Deformation Processes in the Andes - new Data and Problems

Antofagasta, Thursday, July 31st , 1997 - Saturday, August 2nd, 1997

Abstracts

Thrust tectonic controls on late Tertiary sedimentation pattern in the Salar de Antofalla area, southern Puna (NW Argentina) (Poster)

Adelmann, D., Kiefer, E. and Görler, K.
FU Berlin

Since Miocene times the area of the Salar de Antofalla, located east of the present Andean volcanic arc, was an isolated area of continental sedimentation. Compressional tectonics between early Miocene and Pliocene times produced a series of westvergent folds and thrusts.

In particular, faulting during the mid-Miocene (Quechua phase) led to a narrow, elongated broken-foreland basin (Salar de Antofalla basin). The development of frontal thrust ramps caused topographic highs along the eastern edge of this sedimentation area and controlled the facies distribution pattern. Coarse-grained alluvial fans were deposited at the basin margin during the initial stages of the basin infill and graded basinwards into sandstones and mudstones of a playa mudflat and sandflat facies. This alluvial fan sedimentation was followed, during a period of tectonic quiescence, by the sedimentation of lacustrine carbonates and sulfates along the border of the fan bodies. Contemporarily, massive halite deposits developed in the center of the partially flooded basin. Finally, the Pliocene Diaguita phase triggered compressional deformation together with another period of alluvial fan sedimentation. These deposits overlay the older, deformed late Tertiary sediments with an angular unconformity.

Tectonic and Neotectonic Style in the Salar de Antofalla Region of the Southern Puna (NW Argentina) (Poster)

Alten, M., Adelmann, D., Kiefer E. & Görler, K.
FU Berlin

The area is characterised by NNE/SSW to N/S striking reversed faults and overthrusts. Their vergency is predominantly WNW to W. The crystalline basement and Palaeozoic sediments are involved. The style of deformation can be interpreted as "thick skin tectonics", as observed in the adjacent Eastern Cordillera and the Subandin, but with an opposite vergency.

The style of deformation depends strongly on the type of the involved rocks. Subhorizontal and flat overthrusts of the crystalline basement and the Neogene continental red beds and massive halites are observed, whereas in the older sediments steep reverse faults prevail. Especially the Neogene limestones and gypsums reacted on the compressional forces with fold structures.

Normal faults are especially frequent on both rims of the southern Salar de Antofalla and are interpreted as progressive step faults, triggered by gravitational effects. The existence of strike slip faults has been proved at only few positions in the field, but a pattern of WNW/ESE striking lineaments are interpreted as sinistral lateral shifts. There is no field evidence for a pull apart origin of the Salar de Antofalla and the related Salina de Fraile basin. An interpretation as a compressional basin is preferred. The results of ongoing compressional forces are visible in the local deformation of the actual surface of the Salar de Antofalla.

Electrical conductivity anomalies (high and low) in the Central Andes - interpretation and open problems (Poster)

H. Brasse
FU Berlin

Magnetotelluric (MT) investigations revealed several zones of enhanced electrical conductivity at different depth levels in the Central Andean crust and upper mantle. While some of these zones are well studied and believed to be explainable in terms of their petrophysical and tectonic behavior, there remain manifold open questions in understanding their detailed geometry and their relation to other geophysical parameters and geological/tectonic processes.

