Tectonic assembly of the Northern Andes
The present-day Northern Andes occupy the north-western corner of South America. This mountain range is dominated by four principal lithospheric plates: the Cocos, Nazca and Caribbean plates of oceanic affinity, and the continental South American plate (Fig. 2.1). The South America plate is dominated in this region by the Precambrian Guyana Shield.
The Northern Andes differ significantly from the Central Andes (Fig. 2.1) in many aspects, including the nature and age of the underlying basement and continental margin, the nature and evolution of stress regimes during uplift, the nature and age of subducting oceanic crust, and the timing and style of deformation and magmatism.
The Northern Andes are an assemblage of different continental blocks, oceanic plateaux, ridges and intra-oceanic arc complexes, which collided with the northern South American plate since the Jurassic along a variety of newly developed or reactivated fault zones (Fig. 2.2). Because of the complexity of the tectonic relationships between the different blocks, plateaux and ridges, the exact mechanisms and timing of collision remain a matter of debate (Cediel et al. 2003). This is based mainly on different authors (Case et al. 1990; Aleman and Ramos, 2000; Cediel et al. 2003, Montes et al. 2005).
The Northern Andes can be subdivided into the Venezuelan Andes (Mérida Andes), the Colombian Andes and the Ecuadorian Andes (Fig. 2.1). The Colombian and Ecuadorian Andes are typical products of subduction and correlative processes (magmatism, shortening, collisions of allochtonous terranes, etc). In contrast, the Venezuelan Andes is an intra-continental mountain belt resulting from the interaction between the Caribbean, Nazca and South America Plates resulting in Neogene tectonic inversion. The difference in orientation of the Colombian and Ecuadorian ranges with respect to Venezuelan Andes result from variations during the Mesozoic extension (Aleman and Ramos, 2000) and the fact that Western Colombia was located in the area affected by the passage of the Caribbean plate. The Colombian Andes consist of three distinct and separate chains: the Western, Central and Eastern Cordillera (Fig. 2.2) and are bounded to the east by the Romeral fault system. This fault separates the Western and Central Cordilleras and has been interpreted as a suture or subduction zone (Toussaint and Restrepo, 1982). The Western Cordillera is constituted by Cretaceous tholeiitic basalt and deep-water sediments that rest on the oceanic crust. In the Baudó range, located in the Western Cordillera, oph iolites have been reported (Toussaint, 1978).
The Central and Eastern Cordillera are separated by the Magdalena Depression and are underlain by continental crust. The separation of these chains mentioned previously dates back to the Mesozoic extensional phase. A first collision of the Dagua-Piñón Terrane (DAP; Fig. 2.2) during the Late Cretaceous produced the exhumation of the Central Cordillera (Aleman and Ramos, 2000) and the deformation of the Eastern Cordillera. This collision and the Late Cretaceous docking of the DAP also affected the Ecuadorian Andes, causing shortening and thrusting of Mesozoic oceanic crust and the accretion of the Western Cordillera into a single geomorphical unit. The extrusion of the Caribbean plate was associated with middle Eocene and late Miocene accretion in the Western Cordillera. The complex plate interaction prior to the formation of the Caribbean plate produced different domains that were later accreted to the northern part of the Western Colombia. This periodic terrain docking enhanced by oblique convergence has led to strain partitioning and continuously reactivated old suture zones within the Northern Andes.
To south, Ecuador can be divided into three geographic zones from east to west (Fig. 2.2): the eastern basin or « East », the Ecuadorian Andes or « Sierra », and the coastal zone or « Costa » (Amortegui, 2007). The eastern basin is filled with continental red Jurassic sandstones, marls and calcareous marine Cretaceous, Palaeogene deposits topped with patchy continental to brackish, coarse Miocene to Recent deposits. Dextral-slip faults form in depth positive flower structures explored by the oil industry. The Sierra is formed by the Cordillera Real and the Western Cordillera. The Cordillera Real to the east consists of Paleozoic and Mesozoic rocks metamorphosed, partly covered or intersected by intrusive rocks and Tertiary volcanic (Litherland et al. 1994). La Costa is the current zone of fore arc. This consists of a substrate of basic lavas, dolerites and pyroclastic rocks with oceanic affinity plateaus, crowned by island arc type rocks of Upper Cretaceous age (Vanmelle et al. 2008).
