is the Titikaka Group, which consists of estuarine sandstones and supratidal dolomites overlain by transgressive fossiliferous limestones and marls (Middle Pennsylvanian conodonts) (D. Merino, 1992, personal communication). The Titikaka Group, which in this area spans the Early Pennsylvanian-Early Triassic interval, corresponds to the Copacabana and Bopi formations of the northwestern sub-Andean belt and the Tarma, Copacabana, and Mitu (partly) groups of Peru. It is capped either by Triassic sandstones or by Upper Cretaceous or upper Oligocene strata. Because the Ambo Group disconformaby overlies Lower Devonian strata in several outcrops forming a northwest-striking alignment, it postdates a local uplift of Late Devonian or earliest Mississippian age. Furthermore, Pennsylvanian–Lower Permian limestones onlap topography formed during the Late Devonian or Mississippian in the Huarina fold and thrust belt and Cochabamba area. The disconformable relationships at the base of these limestones contrasts with the “Hercynian” deformation interpretation of Martinez (1980). However, on a regional basis, angular unconformities represent a Silurian–Triassic hiatus, reflecting a compressive deformation of inferred Late Devonian– Mississippian age. This local deformation merges laterally into an uplifted area spanning much of the present-day Cordillera Oriental, upon which Pennsylvanian–Upper Permian strata onlap. This uplift was probably related to a dextral motion of the KhenayaniTuruchipa paleostructural corridor (Sempere et al., 1991). The eroded detritus was probably deposited in the Frasnian–Mississippian basin.
Chaco Basin, Sub-Andean, and Tarija-Teoponte Belts In the Chaco basin, sub-Andean, and Tarija-Teoponte belts, the Missis sippian locally overlies the Upper Devonian gradationally. It consists of the Macharetí and Mandiyuti subgroups (Figure 7). In the southernmost sub-Andean belt, there is a gradational contact between the two subgroups. Acritarchs in the Macharetí, marine faunas in the lower Mandiyuti, and sedimentary facies suggest that deposition occurred in marine environments. There are marked lithologic contrasts between mud-dominated and sand-dominated Macharetí facies, whereas the Mandiyuti is predominantly sandy and shows nearshore to shore zone facies affinities near the top. In seismic sections, the lower part of the Mississippian interval is characterized by discontinuous reflection
Phanerozoic Evolution of Bolivia and Adjacent Regions patterns, whereas the upper part is mainly featureless because it is sand prone. This twofold lithologic partition represents a major change in sedimentary dynamics, interpreted as a rapid overall shallowing of the depositional systems. There is also a conspicuous difference in color between the two subgroups: the clay fraction in the Macharetí is dark gray to purplish and locally contains plant debris. In contrast, it is bright brown-red in the uppermost Macharetí and the entire Mandiyuti. This color change is attributed to the Serpukhovian climatic event recorded in northwestern Bolivia. Rapid shallowing of depositional environments in the Chaco basin suggests a contemporaneous sea level drop related to a worldwide Serpukhovian regression (Veevers and Powell, 1987). The Macharetí–Mandiyuti contact thus appears to be of early Serpukhovian age. The late Paleozoic stratigraphy shown in Figure 7 is based on these lithostratigraphic criteria and recent biostratigraphic data. I equate the Saipurú Formation of the Charagua area (Suárez-Soruco and López-Pugliessi, 1983), which has late Famennian–Visean palynomorphs, to the lower Macharetí (Itacua, Tupambi, Itacuami, and Tarija formations) of the Villamontes area. Marine fossils of the Levipustula levis zone, the base of which coincides with the Visean–Serpukhovian boundary in Australia, occur in a regularly stratified, reddish siltstone– mudstone intercalation that is traditionally described as the local “Taiguati Formation.” I prefer, however, to call it the Charagua member because it appears to be located within the lower part of the Mandiyuti (Figures 2E, 7). The Macharetí and Mandiyuti facies are broadly similar and are dominated by resedimented deposits, including rarely stratified diamictites, muddy mass flows, slumped sandstone–mudstone intercalations, grain flows, debris flows, cross-bedded sandstones (commonly conglomeratic), slumped sandstones masses, thick to thin regularly bedded Bouma-type turbidites interbedded with siltstones and mudstones, regularly bedded foss il-bearing fine-grained sandstones and mudstones, rarer laminated mudstones, and local intra basinal olistoliths that reach several meters in thickness. The diamictites, debris flows, grain flows, mud flows, and slumps are clast bearing. The clasts are millimeter to meter size, are well-rounded to angular, and consist of a variety of sedimentary, igneous, and metamorphic rock types. Some of these are faceted and striated, suggesting transport from a glaciated area. Ice rafting explains the presence of some of the larger clasts. Although reworked glacial clasts occur in the Mandiyuti, the most convincing glaciomarine features are seen only in the Macharetí. Most of the resedimented facies were deposited in channels, ranging in width from a few meters to a few kilometers. Several large channels and valleys (several hectometers or kilometers wide and up to several hundred meters deep) are clearly visible in many seismic sections in the Chaco basin. The apparent depth of these channels or valleys increases upsection. Channel stacking and amalgamation of their channel fills has resulted in valley complexes locally more than 500 m deep. The abundance of resedimented facies filling an
217
erosional topography indicates tectonic instability in the Chaco basin. Reorganization of paleocurrents in the latest Devonian–earliest Carboniferous (compare Figures 2D and 2E) demonstrates the growth of the northeasttrending Chaco Boreal high. As in other parts of Bolivia, large-scale slumping shows that sediments were fed from tectonically active highs and hence documents periods of tectonic instability. Regional mapping shows the distribution of these paleogeographic highs: Chaco Boreal, Cordillera Real, Madidi, and Guaporé craton uplifts (Figure 2E), the last two of which were probably controlled by a Cambrian crustal heterogeneity. The distal downwarped areas are identified by the occurrence of a transition between the Chuquisaca and Villamontes supersequences, a lower thickness of the Macharetí subgroup, or a thicker and well-defined Charagua member. Available data suggest that crustal deformation, involving uplift and minor folding, progressed from west to east during the Late Devonian and Mississippian. While structural highs and slopes formed in parts of the Beni-Chaco basin in the late Famennian–Early Mississippian, other areas were downwarped and received slumps and laminated black shales. Compared to the underlying Frasnian–early Famennian shallow marine strata, there was a significant increase in depth. Slopes were tectonically maintained during the Mississippian, controlling sigmoidal geometries of progradational units and substantial variations in thickness. The thicknesses of the Macharetí and Mandiyuti sections crossing the Chaco basin display such relationships. The Chaco basin received a considerable sand supply from the southeast and east, that is, from the Chaco Boreal high. Evolution of the Carboniferous climate in Bolivia, from cool and humid to warm and arid, matches the migration of the south pole toward the southeast or eastsoutheast (Veevers and Powell, 1987). South American climatic gradients were probably high and oriented northeast to north-northeast, defining climatic strips where different facies assemblages were able to accumulate at the same time. Current interpretation of palynozones suggests that the Paraná basin was probably subjected to subglacial erosion during the Mississippian and to glacially influenced marine sedimentation during the Pennsylvanian–Early Permian glacial retreat (França and Potter, 1991). The Amazon basin, Bolivia, and western South America experienced cool to temperate and subsequently subtropical conditions during these periods. The average sedimentation rate for the Mississippian of Bolivia is estimated to be about 35–40 m/m.y., indicating a significant decrease compared to the Devonian. However, because of the inferred increase in depth, it is possible that the maximum subsidence rate increased in the late Famennian–earliest Mississippian before decreasing after that. The Late Devonian–Early Mississippian deformation known in Peru (Mégard, 1978; Dalmayrac et al., 1980) and northern Chile (Sempere, 1989) and the inferred uplifts in the present-day Huarina fold and thrust belt, Chaco Boreal, Cordillera Real, Madidi, and Guaporé
218
Sempere
Figure 8—Map of western Gondwana showing approximate distribution of main Paleozoic geotectonic features. Key: 1, Middle Ordovician deformation; 2, Late Ordovician– Devonian granitoids; 3, Late Devonian–Mississippian deformation; 4, Carboniferous–Permian basins; 5, Late Permian–Early Triassic deformation. In Bolivia, basin is bounded by structural highs that formed during late Famennian–Mississippian sedimentation (Figure 2E). CBH, Chaco Boreal high; GCU, Guaporé craton uplift; MU, Madidi uplift. Data mainly from Dalmayrac et al. (1980), Archangelsky (1987), Ramos (1988a), and Rapela et al. (1992).
craton uplifts provide evidence of synsedimentary insta bility. This evidence clearly points to a time of considerable tectonic activity in the Bolivia-Peru basin in the Late Devonian and especially the Mississippian. The Mississippian sedimentation was the culmination of the Silurian–Devonian cycle of evolution. The uplifts and downwarps observed in Bolivia probably represent the foreland response to the transpressional and compressional processes that affected the western margin of Gondwana (Figure 8).
