Freeze–thaw‐related loess landslide accelerations detection in Ili Basin based on multi‐temporal InSAR analysis

Abstract Intensified anthropogenic activities and the changing climate result in frequent loess geohazards in the Ili basin, accounting for >80% of the total geohazards in Xinjiang. Loess landslides are particularly prevalent in northern Yining County due to the widespread distribution of loess deposits, which are characterized by high collapsibility and soluble salt content. However, the processes governing their spatiotemporal evolution and the quantitative relationships between slope deformation and coupled freeze–thaw and precipitation forcing remain poorly constrained, particularly at the regional scale. To address this gap, we integrate multi‐temporal Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS‐InSAR) observations from ascending and descending Sentinel‐1 data sets to derive time series line‐of‐sight (LOS) deformation. These ascending and descending Sentinel‐1 InSAR measurements are further combined to retrieve vertical and eastward displacement components. The resulting deformation time series are then analysed using tangential angle analysis to characterize displacement stages and wavelet coherence analysis to quantify time–frequency relationships and phase lags between displacement and environmental variables. Over 130 active slopes with maximum displacements exceeding 100 mm/year were identified by both data sets. Kinematic analysis of the Hujierti landslide suggests that its initiation may have been triggered by freeze–thaw cycles in December 2016, with subsequent accelerations attributed to snowmelt and precipitation, culminating in a sliding event in May 2022. Furthermore, displacement rates at the Hujierti landslide remain above 60 mm/year, underscoring the necessity for ongoing monitoring. Time series analysis has revealed that the primary drivers of displacement are freeze–thaw cycles and concentrated precipitation, which result in groundwater accumulation, reduced slope stability and eventual landslide activation. Tangential angles exceeding 85° serve as indicators of impending sliding. Wavelet analysis has unveiled the complex interactions between displacement, precipitation and temperature. Given the challenges associated with monitoring such dynamic systems, InSAR offers invaluable insights into long‐term ground displacement, which are crucial for geohazard prevention and management in the Ili Basin.