Attribution analysis of urban social resilience differences under rainstorm disaster impact: Insights from interpretable spatial machine learning framework

• Published in: Sustainable cities and society 2025-01, Vol.118, Article 106029
• Main Authors: Gu, Tianshun, Zhao, Hongbo, Yue, Li, Guo, Jiaojiao, Cui, Qinyu, Tang, Junqing, Gong, Zhaoya, Zhao, Pengjun
• Format: Article
• Language: English
• Subjects: Geographically weighted random forest; Interpretable spatial machine learning framework; Urban Resilience; Urban social resilience
• ISSN: 2210-6707
• DOI: 10.1016/j.scs.2024.106029

•Developed an interpretable spatial machine learning framework.•Investigated the nonlinear correlation between variables and social resilience.•Described the spatial heterogeneity of variables and social resilience. With the frequent occurrence of extreme rainstorms in global cities, understanding differences in social resilience is crucial for constructing climate-adaptive communities. However, quantitatively analyzing the compound effects and interactions of social resilience determinants remains challenging. Here, we developed an advanced interpretable spatial machine learning framework to analyze social resilience across 2,221 blocks in Zhengzhou City, China, from 2005 to 2022. The framework integrates the Geographically Weighted Random Forest model with the SHapley Additive exPlanations model to address non-linearity, spatial heterogeneity, and interpretability simultaneously. Our findings revealed that the social resilience increased by 64.75% in response to rainstorm disasters, while the differences among blocks widened by 13.19%. We also observed that the central areas of cities presented the high social resilience pattern, while the social resilience in the peripheral areas was relatively low. The probability of blocks maintaining their social resilience levels was 66.93% for low (Ⅰ)-resilience blocks and 52.60% for high (IV)-resilience blocks. Key factors—such as economic vitality, population size, government services, and rainfall intensity—significantly influenced social resilience, with nonlinear relationships and local threshold effects. The impact of factors like land use diversity and facility supply varied spatially. This research deepens the understanding of the compound effects of social resilience determinants and highlights the importance of formulating flood intervention strategies in accordance with local conditions.