The lithosphere of the Central Andes is investigated by an interdisciplinary research project? "Deformation processes in the Andes" carried out as a joint program by geoscientists of the Freie Universitaet Berlin, Technische Universitaet Berlin, the GeoForschungsZentrum Potsdam and the Universitaet Potsdam in close cooperation with geoscientific institutions in Argentina, Bolivia and Chile and other countries.
Along the Andean chain the Central Andes exhibit extreme dimensions as large width (700 km), extreme high elevations and very thick crust (70 km), high seismicity, widespread young volcanism and big ore deposits. Which are the reasons for the anomalies? Although the Andean convergence between the oceanic and continental plates lasts 200 Ma, the formation of the young Central Andes with their extreme dimensions started at about 25-30 Ma. Why did it occur so late?
From the pattern of the magnetic anomalies in the SE-Pacific the spreading and convergence rates for the last 120 Ma can be reconstructed. The eastern flank of the SE-Pacific Rise presents a maximum age of about 45 Ma, whereas the western side shows an oceanic crust of a maximum age of about 120 Ma. This allows a reconstruction of the oceanic crust already subducted beneath the Central Andes. At about 30 Ma ago an oceanic lithosphere was subducted which was formed during the event of the Cretaceous superplume produced at 120 Ma. This event may have caused the interruption of the volcanic activity at about 30 Ma and the begin of a new magmatic cycle. From 25 Ma on an oceanic crust was subducted with high convergence rates but which was produced at slow spreading rates. As a consequence a deeply hydrated lithosphere in combination with a high convergence rates generates a high fluid flux from the lower into the upper plate. This anomalous high fluid input causes a general weakening of the upper plate and consequently its deformation. The climatic environment is another very important parameter controlling the tectonic evolution of the Andes especially the Central Andes. Thus it can be concluded, that the formation of the Central Andes was strongly controlled by the high fluid flux from the lower into the upper plate and the varying climatic conditions along their surface.
The integrated geophysical experiment ANCORP'96 imaged the subduction of the Nazca plate beneath the central Andes and the Andean lithosphere at 21°S (ANCORP research group 1998). Two strong reflective zones occur at depths of 20-30 km and 40-80 km under the Central Andean forearc which shows the deepest so far acquired reflections from a subduction zone. The deep reflective zone (Nazca reflector) is offset by 20 km above the Wadati-Benioff zone and shows increasing downdip intensity before abrupt breakdown below 80 km. Both, the brightest part of the Nazca reflector and the shallow reflection zone (QB bright spot) are exceptionally bright reflectors with estimated apparent reflectivity coefficients higher than 0.2.
The preferred model explains both, the Nazca reflector and the QB bright spot as fluid traps located at fronts of recent hydration of the mantle (Nazca reflector) and crust (QB bright spot), the fluids being supplied by dehydration of the upper layer of the oceanic crust of the subducting Nazca plate. This dehydration is nearly a-seismic. Intermediate-depth earthquakes are associated with dehydration of the lower oceanic crust and the uppermost mantle which contains much less water than the upper crust. This zone is not associated with reflections.
The continental MOHO beneath the Andes does not show up in reflection data but is well imaged by converted phases suggesting a broad transitional character of the crust-mantle boundary in an active arc setting. While the arc itself is more transparent in terms of reflectivity, the backarc area shows some diffuse reflectivity down to 20 s TWT which is identical to the depth to MOHO from wide angle data. The complex structure of both units, but mainly the presence of partial melts beneath the arc and Altiplano plateau, as suggested by magnetotelluric soundings and tomographic data, control the difference in reflectivity structure from that of the forearc.
We suggest that reflectivity in active margin settings indicates active fluid-associated processes rather than structures and that reflection seismology may precisely localize the site of these processes. Interpretation is strongly dependant on associated experiments which have a potential to image fluid-driven changes in petrophysical properties (MT, tomography, receiver functions).
ANCORP research group, Nature, in press, (1998).