The Mérida or Venezuelan Andes are characterized bylate Tertiary uplift and exhumation and high current seismicity. This belt is crossed by different strike-slip faults: the northeast-southwest trending dextral Boconó, Central-Sur Andi no and Caparo faults and the north-south sinistral Icotea, Valera and Burbusay faults. The chain is flanked by two currently active foreland fold-and-thrust belts to the northwest and southeast, which incorporate Plio-Pleistocene sediments (Fig. 2.2). The basement includes Precambrian and Paleozoic metamorphic and igneous rocks, which are overlain by Paleozoic sedimentary rocks, Mesozoic bed reds, marine clastic and carbonate strata, and Tertiary marine and continental deposits with variable thickness. Most of the northeast-trending faults appear relatively steep at shallow crustal depths (Case et al. 1990). The Venezuelan Andes is separated from the Eastern Cordillera of Colombia by the Táchira Depression, along which the latter may overthrust the former to the Northeast (Macellari, 1984). In its central part, the Venezuelan Andes are split symmetrically by the Boconó Fault S ystem. To the east, the Trujillo Block (Backé et al. 2006) displays a more complex tectonic regime imposed by the convergence among the Boconó, Valera and Burbusay faults. Focal mechanism solutions of earthquakes in the northern part of the Trujillo block display compressional to transpressional shearing (Backé et al. 2006). Thus, the Venezuelan Andes present at least three contrasting histories for the Táchira Depression, the Central Venezuelan Andes and the Trujillo Block.
Despite the inherent complexity a simplified reconstruction of this part of South America is given here, in order to discuss the origin of the Northern Andes, and particularly of the Venezuelan Andes.
Pangaea break-up and origin of the Caribbean plate
The evolution of the Caribbean region started with the separation of North America, South America and Africa during Pangaea breakup in the early Jurassic (Wadge and Burke, 1983; Pindell and Kennan, 2001). This evolution has been marked by phases of extension, convergence, translation, subduction and exhumation, resulting in the current plate-tectonic configuration of the Caribbean region (Fig. 2.2; Chicangana, 2005).
Continental rifting and formation of structural discontinuities
Continental rifting creates and can exploit pre-existing weaknesses in the continental crust which can be reactivated during convergence and collision (Vauchez et al. 1997). For example, the break-up of Gondwana has left a system of fractures on the North American and South American continents. These fracture zones were formed during Permo-Triassic super plume activity (Maruyama, 1994) and Jurassic rifting. The Permo-Triassic super plume caused heating of the north-western South American lithosphere, and produce volcanic rocks which was contemporaneous with NE-SW extension along the axis of the Cordillera Central, on the La Guajira Peninsula (Maya, 2001), and in the Sierra Nevada de Santa Marta (Tschanz et al. 1974; Chicangana, 2005; Montes et al. 2005) in Colombia (Fig. 2.2). The super plume activity resulted in rifting and formation of a triple junction which eventually caused the separation of Laurentia from Gondwana (Fig. 2.3). Widespread rifting in north-western South American started during the Early Jurassic. The Uribante or Trujillo Rift and Espino Graben (Fig. 2.3A) in Venezuela are evidences for this process, which is responsible for the development of N30°W striking weaknesses in the cru st and possibly lithosphere of north-western South America. For example, the Romeral Fault System in Colombia (Fig. 2.2) represents today the former margin of South America (Von Estorff, 1946; James, 2000a). Therefore, paleo-structures are important for explaining the current tectonic structures in the Northern Andes (Fig. 2.2, 2.3B).
The modern strike-slip faults systems in northern South America (Fig. 2.2) are also related to the reactivation of these paleodiscontinuities (Figs. 2.3 A and B). The strike-slip systems control the recent exhumation patterns across the Northern Andes as will be discussed in the following chapters. During the Middle to Late Jurassic, a magmatic arc developed along the margin of South America. This magmatic arc crops out today from the Sierra Nevada de Santa Marta to the north of the Cordillera Real in south-eastern Ecuador (Figs. 2.2 and 2.3B). The origin of this magmatic arc is related to subduction of the Pacific plate beneath the South American plate that initiated in the Middle Jurassic (Chicangana, 2005). The regional tectonic development during the Middle to Late Jurassic shows the formation of large shear zones as a result of a transtensive regime, which originated from rifting in a back-arc rift setting, related to subduction of the Pacific plate (Jaillard et al. 1990). Two sets of major pre-Mesozoic crustal discontinuities (Von Estorff, 1946; Mora et al. 1993; James, 2000a, Jacques, 2004) can be recognized in the cratonic areas of northern South America and constitute a rectangular network (Fig. 2.3B). A N 50ºE trend is parallel to Espino and Apure grabens. The other set is oriented N40ºW, parallel to the northeastern continental margin of South America, the El Baúl and Mérida arches, the southeastern projection of the Santa Marta fault and the Sierra La Macarena (Fig. 2.2 and 2.3).
