Overview of the tectonic evolutionThe time-space diagram of figure 1 summarizes the main tectonic events associated with the evolution of the Andes. Volcanism appears to start back in the Triassic while back-arc basin development did not start until the late Jurassic. Deformation and closure of the back-arc basin began in late Jurassic and extended from both extremes toward the center. In consequence, youngest deformation is observed in the central Chile-Argentina segment, becoming older toward the northern and southern ends of the Andes. One exception is the F-segment, which appears to have no deformation at all.
The Altiplano-Puna plateau
General overviewThe region considered in this work spans the western margin of the South American Plate between 10º and 35ºS. The most outstanding feature of this region is the broadening of the main bulk of the Cordillera, embedding the Altiplano and Puna plateaus which have average elevations above 3 km. The Altiplano-Puna plateau is the highest plateau associated with abundant magmatism and is second only to Tibet in height and extent. The high elevations, as represented by the yellow shades in figure 2, is approximately 1800km long and 350-400 km wide.
|Figure 2 - General view of the Central Andes. The names outline the most important features of the region. The thin grey contour lines represent the isodepth curves of the Moho every 25km. Profiles A-H are shown in figure 3.|
|In a general sense, the main features are the afore mentioned Altiplano
and Puna plateau, which are bordered on the west by the present day active
volcanic arc (western Cordillera) and in the east the paleo-arc (Eastern Cordillera),
the Santa Barbara and Sierras Pampeanas system, and the Subandean thin-skinned
thrust belts. Each system describes different aspects of the ongoing (and
past) processes and is characterized by geological and structural differences
that will be detailed shortly.
The slab geometry outlined by the gray thin isodepth curves in the figure, have important variations along the convergence zone. Most important are the flat slab regions north and south of 10º and 25ºS, respectively, where the subducting plate dips with an average angle of only 15º instead of the normal 30º observed below northern Chile and Bolivia. The absence of post-Pliocene volcanism correlates very well with the flat-slab subduction parts. Further, both regions coincide with the onset of the Nazca ridge (in the north) and Juan Fernandez ridge (in the south), although the incidence of these features is, so far, not well understood.
Around 17-18ºS the Andes have a concave bending (seen from the ocean towards the continent) where the angle between the subducting Nazca plate and the Southamerican plate changes from an almost perpendicular convergence in the south to a rather oblique collision in the north.
Given this broad description of the main features, we will summarize the current theories and models inferred from geological, geophysical and geochemical studies to explain the tectonic evolution and present day situation of this orogen. Since the Altiplano-Puna plateau is the most outstanding feature of the Central Andean system, this summary will focus on this topic. However, associated features like the ones mentioned above are intrinsically related and will be considered in the analysis.
Topography and morphological featuresFigure 3 shows the topographic profiles A-H defined in figure 1, making it possible to observe significant differences along the convergence zone. In general terms, the average width of the Andes is maximal in the central part and decreases toward the northern and southern ends of the studied region. At the same time, the average elevation shows a similar trend being maximal below the Altiplano-Puna plateau. Figure 4 shows a related plot, where the topography along the strike of the convergence is shown together with the mantle thicknesses and the slab depth. In the northern segment, which corresponds to the Altiplano-Puna plateau, the average elevation ranges between 3.5 and 4 km, and descends abruptly around 1 km southwards of 29ºS. An interesting aspect of the Andean topography is that, despite the differences in geology and climatic environments, the elevation shows a certain similarity across an axis striking NE/SW-wards at the bend (~18ºS). The pole of this plane coincides roughly with the Nazca-South American plate pole of rotation for the period between 36 and 20 Ma, suggesting that continental scale deformation accommodates essentially the required space given by tectonic driving forces.
On the other hand, important differences can be observed comparing the eastern and western tips of the Andean massif, where climatic caused erosion is the dominant factor producing different topographic roughness.
|Figure 3 - Profiles A-H are defined in figure 2.||Figure 4 - Plot showin the topography and depth to Moho over the NS-profile defined in the lower fiugre. The vertical (E-W lines) define the highly seismically attenuated region.|
|Crustal thickness, rheology and isostacy
Early studies suggested that crustal thicknesses could reach up to 70 km beneath the Western Cordillera (James 1971). More recent active and passive earthquake studies have confirmed these results producing contour maps of the crustal depth as in figure 5 (Ocola & Meyer 1972, Wigger 1994 and Beck et al 1996). In general terms, seismic studies conclude that the average crustal P-wave velocities and Poisson’s ratio are rather low (6.0 km/s and 0.25, respectively), suggesting a felsic composition of the crust. These results are consistent with the seismically attenuated zone outlined in figure 4.
