On January 1, 2024, a Mw7.6 earthquake occurred on the Noto Peninsula in Japan, with a focal depth of 9.6 kilometers. To validate the effectiveness of using a finite fault hybrid slip model to simulate strong ground motion events, this study utilized strong motion records from the KiK-net station network of the National Research Institute for Earth Science and Disaster Resilience (NIED) in Japan. Initially, 29 fundamental stations were selected, and 13 of these stations were chosen for further study using the finite fault method. Constrained by the geometric parameters of the initial model from the strong motion inversion, a hybrid slip distribution model (HSM) was established based on the stochastic finite fault model and semi-empirical relationships between source parameters. This model accurately reflects the spatial and temporal characteristics of the source rupture, and its results are comparable to those obtained by the United States Geological Survey (USGS) using GNSS and other long-period effect data. Additionally, using the dynamic corner frequency-based stochastic ground motion synthesis method (EXSIM), a comparative study was conducted on the pseudo-acceleration response spectra (PSA) of the hybrid slip model, the strong motion inversion model, and the source rupture model provided by USGS at the selected stations. These were then compared with the response spectra of actual station records. It was found that all three models showed good fitting results, and in the short-period range, the simulated values of the HSM were closer to the actual records. Finally, the relative error mean and variance between the HSM and actual records’ PSA at different frequencies were calculated for 29 stations. It reveals that they still had good fitting results. Overall, the findings demonstrate that even without relying on extensive station data, the hybrid slip model efficiently and accurately simulated high-frequency strong ground motions near the fault with a damping ratio of 5%, verifying the reliability and applicability of this method.
Read full abstract