Systematic Evaluation and Development of Long-Period Coda Moment Magnitude (Mw0) for Large Earthquakes

Abstract The coda moment magnitude (Mwo) has recently been proposed based on the long-period and long-lasting coda energy of large earthquakes. In this study, we perform a systematic evaluation of the Mwo method on its robustness and applicability using a large dataset of 132 large earthquakes from January 2000 to March 2023. The results demonstrate that the Mwo is in general agreement with the moment magnitude (Mw) from the Global Centroid Moment Tensor, with the former being slightly larger. For relatively smaller events (Mw<8.3), the residual between the Mwo and Mw increases with decreasing magnitude. Further synthetic tests indicate that the background noise (or the signal-to-noise-ratio) is the primary factor for the deviation of Mwo for smaller events. After comparing different approaches, we propose an empirical relationship to correct the Mwo for events with Mw<8.3. Examination of other potential influences reveals a minimal impact, showing the robustness of the corrected Mwo and the enhanced Mwo method retains the advantages of potentially capturing a broader portion of the seismic energy release. Its stability has been shown to be largely insensitive to variations in epicenter location and the number of stations and to be only mildly sensitive to changes in source focal mechanisms and depth of hypocenter. Furthermore, the improved method is applicable to earthquakes with Mw≥7.3 when a sufficient number of high-quality coda measurements is available, with the recommended and more robust operational range being Mw≥7.5, thereby extending its capability for accurate magnitude determination of large earthquakes.