|Figure 5 - General view of the Central Andes. The names outline the most important features of the region. The thin grey contour lines represent the isodepth curves of the Moho every 25km. Profiles A-H are shown in figure 3.|
|Modeling of the gravitational data done by (Fukao et all 1989) in
the upper margin beneath southern Peru and in by (Götze et al 1994)
below northern Argentina and Chile, suggest slightly lower crustal thicknesses.
This apparent contradiction can be explained by the possibility that lower
crustal material has the signature of crustal seismic velocities but mantle
Isostatic anomalies show that northwards of 24ºS, the eastern flank is flexurally supported, although along strike variations can be observed (Lyon-Caen et al 1985 and Watts et al 1995). Further south, the eastern flank is not flexurally but locally compensated, imposing important implications for the crustal shortening processes of that region.
Lithospheric thicknessDetermining seismic attenuation beneath the plateau, the lithospheric thickness is found to be roughly 150 km below the Altiplano and only 100 km below the Puna plateau (Whitman 1992, 1996). Geochemical sampling of young mafic magmas confirmed that these are largely derived from the melting of continental lithosphere, but fail to provide an accurate thickness determination.
Geologic overview of the Central AndesFigure 6 and 7 show respectively a large scale and more detailed geologic map of the Central Andes. The following table summarizes the main geologic features along (north to south) and across (east to west) the convergence zone.
|NS position||EW position||Description|
|North of 22ºS||Chaco foreland||Flexed but otherwise undeformed foreland. The age of the infill is not well known, but generally considered to have Neogene to Quaternary ages.|
|Subandean thin-skinned thrust belt||Corresponds to a classical thin-skinned fold and thrust belt limited in the east by the undeformed foreland and in the west by the Eastern Cordillera|
|Eastern Cordillera||Complex deformation zone characterized by large thrust faults and relatively high elevations. Ordovician and locally older outcrops dominate the picture. Its western limit is the Altiplano-plateau.|
|Altiplano||Constrained by the Eastern and Western Cordillera (present day volcanic arc), it is characterized by the presence of several large scale salars, Quaternary infill, and locally Late Oligocene to Recent volcanic rocks. Sparse basement outcrops are of Ordovician and Cretaceous ages.|
|Western Cordillera||Corresponds to the present volcanic arc and is characterized by a line of stratovolcanoes overlying older ignimbrite sheets.|
|Chilean precordillera||Characterized by intense shortening, Tertiary and Mesozoic volcanism and exposure of pre-Andean igneous basement rocks.|
|Longitudinal Valley||Forearc depression filled with Quaternary and Miocene strata.|
|Coastal Cordillera||Dominated by Mesozoic magmatism. Missing Mesozoic forearc suggest that considerable tectonic erosion has taken place, truncating the western border of the SouthAmerican plate.|
|22º - 25ºS||Transition zone||This zone is characterized by several changes
in the dominating geology and reflect Precambrian to Mesozoic processes and
changes in the subduction geometry.
The most important structural change is the ending of the thin-skinned thrust belt in the north, which gradually gives raise to the thick-skinned Santa Barbara and Sierras Pampeanas systems.
The arc volcanism undergoes a chemical change attributed to the thrusting of older over younger basement.
Between 22.5º and 24º the arc is displaced eastward by the Atacama basin (AB, figure 1), representing an unexplained anomaly in the Andean arc.
