9:45 AM - 10:00 AM
[SSS07-16] Investigation of the Faulting Process of Tsunami Earthquake Based on Dynamic Rupture Simulations
Keywords:Tsumani Earthquake, Dynamic Rupture Simulation, Accretionary Wedge, Rupture Velocity
One of the most distinct features of the faulting process for the 2011 Off the Pacific Coast of Tohoku earthquake (the Tohoku-oki Earthquake, Mw 9.0) is depth-dependent complex radiation features. The deep parts of the faults (near the coast) hosted smaller slips with strong radiation of short-period seismic waves, whereas shallow parts of the fault (near the trench) hosted large slips and weak long-period seismic wave radiation, respectively. To understand the mechanism to generate such complex features, many studies have been conducted intensely after the Tohoku-Oki earthquake. There are still some issues, especially in the fault behavior on the shallow portions of the faults. As such shallow fault behavior can lead to the issue of tsunami generation that might bring devastating damage, understanding such a mechanism is important. The 1896 Sanriku earthquake is recognized as a typical “tsunami earthquake,” producing a larger size of tsunami relative to its magnitude based on the seismic waves and produced weak ground motions (Kanamori, 1972).
To understand such source mechanisms on the shallow portions of the faults, Tsuda and Shibazaki (2021, AGU) have conducted a physics-based dynamic rupture simulation. The 3D spectral element method (e.g., Galvez et al. 2014) with slip-weakening friction law (Ida, 1972) has been applied. The simple characterized source model with one rectangle asperity on the shallow part of the fault with a gentle dipping angle (10°). The asperity’s length and width are set, referring to the previous studies (e.g., Satake et al., 2017) and the soft material corresponding to the tip of accretionary wedge was included. This model showed the rupture propagated diagonally to the surface from nucleation zone and then propagated laterally along the surface toward the center. Even the overall rupture velocity was compatible to the feature of Tsunami earthquake, the mechanism to generate such lateral propagation was difficult to compare with the observation of the Tsunami earthquake so far. Thus, in this study, we have adjusted the values of critical distance (Dc) on the shallow part of the fault to update the model from our previous one. Though this updated model has generated direct propagation of rupture to the surface, the main rupture directly from nucleation zone still propagates slower than 1.5 km/s (sub-shear) even in the shallow part of the fault (close to the surface). Also, most of the radiated seismic energy are trapped within the accretionary wedge and leading to the weak ground motion, because the material boundary prevented the rupture from propagating to the deeper part. These results suggested that the material boundary played an important role to specify the features of faulting process of tsunami earthquake. We further set the area with dynamic weakening close to the fault, referring to the very low frictional coefficient (e.g., Sawai et al., 2017). This setting leading to additional stress drop such area not only generated large slip compared to the original model, but also showed the rupture propagation on reverse direction from surface. This reverse propagation initiated the rupture over the material boundary even with zero stress drop and propagated to the deeper part. These features might bring physical insight to consider the rupture scenario for the Tsunamic earthquake as well as “Tsunami” genic earthquake such as the Tohoku-Oki earthquake.
To understand such source mechanisms on the shallow portions of the faults, Tsuda and Shibazaki (2021, AGU) have conducted a physics-based dynamic rupture simulation. The 3D spectral element method (e.g., Galvez et al. 2014) with slip-weakening friction law (Ida, 1972) has been applied. The simple characterized source model with one rectangle asperity on the shallow part of the fault with a gentle dipping angle (10°). The asperity’s length and width are set, referring to the previous studies (e.g., Satake et al., 2017) and the soft material corresponding to the tip of accretionary wedge was included. This model showed the rupture propagated diagonally to the surface from nucleation zone and then propagated laterally along the surface toward the center. Even the overall rupture velocity was compatible to the feature of Tsunami earthquake, the mechanism to generate such lateral propagation was difficult to compare with the observation of the Tsunami earthquake so far. Thus, in this study, we have adjusted the values of critical distance (Dc) on the shallow part of the fault to update the model from our previous one. Though this updated model has generated direct propagation of rupture to the surface, the main rupture directly from nucleation zone still propagates slower than 1.5 km/s (sub-shear) even in the shallow part of the fault (close to the surface). Also, most of the radiated seismic energy are trapped within the accretionary wedge and leading to the weak ground motion, because the material boundary prevented the rupture from propagating to the deeper part. These results suggested that the material boundary played an important role to specify the features of faulting process of tsunami earthquake. We further set the area with dynamic weakening close to the fault, referring to the very low frictional coefficient (e.g., Sawai et al., 2017). This setting leading to additional stress drop such area not only generated large slip compared to the original model, but also showed the rupture propagation on reverse direction from surface. This reverse propagation initiated the rupture over the material boundary even with zero stress drop and propagated to the deeper part. These features might bring physical insight to consider the rupture scenario for the Tsunamic earthquake as well as “Tsunami” genic earthquake such as the Tohoku-Oki earthquake.