Research on ground subsidence monitoring in Changsha city based on time-series insar technology

To obtain the spatiotemporal distribution characteristics of ground subsidence in Changsha City in recent years, this study utilized 85 ascending-orbit Sentinel-1 IW images from August 25, 2019, to September 3, 2024, and employed SBAS-InSAR technology to invert surface deformation in the main urban area. InSAR results indicate that between 2019 and 2024, ground deformation rates in the study area ranged from -36.21 mm/a to 26.27mm/a, with cumulative subsidence values distributed between -175.11mm and 117.8mm. The annual average deformation rate in the study area primarily concentrated between -8mm/a and -15mm/a. While the overall deformation level in Changsha's main urban area was relatively low, localized areas still exhibited risks of persistent subsidence. K-means clustering analysis of InSAR deformation revealed three deformation types in the study area: Type I (continuous subsidence), Type II (basically stable), and Type III (slight uplift), accounting for 52.07%, 28.81%, and 19.12% respectively. Furthermore, overlaying InSAR rates with HWSD2 soil types revealed that under a unified threshold, soft soil areas exhibited the highest proportion of significant subsidence, followed by medium soils, with non-soft soils showing the lowest (18.9%, 15.2%, and 6.5%, respectively). This ranking remained consistent across different thresholds, indicating a strong spatial correlation between subsidence and soil type distribution. Regarding deformation studies of the Houzishidong Bridge, the central section accumulated approximately -40mm over five years. Wavelet transform results indicate a strong correlation between nonlinear deformation and surface temperature, highlighting the role of computational techniques in time-series signal analysis and causal inference. Finally, exponential model fitting for subsidence funnels in the study area suggests a stabilization period of approximately 622 days. This study, integrating modern computer algorithms and remote sensing technology, provides effective methodological support for refined monitoring and risk assessment of urban ground subsidence.