Fig. 2.3. A) Early Jurassic plate tectonic reconstruction (modified from Pindell and Kennan, 2001). B) Fault pattern in north-western South America. These (or similar) structures originated from the Pangaea breakup: 1. Paleo-Boconó fault limited the Uribante or Trujillo Rift. 2. El Baúl Arc, 3. Apure-Mantecal Graben, 4. Llanos Graben, 5. Guyana Shield, 6. Western Cordillera, 7. Cordillera de la Macarena, 8. Sierra de Perijá, 9 .Cordillera Central, 10. Eastern Cordillera, 11. Santa Marta-Bucaramanga Fault, 11. Romeral Fault, 12. Cordillera Real and 14. Ibague Fault (modified after Von Estorff, 1946).
According to Mora et al. (1993), the tectonic reconstructions suggest that, depending on the angular relation between the displacement field of the surrounding plates and the orientation of the discontinuities, some of them were activated as extensional systems while others suffered transpressional deformation. Due to changes of the relative displacement fields through time, a given discontinuity may show inversion with respect to its previous behaviour. The network of discontinuities bounds blocks which suffered translation and rotation with respect to their neighbours and, in some cases, minor internal deformation. Thus, these rectangular network controlled the location of the following Mesozoic tectonic features: (a) Jurassic grabens (Espino, Uribante, and Machiques); (b) active mountains belts surrounding the Maracaibo block (Santander massif, Venezuelan Andes and the Perijá range); (c) major offsets of the Caribbean-related frontal deformation (Goajira); and the pull-apart basin development.
Origin of the Caribbean plate, the western and central Cordillera (Ecuador and Colombia)
The Caribbean plate formed from the end of the Late Jurassic to the Early Cenomanian, during the Late Cretaceous (Fig. 2.4). A widely accepted model is that the Caribbean plate moved into its inter-American position from the Pacific (Wilson, 1965; Malfait and Dinkleman, 1972; Pindell and Dewey, 1982; Pindell et al. 1988, and many others). This model forms the foundation for complex tectonic models for the development of the Venezuelan margin. In contrast, other authors (e.g. Ball et al. 1969, 1971; James, 1990; Meschede, 1998; Meschede and Frisch, 1998) proposed that the Caribbean plate most probably formed as part of the North American plate when that plate began separating from Gondwana during the Jurassic. The Caribbean plate subsequently interacted in convergent and strike-slip tectonics with the surrounding North and South American plates. Both models for the origin of the Caribbean plate deserve equal consideration.
The oceanic crust of the Caribbean plate is relatively thick (12 km) near the Beata Ridge, (Diebold, 1995). In contrast, it is only 3 km thick in the south-eastern part of the Venezuela basin. According to Donnelly (1973), Donnelly et al. (1990), and Driscoll and Diebold (1998), the Caribbean plate comprises a dominantly Late Cretaceous oceanic basalt province. It formed during spreading, as North America drifted away from South America, and during a widespread flood basalt event in the Early to middle Cretaceous (Fig. 2.4). A coeval, primitive to calc-alkaline island-arc formed in the periphery of the flood basalt province. More recently, mid-Tertiary to Holocene alkalic basaltic suites formed in Central America and in the Lesser Antilles in response to subduction of the Cocos and North American plates respectively. Because of the over-thickening of Late Cretaceous Caribbean oceanic crust, magnetic data in the Caribbean area do not show a well-defined ocean-floor signature. Ghosh et a1., (1984) reported NE trending magnetic anomalies of possible Jurassic age in the Venezuela basin.
To the west of the Romeral Fault System (RFS, Fig. 2.2), thick oceanic crust was formed during Aptian-Albian plume activity (Fig. 2.4C) of the Pacific super plume in the Pacific Ocean (Aspden et al. 1987 b, 1992 b; Larson, 1991). This oceanic crust was part of the paleo-Pacific plate or paleo-Caribbean plate (Mauffret and Leroy, 1997), which began to be accreted to the western margin of South America in Ecuador during the Campanian (Fig. 2.4 D; Aspden and McCourt, 2002; Kerr et al. 2002). During this epoch uplift of the Eastern and Western Cordillera in Colombia started (Fig. 2.4 E and F).
Figure 2.5 shows the position of crustal blocks in the northern Andean region, which Montes et al. (2005) used for their tectonic reconstruction (Fig. 2.6). These blocks are the allochthonous Bonaire and Falcón terrains (denoted by 1, Fig. 2.5), and eight individual crustal blocks, the Sierra Nevada de Santa Marta block, the Maracaibo block, the Northern Cordillera Central block, the Middle Magdalena Valley block, Southern Cordillera Central, Upper Magdalena Valley, Western Cordillera Oriental, and Eastern Cordillera Oriental (Fig. 2.5 and 2.6) are necessary for this reconstruction.
