A study of the mantle flow field and lithospheric deformation beneath the Kuril-Kamchatka subduction zone using seismic anisotropy

We investigate the flow field and deformation in the mantle wedge and subslab mantle beneath the Kuril-Kamchatka subduction zone using seismological data from a recently deployed seismic network around the Klyuchevskoy Volcanic Group (KVG) complemented by data from previous temporary deployments and permanent stations to reach a total number of 145 seismic stations covering a region defined in the geographic coordinates 150°-167°E and 50°-61°N.We perform splitting analysis of both local and core-refracted (SKS) shear waves to study mantle seismic anisotropy as a proxy for the pattern of the mantle flow field and deformation. Anisotropy in the mantle wedge is studied by shear splitting analysis (SWS) of waveform data from local mantle events that occurred along the subducting slab (Wadati-Benioff-Zone) and in the mantle wedge. Crustal anisotropy is also studied by SWS analysis of crustal events. The combined data set (SKS and local) allows us to discriminate the source of mantle anisotropy (sub-slab, mantle wedge, or crust). Shear-wave splitting measurements from the local shear waves give small delay times independent of the depth of the events suggesting that the mantle wedge is characterized by a weak anisotropic fabric. The fast directions of mantle wedge anisotropy are predominantly parallel to the strike of the slab indicating either a trench-parallel flow or B-type seismic anisotropy in the mantle wedge. The relatively small delay times from local shear waves suggest that SKS waves are less affected by potential anisotropy in the mantle wedge and that the results of the SKS-splitting analysis are mainly representative of the sub-slab anisotropy. Our SKS-splitting measurements indicate a trench-normal mantle flow beneath the eastern edge of the Kamchatka peninsula that converts to a more complex pattern beneath the KVG region. We argue that this pattern of fast polarization direction suggests the rotational mantle flow beneath the slab that may be related to the change in slab geometry at the junction between the Kuril-Kamchatka and Aleutian arcs. The observation of relatively strong sub-slab anisotropy against weak mantle-wedge anisotropy suggests that slab termination causes some disturbance in mantle flow; however, no significant component of an around-slab flow occurs in the mantle wedge.