PENNSYLVANIAN–EARLY TRIASSIC The Pennsylvanian–Early Triassic interval is characterized in western Bolivia by the establishment of a carbonate platform and in southeastern Bolivia by northwestward progradation of littoral and continental depositional systems. This was followed by a Middle–Late Permian restricted marine transgression and deposition
Figure 9—Late Permian–Early Triassic paleogeography. General legend same as in Figure 2. Key: 1, erosional boundary of basin in sub-Andean belt; 2, known remnants of Upper Permian strata in Andean area; 3, presence of clasts of cherty limestones typical of Upper Permian deposits in Miocene strata, suggesting proximity of Upper Permian outcrops in the early Neogene; 4, inferred depositional area before Mesozoic erosion; 5, Late Permian basic magmatism; 6, areas of felsic magmatism; 7, interstratified tuffs; 8, known extent of the Late Permian–Early Triassic deformation, including granitoids; 9, fold strikes suggesting southward continuation of structures. Gondwana-age folds are generally roughly parallel to Andean structures. In the Cordillera Oriental of southern Peru and western Bolivia, some structures traditionally assigned to the Late Permian–Early Triassic may be of Late Triassic age.
of carbonates and mudstones (Figures 2F, 9). This interval was also characterized by a substantial decrease in tectonic activity and by the establishment of a subtropical climate. Depositional environments were predominantly shallow marine and continental, with subordinate eolianites in the Pennsylvanian and Early Permian. Pennsylvanian–Upper Permian limestones onlap the older topography in several areas of the present-day Cordillera Oriental. The Pennsylvanian–Permian deposits in the Amazon basin resemble those in northern Bolivia. The stratigraphy used in this paper implies that Pennsylvanian–Lower Permian carbonate deposits in the northwest grade laterally into fluvial and eolian deposits in the southeast (Figure 7) and may even be in facies with the underlying shallow marine clastics. However, the boundary between the Villamontes and Cuevo supersequences in the Chaco basin is not as sharp as it is in northwestern Bolivia. Note that the fluvial-eolian Cangapi Formation at the base of the Cuevo supersequence in the southern sub-Andean belt was originally included in the Mandiyuti Subgroup. Its lower strata may be continental equivalents of the shallow marine upper San Telmo Formation in some parts of the Chaco basin. In this sense, the Mandiyuti Subgroup may span
Phanerozoic Evolution of Bolivia and Adjacent Regions
Figure 10—Simplified stratigraphic sections for the Middle Triassic–Middle Jurassic. See Figure 11 for locations. ER, Entre Ríos area; R, Ravelo; S, Sayari; T, Tarabuco. Key: 1, carbonates; 2, mudstones; 3, medium to fine–grained sandstones; 5, conglomerates; 6, mainly alluvial facies; 7, mainly eolian facies; 8, evaporites; 9, basalts.
the Late Mississippian–earliest Pennsylvanian. More biostratigraphic work will decide this issue. Sedimentation rates for the Pennsylvanian–Early Permian interval were less than 25 m/m.y. in Peru, about 15–20 m/m.y. in the Lake Titicaca area of western Bolivia, and less than 15 m/m.y. in the Chaco basin. Sedimentation rates were also low for the Late Permian–Early Triassic interval, generally less than 10–15 m/m.y. The intensity of subsidence had thus decreased dramatically after the Mississippian. Although mainly a restricted marine deposit, the Vitiacua Formation marks a major transgression that started in late Kungurian or Kazanian time and covered a wide region. Toward the west, it was the most extensive transgression since the Devonian. The lower Vitiacua includes black limy mudstones resembling those of the Irati Formation in the Paraná basin and the Whitehill Formation in the Karoo basin. These facies represent a highstand that flooded Gondwana as far away as Australia (Sempere et al., 1992). Deposition of the middle Permian–Early Triassic strata was partially coeval with the intense compressional or transpressional intracratonic deformation in the Cordillera Oriental of central and southern Peru (Laubacher, 1978; Soler and Bonhomme, 1987). In Bolivia,
219
this deformation is known from outcrops north of Lake Titicaca (Martinez, 1980) and west of Cochabamba (Kennan, 1993). It might correspond to the so-called Eohercynian deformation in the Cordillera Real; it did not affect the rest of the country. It was partially coeval with intense compressional deformation in the Andes north of 7° S lat (Dalmayrac et al., 1980), the San Rafael deformation of west-central Argentina (Ramos, 1988b), and the transpressional deformation of the Sierra de la Ventana and Cape foldbelts (Cobbold et al., 1986) (Figure 8). Accretion of terranes along the southwestern margin of South America ended in Late Permian time (Mpodozis and Kay, 1992). This widespread Gondwana-age deformation is postdated by a postorogenic calc-alkaline magmatism of Early–Middle Triassic age (Laubacher, 1978; Soler and Bonhomme, 1987; Mpodozis and Kay, 1992). By the late Middle Triassic, continental tholeiitic compositions were more common, indicating regional extension (Soler and Sempere, 1993). Onset of tensional conditions in Bolivia in the late Middle Triassic initiated the Serere sedimentary supersequence. In northern Chile, metamorphism related to terrane accretion ended in the Pennsylvanian (Cordani et al., 1988). This was superseded by widespread magmatism of Pennsylvanian–Triassic age (Breitkreuz et al., 1989; Breitkreuz, 1991; Mpodozis and Kay, 1992). It is this magmatic episode that is believed to have generated the numerous tuffs observed in the Bolivian and eastern Peruvian basin.
MIDDLE TRIASSIC–MIDDLE JURASSIC The Middle Triassic–Middle Jurassic stratigraphy of Bolivia has recently been revised by Oller and Sempere (1990) (Figure 10). Late Middle Triassic rifting developed in several areas as a prelude to the breakup of Pangea. It led to the formation of numerous grabens (e.g., Mitu Group of Peru and Sayari and Ipaguazú formations of Bolivia) accompanied by alkaline and tholeiitic magmatism (Mégard, 1978; Kontak et al., 1985; Soler and Sempere, 1993). In Bolivia, tholeiitic basalts from Entre Ríos and Tarabuco cannot be geochemically distinguished (Soler and Sempere, 1993) and have apparent K-Ar ages of 233 Ma (Sempere et al., 1992) and about 171 Ma, respectively. The Mitu Group of Peru (Figure 11) spans the Late Permian–Middle Triassic interval and appears to consist of two successions containing poorly understood magmatic suites. Mafic volcanism developed as early as Late Permian time in Peru. Basaltic flows, yielding Permian–Early Triassic apparent K-Ar ages, are exposed along the southeastern shore of Lake Titicaca (S. McBride, in Kontak et al., 1985), whereas mafic rocks south of Lake Titicaca have Late Permian–Early Triassic apparent K-Ar ages. Swarms of mafic sills and dikes (0.1–1 m thick) intruding Ordovician–Devonian strata in the Cordillera Oriental (Figure 11) are associated with paleograbens preserving Permian strata. These intrusives
220
Sempere Figure 11—Middle Triassic– Middle Jurassic paleogeography. General legend same as Figure 2. Locations of stratigraphic sections in Figure 10: ER, Entre Ríos; R, Ravelo; S, Sayari; T, Tarabuco. Key: 1, Late Triassic– Jurassic marine deposits; 2, Early–Middle Jurassic deepest facies; 3, Middle Jurassic turbidite fan progradations; 4, Jurassic arc magmatism; 5, edge of Jurassic marine basin in northern Chile; 6, Mitu volcanic deposits; 7, Early–Late Triassic plutons, including the Abancay (a) and Zongo-Yani (zy) cataclastic Late Triassic plutons; 8, Middle Triassic–Early Jurassic fluvial deposits in Bolivia; 9, fluvial paleocurrents; 10, occurrences of evaporites; 11, Middle– Late Triassic continental tholeiites; 12, altered mafic dike and sill swarms (Triassic?); 13, extent of Jurassic desert in Bolivia; 14, Jurassic eolian progradations. Jurassic eolianites in Paraná basin crop out just east of map limit. Data for Peru and Chile from Dalmayrac et al. (1980), Vicente et al. (1982), Soler and Bonhomme (1987), Gröschke et al. (1988), and Muñoz and Charrier (1993).
