[S-SS06] Statistical seismology and underlying physical processes

[SSS06-P07] Inter-plate aseismic slip determined from teleseismic similar earthquakes recorded by the Japanese seismic network

*Toshihiro Igarashi¹, Aitaro Kato¹ (1.Earthquake Research Institute, University of Tokyo)

Similar earthquakes that occur in approximately the same location at different times have the potential to reveal the spatio-temporal changes in aseismic slip rate along plate boundaries. In this study, we identified similar earthquakes with moderate magnitudes in worldwide by calculating the cross-correlations of bandpass-filtered seismograms recorded by the Japanese dense seismic network. As a result, many similar earthquakes were detected along subduction plate boundaries, including the Japan–Kuril–Kamchatka–Aleutian, the Andaman–Sumatra–Java, and the Solomon–Vanuatu–Tonga region. Based on the spatio-temporal evolutions of similar earthquakes, we investigated the spatial distribution and temporal characteristics of inter-plate aseismic slip around the Pacific and Indian oceans. The estimated slip rates in several areas are higher than the relative velocity of the two converging plates. These areas correspond to regions where similar earthquakes with relatively short recurrence intervals were identified following megathrust earthquakes. We infer that these similar earthquakes were induced by postseismic slip along the plate boundary. In other regions, similar earthquake sequences have a constant recurrence interval of several years to several decades. Estimated slip rates are equal to or less than the relative velocities of plate motion, including the back-arc spreading rate. These observations indicate that the plate boundary faults in these regions are strongly locked. These results suggest that spatio-temporal changes in inter-plate aseismic slip around the world can be estimated by analyzing similar earthquake sequences. We believe that additional similar earthquakes will be detected by extending the analysis periods in areas where similar earthquakes were not identified in this study, resulting in slip rates with higher spatio-temporal resolution. A time series of stacked slip rates against the elapsed time since megathrust earthquakes initially increased rapidly following the rupture of each earthquake and then gradually decayed over periods of ~10 years, which correlates with after-slip progressing around the source areas. Around 30 years after megathrust earthquakes, the slip rate begins to increase with approaching the next megathrust rupture. This gradual increase in slip rate during the middle and later stages of the earthquake cycle may be due to an increase in stress levels that accumulate during tectonic loading. A similar earthquake catalog provides a valuable framework for understanding the long-term evolution of slip-rate during megathrust earthquake cycles.