Structure of the subduction system in southern Peru from seismic array data

Abstract

[1] The subduction zone in southern Peru is imaged using converted phases from teleseismic P, PP, and PKP waves and P wave tomography using local and teleseismic events with a linear array of 50 broadband seismic stations spanning 300 km from the coast to near Lake Titicaca. The slab dips at 30 and can be observed to a depth of over 200 km. The Moho is seen as a continuous interface along the profile, and the crustal thickness in the back-arc region (the Altiplano) is 75 km thick, which is sufficient to isostatically support the Andes, as evidenced by the gravity. The shallow crust has zones of negative impedance at a depth of 20 km, which is likely the result of volcanism. At the midcrustal level of 40 km, there is a continuous structure with a positive impedance contrast, which we interpret as the western extent of the Brazilian Craton as it underthrusts to the west. V p /V s ratios estimated from receiver function stacks show average values for this region with a few areas of elevated V p /V s near the volcanic arc and at a few points in the Altiplano. The results support a model of crustal thickening in which the margin crust is underthrust by the Brazilian Shield.

Seismic structure in southern Peru: evidence for a smooth contortion between flat and normal subduction of the Nazca Plate

“…Hayes et al (2012) even acknowledges their difficulty in fitting the true subduction interface in this region, as evidenced by anomalous short-wavelength features and large misfits. The slab surface as determined from receiver functions (Phillips et al 2012;Phillips & Clayton 2014) is also more consistent with the Cahill & Isacks (1992) contours.…”

“…12) in an effort to constrain the structure and morphology of the Nazca slab. The three velocity models tested are: (1) southern Peru velocity model from receiver function study by Phillips et al (2012) (P12; Fig. 11a); (2) simplified southern Peru velocity model from surface wave study by Ma & Clayton (2014) (MC14; Fig.…”

“…11c). Phillips et al (2012)'s receiver function results are used to constrain the depths of the crustal and mantle discontinuities in the MC14 and MC14b models. A simplified 1-D average of S-wave velocities from 2-D cross-sections in Ma & Clayton (2014) is used to generate the MC14 and MC14b models.…”

“…In general, the ISC data form a scatter envelope around the EHB locations. Thus, the dip angle is estimated by visually selecting ISC hypocentre locations that are downdip of the trench and not within the overriding plate [as defined by teleseismic receiver functions (Phillips et al 2012;Phillips & Clayton 2014)], then performing a linear regression of the selected locations. Outliers in this selection process are events that are located significantly deeper than or off-trend from the main concentration (i.e.…”

Theory and Observations - Seismology and the Structure of the Earth: Teleseismic Body-Wave Scattering and Receiver-Side Structure

“…This approach is commonly used for reconnaissance work (e.g., Darbyshire et al, 2007;Finotello et al, 2011;Phillips et al, 2012;Wei et al, 2011) and involves stacking of multiple-scattering modes r, s (consistent with the formulation in eqn [31]) along predicted travel-time trajectories for a grid of Moho depths and Poisson's ratios, to recover an Delayed and summed SV-components of source-normalized teleseismic P recorded at the Gräfenberg array, Germany. Each trace represents a sum of seismograms, from events at different epicentral distances, delayed by a time interval corresponding to the expected arrival of a conversion from a given trial depth.…”