are believed to have been emplaced during the Middle Triassic extension. Other probable counterparts are the Cerro Sapo nepheline syenite and kimberlite-like rocks west of Cochabamba, mafic dikes intruding a Neoproterozoic pluton and Lower Ordovician strata south of Tarija, and an alnoite dike described in northwestern Argentina by Meyer and Villar (1984). The grabens were filled by fluvial and lacustrine red beds and evaporites (Figure 11), which permitted local preservation of upper Paleozoic strata. Clasts of Upper Permian–Middle Triassic granitoids are known in the upper Mitu Group. Extensional structures and magmatic rocks are widespread in western South America (e.g., Caputo, 1991; Ramos and Kay, 1991; Soler and Sempere, 1993). Reorganization of the regional stress field at 225–220 Ma produced local intracratonic transpression, as evidenced by the pervasive cataclasis of the Zongo-Yani (Bolivia) and Abancay (southern Peru) Triassic intrusions and deformation of their enclosing strata (Bard et al., 1974; Dalmayrac et al., 1980; Farrar et al., 1990). At least some grabens or transtensional structures were inverted during this event; examples are found in western Brazil (Caputo, 1991), eastern Peru (M. de Barros and E. Carneiro, 1992, personal communication) and northern Bolivia (YPFB, proprietary data). A 220–200 Ma regional metamorphic-plutonic event is also recognized in southwestern Ecuador (Tahuin Group) (Aspden et al., 1992).
Mild extension persisted in Peru after this event, resulting in transgression of the Norian sea from the northwest into the Cordillera Oriental area (Mégard, 1978; Dalmayrac et al., 1980; Jaillard et al., 1990). The Marine Norian–Liassic limestones locally overlie the Mitu Group with a slight angular unconformity (Mégard, 1978) and are in turn overlain in eastern Peru by fluvial and eolian deposits (Sarayaquillo Formation) that correlate with those of Bolivia (Serere supersequence). In southern coastal Peru, extension in the Yura trough culminated in Bajocian–Bathonian time; this trough was eventually filled in by the latest Jurassic (Vicente et al., 1982; Jaillard et al., 1990). The termination of rifting in Bolivia was probably a consequence of tectonic reorganization that induced local transpressional deformation. Subsequent Late Triassic– Middle Jurassic fluvial and eolian sedimentation was controlled by postrift thermal subsidence, reflected in onlapping relationships (Oller and Sempere, 1990) (Figure 11). Eolian deposits were widespread in western Gondwana during the Jurassic, including the Karoo and Paraná basins and sub-Andean central Peru (Boquerón Formation) (E. Bosc, 1986, personal communication), in agreement with global climatic models (Chandler et al., 1992). In Bolivia, the overall pattern of Jurassic eolian progradations (Figure 11) suggests that many dunes were longitudinal and produced by south-southwesterly winds.
Phanerozoic Evolution of Bolivia and Adjacent Regions
221
LATEST JURASSIC–TURONIAN The Kimmeridgian–Paleocene Puca Group of Bolivia (Figures 12, 13) records the evolution of the distal Andean back-arc basin. Bolivia was part of cratonic South America in the Jurassic, but was subsequently captured by the Andean system in the latest Jurassic during a period of large-scale rifting. This episode is attributed to onset of extensional and transtensional conditions in northern Chile and coastal Peru and was contemporaneous with the South Atlantic extension. In Bolivia, extensional reactivation of the CPKT formed the highly fragmented Potosí basin, which was filled with continental red beds (Figures 13, 14) (Sempere, 1994). Deposits of this age are known only from the Andean domain. Silcretes developed contemporaneously on the Jurassic sands in the sub-Andean-Beni-Chaco region. Rifting created a tilted block structural relief that was onlapped by fine-grained younger sediments. Rifting was accompanied by local alkaline basic volcanism (Soler and Sempere, 1993). The oldest and most important extensional episode took place during deposition of the Condo conglomerates in the lowermost Puca Group. Local inhomogeneities between the Kimmerigian– Hauterivian and Barremian–Turonian paleocurrent patterns (Figure 14) suggest that topography was modified by extensional processes during the Early Cretaceous. The intensity of extension appears to have waned through late Neocomian–Aptian time. By Albian time, fault-controlled subsidence was abandoned. Widespread regional subsidence is reflected in red mudstones that blanketed the old rift depocenter and onlapped the Paleozoic basement beyond. A postrift phase of thermal subsidence is inferred. The shallow marine Miraflores carbonates were deposited during the Cenomanian–Turonian interval (Jaillard and Sempere, 1991) (Figure 14). Organic-rich bioturbated limestones are conspicuous at the base (early Cenomanian) of the Miraflores Formation. This unit is thin and has a fairly uniform thickness, reflecting low sedimentation rates and the absence of synsedimentary tectonism. It is believed that this transgression had a global eustatic origin (Sempere, 1994).
Figure 12—Stratigraphic section for the Late Jurassic–early Eocene from the Miraflores syncline near Potosí, overlying gently folded Upper Ordovician strata. Key: 1, conglomerates; 2, sandstones; 3, predominantly mudrocks; 4, limestones; 5, red color; 6, green or gray color; 7, channels; 8, evaporites; 9, roots. Ranges of thicknesses in the basin are 0–1200 m for the Kimmerigian– Albian, 15–25 m for the Cenomanian–Turonian, and 200–1200 m for the Senonian–middle Paleocene.
An important consequence of this evolution is that Late Triassic tectonic reorganization not only terminated this period of widespread intracontinental extension but also apparently resulted in resumption of subduction along the Pacific margin.
SENONIAN–PALEOCENE The Senonian–Paleocene supersequence is best developed in Andean Bolivia, although some Campanian–Maastrichtian strata are known from the northwestern and central sub-Andean belt and the adjacent Beni-Chaco basin. In the Andean depocenter, this muddy supersequence is up to 1.2 km thick (Figure 12). It was deposited in the external part of a wide, underfilled foreland basin (Figure 15A) (Sempere, 1994). The Andes were beginning to form at that time, producing only a minor flexure of the South American lithosphere; this constitutes the earliest record of a true Andean-age foreland basin. This record consists of two sequences that span the Coniacian–Campanian (about 15-m.y. duration) and
222
Sempere
Figure 13—Restored sedimentary cross sections (after Sempere, 1994). See Figure 15A for locations. Strata are decompacted below the middle Selandian unconformity as reference level. Thicknesses are mainly from YPFB proprietary data and personal observations, but those of nonoutcropping rock units are hypothetical. (A) CPKT, Khenayani-Turuchipa paleostructural corridor (see Figure 1). (B) Thicknesses from Sevaruyo-Tambillo area are possibly exaggerated and need confirmation. This tentative restoration assumes a gross 50% shortening as a minimum and more shortening at major thrusts. Symbols: 1, mostly mudrocks (and limestones in the El Molino Formation); 2, mostly sandstones; 3, evaporites; eO, Early Ordovician; lO, Late Ordovician; eS, Early Silurian; lS, Late Silurian; eD, Early Devonian; lPz, late Paleozoic; R, Ravelo Fm (Late Triassic–Jurassic); Cn, Condo Fm; Ko, Kosmina Fm; LP, La Puerta Fm; Su, Sucre Fm; Ta, Tarapaya Fm; β, basic lava flows; M, Miraflores Fm; Ar, Aroifilla Fm; Ch, Chaunaca Fm; To, Torotoro Fm; Cr, Coroma Fm; EM, El Molino Fm; SL, Santa Lucía Fm (u, upper); Im, Impora Fm; Cy, Cayara Fm.
Maastrichtian–early Selandian (about 16-m.y. duration) intervals, respectively. There are a few organic-rich levels that are among the best hydrocarbon source rocks in the Bolivian Altiplano.
