T51B-4621:
Estimation of seismic velocity in the subducting crust of the Pacific slab beneath Hokkaido, northern Japan by using guided waves

Friday, 19 December 2014
Takahiro Shiina1, Junichi Nakajima1, Genti Toyokuni1, Saeko Kita2 and Toru Matsuzawa1, (1)Tohoku University, Sendai, Japan, (2)National Research Institute for Earth Science, Tsukuba, Japan
Abstract:
A subducting crust contains a large amount of water as a form of hydrous minerals (e.g., Hacker et al., 2003), and the crust plays important roles for water transportation and seismogenesis in subduction zones at intermediate depths (e.g., Kirby et al., 1996; Iwamori, 2007). Therefore, the investigation of seismic structure in the crust is important to understand ongoing physical processes with subduction of oceanic lithosphere.

A guided wave which propagates in the subducting crust is recorded in seismograms at Hokkaido, northern Japan (Shiina et al., 2014). Here, we estimated P- and S-wave velocity in the crust with guided waves, and obtained P-wave velocity of 6.6-7.3 km/s and S-wave velocity of 3.6-4.2 km/s at depths of 50-90 km. Moreover, Vp/Vs ratio in the crust is calculated to be 1.80-1.85 in that depth range.

The obtained P-wave velocity about 6.6km/s at depths of 50-70 km is consistent with those estimated in Tohoku, northeast Japan (Shiina et al., 2013), and this the P-wave velocity is lower than those expected from models of subducting crustal compositions, such as metamorphosed MORB model (Hacker et al., 2003). In contrast, at greater depths (>80 km), the P-wave velocity marks higher velocity than the case of NE Japan and the velocity is roughly comparable to those of the MORB model. The obtained S-wave velocity distribution also shows characteristics similar to P waves.

This regional variation may be caused by a small variation in thermal regime of the Pacific slab beneath the two regions as a result of the normal subduction in Tohoku and oblique subduction in Hokkaido. In addition, the effect of seismic anisotropy in the subducting crust would not be ruled out because rays used in the analysis in Hokkaido propagate mostly in the trench-parallel direction, while those in Tohoku are sufficiently criss-crossed.