Effects of change in slab geometry on the mantle flow and slab fabric in Southern Peru

Abstract

The effects of complex slab geometries on the surrounding mantle flow field are still poorly understood. Here we combine shear wave velocity structure with Rayleigh wave phase anisotropy to examine these effects in southern Peru, where the slab changes its geometry from steep to flat. To the south, where the slab subducts steeply, we find trench‐parallel anisotropy beneath the active volcanic arc that we attribute to the mantle wedge and/or upper portions of the subducting plate. Farther north, beneath the easternmost corner of the flat slab, we observe a pronounced low‐velocity anomaly. This anomaly is caused either by the presence of volatiles and/or flux melting that could result from southward directed, volatile‐rich subslab mantle flow or by increased temperature and/or decompression melting due to small‐scale vertical flow. We also find evidence for mantle flow through the tear north of the subducting Nazca Ridge. Finally, we observe anisotropy patterns associated with the fast velocity anomalies that reveal along strike variations in the slab's internal deformation. The change in slab geometry from steep to flat contorts the subducting plate south of the Nazca Ridge causing an alteration of the slab petrofabric. In contrast, the torn slab to the north still preserves the primary (fossilized) petrofabric first established shortly after plate formation.

Lowermost mantle anisotropy near the eastern edge of the Pacific LLSVP: constraints from SKS–SKKS splitting intensity measurements

“…Most of stations in this study are located in the northern part of the Central Andean Plateau or the southern portion of Peru, which has been shown by several previous studies to exhibit strong and complex anisotropy in the upper mantle associated with subduction-related deformation (Eakin et al 2014(Eakin et al , 2015Antonijevic et al 2016;Long et al 2016). Therefore, we expect that our set of SK(K)S measurements primarily reflects upper-mantle anisotropy beneath the corresponding stations, with only a subtle contribution, if any, from the lowermost mantle.…”

“…Thus, an explanation for SKS-SKKS discrepancies that invokes upper-mantle heterogeneity as a primary cause would require variability on this very short length scale. Upper-mantle anisotropy has been extensively studied beneath the PULSE and CAUGHT arrays, using both shear wave splitting (Eakin et al 2014(Eakin et al , 2015Long et al 2016) and surface wave tomography (Antonijevic et al 2016). These studies found that while anisotropy patterns do vary laterally, the variations are generally inferred to be over considerably longer length scales (particularly for SKS splitting).…”

Modeling the Geometry of Plate Boundary and Seismic Structure in the Southern Ryukyu Trench Subduction Zone, Japan, Using Amphibious Seismic Observations

“…The geometry of the plate boundary is fundamental information for subduction zone seismology. For example, the geometry of the subducting plate controls the thermal structure and the variation of the mantle flow pattern (e.g., Antonijevic et al, ; Kneller & van Keken, ). Gustcher () proposed a relationship between the dip angle of the subducted slab and the interplate coupling along the Andean subduction zone.…”

Crustal structure of north Peru from analysis of teleseismic receiver functions

“…5B) (Scire et al, 2016) and surface-wave tomography (Antonijevic et al, 2016). These low velocities could be produced by increased temperature, the presence of partial melt, and/or the presence of hydrous phases, implying that more buoyant material is present in this area than in the surrounding mantle (Scire et al, 2016;Antonijevic et al, 2016). The correspondence between the 100-and 120-km contours and the edge of the low-velocity anomaly is consistent with the southern margin of the flat slab being partially supported by this material.…”

“…The southeastern corner to the flat slab, in a region well sampled by our array, extends well beyond the southern margin of the Nazca Ridge projection and directly above a prominent sub-slab low-velocity zone image at 130 km depth by teleseismic S-wave tomography (Fig. 5B) (Scire et al, 2016) and surface-wave tomography (Antonijevic et al, 2016). These low velocities could be produced by increased temperature, the presence of partial melt, and/or the presence of hydrous phases, implying that more buoyant material is present in this area than in the surrounding mantle (Scire et al, 2016;Antonijevic et al, 2016).…”