|24ºS – 30ºS||Foreland||The foreland thrust belt is replaced by the thick-skinned Santa Barbara and northern Sierras Pampeanas system. Crustal seismicity down to 30 km depth suggest much deeper activity than in the northern thrust belt.|
|Eastern Cordillera||Dominated by Precambrian rocks with minor Late Cretaceous rocks. Southward, Precambian rocks become increasingly more metamorphosed and intruded by Paleozoic and Precambrian plutons.|
|Puna plateau||Characterized by compressional highly deformed Paleozoic "basins and ranges", giving rise to much rougher topography as opposed to the northern Altiplano plateau. Several NW trending Miocene structures are observed and are attributed to old equally oriented lithospheric weakness zones.|
|Volcanic Arc||Absence of recent volcanism is coincident with the flat-slab geometry and consequent absence of astenospheric wedge.|
|~ 30ºS||Sierras Pampeanas|
|Volcanic arc||Volcanism reappears southwards of 33ºS where the subducting slab resumes its "normal" dipping angle of 30ºS.|
|Crustal shortening and its along-strike variation
Shortening in the foreland subandean fold and thrust belt presents great variations and is correlated with the rainfall and therefore erosion rates. It is greatest in Bolivia north of the bend at 18ºS and averages 135 km. This shortening is distributed over an narrow belt (~70km), defining a steep wedge taper (~7-8º between topographic slope and decollement dip). Towards southern Bolivia, the average shortening reduces to 100 km defining a much gentler wedge (~2º). The precipitation is consistently 2-4 smaller. Farther south, the shortening diminishes where the thrust belt disappears in northern Argentina.
The shortening in the Altiplano and Puna plateaus has been difficult to assess because of several reasons. However, crude estimations of the total shortening have been done revealing that Eastern Cordillera and Altiplano shortening is much smaller than the subandean shortening. South of 24ºS, the amount of Andean shortening is even more poorly known. Some estimates of roughly 70 km in the Santa Barbara System and Eastern Cordillera are a few exceptions. Westwards, in the Puna plateau, the shortening is not known at all. However, the greater relief and abundant exposure of pre-Cenozoic rocks suggest either a younger or larger shortening in the Puna than in the Altiplano plateau.
Magmatic patterns across the Altiplano-Puna plateau
Three types of volcanic centers can be identified. Their spatial and temporal distribution gives important clues of the ongoing tectonic process in the whole area.
The main volcanic chain is composed essentially of stratovolcanic complexes, and is characterized by thick andesitic and dacitic lavas associated with pyroclastic flows, domes and hot avalanche deposits. The highest volcanic complexes of the Earth can be found in this area, reaching elevations of more than 6700 m in some cases). The eruptions at high elevations in association with high plumes and strong west to east winds have influenced the distribution of airfall deposits. As expected, coarse-grained material is localized near the centers while fine-grained particles can be found way into the Chaco foreland.
The plateau (in the back-arc) is characterized by giant andesitic to dacitic ignimbrite sheets, which erupted during the late Miocene-Pleistocene, some of them only observable at satellite image scales. These eruptions occurred from huge calderas aligned in parallel to the main arc or on transverse volcanic chains that cross the plateau. In general terms, these events where associated with large scale crustal melting induced by the intrusion of magmas coming from the mantle into the thickened crust.
Finally, small mafic monogenetic and fissure flows in the back-arc have been found to be of Oligocene to early Miocene or late Miocene to Recent age. In general, these basaltic to mafic andesitic flows are of mantle origin. They are more abundant in the southern Puna plateau and are associated with extensional or strike-slip faults in the area. Some of them present crustal or shoshonitic chemical signatures.
The spatial distribution and temporal evolution of the volcanism suggest, as will be discussed below, that the central part of the slab below the Altiplano-Puna plateaus steeped with time, while the northern and southern flanks remained shallow or become progressively shallow.
The distribution over space and time of the volcanic activity undergoes dramatic changes. During late Oligocene to early Miocene the activity is concentrated between 24º-25ºS and shows a strong activity from 24-21 Ma declining during 20-16 Ma.
Magmatic addition and crustal shortening
The early idea to explain the crustal thickening in terms of magmatic addition failed to account forthe amount of crustal volume required to explain the present bulge (1 or 2 orders of magnitude in difference, which makes even large errors in any estimation less important), unless the crustal thickening has been going on since the Jurassic. This is contradicting the general belief that much of the Andean uplift started 15-25 Ma ago. In general it is thought that magmatic addition during the last 15 Ma can only explain 1.5% of the required crustal volume.
Even though magmatic activity can not be considered to be the main reason for the crustal thickening, it is fundamental to understand the process itself, since the magmas and the heat they transport affect largely the rheology and he mechanical behavior of the crust. Small-scale features may also be controlled essentially by magmas.
In general terms, it is commonly believed that crustal shortening is the principal process of crustal thickening. However, discrepancies of the order of 10-20% persist and must be explained in terms of magmatic addition and potential underplating of tectonically eroded forearc material.