1. In the Coastal Cordillerra anomalies in the middle crust superimpose and even dominate the coast effect (originating from the highly conducting Pacific Ocean). Are these anomalies related to the Atacama fault system; do they hint at graphite depletion on shear zones or increased hydraulic routing in these crustal levels?
2. EW-striking anomalies were recognized in the Forearc and display a certain correlation with aeromagnetic features. Does this indicate a prolongation of prominent fault zones like the El Toro-Calama lineament towards the west?
3. A highly conductive zone (HCZ) characterizes the run of the Falla Oeste in the Precordillera in the upper and lower (!) crust. Beyond the more suggestive interpretation by ore deposits, other mechanisms must be taken into account to explain the deep seated HCZ.
4. The magmatic arc hosts the most prominent HCZ in the Central Andes at depths > 20 km. It correlates with a seismic low velocity zone, high seismic absorption and a gravity low and may be explained by vast amounts of partial melts (cf. contribution by F. Schilling). However, the anomaly strikes obliquely to the Western Cordillera and vanishes in the north at 20°S. Is there thus a correlation to the flattening of subduction towards the south and/or does it indicate recent cooling of the lower crust in the region of the Pica gap?
5. Early models of a backarc conductor below the Altiplano suggest a ramp-like geometry (ascending towards the east) and may hint at the overthrusting of the Andean crust over the Brazilian craton. This important topic will be addressed during the ANCORP MT experiment.

An ascending HCZ - interpreted as a rise of the electrical asthenosphere - was modelled below the Eastern Cordillera in NW Argentina. How does this structure - also evident from seismological observations - relate to the Puna volcanism and/or the development of the Salta Rift?

The Gravity Field of the Continent-Ocean Transition Mapped from Land, Air, and Sea - Western Margin of South America (Poster)

H.-J. Götze¹, M. Araneda², G. Chong D.⁴, J. Fritsch³, A. Kirchner¹, M. Kösters¹, and S. Schmidt¹

1 Institut für Geologie, Geophysik und Geoinformatik, FU Berlin, Malteserstr. 74-100, D-12249 Berlin, Germany.
2 Departamento de Geofísica, Universidad de Chile, Blanco Encalada 2085, Santiago, Chile.
3 Bundesanstalt für Geowissenschaften und Rohstoffe, Postfach 51 01 153, D-30631 Hannover, Germany.
4 Universidad Católica del Norte, Departamento de Ciencias Geológicas, Casilla 1280, Antofagasta, Chile.

INTRODUCTION

From 1993 to 1996 the international MIGRA group with participants from Chile, Argentina and Germany surveyed some 3.500 new gravity observations in an Andean geotraverse covering N-Chile and NW-Argentina between 64°- 71° W and 20° - 29°· S.Including reprocessed older data of Freie Universität Berlin, South American universities, oil and mining industry, now there is a data base of about 15.000 gravity values available, which can be used together with other geophysical and geological information for an interdisciplinary interpretation of the structure and evolution of the Central Andes. In summer 1995 MIGRA took part in the "CINCA" offshore experiment of the german research vessel "Sonne" between the latitudes 20° S to 24° S. The offshore gravity data are connected with the onland survey and linked to the IGSN 71 to draw a complete gravity picture of this ocean-continent transition.

ONSHORE FIELD WORK AND REDUCTIONS

The investigated region covers a 900 km x 1.000 km area in the central part of the Andean orogenic system The young Andean orogen between 20° - 29° S comprises different structures which have evolved on a Precambrian-Paleozoic basement. This belt of ancient rocks was also described as the border of the "Faja Eruptiva Occidental". Two of our gravity surveys obtained structural information with new stations covering the northern and southern edges of this belt, near Calama (Chile) and in the southern Argentinean Puna respectively.Enormous topography, its aridity, low population density and limited infrastructure limiting our field work and the lack of topographic maps and geodetic networks in some regions as well. The spacing of stations amounts to approximately 5 km along all passable tracks aside from some local areas with a higher station density. To complete this data base we included gravity observations from different sources.

OFFSHORE GRAVITY AND DATA ACCURACY

Gravity data at sea were collected along continous survey lines (some 8.000 km) during the CINCA offshore experiment SO - 104 of the german research vessel "Sonne". The average density of observational sites amounts of about 15 observations per km. Gravity survey of cruise SO - 104 was tied to the Chilean National Gravity network of the "Departamento Geofísico, Universidad de Chile, Santiago" at reference stations in Valparaiso, Antofagasta and Iquique. The drift of the onboard gravity sensor was low and amounts to -0.048 mGal/day, or -1.44 mGal/month, respectively.

