The Role of Variable Slab Dip in Driving Mantle Flow at the Eastern Edge of the Alaskan Subduction Margin: Insights From Shear‐Wave Splitting

Abstract

Alaska provides an ideal tectonic setting for investigating the interaction between subduction and asthenospheric flow. Within the span of a few hundred kilometers along strike, the geometry of the subducting Pacific plate varies significantly and terminates in a sharp edge. Furthermore, the region documents a transition from subduction along the Aleutian Arc to strike‐slip faulting along the Pacific Northwest. To better understand mantle interactions within this subduction zone, we conduct an SKS shear‐wave splitting analysis on passive‐source seismic data collected between 2011 and 2018 at 239 broadband seismometers, including those from the Transportable Array. Anisotropic fast directions in the east of our study area parallel the Queen Charlotte and Fairweather transform faults, suggesting that the ongoing development of lithospheric anisotropy dominates the results there. However, our observed delay times (δt = 1–1.5 s) obtained across the study region may also imply an asthenospheric contribution to the splitting pattern. Our splitting observations exhibit slab‐parallel fast directions northwest of the trench and a rotation of fast directions around the northeastern slab edge. These observations suggest the presence of toroidal asthenospheric flow around the edge of the downgoing Pacific plate. We suggest that Wrangell Volcanic Field volcanism might be caused by mantle upwelling associated with this flow. Splitting observations closer to the trench can be explained by fossil anisotropy within the downgoing Pacific‐Yakutat plate combined with entrained subslab mantle. The geometry of the slab, including its variable dip and its abrupt eastern edge, thus plays an important role in governing mantle flow beneath Alaska.

Shear Wave Splitting and Mantle Flow Beneath Alaska

“…Jadamec and Billen () favored the shorter, shallower slab model, which produced horizontal flow consistent with the shear wave splitting patterns shown in Christensen and Abers (), and toroidal flow around the inferred edge of the Pacific plate. The shear wave splitting results of Venereau et al () agree with this model to the east of the slab and are very similar to our results. Our new shear wave splitting results may support this interpretation, in that there appears to be a general circular pattern of fast directions centered between the inferred edges of both the Pacific and Wrangell slabs (pink symbols; Figure ), with many null measurements that may be indicative of vertical anisotropy (vertical flow) (Figure ).…”

“…There are 2,389 results from 386 stations from this study plotted in pink in Figure , and the 842 results from Perttu et al () and Christensen and Abers () are shown in red in Figure . Venereau et al () used a subset of the stations that we do in our study region and find similar large‐scale splitting patterns across Alaska.…”

“…Either the principal flow direction is predominantly along strike, or other processes such as aligned melt below the arc are controlling anisotropy here. Venereau et al () propose that the strike‐parallel fast directions are related to the sharp variability of the Pacific slab dip and strike creating pressure gradients in the mantle wedge.…”

Seismic Anisotropy in the Java‐Banda and Philippine Subduction Zones and its Implications for the Mantle Flow System Beneath the Sunda Plate

“…Moreover, as we suggested above, we attribute the trench‐parallel flow to the edge effect by the eastward subducted South China Sea Plate. It is possible that the flow is wrapping around the slab edge as suggested to occur in other subduction zones like in Alaska (e.g., Venereau et al, ) and Ryukyu (Kuo et al, ) but which needs more observations to draw a conclusion. The interpreted 2‐D corner flow and deflected flow due to edge effect in this area are both displayed schematically in Figure .…”

“…The anisotropy resulting from the 2‐D corner flow or the slab‐entrained flow often exhibits trench‐normal fast directions (more precisely, parallel to the APM of subducting slabs). However, due to the influences from a number of factors such as trench migration, proximity to slab edges, slab morphology, existence of slab window, local thermal anomaly, and volatile distribution, the resultant flow pattern and anisotropic parameters exhibit a large variations worldwide (see a more detailed review by Long, ); specifically, small‐scale flow and 3‐D toroidal flow have been found in subduction zones (e.g., Faccenna & Becker, ; Venereau et al, ), and trench‐parallel flow also exists below the subducting slabs (e.g., Russo & Silver, ). In addition, slab‐slab interaction between two neighboring subducting plates (e.g., Király et al, ) also likely plays a role in complicating the flow fields in subduction zones.…”