Simulation of the 2022 Mw 6.6 Luding, China, earthquake by a stochastic finite-fault model with a nonstationary phase

Abstract

We adopted stochastic finite-fault simulations considering the nonstationary frequency characteristics for the 2022 Mw 6.6 Luding, China, earthquake at 16 strong-motion stations using the stochastic finite-fault model based on a dynamic corner frequency (EXSIM) code. The average simulated response spectra of earthquake ground motions were obtained and then compared with response spectra of 5% damped single degree of freedom system under recorded ground motion excitation to verify whether the finite-fault model that considers the nonstationary frequency characteristics can reproduce the critical characteristics of ground motions and improve the simulation accuracy. The high-frequency spectral decay is calculated by the real recordings and the stress drop is determined by the trial-and-error method. In addition, the local site amplifications are estimated from the horizontal-vertical spectral ratio (H/V). An improved EXSIM that considers the slip-related corner frequency is also used in this study. The results indicate that the nonstationary phase can significantly improve the simulation accuracy of EXSIM in the short period and has little effect on the long-period simulation results. However, the effect of the nonstationary phase on the simulation results obtained from the improved EXSIM is not significant. Regarding the peak ground acceleration (PGA), the nonstationary phase can produce waveform characteristics that are more similar to the real waveform, and the synthetic PGA is closer to the observed ones.