The base of the supersequence records a period of intense rifting, accompanied by relatively widespread alkaline basaltic volcanism (Soler and Sempere, 1993) and locally high subsidence rates. This extension is inter-
Phanerozoic Evolution of Bolivia and Adjacent Regions
223
An increase in the sedimentation rate and an influx of clean sand from the east near the Campanian–Maastrichtian boundary are attributed to reactivation of thrusting in the coastal orogen, accompanied by uplift of a forebulge in the sub-Andean-Beni-Chaco region. This episode coincides with a major transgression from the north or northwest in the Maastrichtian (Gayet et al., 1993). Restricted marine to lacustrine marl and carbonate facies locally rich in organic matter were deposited until the early Paleocene. Lacustrine and alluvial depositional environments characterized much of the Paleocene. Apart from early Selandian reactivation of the CPKT, there was no major tectonic activity. Uniform and steady subsidence suggests that flexural loading by thrust sheet encroachment was important in the coastal orogen (Sempere, 1994). A major modification of this tectonic system occurred in the middle Selandian when a widespread unconformity was formed in Andean Bolivia (Marshall and Sempere, 1991).
LATEST PALEOCENE–RECENT
Figure 14—Kimmeridgian–Turonian paleogeography in Bolivia (after Sempere,1994). The main depositional area is the Potosí basin. Tight lines are CPKT. Circled numbers: 1, Tawarreja-Thokori graben; 2, San Lorenzo-Tupiza graben; 3, Sevaruyo subbasin; 4, “Norte del Lago” basin. Kimmerigian–Hauterivian interval: 1, conglomeratic facies; 2, sandy facies; 3, basaltic flows; 4, paleocurrents. Barremian– Albian interval: 5, paleocurrents; 6, gypsum in Tarapaya Fm. Cenomanian–Turonian interval: 7, extent of Miraflores Fm; 8, paleocurrents obtained from orientation of gastropods or ripple marks.
preted as the foreland reaction to the onset of eastward encroachment of the Andes. In Bolivia, it reactivated Late Jurassic–Early Cretaceous extensional structures and produced new normal faults. Coniacian playa lakes formed in the structural lows. Santonian–Campanian deposits include red mudstones and subordinate sandstones and evaporites; there are two thin, organic-rich, restricted marine carbonate intercalations. This stratigraphy onlaps toward the margins, reflecting the distal foreland response to “Peruvian” age deformation affecting the coastal regions at that time (Jaillard, 1994, for Peru; Aspden et al., 1992, for Ecuador). A decrease in the rate of sedimentation and rapid progradation of immature sands from the west in the middle–late Campanian suggest erosion and relative quiescence of the orogen.
The Cenozoic evolution of Bolivia is subject to conflicting interpretations. However, there is a growing consensus that the central Andes were built by considerable crustal shortening (Roeder, 1988; Sempere et al., 1990a). All the basins that developed in the Bolivian mountains or plains during this time were related to the building of the Andes. Cenozoic basin evolution is expressed in a twofold record that contains latest Paleocene–early Oligocene (about 30-m.y. duration) and late Oligocene–Recent (about 27-m.y. duration) sequences of unlike characteristics. During the latest Paleocene–early Oligocene interval, Andean Bolivia was a foreland basin east of the Andes (Figure 15B). Coarsening- and thickening-upward continental red beds, 2–5 km thick, accumulated in what is now the Altiplano and Cordillera Oriental. This basin probably resembled the present-day Beni-Chaco plain. The existence of a continuously subsiding basin indicates that thrusting and tectonic loading were active west of the Altiplano. The base of the sequence records a significant change in sedimentation, approximately coinciding with the end of emplacement of the Peruvian coastal batholith, marking a turning point in the pattern of subduction (Sempere et al., 1989; Sempere, 1991b; Marshall and Sempere, 1991). This change is related to the onset of orthogonal convergence along the margin. A late Eocene or Oligocene uplift in the Cordillera Oriental is reflected in deposition of the Camargo conglomerates (Figure 15B). A tectonic, sedimentary, and magmatic upheaval developed in Bolivia during the late Oligocene and early Miocene (Figure 16) (Sempere et al., 1990a), involving a “jump” of the foreland basin from its Altiplano– Cordillera Oriental position to the present-day subAndean domain and Beni-Chaco plain. This major reorganization of geologic domains occurred only in the Bolivian orocline area (Figure 15B), suggesting that at
224
Sempere
Figure 15—Late Cretaceous–Cenozoic paleogeography. General legend same as in Figure 2. (A) Synthetic Senonian–Danian paleogeography. 1, rift axes; 2, marine-influenced Maastrichtian–Danian basin (maximum extent of Senonian–Danian basin); 3, areas undergoing deformation; 4, magmatic arc; 5, continental foreland sedimentation; 6, approximate extent of the mainly fluvial Cajones Fm (yielded dinosaur bones and an endemic fish known otherwise in Maastrichtian–Paleocene strata, but might be partially older); 7, Maastrichtian paleocurrents; 8, locations of Figure 13 cross sections. Bolivia-Peru basin was connected to open sea through Venezuela and possibly southern Ecuador (Gayet et al., 1993). Data partially from Marquillas and Salfity (1988) and Jaillard et al. (1993). (B) Distribution of main geologic features after the middle Paleocene. External foreland basins: hatchured, latest Paleocene–early Oligocene, gray, late Oligocene–Recent. Paleocurrents: 1, latest Paleocene–Eocene; 2, late Eocene or Oligocene (Camargo conglomerates); 3, late Oligocene–early Miocene. Magmatic arcs: 4, Eocene–early Oligocene; 5, late Oligocene–Recent.
least its eastern part formed at that time (Sempere et al., 1990a). Emergence of thrust deformation in the Cordillera Oriental split the basin into the coeval Altiplano (intermontane) and sub-Andean-Beni-Chaco (foreland) basins (Figure 16). The Bolivian Andes grew as a mountain belt from that time onward. This first major tectonic episode lasted about 8 m.y., starting at about 27 Ma. After a period of relative tectonic and magmatic quiescence between about 19 and 11 Ma, shortening and magmatism resumed. This produced accelerated subsidence of the sub-Andean-Beni-Chaco basin. In the late Miocene, restricted marine waters flooded the axial groove of the underfilled Chaco basin from the south. This transgression is reflected in deposition of the locally organic-rich Yecua Formation (Marshall et al., 1993). The late Miocene–Recent terrestrial fill of the sub-AndeanBeni-Chaco foreland basin is 3–5 km thick and coarsens and thickens upward. Seal rocks occur near its base and above the Yecua Formation. The present-day structure of the Andean fold and thrust belt of Bolivia involved décollement levels in Ordovician, Silurian, and Devonian shale-dominated units. Within the mountain belt, sedimentation continued in a variety of piggyback basins. The largest and most famous of these basins is the Altiplano, which has a complex structure (Sempere et al., 1990b). Magmatism has been active since the late Oligocene in the Cordillera Occidental, which is a subduction-related arc, and in the Altiplano and western Cordillera Occi-
dental, with some periods of increased activity (Sempere et al., 1990a,b). Preexisting structures have largely controlled the geometry and style of Andean deformation (Sempere et al., 1989, 1991; Hérail et al., 1990). Ordovician deformation and thickening in the Puna aulacogen area created a significant crustal heterogeneity, which probably influenced the formation of the Bolivian orocline (Sempere, 1990).
PETROLEUM GEOLOGY Oil and gas are produced in the central and southern sub-Andean belt and adjacent lowlands of Bolivia. Many oil seeps are known in the northwestern sub-Andean belt, as well as in the Altiplano. The stratigraphy of the productive sub-Andean and Chaco regions is divided into a lower Paleozoic–Jurassic pericratonic Gondwana section and an upper Oligocene–Recent Andean section. The sub-Andean fold and thrust belt is of late Neogene– Recent age. The principal source rocks in Bolivia are Devonian shales and Mississippian mudstones. Other potential source rocks include Silurian shales, Pennsylvanian– Permian shales and limestones, Upper Cretaceous–lower Paleocene limestone–mudstone units in the Altiplano, and possibly dark shales of the upper Miocene Yecua Formation in the central and southern sub-Andean foothills and adjacent Chaco basin.
Phanerozoic Evolution of Bolivia and Adjacent Regions
225
Figure 16—Synopsis of data concerning the late Oligocene–early Miocene tectonic crisis in Andean Bolivia (after Sempere et al., 1990a). BI, Illimani batholith; BQC, Quimsa Cruz batholith; CANP, main Andean thrust; CCR, Cordillera Real thrust; FCC, Coniri thrust front. (A) 1, Composite schematic section for northern Altiplano; arrows, azimuths of mean paleocurrents; 2, litho- and magnetostratigraphic sections at Salla, correlation to absolute time scale, and ages of nearby intrusions in BI and BQC; 3, favored age of Cenozoic heating of Zongo-Yani pluton; 4, stratigraphic section 60 km west-northwest of Santa Cruz, evidencing age of onset of sub-Andean-Beni-Chaco foreland basin. (B) Cross section illustrating activity of main Andean thrust system (CANP) and synchronous development of sub-Andean and Altiplano basins at ~20 Ma. Vertical arrows show location of data displayed above. Wavy pattern, pre-Ordovician rocks.
