Investigating overlapping deformation patterns of the Beijing Plain by independent component analysis of InSAR observations

• Interpretate heterogeneous ground deformations under the new water conditions. • Decompose overlapping ground deformation patterns using ICA. • Explore influencing factors of different deformation patterns. • Great potential of InSAR and ICA in hydrogeological applications. Due to policies such as groundwater extraction restrictions, water diversion, and water replenishment, groundwater levels in the Beijing Plain have generally risen. This has effectively alleviated ground subsidence, with some regions even experiencing uplift. Under these new water conditions, strata deformation also shows spatiotemporal heterogeneity, and the overlap of these multiple deformation patterns hinders the interpretation of ground deformation. In this study, we employed Satellite Interferometric Synthetic Aperture Radar (InSAR) to investigate the spatiotemporal heterogeneous deformation patterns and their influencing factors in the Beijing Plain. We utilized Independent Component Analysis (ICA) to effectively separate overlapping ground deformation patterns from InSAR-derived ground surface deformation. Four ground deformation patterns—approximately linear subsidence, approximately linear uplift, decelerating subsidence, and periodic deformation—were extracted. Their overlapping status and influencing factors were explored based on external hydrogeological data. Approximately linear subsidence represents the continuous compression of deep strata in areas with severe subsidence. Approximately linear uplift represents the rebound of shallow strata caused by water replenishment. Periodic deformation represents the ground’s response to precipitation and irrigation pumping. The ICA results benefit the estimation of the time lag between groundwater rise and strata rebound in the Beijing Plain, as well as the estimation of the increase in groundwater storage capacity in the Miyun-Huairou-Shunyi region. This study provides a new perspective for extracting deformation information from InSAR observations, gaining a deeper understanding of deformation patterns related to hydrogeological phenomena, and assessing the effectiveness of groundwater management and land subsidence control policies.