Detecting subtle nonlinear changes in slow-moving landslides with a hybrid InSAR change point detection framework: Methodological assessment and validation
Citation
Li, Z., Chen, J., Cao, C., Zhang, W., Li, Y., Hu, J., Li, Z. (2025). Detecting subtle nonlinear changes in slow-moving landslides with a hybrid InSAR change point detection framework: Methodological assessment and validation. International Journal of Applied Earth Observation and Geoinformation, 136: 104321. Link to paper
Abstract
InSAR technology provides high-resolution time-series observation capabilities for monitoring surface slow-moving landslide deformation, with a key challenge being accurately identifying subtle yet critical change points—acceleration, destabilization, or stagnation—to improve timeliness and reliability of geological hazard early warning systems. This study employed Bayesian optimization to quantitatively delineate the applicability boundaries of three algorithms—BFAST, BEAST, and PAE. Results reveal that BEAST achieves superior precision (0.93) in noise-free, seasonally dominated environments, whereas PAE demonstrates robust recall (0.62) under real-world conditions, while BFAST prioritizes sensitivity to trend breaks but exhibits higher false positive rates. The study provides the first evidence that local earthquake swarms (Mw 4.6–5.3) can trigger abrupt landslide displacement.