10:45 AM - 12:15 PM
[SCG45-P41] Effects of spray faults on dynamic rupture of the 2016 Mw7.0 Kumamoto, Japan, earthquake: a simulation study
Keywords:2016 Mw7.0 Kumamoto earthquake, 3-D dynamic rupture simulation, The Aso caldera, Spray fault
The rupture in the 2016 Mw7.0 Kumamoto, Japan, earthquake (mainshock) started at depth on the fault plane along the Hinagu fault, propagated northeastward along the Futagawa fault, and stopped in the Aso caldera (Asano and Iwata 2016; Yagi et al. 2016). The distinguishing subsurface structure of the Aso caldera may have affected the rupture process of the mainshock. Kobayashi et al. (2023, Submitted) estimated the complex distribution of multiple spray faults branching at the eastern end of the Futagawa Fault in the Aso Caldera and amount of fault slip based on the surface displacement estimated by InSAR, and conducted a dynamic rupture simulation considering two major spray faults among them. In this study, we perform dynamic rupture simulations considering small spray faults, which were excluded in Kobayashi et al. (2023). Furthermore, we compare the results of the slip inversion analysis with the simulation results to examine the reproducibility of the slip distribution and to investigate the effect of the branching geometry of the fault on the slip distribution.
We use FDP-BIEM (Ando et al., 2017) as the simulation method. As the fault geometries, we consider F6, which corresponds to the Futagawa fault, and five spray fault planes F1-F5 in the Aso caldera, among the non-planar fault geometries obtained by Kobayashi et al. (2023). We use the linear slip weakening friction law (Ida, 1972) as the friction law, and the regional stress field is based on Matsumoto et al. (2015). In the Aso caldera, two initial stress field models (stress models I and II) are prepared following Kobayashi et al. (2023). In Model I, the initial stress field is continuous inside and outside the Aso caldera, and in Model II, the stress state inside the Aso caldera is distinctively lower than that outside. Rupture process is initiated at the depth of the western margin of F6, and the time evolution is calculated until the complete cessation of slip on the fault planes to obtain the spatial distribution of the final slip distribution.
The final slip distribution on F1 obtained by the simulation shows that the slip amounts of both stress models I and II are particularly large at the shallow part of the western margin of F1 (near the junction with the main body of the Futagawa Fault). In addition, focusing on the intersection of F1 and F5, which is conjugate to F1, the west side of the intersection line showed a larger slip than the east side. A similar spatial pattern was also observed in the final slip distribution obtained by Kobayashi et al. (2023). In the shear stress variation during the simulation, the spatial pattern of the shear stress variation on F1 during the rupture propagation on F5 is different between the east side and the west side of the intersection with F5. It can be considered that the rupture propagation on F5 created an asymmetric stress field in the surrounding area, resulting in a different mode of rupture propagation on F1 between the east and west sides of the intersection line and therefore the aforementioned spatial pattern of the final slip distribution. This indicates that the spatial distribution of slip on the fault plane is determined by the dynamic interaction with the surrounding spray faults, and that InSAR data with high spatial resolution may be able to resolve the peculiar spatial distribution caused by such a process.
We use FDP-BIEM (Ando et al., 2017) as the simulation method. As the fault geometries, we consider F6, which corresponds to the Futagawa fault, and five spray fault planes F1-F5 in the Aso caldera, among the non-planar fault geometries obtained by Kobayashi et al. (2023). We use the linear slip weakening friction law (Ida, 1972) as the friction law, and the regional stress field is based on Matsumoto et al. (2015). In the Aso caldera, two initial stress field models (stress models I and II) are prepared following Kobayashi et al. (2023). In Model I, the initial stress field is continuous inside and outside the Aso caldera, and in Model II, the stress state inside the Aso caldera is distinctively lower than that outside. Rupture process is initiated at the depth of the western margin of F6, and the time evolution is calculated until the complete cessation of slip on the fault planes to obtain the spatial distribution of the final slip distribution.
The final slip distribution on F1 obtained by the simulation shows that the slip amounts of both stress models I and II are particularly large at the shallow part of the western margin of F1 (near the junction with the main body of the Futagawa Fault). In addition, focusing on the intersection of F1 and F5, which is conjugate to F1, the west side of the intersection line showed a larger slip than the east side. A similar spatial pattern was also observed in the final slip distribution obtained by Kobayashi et al. (2023). In the shear stress variation during the simulation, the spatial pattern of the shear stress variation on F1 during the rupture propagation on F5 is different between the east side and the west side of the intersection with F5. It can be considered that the rupture propagation on F5 created an asymmetric stress field in the surrounding area, resulting in a different mode of rupture propagation on F1 between the east and west sides of the intersection line and therefore the aforementioned spatial pattern of the final slip distribution. This indicates that the spatial distribution of slip on the fault plane is determined by the dynamic interaction with the surrounding spray faults, and that InSAR data with high spatial resolution may be able to resolve the peculiar spatial distribution caused by such a process.