Reservoir intervals occur in numerous Silurian, Devonian, Carboniferous, Permian, Triassic, Jurassic, and upper Oligocene–middle Miocene sandstone units in the sub-Andean-Beni-Chaco region. Lower Silurian, Pennsylvanian, and Permian fractured limestones are also possible reservoirs. In the Altiplano, where no discoveries have been made, possible reservoirs include Upper Silurian–Lower Devonian sandstones, Upper Cretaceous–lower Paleocene limestones, and numerous Tertiary sandstone units. Although several seal rocks are interstratified in the Devonian–Mississippian succession, the thick and extensive upper Miocene mudstone unit is the important seal horizon in the sub-Andean-Beni-Chaco region. In the Altiplano, potential seals include several Upper Cretaceous–Eocene mudstone units. The timing of hydrocarbon maturation, migration, and trapping is poorly constrained because of the absence of an established burial history. Sediment accumulation was nearly continuous from the early Paleozoic to Jurassic in most of the Bolivia-Peru basin, albeit with Late Devonian–Mississippian uplift in some parts of the BeniChaco basin and Permian–Triassic uplift in Peru. In most
of the sub-Andean-Beni-Chaco region, subsidence was negligible from the Late Jurassic to Paleogene, resumed in the late Oligocene, and increased considerably in the late Miocene, resulting in rapid burial of Paleozoic source rocks beneath the Andean foreland basin. Tectonic burial occurred during propagation of Andean thrusting. However, there are differences in the style of foreland basin development in latest Paleocene– early Oligocene and late Oligocene–Recent time. Migration and trapping of hydrocarbons in the productive areas is believed to be related to these phenomena. Furthermore, earlier hydrocarbon accumulations may have been remobilized. Hydrocarbon generation may have occurred in Permian–Triassic time due to locally high geothermal gradients. In the still poorly explored Altiplano where only five exploration wells have been drilled, the burial of Paleozoic and Upper Cretaceous source rocks occurred during development of the latest Paleocene–early Oligocene foreland basin, that is, earlier than in the subAndean-Beni-Chaco region. Neogene magmatism probably produced locally high geothermal gradients in this area.
226
Sempere
DISCUSSION The Phanerozoic geology of Bolivia preserves a long record that includes the effects of the dispersal of the Late Proterozoic supercontinent, the initiation and evolution of the western Gondwana margin, the aggregation and consolidation of Pangea, its subsequent fragmentation, and the convergence effects of the westward motion of South America. In this history, the building of the central Andes mountains appears as a comparatively subordinate phenomenon, albeit of impressive magnitude, related to the dispersal of Pangea elements. There are several milestones that summarize the evolution of the Bolivia-Peru basin: • Extension: Late Cambrian–Early Ordovician, Ashgill–Llandovery, late Middle Triassic, latest Jurassic–Early Cretaceous, early Senonian • Compression and uplift (regional or local): Llanvirn–Caradoc, Late Devonian–Mississippian, Late Permian, Late Triassic, late Paleocene, late Oligocene–Neogene • Magmatism (in specific areas only): Early–Late Triassic, late Oligocene–Neogene • High subsidence rates: Late Cambrian– Mississippian, Senonian–Recent • Transgressions are grouped into three time intervals, with a decrease in amplitude of inundation: Ordovician–Mississippian, Pennsylvanian– Late Permian, Late Cretaceous–Danian Some points of interest for global comparisons are as follows: • The Middle Ordovician deformation in the Andes was coeval with the Taconic deformation of the inferred conjugate margin of eastern North America (Keppie, 1993). Development of approximately synchronous compressional deformation on both sides of the southern Iapetus ocean and in the western United States (Cotkin et al., 1992) suggests that a major plate reorganization occurred in the Middle Ordovician. • The Silurian–Devonian to Jurassic stratigraphies of Bolivia and the Paraná basin are similar. • The Late Devonian–Mississippian orogenicepeirogenic activity in the central Andean area was contemporaneous with the Antler orogeny in western North America (Smith et al., 1993). • Eolianites occur in the Pennsylvanian, Early Permian, and Jurassic of Bolivia, similar to the western United States (Blakey et al., 1988), which occupied a symmetric geotectonic location relative to the equator. • The Senonian–early Paleocene underfilled foreland basin of the Andes shows similarities with the Western Interior seaway of the United States, reflecting similarities in early Cordilleran-type evolution.
• Stromatolites are common in strata that span the Permian–Triassic (Sempere et al., 1992) and Cretaceous–Tertiary (Gayet et al., 1993) boundaries. The main aspects of the evolution of Bolivia and the central Andes can be grouped into three periods: Late Cambrian–Mississippian, Pennsylvanian–Middle Jurassic, and latest Jurassic–Recent.
Late Cambrian–Mississippian (≥ 205 m.y. duration) Interpretation of the evolution of the central Andean region is illustrated in Figures 4, 6, and 8. While North America separated from South America in the latest Proterozoic–Early Cambrian (Bond et al., 1984), the western margin of Gondwana evolved as a passive margin that was subjected to large-scale dextral shear. This led to the Middle–Late Cambrian formation of the Puna aulacogen above a thinned crust. Regional extension also favored the formation along the protoAndean domain of a wide marine epicontinental basin which was the locus of thick sedimentation. Closure and deformation of the Puna aulacogen propagated from the west in the Middle Ordovician, probably reflecting the onset of subduction along the western margin of Gondwana after an inferred plate reorganization. The rest of the basin evolved as the marine foreland basin of this deformation. High subsidence rates during latest Ordovician– earliest Mississippian time resulted from the transcurrent state of a wide tract running along the margin (Figures 6, 8), induced by oblique subduction. This foreland-type basin was filled with southwest-thickening, northeastonlapping sequences. It was eventually affected by local deformation and uplift in the Late Devonian–Mississippian. The source rocks associated with known hydrocarbon reserves in Bolivia were deposited during this period.
Pennsylvanian–Middle Jurassic (~165 m.y. duration) The consolidation and subsequent fragmentation of Pangea occurred during the Pennsylvanian–Middle Jurassic interval, with a pivotal event in the Carnian when rifting developed worldwide (Veevers, 1988). In the central Andean region, active subduction is believed to have ceased in the Permian with consolidation of Pangea (Kay et al., 1989). The central Andean geologic evolution includes a complex suite of tectonic, magmatic, and sedimentary events. Amalgamation of Pangea formed an enormous thermal blanket, inducing heat anomalies in the mantle. Deep-rooted thermal anomalies probably resulted in fracturing the relatively nonrefractory young Andean crust along preexisting zones of weakness (Mpodozis and Kay, 1992). This dislocation was expressed in extensional and transtensional structures at the surface. Voluminous silicic magmatism started as early as Pennsylvanian time in northern Chile (Breitkreuz et al.,
Phanerozoic Evolution of Bolivia and Adjacent Regions 1989). In the Peruvian Cordillera Oriental, a still poorly understood but probably transpressional deformation of Late Permian age created an intracratonic foldbelt that was subsequently intruded by calc-alkaline, posttectonic granites of Early–Middle Triassic age. This magmatism evolved to tholeiitic composition at the same time as several grabens formed, indicating the onset of extensional stress fields in the late Middle Triassic. A subsequent reorganization of regional stresses, attributed to a resumption of subduction beneath western Gondwana, terminated rifting in the Late Triassic and resulted in local transpression and basin inversion. This period was followed by renewed extension in western Andean Peru and a Norian transgression in eastern Andean and subAndean Peru.