According to Montes et al. (2005), the collision of the Caribbean plate with the South America plate between the Late Cretaceous and the Late Paleocene caused northward translation and 20° of internal deformation (rotati on) of the rigid Cordillera Central Block (Fig. 2.6 A, B). At the same time oblique accretion of oceanic sequences west of the Northern Cordillera Central occurred, and dextral transpression of the Middle Magdalena Valley to the Eastern Cordillera Oriental started.
In Colombia, accretion moved gradually toward the north until the Middle Eocene, when the Provincia Litosferica Oceánica de la Cordillera Occidental (PLOCO) and the San Jacinto accretionary wedge were accreted (Duque, 1984, 1980; Nivia, 1996). Contemporaneously, clockwise rotation of the Sierra Nevada de Santa Marta and Maracaibo blocks started, sinistral slip on Santa Marta Bucaramanga fault was initiated, the Northern Cordillera Central fragmented block and 20° clockwise rotation of Midd le Magdalena Valley block occurred (Fig. 2.6C). Oblique collision and dextral displacement of the Caribbean plate toward the NE along the South American margin during the Paleogene resulted in surface uplift and a regional unconformity during gradual exhumation of Romeral Fault System (Fig. 2.2) rocks. In several sectors mylonite belts were developed, mainly in faults of the former subduction zone and the Early – Late Cretaceous volcanic arc.
During the Late Eocene the blocks of the Southern Cordillera Central and the Eastern Cordillera Oriental experienced 30° internal deform ation, and the Sierra Nevada de Santa Marta and Maracaibo blocks rotated clockwise by 20° (Fig. 2.6D). In addition, the Northern Cordillera Central and the Middle Magdalena Valley blocks were translated northward, and extrusion of the southern tip of the Falcon terrain started (Montes et al. 2005).
Fig. 2.6. Schematic sequential reconstruction of the movement of northern Andean blocks. (a) Predeformational state. (b) Northward translation of the rigid Cordillera Central block. (c) Fragmentation of the rigid Cordillera Central block as activity along the sinistral Santa Marta–Bucaramanga fault, and rotation of the Marac aibo block begin. (d) Initiation of significant dextral transpressional deformation in the Cordillera Oriental–Up per Magdalena block, further rotation of the Maracaibo block, causing the emplacement of the Villa del Cura rocks on the South American margin. (e) Rotation of the Maracaibo block is almost complete and extensional opening of the Falcon–Bonaire basin starts as the Caribbean deformation front continues to migrate east, result of a right-lateral, releasing bend (Muessig, 1984). (After Montes et al. 2005).
Maracaibo block rotation, reactivation of paleo-structures
The Maracaibo block rotated clockwise from the Middle Eocene to the Oligocene (Fig. 2.6), but it was not until the Miocene that rotation of the Maracaibo block triggered reactivation of faults associated with Jurassic graben structures, in what are today the Venezuelan Andes. This rotation is supported by the paleomagnetic studies of Hargraves and Shagam (1969) for the La Quinta Formation and for the Santa Marta-Bucaramanga and Maracaibo block, according to Bayona et al. (2008).
The Mesozoic graben structures (Fig. 2.3B) were inverted during the Neogene as a result of the collision of the Panamá arc with South Americain the west, and by continuous oblique collision of the Caribbean plate with South America in eastern Venezuela (Ostos et al. 2005). During the late Oligocene-early Miocene (Fig. 2.6 E, F), the northern part of the Falcón basin, the Maracaibo basin, and the Andean foreland basins were formed (Montes et al. 2005). With increasing rotation of the Maracaibo block during the Miocene, according to James (2000a), the interaction between this block and South America controlled transpression in the Venezuelan Andes. The Maracaibo block continued its rotation during the Pliocene (Montes et al. 2005), causing tranpressional movement on the Boconó fault and associated strike slip faults systems (see Fig. 2.2; Valera, Burbusay, Burro Negro, and Icotea fault systems), and along the NW and SE thrusts. Inside the Maracaibo block small fragments suffered rotation during Jurassic to Eocene times (Lugo and Mann, 1995). Probably these rotations could be extrapolated to other small tectonic blocks across the Venezuelan Andes, which may complicate the exhumations patterns across the Venezuelan Andes.