The continental crust in the Central Andes has been shortened and thickened in the last 60 Ma. Tectonism and uplift is documented in phases of deformation and the sedimentary record which we differentiated into time and space from the forarc to the Subandean Ranges. While tectonism mainly transports material within the crust, magmatism contributes directly to crustal growth. The volumes of displaced crust involved in Central Andean orogeny are estimated from tectonic evidence for the amount of shortening, and uplift histories through time. The amount of magma added through time is estimated from exposed and reconstructed volumes of igneous rocks, including volcaniclastic sediments, and their geochemical and isotopic composition used to estimate differentiation and assimilation.Juvenile magmatic input ranges between >30 km3/per km arc and Ma in the Jurassic to average values of 20 km3/per km arc and Ma to periods of magmatic quiescence. Intra-crustal recycling by assimilation of mantle magmas (andesites) and crustal melting (ignimbrites) increases with crustal thickness and time up to 8 km3/ per km arc and Ma in the Upper Miocene to Recent. Shortening and uplift at 30 to 25 Ma correlates with a magmatic lull. From 25 Ma on, mafic magmatism occurs on the Altiplano while extensive uplift and erosion is indicated at the Western Altiplano Escarpment by up to 800 m of coarse-grained gravels. Up to 900 m of ignimbrites at 19 Ma indicate crustal melting in response to thickening following thermal relaxation of about 8 to 10 Ma. Tectonism, uplift and absence of magmatism at 25 to 30 Ma is interpreted to result from flat-slab subduction and enhanced mechanical coupling between plates. Steepening of the slab at 25 Ma caused hot mantle to enter the wedge and the formation of basalts. Both processes contributed to thickening and uplift, respectively. Erosion and crustal melting follow at 25 to 19 Ma. Clearly, the volume of juvenile magmatic rocks added to the crust cannot account for all the observed crustal thickening. Local crustal shortening is also insufficient to explain the observed crustal thickening (ca. 30 km) beneath the Altiplano. This discrepancy could be explained by some combination of (1) underplating of material, derived by tectonic erosion from the forearc region, further west, and (2) lower crustal flow. Since the Miocene, subduction of trench sediments certainly cannot account for the missing crustal volumes because trench sedimentation decreased significantly as the western margin of the Andes became progressively more arid. Large scale lower crustal ductile flow, both along and across the width of the mountain belt, may have been the most important mechanism of crustal thickening beneath the Altiplano, resulting in the transport of excess material from further east.
The La Pacana caldera (60x35 km) is located in the Central Volcanic Zone of the Andes in a large ignimbrite province known as the Altiplano-Puna Volcanic Complex (APVC). The ignimbrites of the APVC can be broadly characterised as one of two kinds. The dominant type is homogeneous, crystal-rich and dacitic, similar to the 'monotonous intermediates' typical of regions of large-scale crustal melting. Also present are smaller volume, relatively homogeneous, crystal-poor, rhyolitic ignimbrites. The La Pacana caldera is the only known caldera in the APVC to have sourced both of these ignimbrite types, the Atana and Toconao ignimbrites respectively. The caldera-forming dacitic Atana ignimbrite occurs as a homogenous single flow unit, and lacks a basal plinian. It is crystal-rich (50% crystals) and contains variable amounts of crystal-rich pumices. K-Ar ages of biotites from pumices show that this ignimbrite erupted 4.1 Ma ago. The Atana immediately overlies the crystal poor (<10% crystals), rhyolitic Toconao ignimbrite, which contains abundant glassy, crystal-poor, tubular pumice, and which has a thin plinian deposit at its base. K-Ar ages for the Toconao ignimbrite are inconclusive, ranging from 4-5 Ma.
These two ignimbrites both represent very different but internally relatively homogenous units, yet have thickness variations and lateral distributions that point to the same source caldera. Age, stratigraphy and geochemistry are consistent with these units representing two homogeneous zones in a vertically-stratified, large-volume, silicic magma chamber. Calculations of temperatures and water contents reveal that the Toconao melt was cooler (735°C) and more volatile rich (6-7 wt.% H2O) than the Atana (790°C; 3-4 wt.% H2O), consistent with it representing the evolved cap of the Atana magma chamber. Volume estimates based on caldera geometry reveal that the Toconao and Atana ignimbrites together comprise some 4,000 km3 of erupted material. Other studies indicate that large-volume, crystal-rich, dacite ignimbrites such as the Atana erupt from homogeneous magma chambers. The La Pacana ignimbrites show that these can also fractionate and develop a zoned magma chamber.
