4:30 PM - 4:45 PM
[SSS11-28] Source model of the 2024 Noto Peninsula earthquake using the fault planes based on the precisely relocated aftershocks
Keywords:The 2024 Noto Peninsula earthquake, Source inversion analysis, Empirical Green's function, Strong ground motion, Aftershock distribution
In the estimation of the source model for the January 1, 2024 Noto Peninsula earthquake (Mj 7.6), the location and geometry of the main shock source fault planes were determined on a preliminary basis from aftershock distribution information based on the JMA unified hypocenters (Shiba: 2024a, b). Subsequently, high-density aftershock observations in both land and sea areas around the source region were conducted by several research institutes, and highly accurate aftershock information was obtained (e.g., Takahashi, et al., 2024, Shinohara, et al., 2024). These aftershock distributions are determined to be systematically shallower than those reported immediately after the main shock occurrence, and the trend of the dip direction suggesting the main shock fault also shows some difference. In this study we reflect these systematic differences in aftershock distribution in the initial fault plane model for the main shock and re-estimate the slip distribution on it.
Based on the relocated aftershock hypocenters by Takahashi et al. (2024), we redefined the fault planes of the main shock in the land area. The lower limit of the fault plane based on the aftershock distribution was shallowened from about 20 km to about 13 km, and the dip angle of the fault including hypocenter was redefined from 50 to 35 degrees. The upper depth of the faults in the land area were uniformly fixed at 3 km, in accordance with the aftershock observation by Takahashi et al. (2024). For the fault plane of the foreshock, which occurred about 13 seconds before the main shock, the rupture is assumed to propagate on the same plane as the main shock, whereas it was set to another parallel fault in the previous study. Furthermore, the northwest-dipping fault plane off the northeast coast of the peninsula was relocated to be shallower with the lower limit at about 16 km, referring to Shinohara et al. (2024) by the seafloor seismic observations.
For the source inversion method, we used an optimization algorithm combining the empirical Green’s function method and the fast simulated annealing by Shiba and Irikura (2005), the same as in the previous report. Different small events were applied to each fault plane as the empirical Green’s function following the previous report, but only the focal depths were moved shallower according to the relocation of the fault plane model. As for the rupture scenario during the inversion procedure, the rupture is initiated at the origin time of the foreshock and propagates within a wider time delay including the origin time of the main shock, ensuring causality of rupture through all fault planes. In the analysis 25 stations were selected from K-NET, KiK-net and F-net in and around the source region on the condition that no significant soil nonlinearity effect was observed. Velocity waveform records of two horizontal components were used for the inversion.
The slip distribution model obtained from the inversion analysis is shown in the Figure. The estimated seismic moment is 2.07E20 Nm with Mw 7.5. The peak slip is 8.26 m. Compared to the previous report, the seismic moment is almost the same, but the peak slip was estimated to be slightly larger. This is due in part to the smaller fault area and width of the newly defined fault planes. There shows almost no slip on the southern part of the southeastern end of the fault plane sequence, which corresponds to the Ama-Misaki Offshore Fault Zone. This feature is also recognized in the previous result. On the other hand, the slip on the northwest dipping fault at the northern end became relatively large. The spatial distribution of slip in the central part corresponding to the segments from Off-Suzu to Off-Saruyama shows almost the same characteristics as in the previous report and is in good agreement with the distribution of crustal uplift obtained from the synthetic aperture radar interferometric analysis.
Based on the relocated aftershock hypocenters by Takahashi et al. (2024), we redefined the fault planes of the main shock in the land area. The lower limit of the fault plane based on the aftershock distribution was shallowened from about 20 km to about 13 km, and the dip angle of the fault including hypocenter was redefined from 50 to 35 degrees. The upper depth of the faults in the land area were uniformly fixed at 3 km, in accordance with the aftershock observation by Takahashi et al. (2024). For the fault plane of the foreshock, which occurred about 13 seconds before the main shock, the rupture is assumed to propagate on the same plane as the main shock, whereas it was set to another parallel fault in the previous study. Furthermore, the northwest-dipping fault plane off the northeast coast of the peninsula was relocated to be shallower with the lower limit at about 16 km, referring to Shinohara et al. (2024) by the seafloor seismic observations.
For the source inversion method, we used an optimization algorithm combining the empirical Green’s function method and the fast simulated annealing by Shiba and Irikura (2005), the same as in the previous report. Different small events were applied to each fault plane as the empirical Green’s function following the previous report, but only the focal depths were moved shallower according to the relocation of the fault plane model. As for the rupture scenario during the inversion procedure, the rupture is initiated at the origin time of the foreshock and propagates within a wider time delay including the origin time of the main shock, ensuring causality of rupture through all fault planes. In the analysis 25 stations were selected from K-NET, KiK-net and F-net in and around the source region on the condition that no significant soil nonlinearity effect was observed. Velocity waveform records of two horizontal components were used for the inversion.
The slip distribution model obtained from the inversion analysis is shown in the Figure. The estimated seismic moment is 2.07E20 Nm with Mw 7.5. The peak slip is 8.26 m. Compared to the previous report, the seismic moment is almost the same, but the peak slip was estimated to be slightly larger. This is due in part to the smaller fault area and width of the newly defined fault planes. There shows almost no slip on the southern part of the southeastern end of the fault plane sequence, which corresponds to the Ama-Misaki Offshore Fault Zone. This feature is also recognized in the previous result. On the other hand, the slip on the northwest dipping fault at the northern end became relatively large. The spatial distribution of slip in the central part corresponding to the segments from Off-Suzu to Off-Saruyama shows almost the same characteristics as in the previous report and is in good agreement with the distribution of crustal uplift obtained from the synthetic aperture radar interferometric analysis.