3:00 PM - 3:15 PM
[SCG45-37] Slow slip events and complex slip dynamics in uniform fault: effect of evolution laws
Keywords:RSF law, Evolution law, Slow slip
Various laboratory experiments have shown that a very slow creep-like slip called “earthquake nucleation process” appears before a fast fault slip occurs (For example, Latour et al., 2013). However, there is no case that the nucleation process has been clearly observed in actual earthquakes.
Is nucleation observed in the laboratory not observable in natural faults? The answer to this question is still unclear, and many theoretical studies have been conducted to estimate whether the nucleation scale is realistically observable or not. For example, the scale has been estimated from the theory of linear stability (Rice and Ruina, 1983), the spring slider model (Dietrich, 1992), and calculations using fracture mechanics (Rubin and Ampuero, 2005). However, these theoretical predictions are not entirely correct, and it has been pointed out that the nucleation process is in fact extremely complex (Fang et al. 2010), and a clear understanding of the process has not yet been achieved.
In this study, we revisit this problem from the viewpoint of how the characteristics of the friction law affect the nucleation process. As in other studies, we solve a quasi-static one-dimensional semi-infinite elastic fault model using the boundary integral equation method (BIEM). For the friction law, we use the rate-state-dependent friction law. Especially for the evolution law, we use two new laws which are expected to be suitable for the quasi-static case (Mizushima and Hatano, 2022). These laws are improvements to the aging law, one is changed to the power type, and the other introduces a new parameter.
The results show that the use of these evolution laws leads to a peculiar behavior in that the slip does not accelerate up to the seismic slip rate even in the velocity weakening cases. These results are significantly different from the case where the aging law is used. Also, they suggest that the possibility of earthquake occurrence cannot be determined solely from the velocity dependence of friction in a steady state.
We also investigate the behavior of slip propagation and seismic moments. The results show that when the friction parameters are near the instability point, the rate of propagation is much slower than in the case of aging law. In addition, the seismic moment increases like a slow slip for each case. These results are consistent with the fact that slow earthquakes occur around the stability of slip switches.
Is nucleation observed in the laboratory not observable in natural faults? The answer to this question is still unclear, and many theoretical studies have been conducted to estimate whether the nucleation scale is realistically observable or not. For example, the scale has been estimated from the theory of linear stability (Rice and Ruina, 1983), the spring slider model (Dietrich, 1992), and calculations using fracture mechanics (Rubin and Ampuero, 2005). However, these theoretical predictions are not entirely correct, and it has been pointed out that the nucleation process is in fact extremely complex (Fang et al. 2010), and a clear understanding of the process has not yet been achieved.
In this study, we revisit this problem from the viewpoint of how the characteristics of the friction law affect the nucleation process. As in other studies, we solve a quasi-static one-dimensional semi-infinite elastic fault model using the boundary integral equation method (BIEM). For the friction law, we use the rate-state-dependent friction law. Especially for the evolution law, we use two new laws which are expected to be suitable for the quasi-static case (Mizushima and Hatano, 2022). These laws are improvements to the aging law, one is changed to the power type, and the other introduces a new parameter.
The results show that the use of these evolution laws leads to a peculiar behavior in that the slip does not accelerate up to the seismic slip rate even in the velocity weakening cases. These results are significantly different from the case where the aging law is used. Also, they suggest that the possibility of earthquake occurrence cannot be determined solely from the velocity dependence of friction in a steady state.
We also investigate the behavior of slip propagation and seismic moments. The results show that when the friction parameters are near the instability point, the rate of propagation is much slower than in the case of aging law. In addition, the seismic moment increases like a slow slip for each case. These results are consistent with the fact that slow earthquakes occur around the stability of slip switches.