The sharp contrast in physical characteristics between these two ignimbrites and lack of intermediate components indicate that the two magmas did not mix during eruption. This separation may be envisioned as an 'eruptability' barrier, above which a volatile-saturation driven eruption is possible. In this scenario, the eruption that emptied the Toconao magma chamber did not tap the underlying Atana magma, as the crystal-rich and volatile-poor nature of the Atana places it below this eruptablitiy barrier. Implicit in this reasoning is that the Atana magma necessitates an external trigger in order to erupt. Caldera geometry and the distribution of regional faults are consistent with a tectonic trigger to instigate eruption.
Late Cretaceous to late Eocene magmatic arc rocks from the Central Andes (north Chilean Precordillera between 21.5 and 26°S) reveal several temporally evolving suites showing increasing Sr/Y, La/Yb, Hf/Lu ratios and Al2O3 contents with increasing arc maturity. Granodiorites (Sr/Y 54-69, La/Yb 25-27, Al2O3 16.9-17.1 wt%) and granites similar to that of trondjiemites-tonalites-granodiorites (TTG) mark the end of arc magmatism in the Chilean Precordillera. The timing of pluton emplacement overlaps the Incaic transpressional deformation (38.5 Ma) and is associated with initial crustal thickening in the Andes.This tectonomagmatic setting suggests that the late Eocene TTG-like melts are derived from partial melting of a garnet-bearing amphibolite (<10 vol% gt) and/or extensive interaction of mantle-derived magmas with such crustal melts in the keel of the thickened continental crust. Syn/posttectonic late Eocene granodiorites in the north Chilean Precordillera therefore reflect the onset of garnet stability in residuum mineralogies of Andean magmatic arc systems at crustal thicknesses exceeding 40-45 km after Incaic deformation. Prior to tectonic crustal thickening, between late Cretaceous and late Eocene, the increasing Sr/Y, La/Yb, Hf/Lu ratios and Al2O3 contents demand a different explanation for the older suite. The enriched geochemical signature of the older rocks evolved in a largely extensional tectonic setting. Extension oblique to the arc axis caused mantle upwelling and crustal growth by magmatic underplating. Increased La/Yb with time then reflect mantle dynamics rather than crustal thickening. Sr, Nd and Pb initial isotopes of Chilean Precordillera magmatic rocks (87Sr86Sr 0.7038 to 0.7064, 143Nd144Nd 0.51261 to 0.51279, 206Pb204Pb 18.19-18.77) lack correlations with SiO2 and simple crustal contamination by AFC with old crustal material can be ruled out. Most 87Sr86Sr ratios, however, range between 0.7038 and 0.7048 and are increasing slightly towards higher 87Sr86Sr ratios in younger rocks. Enrichment through time possibly took place at lower crustal levels where the residual mineralogy changed through time towards a higher pressure garnet-bearing mineral assemblage. Comparison with paleo-Arc settings in the Andean system (Coastal Cordillera, Longitudinal Valley) and late Cenozoic rocks in the Central Andes shows that these mechanisms operate in all Andean magmatic arc systems and control the trace element signature through time.