Further information about the accuracy of the offshore survey provide the so called "misties". Misties are crossover errors if the gravity readings of two crossing lines are compared.

RESULTS

We present a three dimensional density model in the Central Andes explaining the observed geoid and gravity without offset between the calculated and measured signals. Except the downgoing slab the model does not take into account other density inhomogeneities below 220 km. The density distribution in forearc and arc regions was calibrated to the generalized geometry of 2D ray tracing models. Crustal densities were preliminarily determined according to common velocity-density relations and slightly changed under the need to fit curves with fixed geometry. The model's Moho discontinuity, deepest below the recent arc and the Altiplano-Puna zone, shows an along-strike change in depth ranging from 56 km at 24° S to 65 km at 20° -21° S. Density contrasts at the "gravimetric" Moho vary between 0.36 - 0.37 g cm-3 in arc and backarc areas and 0.45 g cm-3 in the Coastal Cordillera due to low density bodies directly above the subducting slab. A Vening-Meinesz isostatic Moho derived from topographic loads with anormal crustal thickness of 35 km, a crust-mantle density contrast of 0.35 g cm-3 and a lithospheric rigidity of 1023 Nm is situated well below the model Moho in the south at about 24°·, but coincides well with it at 21°S. This possibly reflects concepts of lithospheric delamination south of 24°S.

The isostatic state of the model, derived by comparing the local density distribution relative to the reference model with the topographic load, shows a significant overcompensation (mass deficit below sea level) in the forearc between 18°S and 23°S. This coincides with a region where modelled gravity and geoid trend a little lower than the observed signals. This could be explained by the fact that the modelled slab's top is situated too low thus leaving too much room for light weighted crustal material. A slab subduction at a lower angle is suggested in this zone.

Vp, Vp/Vs and hypocenters in the southern central Andes from local earthquake tomography (Poster)

Graeber, F.
GFZ Potsdam

A subset of high quality travel time data from the seismological projects PISCO'94 and CALAMA was used to determine the velocity structure between the Longitudinal Valley and the Western Cordillera by a simultaneous inversion. The most prominent features in the final results are very high velocities along the Wadati-Benioff-Zone (WBZ), low velocities beneath the recent magmatic arc, a high VpNs-ratio above the WBZ and, on average, a low Vp/Vs-ratio in the continental crust. The majority of hypocenters in the WBZ are related to the high velocity anomalie, thus supporting the idea of phase transitions within the subducting plate. Low values of Vp and high values of Vp/Vs beneath the crust of the Western Cordillera can be explained by the presents of an asthenospheric wedge.

Structural and Magmatic Evolution of the Late Cretaceous-Early Tertiary Precordilleran Arc System, Northern Chile, 21-26°S (Poster)

A. Günther, M. Haschke, K.J. Reutter, E. Scheuber
FU Berlin

The Chilean Precordillera between 21-26°S displays two contrasting tectonic styles along its N-S trending axis. We distinguish a northern section (north of 22,5°S), marked by intense folding and faulting in the arc (38,5 Ma, "Incaic phase"), and, south of Calama (22,5°S), a southern section where deformation occurred at the arc-backarc transition. This subdivision coincides with a transitional N-S pattern of dated magmatic rocks, such that the occurrence of volcanic rocks migrates, from the southernmost locality towards the northern section during time, whereas intrusives vary widely and concentrate in the northern section after the onset of Incaic deformation. We suggest that the spatially and temporarily contrasting tectonism and magmatism along the arc axis are controlled by a suite of NW-SE/NE-SW and N-S trending fault zones ("Megafallas") that separate individual crustal blocks from each other. These faults have presumably been established during the Paleozoic and, due to repeated reactivation, still control the tectonic development of the Central Andes (e.g. Falla Olacapato-El Toro).

