基于SSA-RF的东北三省洪涝灾害韧性评估

Flood Disaster Resilience Assessment in the Three Northeastern Provinces Based on SSA-RF

摘要: 为提升东北三省城市洪涝灾害韧性评估的科学性，本研究基于“驱动力–状态–响应”(DSR)模型构建动态评价体系，融合随机森林算法优化指标权重，并采用麻雀搜索算法加速参数寻优。2013~2022年东北三省35个地级行政区数据分析表明：时序上，黑龙江、吉林和辽宁省韧性指数总体呈上升趋势，但因极端降水和经济波动呈现阶段性波动；空间上，形成显著的“核心–边缘”梯度格局，沈阳、大连、哈尔滨、长春等高韧性核心城市对周边产生辐射效应，但铁岭、鹤岗等边缘城市受制于排水设施和信息基础设施等短板；驱动机制上，经济指标和信息能力是韧性分异的核心驱动力，而极端灾害下基础设施的“隐性短板”效应凸显。研究揭示通过强化经济支撑、信息赋能和区域协同机制促进韧性均衡发展，为东北洪涝治理提供决策依据。

Abstract: In order to improve the scientificity of flood disaster resilience assessment in cities in the three northeastern provinces, this study constructs a dynamic evaluation system based on the “driving force-state-response” (DSR) model, integrates the random forest algorithm to optimize the index weights, and uses the sparrow search algorithm to accelerate parameter optimization. Data analysis of 35 prefecture-level administrative regions in the three northeastern provinces from 2013 to 2022 shows that: in terms of time series, the resilience index of Heilongjiang, Jilin and Liaoning provinces generally shows an upward trend, but it fluctuates periodically due to extreme precipitation and economic fluctuations; in terms of space, a significant “core-edge” gradient pattern is formed, and high-resilience core cities such as Shenyang, Dalian, Harbin and Changchun have a radiation effect on the surrounding areas, but marginal cities such as Tieling and Hegang are constrained by shortcomings such as drainage facilities and information infrastructure; in terms of driving mechanism, economic indicators and information capabilities are the core driving forces of resilience differentiation, and the “hidden short board” effect of infrastructure under extreme disasters is prominent. The study reveals that strengthening economic support, information empowerment and regional coordination mechanisms can promote balanced development of resilience, providing a decision-making basis for flood control in Northeast China.