Affiliations:
College of Resources, Sichuan Agricultural University, Chengdu 611130, China
Earthquake prediction remains a major challenge in seismology, and no reliable method has yet been established. Previous studies suggest that large earthquakes (triggered earthquakes, TDEs) are often preceded by smaller events (triggering earthquakes, TGEs), but their quantitative relationship is unclear. In this study, we developed a prediction method based on four quadratic regression equations linking magnitude and distance between TGEs and TDEs, and three linear regression equations linking magnitude and time interval. The method was tested on 87 large earthquakes (M ≥ 6.6) from the past 100 years, including the 2023 Turkey (M7.8) and 2025 Myanmar (M7.7) events. Results show that 89% of epicenters and 63% of occurrence times were successfully predicted. For earthquakes of M ≥ 9.0 and those causing major casualties, the model achieved an accuracy within 100 km and 3 days, outperforming most existing approaches. Based on these results, we also propose a prediction for a future earthquake in Japan.
Earthquake prediction, Triggering earthquakes, Triggered earthquakes, Regression model, Seismology
https://doi.org/10.21833/ijaas.2026.03.003
Fu, Y. F., Zhang, Z. W., Yang, X. Y., & Yuan, S. (2026). Prediction of large-scale earthquakes using precursor earthquakes: A regression study. International Journal of Advanced and Applied Sciences, 13(3), 21–32. https://doi.org/10.21833/ijaas.2026.03.003