Evolution of a Miocene / Pliocene Trench-Slope Basin in the Peru-Chile-Trench: The Iquique Basin, Chile (Poster)

E.Kiefer¹ and G. Wissmann²

1 Institut für Geologie, Geophysik und Geoinformatik, FU Berlin
2 Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover

The German-Chilean offshore seismic survey CINCA'95 provided new data about geology of the Peru-Chile Trench between 19.0° and 26.0° S. First analyses prooved gravitative tectonics and tectonic erosion by the subducting Nazca Plate as major controls of the continental margin development. According to the seismic data, the continental margin is characterized by three major seismic horizons. A base reflector can be traced from the Coastal Cordillera down to the trench. It is interpreted as top of the Andean basement formed by the Jurassic La Negra Formation and subcrops of pre-Andean plutons and metamorphics. The basement is covered by Miocene sediments, accumulated in antithetically rotated half-grabens. They are disconformably topped by Pliocene continental slope sediments, which grade into Pleistocene to recent turbidites. The half-graben master faults seem to merge into intra-crustal detachments, on which the tilted blocks slide down to the trench in a conveyor belt style. The seismic pattern of the basin fill is characterized by prograding wedges and onlap structures. The seismic sequences of the lquique Basin point to sedimentation in a neritic marine environment, controlled by a fluctuating sea-level and rapid subsidence in Miocene. Pliocene sedimentation completed the basin fill. Subsidence and half-graben formation is interpreted to be the result of thinning and weakening of the forearc due to basal tectonic erosion by the subducting Nazca Plate. Prominent features are trench-slope basins. Their structures and subsidence histories offer clues to the dynamics of basins at convergent margins, Subsidence and sedimentation of the lquique Basin offshore northern Chile is modelled and restored by computerbased techniques:
1. The complex trench-slope basin architecture is digitized by transformation of seismic time sections into geological cross-sections by a graphic system.
2. The vertices of the layer polygons are filed in a non-graphical 3D-modelling system in order to calculate a real depth model and to balance volumes and masses.
3. The volume balance and dredge sample control provide the data base for calculation of subsidence curves. The resulting sedimentation rates are compared with published data to improve calibration of the subsidence model.
4. Finally, the basin and fault geometries are compiled in a dynamic 2D-model in order to trace the spatial evolution of the basin.

The Cretaceous to Palaeocene Potosi Basin (Eastern Cordillera, Southern Bolivia) (Poster)

Dorothee Mertmann, Kerstin Fiedler and Volker Jacobshagen
FR Allgemeine Geologie, Institut for Geologie, Geophysik und Geoinformatik, FU Berlin, Malteserstr. 74-100, D-12249 Berlin

The Potosi Basin is located in the Eastern Cordillera of Southern Bolivia. It was a narrow, N-S orientated basin-branch, which connected the Peruvian/Northern Bolivian back-arc basin with the Salta rift system in northwestern Argentina. The basin fill (Puca Group) is composed mainly of fluvial to lacustrine clastics, lagoonal and marine carbonates, and subordinated mafic volcanics. Sedimentation patterns reveal a synrift phase during the Early to lower Late Cretaceous and a postrift phase until the Palaeocene.

The synrift deposits are confined to a graben segment between Potosi and Tupiza. The continental sediments developed in a semiarid climate. During the Cenomanian and Turonian a marine ingression reached the northern edge of the basin. A major change during the basin evolution took place, when widespread thermal subsidence integrated previous areas of non-deposition during the postrift phase. Presumably Santonian/Campanian sediments considerably onlap onto the Palaeozoic basement. The basin shows an asymmetrical cross-section with a steeper gradient in the west relative to the east. During the Maastrichtian and Palaeocene, fluvial and lacustrine environments prevailed. Palaeocene movements and subsequent erosion occured only along the western margin, e.g. Atocha. An angular unconformity is present there, but is missing in the Camargo area.

