The spatiotemporal distribution and magnitude of seismicity collected over decades are crucial for understanding the stress interactions underlying large earthquakes.In this study,machine learning(ML)explainers identi...The spatiotemporal distribution and magnitude of seismicity collected over decades are crucial for understanding the stress interactions underlying large earthquakes.In this study,machine learning(ML)explainers identify and rank the features that distinguish Large Earthquake Occurrence(LEO)from non-LEO spatiotemporal windows.Seventy-eight statistics related to time,latitude,longitude,depth,and magnitude were extracted from the earthquake catalog(Global Centroid Moment Tensor)to produce 202,706 spatiotemporally discretized windows.ML explainers trained on these windows revealed the maximum magnitude(Mmax)as the most influential feature.Classification performance improved when the maximum inter-event time,the average interevent time,and the minimum ratio of focal depth to magnitude were jointly trained with Mmax.The top five features showed weak-to-moderate correlations,providing complementary information to the ML explainers.Our explainable ML framework can be extended to different earthquake catalogs,including those with focal mechanisms and smallmagnitude events.展开更多
基金supported by the National Research Foundation of Korea(RS-2025-02293161,NRF-2022R1A2C1009742,and No.2019R1A6A1A03033167)awarded to B.So.J.Jang is supported by the Ministry of the Interior and Safety,as the Human Resource Development Project in Disaster Managementsupported by Korea Institute of Marine Science&Technology Promotion(KIMST)funded by the Ministry of Oceans and Fisheries(RS-2023-00254680 and RS-2023-00259686).
文摘The spatiotemporal distribution and magnitude of seismicity collected over decades are crucial for understanding the stress interactions underlying large earthquakes.In this study,machine learning(ML)explainers identify and rank the features that distinguish Large Earthquake Occurrence(LEO)from non-LEO spatiotemporal windows.Seventy-eight statistics related to time,latitude,longitude,depth,and magnitude were extracted from the earthquake catalog(Global Centroid Moment Tensor)to produce 202,706 spatiotemporally discretized windows.ML explainers trained on these windows revealed the maximum magnitude(Mmax)as the most influential feature.Classification performance improved when the maximum inter-event time,the average interevent time,and the minimum ratio of focal depth to magnitude were jointly trained with Mmax.The top five features showed weak-to-moderate correlations,providing complementary information to the ML explainers.Our explainable ML framework can be extended to different earthquake catalogs,including those with focal mechanisms and smallmagnitude events.