Table of contents :
I. GENERAL INTRODUCTION
1.1 Overview of chapters and results
1.2 Publications and abstracts from this dissertation
I. INTRODUCTION GÉNÉRALE
1.1 Résumé des chapitres et principaux résultats
1.1 Distribución de los capítulos y resultados
II. PLATE TECTONIC EVOLUTION OF THE NORTHERN ANDEAN REGION
2.1 Tectonic assembly of the Northern Andes
2.2 Pangaea break-up and origin of the Caribbean plate
2.2.1 Continental rifting and formation of structural discontinuities
2.2.2 Origin of the Caribbean plate, the western and central Cordillera (Ecuador and Colombia)
2.2.3 Maracaibo block rotation, reactivation of paleo-structures
2.3 Current tectonic models for the origin of the Venezuelan Andes
2.3.1 Symmetric models
2.3.2 Asymmetric models
.3.3 Present-day tectonic
III. APATITE FISSION-TRACK THERMOCHRONOLOGY
3.1 Fission-track formation theory
3.2 Fission-track age equation
3.2.1 External detector method and z calibration method or Z factor
3.3 Conventional statistics for in-situ FT thermochronology
3.3.1 Estimators for R
220.127.116.11 Isochron fitting
18.104.22.168. Mean ratio
22.214.171.124 Pooled Mean
126.96.36.199. Central Age
3.3.2 Standard error on fission-track age
3.3.3 Galbraith test
3.4 Fission-track annealing and modeling
3.5 Length measurements
3.6 Analytical procedure
3.7 Detrital apatite fission-track thermochronology
3.8 Conventional statistics for detrital FT thermochronology
3.9 Data interpretation and quantitative thermochronology methods
3.9.1 Quantitative thermochronology
IV. SPATIAL AND TEMPORAL PATTERNS OF EXHUMATION ACROSS THE VENEZUELAN ANDES: IMPLICATIONS FOR CENOZOIC CARIBBEAN GEODYNAMICS
4.2 Geodynamic setting and structure of the Venezuelan Andes
4.3 Methods and analytical procedures
4.3.1 Topographic characteristics
4.3.2 Apatite fission-track thermochronology
4.3.3. Thermal history modeling
4.4.1 Topographic characteristics
4.4.2. Apatite fission-track data
V. THERMOCHRONOLOGIC EVIDENCE FOR KM-SCALE VERTICAL OFFSET ACROSS THE BOCONÓ STRIKE-SLIP FAULT, CENTRAL VENEZUELAN ANDES
5.2 Tectonic setting
5.3 Thermochronology data
5.4 Numerical modeling
5.5 Discussion and conclusions
VI. TECTONIC VERSUS CLIMATIC CONTROLS ON EXHUMATION IN THE VENEZUELAN ANDES
6.2 Tectonic, geomorphic and climatic setting
6.2.1 Geological evolution and exhumation history
6.2.4 Precipitation pattern
6.3 Detrital apatite fission-track thermochronology
6.3.1. Data collection and discrimination of age components
6.3.2 Comparison of detrital and bedrock apatite FT ages
6.3.3 Implications for sediment provenance
6.4 Predicted exhumation and erosion patterns
6.4.1 Long-term exhumation rates
6.4.2 Short-term erosion patterns
6.5.1 Relations between tectonics, climate and erosion in the VA
VII. STRATIGRAPHY AND PROVENANCE OF THE MÉRIDA ANDES PRO- AND RETRO-SIDE FORELAND BASIN DEPOSITS: INSIGHTS FROM DETRITAL APATITE FISSION-TRACK THERMOCHRONOLOGY, PALYNOLOGY, AND SEDIMENT PETROLOGY.
7.2 Geologic setting
7.3 Stratigraphic analysis and sample collection
7.4.1 Detrital apatite fission-track thermochronology
7.4.2 Palynological analysis
7.4.3 Characterization of organic matter
188.8.131.52 Characterization by Rock-Eval pyrolysis
184.108.40.206 Determination of total carbon content by elementary analysis (LECO)
220.127.116.11 Calcimetric measure
7.4.4 Sediment petrology
7.5.1 Detrital apatite fission-track thermochronology
7.5.2 Palynological analysis
7.5.4 Sediment petrology
7.5.5 The Río Hoyos-Río Vichú stratigraphic section
7.5.6 Parángula River stratigraphic section
7.6.1 Stratigraphy and provenance of the two foreland basins
18.104.22.168 Depositional environments
22.214.171.124 Depositional ages
126.96.36.199 Sediment provenance
7.6.2 Neogene exhumation history of the Mérida Andes
VIII. GENERAL CONCLUSIONS
8.3 Future perspectives
VIII. CONCLUSIONS GÉNÉRALES
8.3 Perspectives et avenir
VIII. CONCLUSIONES GENERALES
8.3 Perspectivas a futuro