Cross section at 22.5° S showing fault plane solutions at 100 km depth. (Poster)

Alexander Rudloff
GFZ Potsdam

The picture above left shows backround seismicity taken for half-a-degree width, active volcanoes and the shape of the WBZ as derived by Cahill & Isacks (1992). picture above right shows sorted events, taking the rake-value as marker: If the rake is positive, the event is marked light grey - indicating compression -,if the rake is negative, the event is marked light grey - indicating extension.This sorting is done taking the lower hemispheric data. The picture below: Here the region of activity is zoomed and the fault plane solutions are rotated to a backhemispheric projection (sized by magnitude). One can see that only three events remain as clear compressive events, while the most events become/remain extensive. The dip angle of the Wadati-Benioff-Zone reaches 22°S. Some words about data treatment: the fault plane solutions were estimated using the FPFIT program by Reasenberg and Oppenheimer (1985). Stress inversion is calcultated using the FMSI program by Gephart & Forsyth (1984), Gephart (1990); an inversion for the dataset above shows mainly slab parallel extension (e.g. slab pull forces).The dataset is not as homogeneous as the data from CMT solutions, covering a higher magnitude range (the PISCO '94 data covers magnitudes between 2.0 and 5.0.

Estimates of magmatic pressure, temperature and water concentrations in late Neogene ignimbrites of the Altiplano-Puna Volcanic Complex: evidence for deep-seated magma chambers (Poster)

Axel Schmitt, Jan Lindsay, Rainer Thomas, Robert Trumbull, Wolfgang Siebel
GFZ Potsdam

Starting about 10 Ma ago, a major phase of silicic volcanism occurred in the Central Volcanic Zone at latitudes of 21° to 24° S. This resulted in widespread ignimbrites sheets, which are well represented by the Aliplano-Puna Volcanic Complex (deSilva, 1989). Isotopic evidence and other geochemical arguments suggest that the ignimbrite magmas are dominantly crustal melts, in contrast to the andesite-dacite magmas. We are working to constrain the P-T-XH20 conditions of the ignimbrite magmas in order to reveal the position of magma chambers in the crust and to place limits on the source region and melting process which gave rise to the magmas. The current data comes from samples of the Purico (1Ma) and Atana (4Ma) ignimbrites of the APVC.The rocks are rhyodacitic, crystal-rich, pumaceous tuffs with phenocrysts of plagioclase, quartz, biotite, hornblende, magnetite, and accessory ilmenite, zircon and titanite. Some samples also contain orthopyroxene or K-feldspar. Conventional geothermobarometry yields temperatures of 780 -820 C (magnetite-ilmenite, horneblende-plagioclase) and pressures of 3-6 kbar (Al- in- hornblende and silica activity in Qtz-Opx-Mt assemblages). These P-T values represent the conditions at which phenocrysts formed or last equilibrated, and therefore they probably relate to the level of magma storage and not the level of melting. Using a special electron microprobe technique, water contents of 4-5 wt.% were determined in glassy melt inclusions in quartz and plagioclase. Based on phase relations in the H2O-haplogranite system, we conclude that the melts were strongly undersaturated in water and that melt inclusions were trapped at a minimum pressure of 5 kbar. These first results suggest that silicic magma was present under the APVC at a level of about 15 kilometers. Therefore we suggest that, like the Fish Canyon Tuff in Colorado, the large ignimbrites of the APVC were not erupted from high-level magma chambers. deSilva, S. (1989) Geology, 17, 1102-1106.

