SEISMIC VELOCITIES AND CRUSTAL STRUCTURE IN THE FLAT SLAB REGION OF CHILE AND ARGENTINA AS REVEALED BY LOCAL RECEIVER FUNCTIONS AND SURFACE WAVE DISPERSION

The South Central Andes are situated above the zone where the subducting Nazca plate transitions from a normal angle of subduction (~30°) south of -32.5° latitude to a nearly horizontal attitude at 100 km depth over approximately 2° of longitude in the north. This geometry, potentially linked to subduction of the aseismic Juan Fernandez ridge, is thought to play a large role in controlling arc volcanism, upper mantle hydration, and the style of deformation in the back arc. Between 2000 and 2002, researchers with the Chile Argentina Geophysical experiment (CHARGE) collected broadband earthquake data for a period of 18 months at 22 sites in Central Chile and Western Argentina. Results of tomographic (Wagner et al., 2004) and receiver function (Gilbert et al., 2005) analyses of the CHARGE data have revealed unexpected variations in upper mantle velocities, crustal thickness, and sharpness of the Moho discontinuity across the Andes and the Sierras Pampeanas. Interpretations of these results have suggested the presence of eclogite in the lower crust and Mg-rich compositions in the upper mantle. Both of these lithologies have distinct seismic velocity signatures (high Vp and Vs in the case of eclogite and low Vp/Vs, high Vs for Mg-rich mantle rocks) that will be apparent in dispersion of surface waves that traverse the region.

We present a comparison of receiver functions calculated with teleseismic and local events along with preliminary analysis of surface wave dispersion measurements. High frequency receiver functions calculated using intermediate depth local earthquakes reveal crustal structure that is obscured in the lower resolution teleseismic data as well as discrepancies in the arrival time of the converted phase from the Moho. In order to place further depth constraints on receiver function stacks, we impose velocity models consistent with surface wave dispersion curves. The two techniques complement each other well as the average shear velocity structure revealed by surface wave analysis helps to resolve the inherent ambiguity between depth and velocity that can be problematic in receiver function studies.