Co‐Seismic Landslides and a Massive Lateral Spread of the 2023 Jishishan Earthquake in China Characterized by Intelligent Remote Sensing Analysis

Abstract Understanding earthquake‐induced secondary hazards is critical for effective disaster mitigation and risk reduction. However, while most research focuses on hazards triggered by major earthquakes, moderate‐magnitude events can also cause severe disasters in specific geo‐environmental configurations. The 2023 Ms 6.2 (Mw 6.0) Jishishan earthquake, occurring in the transition zone between the Qinghai‐Tibet Plateau and the Loess Plateau in China, serves as an ideal case to investigate this “moderate quake but numerous hazards” phenomenon. To systematically characterize its impacts, we generated a comprehensive inventory of 16,544 co‐seismic landslides through visual interpretation utilizing high‐resolution spaceborne and airborne remote sensing images, and illuminated the driving factors of landslide evolution by machine learning methods. Random Forest and Light Gradient Boosting Machine models, which achieve an area under the receiver operating characteristic curve (ROC AUC) >0.96, are effective in landslide susceptibility assessment. Our results demonstrate that the occurrence of co‐seismic landslides is primarily controlled by large peak ground acceleration and the distribution of sedimentary lithology (loess). The proxy for ground disturbance derived from interferometric synthetic aperture radar (InSAR) data succeeds in highlighting the spatial pattern of earthquake damage. Furthermore, multi‐source remote sensing analysis of an enormous lateral spread disaster triggered by this earthquake revealed that flood irrigation over sandy soils, indicated by Sentinel‐2 soil moisture indices, is assumed to induce a liquefied process under ground shaking, subsequently leading to lateral spread. Our findings advance the understanding of landslide occurrence and multi‐hazard interaction in moderate seismic events within vulnerable landscapes.