The Puna plateau, located in northwestern Argentina, is the southernmost segment of the bolivian Altiplano. It results from the oblique convergence between the oceanic Nazca plate and continental South America along an ENE-WSW direction which remained stable since about 49 Ma. The plateau is characterized by an average elevation ~4400 m and by a crust which has been thickened up to 70 km. According to digital topographic mapping, satellite imagery, surface mapping, and seismic reflexion data, the Puna is a composite feature made of Precambrian to Paleozoic basement ranges bounded by high angle reverse faults, which alternate with continental cenozoic intermontane basins. Major thrusts trend NNE-SSW. They are distributed all over the plateau and verge alternatively Westward or Eastward. According to field mapping and fault-slip data, these thrusts are mainly characterized by dip-slip components and do not show significant right-lateral strike-slip components as classically expected for this part of the orogen. The onset of shortening and thickening of the Puna has long been considered to be Late Miocene in age. However, stratigraphic data and geometries of basin filling indicate that internal deformation of this back-arc region started during the Late Paleogene and probably during the Incaic phase of the andean orogeny, which is Late Eocene in age. Kinematic analysis of fault slip data indicate that the orientation of principal strains generated by cenozoic deformation is rather homogeneous over the area. Main extension axes are subvertical, and main shortening axes are subhorizontal and trend WNW-ESE to NW-SE. This strike is strongly oblique with respect to the vector of relative motion between the two plates. Strain directions must have been substantially rotated clockwise as a result of cenozoic bending of the bolivian orocline. Anisotropy of Magnetic Susceptibility measured within cenozoic sediments of the Puna basins and compiled throughout the Southern border of the plateau and the northern bolivian Altiplano, show well-developped magnetic lineations which have been partly attributed to the regional horizontal shortening. These data, restored from local block rotations inferred from paleomagnetic analysis, provide initial shortening directions recorded by the sediments during their magnetization. These shortening directions define a radial pattern around the bolivian orocline. This we attribute to the eastward convex geometry of the deformation front flanking the eastern side of Central Andes.
The Sierras Pampeanas in the northwest Argentine Andean foreland between 26 and 33°S are tectonically active reverse-fault bounded crystalline basement blocks which partly preserve a basement peneplain at elevations in excess of 5000 m. Remnants of this surface are observed on portions of the Aconquija and Cumbres Calchaquies ranges. Based on radiometrically dated intramontane sediments, uplift of these ranges is inferred to have started in the Pliocene. The adjacent Santa Maria basin contains Miocene sediments apparently deposited on the same erosion surface. In order to constrain the uplift and the long-term development of this erosion surface, samples were collected for apatite fission track thermochronology (AFTT) from Paleozoic granite just above the basal thrust at Sierra Aconquija (27°S) and from a vertical transect through granite and Paleozoic metasediments at the Cumbres Calchaquies range (26°S). The relief of the northern and southern localities are 1300 m and 2500 m, respectively.
AFTT analysis from the base of Sierra Aconquija yields long confined track lengths indicative of rapid cooling through the partial annealing zone. Although single-grain ages are all young, with a central age of 6.3±0.8 Ma, the sample fails the chi squared test. The sample likely resided at low temperatures prior to a maximum Neogene burial temperature of 115±15°C followed by exhumation beginning around 6 Ma. The sample's maximum Neogene burial was likely about 4 km (2.5 km modern relief plus approximately 1.5 km of presently-eroded bedrock and Neogene sediment deposited above the erosion surface), assuming a paleogeothermal gradient of 25°C/km.
All analyzed samples from the Cumbres Calchaquies transect have mean track lengths between 13.1 and 12.3 µm. Single grain ages range from Paleocene to late Cretaceous; all samples pass the chi square test. Corrections for track length reductions yield exclusively late Cretaceous ages. As the highest elevation sample was collected at the erosion surface, the surface was likely cut no earlier than during this cooling event, which may be linked to the tectono-magmatic evolution of the adjacent Cretaceous Salta Rift. Track-length distributions require late-stage cooling from low temperatures. As 11 Ma sediments overlie the erosion surface in the neighboring intramontane basin, the simplest interpretation for this data set is that late Cretaceous exhumation created the erosion surface. Around 11 Ma, a basin extending over the Santa Maria valley and the adjacent ranges began to subside, reaching a maximum temperature of perhaps 80°C at the position of the fission track samples, followed by thrust-driven exhumation of the range beginning around 6 Ma.