Latest Jurassic–Recent (≥ 145 m.y. duration) Mild extensional conditions prevailed in Bolivia from Middle Triassic to Coniacian time, although local transpressional deformation occurred in the Late Triassic. Along the margin, Triassic–Albian transtensional conditions were subsequently replaced by transpression (Jaillard, 1994). Emplacement of the coastal batholith of Peru began in middle–late Albian time (Soler and Bonhomme, 1990). Six transgressions of marine to restricted marine waters are recorded: Cenomanian–Turonian, early Santonian, middle Campanian, early Maastrichtian, Danian, and late Miocene. The first five entered Bolivia from the north and the last one from the south. Development of the last five was related to foreland basin dynamics. Compressional building of the central Andes started in the late Turonian or early Coniacian in the west and subsequently propagated discontinuously toward the east. Thrust deformation reached the Paleozoic basin of Bolivia in late Oligocene time and has since propagated great distances through this succession in the southwestern part of the basin. Andean evolution is a consequence of convergence directly related to the break-up of Pangea and subsequent plate motions. The style of deformation in the central Andes appears to be controlled by preexisting structures and by the thickness and lithologies of basin fills. The productive Beni-Chaco basin of Bolivia is characterized by a twofold history: first, an early Paleozoic– Jurassic pericratonic Gondwana evolution during which time the principal source and reservoir rocks were deposited, and second, a late Oligocene–Recent Andean episode of tectonic evolution and burial.
Acknowledgments This work is a result of an extensive program on the geology of Bolivia initiated in 1983 and funded by UR 1H of Orstom. Field and stratigraphic studies have been conducted since 1984 in close collaboration with Yacimientos Petrolíferos Fiscales Bolivianos (YPFB). I particularly thank the Gerencia de Exploración and staff and field geologists of YPFB
227
for their help and constant support. The manuscript benefited from comments by N. Eyles and P. E. Isaacson, and from editing work by A. J. Tankard. Thanks to J. Oller, E. Aguilera, Y. Ardaya, G. Beccar, M. G. Bonhomme, R. F. Butler, G. Carlier, L. Coudert, E. Díaz-Martínez, G. Hérail, E. Jaillard, O. López, R. Marocco, L. G. Marshall, D. Merino, G. Montemurro, M. Pérez, E. Requena, A. Sadud, E. Salinas, P. Soler, H. Toledo, E. Vargas, D. Zubieta, and many others for fruitful discussions or additional information.
REFERENCES CITED Allmendinger, R. W., V. A. Ramos, T. E. Jordan, M. Palma, and B. L. Isacks, 1983, Paleogeography and Andean structural geometry, northwest Argentina: Tectonics, v. 2, p. 1–16. Archangelsky, S., ed., 1987, El sistema Carbonífero en la República Argentina: Academia Nacional de Ciencias, Córdoba, 383 p. Aspden, J. A., S. H. Harrison, and C. C. Rundle, 1992, New geochronological control for the tectono–magmatic evolution of the metamorphic basement, Cordillera Real and El Oro Province of Ecuador: Journal of South American Earth Sciences, v. 6, p. 77–96. Avila, W. A., 1992, El magmatismo cámbrico-ordovícico en Bolivia, in J. G. Gutiérrez Marco, J. Saavedra, and I. Rábano, eds., Paleozoico Inferior de Ibero-América: Universidad de Extremadura, Spain, p. 241–253. Bahlburg, H., 1990, The Ordovician basin in the Puna of NW Argentina and N Chile: geodynamic evolution from backarc to foreland basin: Geotektonische Forschungen, v. 75, 107 p. Bard, J. P., R. Botello, C. Martinez, and T. Subieta, 1974, Relations entre tectonique, métamorphisme et mise en place d’un granite éohercynien à deux micas dans la Cordillère Real de Bolivie (massif de Zongo-Yani): Cahiers Orstom, série Géologie, v. 6, p. 3–18. Bell, C. M., 1987, The Late Paleozoic evolution of the Gondwanaland continental margin in northern Chile, in G. D. MacKenzie, ed., Gondwana six: structure, tectonics, and geophysics: American Geophysical Union, Geophysical Monographs, v. 40, p. 261–270. Benedetto, J. L., T. M. Sánchez, and E. D. Brussa, 1992, Las cuencas silúricas de América Latina, in J. G. Gutiérrez Marco, J. Saavedra and I. Rábano, eds., Paleozoico Inferior de Ibero-América: Universidad de Extremadura, Spain, p. 119–148. Blakey, R. C., F. Peterson, and G. Kocurek, 1988, Synthesis of late Paleozoic and Mesozoic eolian deposits of the Western Interior of the United States: Sedimentary Geology, v. 56, p. 3–125. Bond, G. C., P. A. Nickeson, and M. A. Kominz, 1984, Breakup of a supercontinent between 625 Ma and 555 Ma: new evidence and implications for continental histories: Earth and Planetary Science Letters, v. 70, p. 325–345. Bourgois, J., P. Huchon, and G. Pautot, 1990, Geología de la margen activa del Perú entre los 3° y 12° de latitud Sur: Bulletin de l’Institut Français d’Etudes Andines, v. 19, p. 241–291. Branisa, L., G. A. Chamot, W. B. N. Berry, and A. J. Boucot, 1972, Silurian of Bolivia, in W. B. N. Berry and A. J. Boucot, eds., Correlations of the South American Silurian Rocks: GSA Special Paper 133, p. 21–31.
228
Sempere
Breitkreuz, C., 1986, Das Paläozoikum in den Kordilleren Nordchiles (21°–25°S): Geotektonische Forschungen, v. 70, 88 p. Breitkreuz, C., 1991, Fluvio-lacustrine sedimentation and volcanism in a Late Carboniferous tensional intra-arc basin, northern Chile: Sedimentary Geology, v. 74, p. 173–187. Breitkreuz, C., H. Bahlburg, B. Delakowitz, and S. Pichowiak, 1989, Volcanic events in the Paleozoic central Andes: Journal of South American Earth Sciences, v. 2, p. 171–189. Caputo, M. V., 1991, Solimões megashear: intraplate tectonics in northwestern Brazil: Geology, v. 19, p. 246–249. Chandler, M. A., D. Rind, and R. Ruedy, 1992, Pangean climate during the Early Jurassic: GCM simulations and the sedimentary record of paleoclimate: GSA Bulletin, v. 104, p. 543–569. Cobbold, P., A. C. Massabié, and E. A. Rossello, 1986, Hercynian wrenching and thrusting in the Sierras Australes foldbelt, Argentina: Hercynica, v. 2, p. 135–148. Cordani, U., K. Kawashita, L. Baeza, M. Díaz, and D. Evangelista, 1988, Geocronología de la Sierra Limón Verde, Antofagasta, Chile: Fifth Congreso Geológico Chileno, Santiago, v. 2, p. E63–E74. Cotkin, S. J., M. L. Cotkin, and R. L. Armstrong, 1992, Early Paleozoic blueschist from the schist of Skookum Gulch, eastern Klamath Mountains, northern California: Journal of Geology, v. 100, p. 323–328. Dalmayrac, B., G. Laubacher, and R. Marocco, 1980, Caractères généraux de l’évolution géologique des Andes péruviennes: Travaux et Documents de l’Orstom, Paris, v. 122, 501 p. Damm, K. W., S. Pichowiak, R. S. Harmon, W. Todt, S. Kelley, R. Omarini, and H. Niemeyer, 1990, Pre-Mesozoic evolution of the central Andes; the basement revisited, in S. M. Kay and C. W. Rapela, eds., Plutonism from Antarctica to Alaska: GSA Special Paper 241, p. 101–126. Díaz-Martínez, E., 1991, Litoestratigrafía del Carbonífero del Altiplano de Bolivia: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 295–302. Dyck, M., 1992, Stratigraphy, fauna, sedimentology and tectonics in the Ordovician and Silurian of east-Paraguay and the comparison with the Argentinian and Bolivian Andes (abs.): Conferencia Internacional sobre el Paleozoico Inferior de Ibero-América, Publicaciones del Museo de Geología de Extremadura, Spain, v. 1, p. 64–65. Farrar, E., A. H. Clark, and S. M. Heinrich, 1990, The age of the Zongo pluton and the tectonothermal evolution of the Zongo–San Gabán Zone in the Cordillera Real, Bolivia, in First International Symposium on Andean Geodynamics, Grenoble: Paris, Editions de l’Orstom, p. 171–174. França, A. B., and P. E. Potter, 1991, Stratigraphy and reservoir potential of glacial deposits of the Itararé Group (Carboniferous–Permian), Paraná basin, Brazil: AAPG Bulletin, v. 75, p. 62–85. Gagnier, P. Y., A. Blieck, C. C. Emig, T. Sempere, D. Vachard, and M. Vanguestaine, in press, New data on the Ordovician and Silurian of Bolivia: Journal of South American Earth Sciences. Gayet, M., T. Sempere, H. Cappetta, E. Jaillard, and A. Lévy, 1993, La présence de fossiles marins dans le Crétacé terminal des Andes centrales et ses conséquences paléogéographiques: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 102, p. 283–319. Götze, H. J., S. Schmidt, and S. Strunk, 1988, Central Andean gravity field and its relation to crustal structures, in H. Bahlburg, C. Breitkreuz, and P. Giese, eds., The Southern
