Regional Flood Risk Assessment and Prediction Based on Environmental Attributes and Pipe Operational Characteristics

Urban flood risk assessments play a crucial role in urban resilience and disaster management. This paper proposes a comprehensive method for urban flood risk assessment and prediction that is based on environmental attributes and the operational characteristics of pipe networks. Using the central urban area of Zhengzhou as a case study, an integrated urban flood risk evaluation index system was developed, and the entropy weight method was applied to quantify risk indicators. A loosely coupled RF-XGBoost model was constructed to predict the flood risk of different rainfall scenarios. The results indicate that (1) the overall flood risk in the study area exhibits an increasing trend from the northeast to the southwest, with medium- to high-risk zones being predominant; (2) the spatial distribution pattern of the comprehensive flood risk closely aligns with that of the environmental risk but shows slight variations under the influence of pipe network operational risks; (3) the RF-XGBoost model demonstrates superior predictive accuracy under multi-factor coupling scenarios. When rainfall characteristics, environmental attributes, and pipe network operational risks are comprehensively considered, the Nash–Sutcliffe Efficiency (NSE) of the predictions improves from 0.85 (when using only rainfall characteristics) to 0.94. This study provides valuable insights and technical support for mitigating urban flood risks.