The late Cenozoic Argentine Sierras Pampeanas are a prime example of the close relation between tectonics, the creation of important orographic contrasts, and climatic evolution. At the northeastern margin of this Andean foreland province, the 4000 to 5500 m high Cumbres Calchaquies and Sierra Aconquija (26 to 27°S) act as an efficient barrier to easterly moisture-bearing winds and sharply delimit a subtropically humid environment in the eastern lowland from a semidesert in the western intermontane Santa Maria Valley. The Santa Maria basin comprises a marine/terrigenous clastic section spanning the last 11 Ma. Sedimentation patterns, paleosols, and structural data show that the final and most rapid uplift of the Calchaquies and Aconquija ranges began in late Pliocene time after 3.4 Ma. Since then, thick pedogenic carbonate accumulation on gravelly pediment surfaces attests to arid climate conditions in the western intermontane basins. However, our observations show that initial uplift and formation of an orographic barrier began in late Miocene time.
After 10 Ma, fluvial deposition superseded marine conditions related to the Parana transgression and resulted in a sedimentary section, locally over 4000 m thick. Paleosols in these sediments help to assess the evolution of the rain shadow and temperature/moisture regimes in the course of Andean foreland deformation. In the mid-Miocene Chiquimil formation (about 9 to 8 Ma), paleosols developed under slow sedimentation rates and are well horizoned. Abundant gypsum and halite precipitation, preservation of organic matter, and orange to gray-green mottles are diagnostic of a fluctuating groundwater table under a hydrologic regime of strongly seasonal water supply. In contrast, soil development in the late Miocene/ early Pliocene Andalhuala formation (7.1 to 3.4 Ma) was hampered by high clastic sediment input with upward-coarsening grain size, probably due to increased relief contrasts in the easterly source area. Paleosols in these sectors lack indications of waterlogging. Instead, calcic rhizoliths and pedogenic carbonate nodules are indicative of a fundamentally different climatic regime and suggest increased aridization. The most likely inference is the presence of a moderately elevated and actively rising eastern range at least by about 7 Ma.
This interpretation is further supported by apatite fission-track thermochronology results obtained on samples from the crystalline basement blocks, thrust over the Neogene basin fill along high-angle reverse faults. Central ages and track-length distributions document thrust-driven exhumation beginning at about 6 Ma. The results provide new insight into the timing of eastward propagation of thick-skinned Andean foreland deformation. The pronounced climatic contrast between the humid eastern Argentine lowland and the arid Andean interior, although a more recent phenomenon in its present drastic expression, was already manifest after the mid-Miocene uplift of the Puna plateau to the west.
Until the end of the last century several vein-hosted selenide mineralizations in the area of the Sierra de Cacho (Sierra de Umango, Western Sierras Pampeanas, Prov. La Rioja, central W-Argentina) have been mined mainly for copper. Selenides, predominantly umangite (Cu3Se2) and tiemannite (HgSe), occur within NE-SW oriented carbonate veins, which are generally related to large brittle faults. Within the area of investigation brittle deformation shows a three step evolution: D1: The first event indicates subhorizontal NW-SE compression, which is responsible for the uplift of Sierras Pampeanas basement block. D2: This event was followed by a period of subhorizontal NE-SW compression coincident with the orientation of mineralized veins. D3: The last event displays subhorizontal E-W compression and is in accordance to the orientation of the recent stress field within the Central Andes. Structural data from Sierra de Cacho Area and comparison with age estimates of stress field conversion at other areas of the Central Andes indicate a very young, probably Late Miocene to Pliocene age for the mineralization.
Located within the western Peruvian cordillera (16°13'S, 71°51'W), the Nevado Sabancaya volcano presents a persistant explosive volcanic activity since the eruption of May 28, 1990, which ended a dormant stage of about 200 years. The behaviour was increasingly explosive during the two first years, and was rather constant until late 1994, with hydromagmatic and moderate-magnitude vulcanian activity (Thouret et al., 1994). Since 1995, the frequency of explosions decreased gradually so that the time intervals between two events increased from 30 minutes to several hours. Since late 1997, the activity has been mainly phreatic.
The erupted juvenile products consist of highly porphyritic lavas of andesitic to dacitic compositions. They are charaterized by high alkali-contents, and plot in the field of the high-K calc-alkaline series. The silica-content evolution is not linear through time: the 1990 lavas are the most evolved (61 to 64 wt% SiO2), then the composition drop down to 60-61.5 wt% SiO2, and from 1994, it remains rather constant (61.5-62.5 wt% SiO2). The juvenile faciès emitted during the 1992 to 1995 eruptions contain rare nearly aphyric enclaves. These small enclaves exhibit a sharp boundary and some radial fractures. They are plagioclase- and amphibole-rich with a microdoleritic texture and needle-shaped crystals. They are characterized by lower silica-content (about 57 wt% SiO2).