Central Andes: Lecture Notes in Earth Sciences: Berlin, Springer–Verlag, v. 17, p. 199–208. Gröschke, M., A. von Hillebrandt, P. Prinz, L. A. Quinzio, and H. G. Wilke, 1988, Marine Mesozoic paleogeography in northern Chile between 21°–26°S, in H. Bahlburg, C. Breitkreuz, and P. Giese, eds., The southern central Andes: Lecture Notes in Earth Sciences, Berlin, Springer–Verlag, v. 17, p. 105–117. Helwig, J., 1972, Stratigraphy, sedimentation, paleogeography, and paleoclimates of Carboniferous (“Gondwana”) and Permian of Bolivia: AAPG Bulletin, v. 56, p. 1008–1033. Hérail, G., P. Baby, J. Oller, M. López, O. López, R. Salinas, T. Sempere, G. Beccar, and H. Toledo, 1990, Structure and kinematic evolution of the sub-Andean thrust system of Bolivia, in First International Symposium on Andean Geodynamics, Grenoble: Paris, Editions de l’Orstom, p. 179–182. Isaacson, P. E., 1975, Evidence for a western extracontinental land source during the Devonian period in the central Andes: GSA Bulletin, v. 86, p. 39–46. Isaacson, P. E., B. Antelo, and A. J. Boucot, 1976, Implications of a Llandovery (Early Silurian) brachiopod fauna from Salta Province, Argentina: Journal of Paleontology, v. 50, p. 1103–1112. Jaillard, E., 1994, Tectonic evolution of the Peruvian margin between Kimmeridgian and Paleocene times, in J. A. Salfity, ed., Cretaceous tectonics in the Andes: Earth Evolution Sciences Monograph Series, Wiesbaden, Vieweg Publications, p. 101–167. Jaillard, E., and T. Sempere, 1991, Las secuencias sedimentarias de la Formación Miraflores y su significado cronoestratigráfico: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 257–264. Jaillard, E., P. Soler, G. Carlier, and T. Mourier, 1990, Geodynamic evolution of the northern and central Andes during early to middle Mesozoic times: a Tethyan model: Journal of the Geological Society of London, v. 147, p. 1009–1022. Jaillard, E., V. Carlotto, J. Cárdenas, R. Chávez, and W. Gil, 1993, La “Nappe des Couches Rouges” de Cuzco (Sud du Pérou): mise en évidence stratigraphique, interprétations tectoniques et paléogéographiques: Comptes Rendus à l’Académie des Sciences de Paris, série II, v. 316, p. 379–386. Kay, S. M., V. A. Ramos, C. Mpodozis, and P. Sruoga, 1989, Late Paleozoic to Jurassic silicic magmatism at the Gondwana margin: analogy to the middle Proterozoic in North America?: Geology, v. 17, p. 324–328. Kennan, L., 1993, Cenozoic evolution of the Cochabamba area, Bolivia, in Second International Symposium on Andean Geodynamics, Oxford: Paris, Editions de l’Orstom, p. 199–202. Keppie, J. D., 1993, Synthesis of Paleozoic deformational events and terrane accretion in the Canadian Appalachians: Geologische Rundschau, v. 82, p. 381–431. Kontak, D. J., A. H. Clark, E. Farrar, and D. F. Strong, 1985, The rift associated Permo-Triassic magmatism of the eastern Cordillera: a precursor to the Andean orogeny, in W. S. Pitcher, M. P. Atherton, J. Cobbing and R. D. Beckinsale, eds., Magmatism at a plate edge: the Peruvian Andes: London, John Wiley, p. 36–44. Laubacher, G., 1978, Géologie de la Cordillère Orientale et de l’Altiplano au nord et nord-ouest du lac Titicaca (Pérou): Travaux et Documents de l’Orstom, Paris, v. 95, 217 p. Lobo-Boneta, J., 1975, Sobre algunos palinómorfos del Devónico superior y Carbónico inferior de la zona subandina sur de Bolivia: Revista Técnica de YPFB, v. 4, p. 59–176.
Phanerozoic Evolution of Bolivia and Adjacent Regions López, O., G. Montemurro, and H. Trujillo, 1982, Estratigrafía del Paleozoico medio en las Serranías de San José y Santiago de Chiquitos, Bolivia: 5th Proc. Congreso Latinoamericano de Geología, Buenos Aires, v. 1, p. 283–292. Marquillas, R. A., and J. A. Salfity, 1988, Tectonic framework and correlations of the Cretaceous–Eocene Salta Group, Argentina, in H. Bahlburg, C. Breitkreuz, and P. Giese, eds., The southern central Andes: Lecture Notes in Earth Sciences, Berlin, Springer–Verlag, v. 17, p. 119–136. Marshall, L. G., and T. Sempere, 1991, The Eocene to Pleistocene vertebrates of Bolivia and their stratigraphic context: a review, in R. Suárez-Soruco, ed., Fósiles y Facies de Bolivia: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 631–652. Marshall, L. G., T. Sempere, and M. Gayet, 1993, The Petaca (late Oligocene–middle Miocene) and Yecua (late Miocene) formations of the sub-Andean–Chaco basin, Bolivia, and their tectonic significance: Documents du Laboratoire de Géologie de Lyon, France, v. 125, p. 291–301. Martinez, C., 1980, Structure et évolution de la chaîne hercynienne et de la chaîne andine dans le nord de la Cordillère des Andes de Bolivie: Travaux et Documents de l’Orstom, Paris, v. 119, 352 p. McGregor, D. C., 1984, Late Silurian and Devonian spores from Bolivia: Academia Nacional de Ciencias de Córdoba, v. 69, p. 1–57. Mégard, F., 1978, Etude géologique des Andes du Pérou central: Travaux et Documents de l’Orstom, Paris, v. 86, 310 p. Merino, D., 1987, Los conodontos permocarboníferos de la Formación Copacabana en Yaurichambi, Departamento de La Paz: Tesis de Grado de la Universidad Mayor de San Andrés, La Paz, 135 p. Merino, D., 1991, Primer registro de conodontos silúricos en Bolivia: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 271–274. Merino, D., and J. Blanco, 1990, Conodontos de la Formación Copacabana (Carbonífero superior-Pérmico inferior) en la sección de Huarachani-Pacobamba, Departamento de La Paz, Bolivia: Revista Técnica de YPFB, Cochabamba, v. 11, p. 109–116. Meyer, H. A. O., and Villar, L. M., 1984, An alnoite in the Sierras Subandinas, northern Argentina: Journal of Geology, v. 92, p. 741–751. Moya, M. C., 1988, Lower Ordovician in the southern part of the Argentine eastern Cordillera, in H. Bahlburg, C. Breitkreuz, and P. Giese, eds., The southern central Andes: Lecture Notes in Earth Sciences, Berlin, Springer–Verlag, v. 17, p. 55–69. Mpodozis, C., and S. M. Kay, 1992, Late Paleozoic to Triassic evolution of the Gondwana margin: evidence from Chilean frontal Cordilleran batholiths: GSA Bulletin, v. 104, p. 999–1014. Mukasa, S. B., and D. J. Henry, 1990, The San Nicolás batholith of coastal Peru: early Paleozoic continental arc or continental rift magmatism?: Journal of the Geological Society of London, v. 147, p. 27–39. Muñoz, N., and R. Charrier, 1993, Jurassic–Early Cretaceous facies distribution in the western Altiplano (18°–21°30’ S.L.)—implications for hydrocarbon exploration, in Second International Symposium on Andean Geodynamics, Oxford: Paris, Editions de l’Orstom, p. 307–310. Murphy, J. B., and R. D. Nance, 1991, Supercontinent model for the contrasting character of Late Proterozoic orogenic belts: Geology, v. 19, p. 469–472. Niemeyer, H., F. Urzúa, F. G. Aceñolaza, and C. González, 1985, Progresos recientes en el conocimiento del Paleozoico