Relationships between age and composition of Miocene to Recent arc magmas from the central Andes, 25-26°S and implications on crustal contamination

R. Trumbull, K. Hahne, R. Emmermann
GFZ Potsdam

W. Büsch
FU Berlin

H. Gerstenberger, W. Siebel
GFZ Potsdam

R. Wittenbrink
FU Berlin

The Neogen volcanic chain in the central Andes (Central Volcanic Zone, CVZ) is an example of subduction-related, calc-alkaline magmatism in a setting of anomalously thick continental crust. It is well established that the chemical and isotopic compositions of the Recent andesite-dacite volcanoes in this area reflect a high degree of crustal contamination (high LILE contents, radiogenic Sr and Pb, non-radiogenic Nd) which distinguish them from older Andean magmatic rocks and from time-equivalent rocks in other parts of the Andean cordillera, where the crust is not so thick. Most workers seem to agree that the crustal signature of CVZ volcanoes is caused by assimilation of material within the crust. Ideally, the volcanics can be used as geochemical probes of the crust. It has been shown already that differences in the Pb-isotopic composition of Recent arc volcanic rocks can indicate the existence of different crustal domains beneath the arc (Wörner et al., 1992).

The correlation between crustal thickness and the geochemical signature of central Andean magmas suggests a simple connection between thickening and enhanced crustal assimilation by arc magmas. However, the contribution of subducted continental material (subduction erosion) below the arc has also been suggested (Stern, 1991). The present study attempts to shed more light on the nature and origin of the crustal signature of arc volcanics in the CVZ by looking at the changes in the composition of the andesitic and dacitc rocks as a function of age, i. e. from early Miocene to present. The study area is restricted to a small segment of the arc at the southern end of the CVZ (25-26°S), aloowing us to assume that the volcanoes of different ages erupted through the same crustal domain. The focus is on the period from about 20 Ma to the present, a time span which brackets the main episode(s) of crustal thickening in the central Andes as deduced from geological constraints. The rocks analysed here are from large strato-volcanoes and associated lava fields (e. g. Leon Muerto, Colorado del Azufre, Chaco, Cerro Blanco, Cordon del Azufre, Aguas Blancas, Bayo and Lastarria). Their radiometric ages are approximately 20 Ma, 16 Ma, 11 Ma, 7-8 Ma, 5 Ma, 1 Ma and less than 1 Ma.

Within a given SiO²-range, the chemical composition of the magmas erupted at different times is very similar for most major and trace elements, but significant and systematic differences exist for a limited number of large-ion litophile elements (LILE) and for the Sr-, Nd- and Pb-isotopes. The changes in magma chemistry correlate with decreasing ages as follows:

Rb, Cs, Th and U concentrations of the magmas progressively increase. Fig. 1 shows similar shifts in Rb for low-SiO² samples and for high-SiO² samples. This clearly indicates that the elemental enrichments do not simply reflect increasing degrees of fractionation. Sr-isotopic ratios increase gradually from 0.7053 to 0.7075; eNd values decrease from 0 to -5 (Fig. 2).The magmas show increasing ²°6Pb/²°4Pb- (18.60 to 18.87), ²°7Pb/²°4Pb- (15.60 to 15.68) and ²°6Pb/²°4Pb- (38.48 to 38.94) isotopic ratios (Fig. 3).

Fig. 1.

Rb content versus age for Miocene to Recent volcanic rocks from the southern CVZ (25-26°S).Diagram on the left side includes samples with SiO² < 60 wt%, that on the right shows samples with SiO² > 60 wt%.

These differences can be explained by assimilation of increasing amounts of crustal material by younger samples. Geochemical modelling suggests that the assimilation took place by incorporation of felsic melts. The timing of enhanced crustal assimilation (ca. 10 Ma to present) closely corresponds with the eruption of voluminous dacite-rhyolitic ignimbrite sheets in the CVZ, and the composition of these ignimbrites is compatible with the felsic component called for by the mixing model (Fig. 3). Therefore, we argue that the trends displayed by the Miocene to Recent andesitic and dacitic arc magmas is mainly caused by incorporation of increasing amounts of intra-crustal melts.

References

WÖRNER, G., MOORBATH, S. and HARMON, R.S. (1992), Andean Cenozoic volcanic centers reflect basement isotopic domains. Geology 20: 1103-1106.

STERN, C.R. (1991), Role of subduction erosion in the generation of Andean magmas. Geology 19: 78-81.