Tracking 30-year evolution of subsidence in Shanghai utilizing multi-sensor InSAR and random forest modelling

Land subsidence is a significant issue in many coastal megacities, including Shanghai, where it poses risks to infrastructure and economic stability. Although numerous studies have used SAR datasets to monitor land subsidence in Shanghai, multi-decadal displacement measurements obtained from multi-sensor SAR data remain unavailable. Moreover, the contributions and variations of driving factors behind the evolution of land subsidence remain poorly understood. This study employs multi-sensor SAR fusion method and a Random Forest model, along with Shapley Additive exPlanations (SHAP), to examine subsidence evolution and assess the influence of key drivers over the past 30 years. The results show that severe subsidence has spread from central urban areas to surrounding suburban regions, particularly in the eastern coastal and southern industrial zones in Shanghai. SHAP analysis identified that evapotranspiration, sediment thickness, and groundwater extraction were the dominant factors in the early stage of subsidence, while recent groundwater management and recharge practices have significantly mitigated the subsidence rate. These findings demonstrate the shifting importance of different subsidence factors over time and provide valuable insights for long-term prevention and control measures.