229
de la Región de Antofagasta: 4th Congreso Geológico Chileno, Antofagasta, v. 1, p. 439–457. Oller, J., and T. Sempere, 1990, A fluvio-eolian sequence of probable Middle Triassic–Jurassic age in both Andean and sub-Andean Bolivia, in First International Symposium on Andean Geodynamics, Grenoble: Paris, Editions de l’Orstom, p. 237–240. Pérez, M., 1990, Miospores du Dévonien moyen et supérieur de la coupe de Bermejo-La Angostura (sud-est de la Bolivie): Annales de la Société Géologique de Belgique, v. 113, p. 373–389. Racheboeuf, P. R., A. Le Hérissé, F. Paris, C. Babin, F. Guillocheau, M. Truyols-Massoni, and R. Suárez-Soruco, 1993, Le Dévonien de Bolivie: Biostratigraphie et chronostratigraphie: Comptes Rendus à l’Académie des Sciences de Paris, série II, v. 317, 795–802. Ramos, V. A., 1988a, Late Proterozoic–early Paleozoic of South America—a collisional history: Episodes, v. 11, p. 168–174. Ramos, V. A., 1988b, The tectonic of the central Andes, 30°–33°S latitude, in S. P. Clark, Jr., B. C. Burchfiel, and J. Suppe, eds., Processes in continental lithospheric deformation: GSA Special Paper 218, p. 31–54. Ramos, V. A., and S. M. Kay, 1991, Triassic rifting and associated basalts in the Cuyo basin, central Argentina, in R. S. Harmon and C. W. Rapela, eds., Andean magmatism and its tectonic setting: GSA Special Paper 265, p. 79–91. Rapela, C. W., B. Coira, A. Toselli, and J. Saavedra, 1992, El magmatismo del Paleozoico inferior en el Sudoeste de Gondwana, in J. G. Gutiérrez-Marco, J. Saavedra, and I. Rábano, eds., Paleozoico Inferior de Ibero-América: Universidad de Extremadura, Spain, p. 21–68. Rodrigo, L. A., A. Castaños, and R. Carrasco, 1977, La Formación Cancañiri, sedimentología y paleogeografía: Revista de Geociencias de la Universidad Mayor de San Andrés, La Paz, v. 1, p. 1–22. Roeder, D., 1988, Andean-age structure of eastern Cordillera (Province of La Paz, Bolivia): Tectonics, v. 7, p. 23–39. Rösler, O., R. Ianuzzi, and R. Suárez-Soruco, 1989, A flora carbonífera da Formação Kasa em Belén, península de Copacabana, Altiplano boliviano, e a importancia das formas trifoliadas: 11th Congresso Brasileiro de Paleontologia, Curitiba, Abstracts, p. 45–46. Saunders, W. B., and W. H. C. Ramsbottom, 1986, The midCarboniferous eustatic event: Geology, v. 14, p. 208–212. Sempere, T., 1989, Paleozoic evolution of the central Andes (10°–26° S): 28th International Geological Congress, Extended Abstracts, Washington, D.C., v. 3, p. 73. Sempere, T., 1990, Cuadros estratigráficos de Bolivia: Propuestas nuevas: Revista Técnica de YPFB, Cochabamba, v. 11, p. 215–227. Sempere, T., 1991a, Evolución de la cuenca centro-andina (10°–26° S) del Cámbrico superior al Silúrico inferior: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 221–223. Sempere, T., 1991b, Cenozoic tectonic “phases” in Bolivia: some needed clarifications: 6th Congreso Geológico Chileno, Viña del Mar, v. 1, p. 877–881. Sempere, T., 1994, Kimmeridgian? to Paleocene tectonic evolution of Bolivia, in J. A. Salfity, ed., Cretaceous tectonics in the Andes: Earth Evolution Sciences Monograph Series, Wiesbaden, Vieweg Publications, p. 168–212. Sempere, T., G. Hérail, J. Oller, and P. Baby, 1989, Geologic structure and tectonic history of the Bolivian orocline: 28th International Geological Congress, Extended Abstracts, Washington, D.C., v. 3, p. 72–73.
230
Sempere
Sempere, T., G. Hérail, J. Oller, and M. G. Bonhomme, 1990a, Late Oligocene–early Miocene major tectonic crisis and related basins in Bolivia: Geology, v. 18, p. 946–949. Sempere, T., G. Hérail, J. Oller, P. Baby, L. Barrios, and R. Marocco, 1990b, The Altiplano: a province of intermontane foreland basins related to crustal shortening in the Bolivian orocline area, in First International Symposium on Andean Geodynamics, Grenoble: Paris, Editions de l’Orstom, p. 167–170. Sempere, T., P. Baby, J. Oller, and G. Hérail, 1991, La napa de Calazaya: una prueba de acortamientos importantes controlados por elementos paleoestructurales en los Andes bolivianos: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 229–234. Sempere, T., E. Aguilera, J. Doubinger, P. Janvier, J. Lobo, J. Oller, and S. Wenz, 1992, La Formation de Vitiacua (Permien moyen à supérieur–Trias ?inférieur, Bolivie du Sud): stratigraphie, palynologie et paléontologie: Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen, v. 185, p. 239–253. Shackleton, R. M., A. C. Ries, M. P. Coward, and P. R. Cobbold, 1979, Structure, metamorphism and geochronology of the Arequipa Massif of coastal Peru: Journal of the Geological Society of London, v. 136, p. 195–214. Smith, M. T., W. R. Dickinson, and G. E. Gehrels, 1993, Contractional nature of Devonian–Mississippian Antler tectonism along the North American continental margin: Geology, v. 21, p. 21–24. Soler, P., and M. G. Bonhomme, 1987, Données radiochronologiques K-Ar sur les granitoïdes de la Cordillère Orientale des Andes du Pérou central: implications tectoniques: Comptes Rendus à l’Académie des Sciences de Paris, série II, v. 304, p. 841–844. Soler, P., and M. G. Bonhomme, 1990, Relations of magmatic activity to plate dynamics in central Peru from Late Cretaceous to present, in S. M. Kay and C. W. Rapela, eds., Plutonism from Antarctica to Alaska: GSA Special Paper 241, p. 173–191. Soler, P., and T. Sempere, 1993, Stratigraphie, géochimie et signification paléotectonique des roches volcaniques basiques mésozoïques des Andes boliviennes: Comptes Rendus à l’Académie des Sciences de Paris, série II, v. 316, p. 777–784.
Suárez-Riglos, M., M. A. Hünicken, and D. Merino, 1987, Conodont biostratigraphy of the Upper Carboniferous–Lower Permian rocks of Bolivia, in R. L. Austin, ed., Conodonts: Investigative Techniques and Applications: Amsterdam, Elsevier, p. 316–322. Suárez-Soruco, R., 1976, El sistema ordovícico en Bolivia: Revista Técnica de YPFB, v. 5, p. 111–223. Suárez-Soruco, R., and J. M. López–Pugliessi, 1983, Formación Saipurú, nuevo nombre formacional para representar a los sedimentos superiores del Ciclo Cordillerano (Devónico superior-Carbónico inferior): Revista Técnica de YPFB, v. 9, p. 209–213. Toro, M., C. Vargas, and R. Birhuet, 1992, Los trilobites ashgillianos de la Formación Cancañiri (región de Milluni, Cordillera Real, departamento de La Paz): 10th Congreso Geológico Boliviano, Extended Abstracts, La Paz, p. 188–190. Vavrdová, M., P. E. Isaacson, E. Díaz-Martínez, and J. Bek, 1991, Palinología del límite Devónico–Carbonífero en torno al Lago Titikaka, Bolivia: resultados preliminares: Revista Técnica de YPFB, Santa Cruz, v. 12, p. 303–313. Veevers, J. J., 1988, Gondwana facies started when Gondwana merged in Pangea: Geology, v. 16, p. 732–734. Veevers, J. J., and C. McA. Powell, 1987, Late Paleozoic glacial episodes in Gondwana reflected in transgressive-regressive depositional sequences in Euramerica: GSA Bulletin, v. 98, p. 475–487. Vicente, J. C., B. Beaudoin, A. Chávez, and I. León, 1982, La cuenca de Arequipa (Sur Perú) durante el JurásicoCretácico inferior: 5th Congreso Latinoamericano de Geología Proceedings, Buenos Aires, v. 1, p. 121–153. Zubieta-Rossetti, D., P. Huyghe, G. Mascle, J. L. Mugnier, and P. Baby, 1993, Influence de l’héritage anté-dévonien au front de la chaîne andine (partie centrale de la Bolivie): Comptes Rendus à l’Académie des Sciences de Paris, série II, v. 316, p. 951–957.
Author’s Mailing Address
Thierry Sempere Consulting Geologist 13 rue Geoffroy l’Angevin 75004 Paris France
