Abstract

Shear wave splitting parameters of local and teleseismic S waves from intermediate and deep earthquakes in the southern Kurile and Japan subduction zones are combined with splitting parameters obtained from SKS and SKKS waves to determine depth variation in shear wave splitting both above and below the earthquake. Local S wave splitting results measured at station MAJO (Matsushiro, Japan) show fast directions, from NNE to NE, parallel to the extension directions measured from Quaternary fault and geodetic data. The inferred finite strain field from shear wave splitting is consistent with the closing of the Sea of Japan. At YSS (Yuzhno Sakhalinsk, Russia) the shear wave splitting parameters obtained from local S waves are approximately oriented N‐S and appear to be controlled by deformation of the upper plate rather than the subducted slab. SKS phases recorded at YSS, located on the southern tip of Sakhalin Island, show an approximately N‐S oriented fast direction and lag time of 1.3±0.3 s. Local S phases recorded at station YSS yield fast directions similar to the SKS results and the magnitude of splitting varies systematically with depth. These results combine to indicate that very little splitting occurs in the asthenosphere below the southern Kurile slab at about 450 km depth. S waves from deep earthquakes beneath MDJ (Mudanjiang, Heilongjiang Province, China) show ∼1.0 s of lag time, while SKKS results show ∼1.6 s of lag time. One SKS phase, however, shows similar lag times as those observed by the split local S waves. This difference in shear wave splitting lag times indicates significant variation in azimuthal anisotropy beneath a minimum depth of 350 km. This inference is consistent with the source‐side splitting of ∼0.8 s lag time observed from deep teleseismic S waves which traverse the lower parts of the upper mantle and the upper mantle/lower mantle boundary. A plausible explanation for the presence of deep seismic anisotropy is that shear wave splitting is occurring in the metastable olivine in the flattened and broadened southern Japan slab. Another explanation for these observations is the presence of an anisotropic layer composed primarily of highly anisotropic β‐spinel at the base of the 410‐km discontinuity.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call