Broadband Seismic Source Characterization Based on Composite Source Time Functions: A Case Study of the 2022 Luding Earthquake

Abstract Obtaining broadband ground motion simulations consistent with observed records is essential for seismic modeling, and the key to achieving this lies in a physically reasonable source representation. Only a scientifically constrained broadband source can effectively generate broadband ground motions. This study proposes a broadband source construction method based on composite source time functions within the theoretical Green's function framework. Multiple individual source time functions were combined into a composite function, and the corresponding source parameters were inverted using a simulated annealing algorithm, thereby enabling broadband source characterization in deterministic simulations and achieving broadband ground motion modeling in the 0–40 Hz frequency band. The method was applied to the 2022 Luding Ms 6.8 earthquake. Considering topographic amplification effects on three ground‐motion components, response spectra from ten strong‐motion stations at near‐, intermediate‐, and far‐field distances were used as constraints to invert for the optimal broadband source parameters. The results indicate that the simulated spatial distribution of peak ground acceleration is generally consistent with the seismic intensity survey released by China's Ministry of Emergency Management (2022, https://www.mem.gov.cn/xw/yjglbgzdt/202209/t20220911_422190.shtml ), thereby validating the effectiveness of the proposed approach. Furthermore, quantitative relationships among source time function parameters were established, enabling rapid determination of broadband source parameters and providing a useful reference for similar studies. In summary, the broadband source characterization method developed in this study not only facilitates deterministic broadband ground motion simulation but also incorporates topographic effects and medium properties, offering technical support for broadband ground motion modeling under complex surface conditions.