Assessing flood and waterlogging vulnerability and community governance in urban villages in the context of climate change: A case study of 89 urban villages in Shanghai

• An ensemble learning model is proposed for waterlogging vulnerability assessment. • Future waterlogging vulnerability is projected according to SSP-RCP scenarios. • Suburban growth and ecological agricultural zones exhibit greater vulnerability. • Policy analysis shows a mismatch between policy focus and urban village needs. • The study offers a general framework for climate adaptation in similar communities. Climate change has increased the vulnerability of urban villages to flooding and waterlogging, making it a major challenge for government community management. This study assesses flood-waterlogging vulnerability in Shanghai's urban villages under current conditions and long-term SSP-RCP scenarios using data from 89 redevelopment communities. We identify development patterns based on clustering models, evaluate vulnerability using an ensemble learning model, and analyze climate policy attention with Python-based text mining. Results highlight significant variations in infrastructure and flood risk across urban villages with different development patterns. Suburban Growth Zones and Ecological Agricultural Zones show higher vulnerability. In the long term, most villages become more vulnerable under both SSP245 and SSP585 scenarios, and the uncertainties and risks will increase due to cumulative effects. The study emphasizes the necessity of thorough governmental management in mitigating climate-induced waterlogging risks in urban villages and suggests specific policy recommendations customized for various categories of urban villages.