Mantle Flow in the Vicinity of the Eastern Edge of the Pacific‐Yakutat Slab: Constraints From Shear Wave Splitting Analyses

Abstract

To investigate the effects of a slab edge and varying slab geometry on the mantle flow systems beneath south central Alaska, a total of 971 pairs of teleseismic shear wave (SKS, SKKS, and PKS) and 65 pairs of local S wave splitting parameters (fast orientations and splitting times) are measured using data from the USArray and other networks. The Pacific‐Yakutat slab edge separates two regions with different characteristics of the splitting measurements. The area to the west of the slab edge has greater splitting times and mostly trench parallel fast orientations, and the area to the east is dominated by smaller splitting times and spatially varying fast orientations. The spatial distribution of the splitting parameters and results of anisotropy layering and depth analyses can be explained by a model involving three flow systems. The sub‐slab flow initially entraining with the shallow‐dipping Yakutat slab deflects to a trench‐parallel direction due to slab retreat and an increase in slab dip, and flows northeastward toward the slab edge, where it splits into two branches. The first branch enters the mantle wedge as a toroidal flow and flows southwestward along the slab, and the second branch continues approximately eastward. The flowlines of the toroidal and continued flow systems are approximately orthogonal to each other in the vicinity of the slab edge, producing the observed small splitting times and spatially varying fast orientations.

Panoptic View of Mantle Flow Beneath Trans‐Continental Northeast Asia: Distinct Variation Detected From ∼2,000 km Shear Wave Splitting Profile

“…Abaga volcanic field, one of the largest intraplate volcanic areas in the late Cenozoic in eastern Asia, has complex geological structures and ambiguous geodynamic causes (Xu et al, 2019;Zhang & Guo, 2016). It has been shown that events from different back-azimuth directions have the 10.1029/2021GL097116 of 11 possibility of identifying "piercing point dependent" anisotropy, as recently highlighted by several shear wave splitting studies in Alaska (Yang et al, 2021), Sichuan Basin (Jia et al, 2021) and Central Myanmar areas (Fan et al, 2021). Therefore, it must be mentioned that the existence of complex anisotropy around this intraplate volcano cannot be entirely ruled out and it could be left for future detailed investigation.…”

Classification of Teleseismic Shear Wave Splitting Measurements: A Convolutional Neural Network Approach

“… (a) Comparison of human‐determined (blue bars; Yang et al., 2021) and convolutional neural network (CNN)‐selected (red bars) shear wave splitting measurements in south central Alaska. Green bars are measurements accepted by both CNN and human operators.…”

“…We next apply the trained CNN to broadband seismic data in south central Alaska recorded by 127 stations with variable recording period from 1988 to October 2019. The procedures to request (from the Data Management Center of the Incorporated Research Institutions for Seismology) and preprocess the XKS data follow those described in Liu and Gao (2013) and are identical to those used by Yang et al (2021). In total 19,960 pairs of splitting parameters are obtained at 127 Stations (Figure 3a).…”

“…In this method, a grid search procedure is applied to find the optimal pair of splitting parameters ( ϕ , δt ) corresponding to the minimum XKS energy on the corrected transverse component. Numerous SWS studies suggest that in order to obtain reliable splitting measurements, an essential step is to visually verify all the measurements (e.g., Liu & Gao, 2013), as demonstrated in recent studies in North America (e.g., Liu et al., 2014; Yang et al., 2016, 2017; Yang et al., 2021). Due to the ever‐increasing number of stations established around the world and the resultant exponential increase in the amount of data available for SWS analysis, this laborious task is increasingly time‐consuming, and therefore alternate time‐efficient yet reliable approaches are needed.…”

Potassic Volcanism Induced by Mantle Upwelling Through a Slab Window: Evidence From Shear Wave Splitting Analyses in Central Java

“…Numerous studies using shear wave splitting (SWS, a tool that the current study employs) have demonstrated that trench retreat and the accompanying slab rollback can induce trench‐parallel entrained mantle flow in the subslab region and trench‐normal corner flow in the mantle wedge above the slab (e.g., Becker & Faccenna, 2009; Long & Becker, 2010; Schellart, 2004; Sternai et al., 2014). In addition, driven by the difference of pressure across the subducted slab, part of mantle flow in the subslab may enter the mantle wedge through lateral slab edges or slab windows (e.g., Faccenda & Capitanio, 2013; Fan et al., 2021; Yang et al., 2021). The horizontal component of the various flow systems could be readily characterized and constrained by the strength and orientation of seismic azimuthal anisotropy (Silver & Chan, 1991; Zhang & Karato, 1995).…”