Understanding “8.12” flash flood in Suizhou, China: A meteorological analysis and implications for multi-scale prevention strategies

Abstract

On August 12, 2021, Liulin Town in Suizhou, China, experienced a catastrophic flash flood from unprecedented rainfall. This study examines the dynamic mechanism using multi-source datasets. The flash flood was induced by a quasi-stationary convective system embedded within the Yangtze-Huaihe Meiyu Circulation, which was further intensified by an anomalous configuration of the Western Pacific Subtropical High (WPSH). Between 04:00 and 06:00 local time, convective cells persisted over Liulin Town, with rainfall exceeding 100 mm/h. The storm's path aligned with the local natural flood channels, intensifying the disaster's impact. Notably, the primary moisture source differed from typical southerly monsoonal transport. Backward trajectory analysis using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model reveals that an easterly jet over the East China Sea served as the dominant moisture supplier, marking a significant deviation from historical patterns. This deviation highlights the role of large-scale atmospheric circulation anomalies in shaping extreme precipitation. Based on these insights, we conceptually propose a multi-scale framework for flash flood prevention that could potentially integrate real-time precipitation tracking across different altitudes and timescales. This theoretical approach suggests a pathway toward more proactive flood risk mitigation by leveraging advancements in rainfall forecasting, hydrometeorological monitoring, and adaptive response systems, though further validation is needed. By shifting from passive response to active defense, this framework provides a scientific foundation for enhancing early warning systems in flood-prone regions. The findings contribute to a deeper understanding of extreme rainfall and offer practical implications for disaster risk reduction in the changing climate environment.