2:00 PM - 2:15 PM
[SSS08-11] Spatial correlation of the rate of the shear strain energy density change and the number of declustered crustal earthquakes in Japan
The spatial correlation of geodetically estimated strain rate and the background seismicity rate has been discussed in several areas in the world (e.g., Kreemer and Young, 2022). In Japan, however, there are azimuthal differences between the geodetically estimated maximum contraction strain rate and the seismologically estimated maximum compression stress (Kosugi and Mitsui. 2022). Thus, the increase in strain may not necessarily represent stress accumulation. In this study, we compare the number of background events for the crustal seismicity in Japan with the rate of the shear strain energy density change (Saito et al., 2018) which considers both the background stress field and the current crustal deformation.
The earthquake data were taken from the Japan Meteorological Agency (JMA) unified catalog. We used Mj > 1.9 shallow (< 30 km depth) crustal earthquakes. We applied the nearest neighbor distance approach (Zaliapin & Ben-Zion, 2020) to decluster the catalog during 2002–2010. For the strain rate estimation, we used GNSS displacement data archived at the Geospatial Information Authority of Japan (GSI), the Japan Coast Guard (JCG), Kyoto University, the International GNSS Service (IGS), and UNAVCO. The daily coordinates of GNSS stations during 2002–2010 were estimated using the Gipsy X Ver. 1.4 software (Bertiger et al., 2020). We estimated secular velocities with seasonal and significant step components using the DISSTANS python package (Köhne et al., 2023). We estimated the strain rate distribution using the basis function expansion (Okazaki et al., 2021). We calculated the stress change using the estimated strain rate tensor assuming the state of plane stress and evaluated the rate of the shear strain energy density change considering the background stress field estimated by Uchide et al. (2022) at 0.2-degree intervals.
The number of background earthquakes is weakly positively correlated with the rate of the shear strain energy density change and this correlation is statistically significant. Especially in SW Japan, this correlation is significantly better than the correlation between the strain rate and seismicity. The number of background earthquakes is < 20% of the number of earthquakes expected from the geodetic moment rate calculated using the rate of the shear strain energy density change, suggesting that the geodetic strain rate includes much amount of inelastic and/or aseismic strain rate. In addition, the logarithm of the number of background earthquakes correlates better with the rate of shear strain energy density change than a linear correlation. These results might indicate that a more complex model should be considered, but the possibility that better correlations might be apparent should also be considered.
Acknowledgements: The earthquake catalog was obtained from JMA, and GNSS data was obtained from GJI, JCG, IGS, and UNAVCO. Dr. Okazaki provided the code for the estimation of strain rate distribution (Okazaki et al., 2021). Dr. Uchide provided the stress map data at a 0.2-degree mesh (Uchide et al., 2022). This study was supported by JSPS KAKENHI Grant Numbers JP22KJ1770.
The earthquake data were taken from the Japan Meteorological Agency (JMA) unified catalog. We used Mj > 1.9 shallow (< 30 km depth) crustal earthquakes. We applied the nearest neighbor distance approach (Zaliapin & Ben-Zion, 2020) to decluster the catalog during 2002–2010. For the strain rate estimation, we used GNSS displacement data archived at the Geospatial Information Authority of Japan (GSI), the Japan Coast Guard (JCG), Kyoto University, the International GNSS Service (IGS), and UNAVCO. The daily coordinates of GNSS stations during 2002–2010 were estimated using the Gipsy X Ver. 1.4 software (Bertiger et al., 2020). We estimated secular velocities with seasonal and significant step components using the DISSTANS python package (Köhne et al., 2023). We estimated the strain rate distribution using the basis function expansion (Okazaki et al., 2021). We calculated the stress change using the estimated strain rate tensor assuming the state of plane stress and evaluated the rate of the shear strain energy density change considering the background stress field estimated by Uchide et al. (2022) at 0.2-degree intervals.
The number of background earthquakes is weakly positively correlated with the rate of the shear strain energy density change and this correlation is statistically significant. Especially in SW Japan, this correlation is significantly better than the correlation between the strain rate and seismicity. The number of background earthquakes is < 20% of the number of earthquakes expected from the geodetic moment rate calculated using the rate of the shear strain energy density change, suggesting that the geodetic strain rate includes much amount of inelastic and/or aseismic strain rate. In addition, the logarithm of the number of background earthquakes correlates better with the rate of shear strain energy density change than a linear correlation. These results might indicate that a more complex model should be considered, but the possibility that better correlations might be apparent should also be considered.
Acknowledgements: The earthquake catalog was obtained from JMA, and GNSS data was obtained from GJI, JCG, IGS, and UNAVCO. Dr. Okazaki provided the code for the estimation of strain rate distribution (Okazaki et al., 2021). Dr. Uchide provided the stress map data at a 0.2-degree mesh (Uchide et al., 2022). This study was supported by JSPS KAKENHI Grant Numbers JP22KJ1770.