[S-CG55] Dynamics in mobile belts

[SCG55-P07] Spatial variation in the origin of S-wave polarization anisotropy in the upper crust of the Tohoku region, NE Japan.

*Ryotaro Fujimura¹, Tomomi Okada¹, Tatsuya Mizuta¹, Savage Martha², Ryota Takagi¹, Keisuke Yoshida¹, Shinichi Sakai³, Kei Katsumata⁴, Mako Ohzono^4,3, Masahiro Kosuga⁵, Takuto Maeda⁵, Yoshiko Yamanaka⁶, Hiroshi Katao⁷, Takeshi Matsushima⁸, Hiroshi Yakiwara⁹, Shuutoku Kimura¹, Satoshi Hirahara¹, Toshio Kono¹, Toru Matsuzawa¹, Ayaka Tagami¹ (1.Research Center for Prediction of Earthquakes and Volcanic Eruptions, Graduate School of Science, Tohoku University, 2.Victoria University of Wellington, 3.ERI, University of Tokyo, 4.Hokkaido University, 5.Hirosaki University, 6.Nagoya University, 7.DPRI, Kyoto University, 8.Kyushu University, 9.Kagoshima University)

Keywords:S-wave splitting, the S-wave polarization anisotropy

When seismic waves propagate in an anisotropic elastic medium, the shear wave splits into two S-waves with orthogonal oscillation directions and propagates at different speeds (S-wave splitting). In general, the direction of oscillation of the fast waves originates from the orientation of the maximum horizontal compressive stress axis (SHmax) and the lattice of minerals. In the upper crust, the anisotropy is generally observed in the direction parallel to SHmax. However, anisotropy in directions different from the stress field has also been observed in areas where faults and volcanic-related crustal structures exist.
In Mizuta (2021), anisotropy analysis using S-wave splitting was widely performed throughout the Tohoku region, especially examining changes in the direction and amplitude of anisotropy before and after the Tohoku-Oki earthquake. For this analysis, they used routine and temporary seismic data from Dec. 2008 to Apr. 2016 and applied the Multiple Filter Automatic Splitting Technique (MFAST; Savage et al., 2010) to determine the parameters of S-wave splitting. In this study, I used the anisotropic direction data obtained at Mizuta (Tohoku Univ. Master Thesis, 2021) in the upper crust and after the Tohoku-Oki earthquake to examine the distribution of characteristic anisotropy directions and to estimate the cause of the anisotropy by comparing it with the regional stress field. We used the P-axis of the focal mechanism estimated for the Tohoku region in Okada et al. (2022) as the reference for the stress field.I examined the distribution of anisotropy parallel to the P-axis direction and anisotropy in different directions by local comparison of the anisotropy direction and stress field. In the Oritsume Fault near the Aomori-Iwate border, I confirmed anisotropy of the fault parallel, which is different from the P-axis direction. This anisotropy was strongly detected in seismic waves that traveled along a ray path that passed through the fault. It suggests that a fault causes the anisotropy. Near the Fukushima-Yamagata border, we detected anisotropic heterogeneity at the location of the Otoge caldera and anisotropy in a direction different from the P-axis near the Nekoma volcano. Studies in other regions (Honda et al., 2022; Johnson et al., 2011) have confirmed anisotropic heterogeneity in volcanic zones, and the present results are consistent.
In other studies, anisotropic heterogeneity is attributed to a combination of factors, including stress fields caused by magma reservoirs and faults, and its cause is speculated by the estimation of the spatial distribution of anisotropy. In the future, I expect to obtain more detailed spatial distribution and estimation of the origin in this research area with the methods.