Plagioclase phenocrysts from the porphyritic lavas exhibit complex chemical zoning : a large core of andesine composition (An30-40), a glass inclusion-rich zone of labrador composition (An50-60) and a tiny andesine rim. Plagioclases from the microdoleritic enclaves also have a labrador composition (An45-60); but, they. are mostly unzoned. In the lavas, zoning of the plagioclase phenocrysts can be explained by some inputs of more mafic magmas in the reservoir. Some of these mafic magmas may crystallize in the plumbing system. Microdoleritic enclaves represent fragments of these hypovolcanic rocks carried up by magmas erupted between 1992-1995. Thus these enclaves do not give any evidence of a new magma injection that may have triggered off the actual eruption. The overall series displays major and trace element trends and REE patterns, which seem compatible with an evolution through fractional crystallization. Nevertheless ligth rare earth enrichments may suggest participation of other processes, such as partial melting of hydrated oceanic crust leading to the inputs of an adakitic-type source component (Monzier et al., 1997), or most probably assimilation of various amounts of crustal materials from the thick andean continental lithosphere. Preliminary Sr-isotope ratios on the 1990-1998 lavas are changeable and may indicate various amounts of crustal assimilation.
Thouret J-C, Guillande R, Huaman D, Gourgaud A, Salas G, & Chorowicz J, Bull. Soc. Geol. France, 165, 49-63, (1994).
Monzier M, Robin C, Hall ML, Cotten J, Mothes P, Eissen JP, Samaniego P, C. R. Acad. Sci. Paris, 324, 545-552, (1997).
Volcanic rocks from Patagonia, 41°S, along an E-W profil (300 km), consist of trachybasalts and tephrites/basanites of Pliocene to Quaternary age. Older tholeiitic basalts and basaltic andesites of Oligocene to Miocene age are outcropping in the western part of the profile (Comallo and Jacobacci). Trachybasalts and tephrites/basanites carry mantle and deep crustal derived xenoliths, whereas the older lavas (W) are free of xenoliths. Deep crustal xenoliths (granulites and pyroxene-amphibolites) are more common to the east (Queupuniyeu) and, arguably, indicate different genetic conditions.
Primitive mantle normalized incompatible elements of 200 samples show four different major trends from W to E: 1) Lavas from W exhibit significant negative Nb anomalies and ratios of K/Nb and Ba/La that vary between 417 and 730 and 17-23 respectively. 2) Oldest lavas, E of Comallo, which contain no xenoliths, inherit the least contents of incompatible elements strong Nb anomalies and high Ba/La ratios of 24-53. 3) Lavas from the central part are enriched in incompatible elements and show patterns with steep gradients. No distinct anomalies are observed. Ba/La ratios are uniform between 15 and 20.4) Lavas from E (Queupuniyeu) do not show Nb-anomalies. Three trends are identified: A) Lavas similar to the central part but with pronounced negative K-anomalies inherit low Ba/La ratios of 11-14. B) Lavas with lower abundances of incompatible elements relative to A and minor K-anomalies show Ba/La ratios of 19-24. C) Lavas with low abundances of incompatible elements and strong K-anomalies have Ba/La ratios between 13-18.
The results show that the lavas, which erupted in the W part of the profile, in a back-arc situation, have been modified by subduction related processes. Their low contents of incompatible elements indicate high degrees of partial melting. Central part lavas have OIB-like characteristics and high abundances of incompatible elements, suggesting low degree of partial melting of asthenosphere. The three trends observed in E part lavas indicate variable degrees of partial melting. In addition, the negative K-anomaly suggests that their magma source was different of that from the central part. The patterns of mantle normalized incompatible elements are consistent with (Stern et al., 1990), who investigated a larger area, however with a limited number of samples. This implies that the subcontinental mantle beneath Patagonia has experienced similar magmatic processes on a large scale.
