Long-term postseismic deformation at the eastern margin of the Japan Sea implied by temporal changes in the strain rates and background seismicity rates

Abstract:

Magnitude-8 class earthquakes have occurred on the back-arc side of the Tohoku region, especially on the eastern margin of the Japan Sea. For the eastern margin of the Japan Sea, there had been a lot of discussion about its tectonics. For example, Nakamura (1983) hypothesized that the eastern margin of the Japan Sea is a convergence boundary (and beginning trench) between the North American Plate and the Eurasian Plate. Okamura (2002) also hypothesized that the eastern margin of the Japan Sea is a strain concentration zone where a lot of faults had been formed within the plate. In this study, we analyze crustal deformation and seismicity using geophysical data from recent observation system, to investigate the recent tectonic activity in the Tohoku region.

In terms of crustal deformation analysis, we investigate temporal changes in geodetic strain of the Tohoku region using the F3 solution of GNSS data provided by Geospatial Information Authority of Japan. We estimate the dilatational strain rates and the maximum shear strain rates based on the method of Shen et al. (2015). This method allows us to treat the smoothing distance D as an optimal weighting variable. We find both the dilatation and the maximum shear rates had gradually converged into steady values until 2010 and increased after the 2011 Tohoku megathrust earthquake in the eastern margin of the Sea of Japan. In terms of seismicity analysis, we estimate stress loading rates within the eastern margin of the Japan Sea by applying the ETAS model (Ogata, 1988) to the earthquake catalogue provided by Japan Meteorological Society. These analyses imply that long-term postseismic deformation of two major earthquakes (the 1983 Nihonkai-Chubu earthquake and the 1993 southwest-off Hokkaido earthquake) had continued until around 2010 in the eastern margin of the Japan Sea.

We find the maximum shear strain rates and the background seismicity rate varied uncorrelatedly by comparing the results of these analyses. This may be due to differences between the azimuth of the maximum shear strain and the fault orientation.