Risk Identification and Resilience Enhancement for Rainstorm Disaster Chains: An Event Evolutionary Graph–Driven Approach

Abstract

Under the influence of global warming, the frequency of rainstorm disaster chains continues to increase with more severe impacts, posing substantial threats to socioeconomic development and public safety. Early identification and prevention of rainstorm disaster chains are crucial for reducing disaster losses. To address this challenge, this study introduces an event evolutionary graph–based framework for risk identification and resilience enhancement of such disaster chains. The framework involves three key steps. First, an event evolutionary graph–based risk network for rainstorm disaster chains is constructed by extracting causal relationships from historical disaster data. Next, a historically calibrated topological metric is developed to identify critical risk nodes and high-risk propagation pathways. Finally, the proposed method was validated using a case study of Sichuan Province by comparing the effectiveness of resilience enhancement strategies guided by different risk identification approaches. The results demonstrate that the proposed event evolutionary graph–driven framework significantly outperforms conventional methods in both accurately identifying critical risk nodes and enhancing systemic resilience. This study thus provides a scientific basis for formulating targeted prevention strategies and offers effective decision support for improving risk management of the rainstorm disaster chain.