Stern CR, Frey FA, Futa K, Zartman RE, Peng Z & Kyser TK, Contrib. Mineral. Petrol., 104, 294-308, (1990).
The Altiplano-Puna Plateau of the Central Andes is one of the largest plateaus on earth. This orogenic feature is part of an active continental margin where terrane accretion is lacking and tectonic erosion removes material prior and even during plateau uplift.
There is a vivid discussion about uplift and thickening mechanisms of the N-Chilean Forearc region. Studying the kinematic evolution of the western border of the Altiplano plateau from the onset of uplift in the Miocene until the present may give clues to the mechanisms and also to the extent to which the possible mechanisms are responsible for uplift and thickening. Our methods comprise the analysis of tectonic features and cross cutting relationships in the field and from satellite and aerial photographs as well as the timing of deformation events by dating ignimbrite flows in the area. Furthermore morphological properties of streams draining the western flank of the Altiplano are examined to record the recent uplift of the area. Computer supported tectonic modelling is used to quantify and visualize the processes.
The first results of our study indicate that processes leading to uplift and thickening are not continuous. Strain in the upper plate is not partitioned into arc normal and arc parallel components to the same proportions throughout the Neogene and Quaternary. Although velocity and direction of plate convergence have not changed since 26 Ma we observe two different deformation regimes in the Precordillera of N-Chile. Arc normal compression lead to minor tectonic shortening in Miocene to Pliocene time. Arc parallel dextral shear in a transtensional regime lead to subsidence of extensive pull-apart basins on the western border of the Altiplano during the Quaternary. Discontinuity of uplift is indicated by cycles of syn-uplift sediments (Miocene-Pleistocene) with intercalation of ignimbrites. These clastic sediments and ignimbrites form repeated sequences deposited on the western flank of the Altiplano, where the topographic gradient at present is as high as 6°. Each sediment/ignimbrite couple is correlated with a cycle of ceasing uplift rate. Also there is a clear correlation of ignimbrite eruption with the arc normal compressive deformation event. Since the switch to the transtensional deformation event no eruptive magmatism is observed in the area.
Since the plate kinematic parameters are stable since 26 Ma we have to infer that additional mechanisms control the complex evolution of the N-Chilean Forearc and western Altiplano. Its evolution seems to be strongly dependent on the degree of plate coupling and thermal weakening by pulses of active arc magmatism.
Horizontal crustal shortening, tectonic underplating, addition of voluminous magmas and climatic relationships are currently considered as prime mechanisms controlling the formation of the central Andean plateau. A field-based study of first-order structures in the area of the upper Calchaquies River was complemented by an analysis of air photos and Landsat images to assess the tectonic significance in the formation of the Puna Plateau. In this area, Precambrian to Cambrian low-grade metasedimentary rocks form the basement to Cretaceous/Paleocene rift sediments and Miocene continental red beds. Buckle folding of the rift sediments around dominantly north-south trending, orogen-parallel axes involved ignimbrite layers, most likely 12 to 8 Ma in age, and accounts for subhorizontal shortening during Andean (post mid Miocene) deformation. The unconformity at the base of overturned rift sediments is unstrained and indicates that Precambrian basement rocks were folded along with their cover rocks. Amplitudes of first-order folds are in the order of 0.5 to 1.5 km. This provides strong evidence for thick-skinned tectonism and consequently substantial crustal thickening during an early episode of Andean deformation. By contrast, the contact between basement rocks and Miocene red beds is characterized by high-angle reverse faults which formed at a later stage of Andean deformation. Displacement magnitudes on the fault surfaces as indicated by drape folds in the red beds are generally less than 50 m and suggest that reverse faulting was of minor importance to Andean crustal thickening. Structures indicating horizontal extension are seen sporadically but appear to be kinematically linked to the Olacapato - El Toro strike-slip fault, a NW-SE striking Neogene master dislocation in the central Andes. The observations confirm that the eastern margin of the Puna Plateau is still undergoing horizontal compression. However, major crustal thickening was achieved by thick-skinned tectonism during an early stage of Andean deformation.
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications