Seismic Detection and Characterization of the Altiplano-Puna Magma Body, Central Andes

Abstract

— The Altiplano-Puna Volcanic Complex (APVC) in the central Andes is the product of an ignimbrite “flare-up” of world class proportions (de Silva, 1989). The region has been the site of large-scale silicic magmatism since 10 Ma, producing 10 major eruptive calderas and edifices, some of which are multiple-eruption resurgent complexes as large as the Yellowstone or Long Valley caldera. Seven PASSCAL broadband seismic stations were operated in the Bolivian portion of the APVC from October 1996 to September 1997 and recorded teleseismic earthquakes and local intermediate-depth events in the subducting Nazca plate. Both teleseismic and local receiver functions were used to delineate the lateral extent of a regionally pervasive ∼20-km-deep, very low-velocity layer (VLVL) associated with the APVC. Data from several stations that sample different parts of the northern APVC show large amplitude Ps phases from a low-velocity layer with Vs ≤ 1.0 km/s and a thickness of ∼1 km. We believe the crustal VLVL is a regional sill-like magma body, named the Altiplano–Puna magma body (APMB), and is associated with the source region of the Altiplano–Puna Volcanic Complex ignimbrites (Chmielowski et al., 1999).¶Large-amplitude P–SH conversions in both the teleseismic and local data appear to originate from the top of the APMB. Using the programs of Levin and Park (1998), we computed synthetic receiver functions for several models of simple layered anisotropic media. Upper-crustal, tilted-axis anisotropy involving both Vp and Vs can generate a “split Ps” phase that, in addition to the Ps phase from the bottom of a thin isotropic VLVL, produces an interference waveform that varies with backazimuth. We have forward modeled such an interference pattern at one station with an anisotropy of 15%–20% that dips 45° within a 20-km-thick upper crust. We develop a hypothesis that the crust above the “magma body” is characterized by a strong, tilted-axis, hexagonally symmetric anisotropy. We speculate that the anisotropy is due to aligned, fluid-filled cracks induced by a “normal-faulting” extensional strain field associated with the high elevations of the Andean Puna.