Abstract
Because of its centimetre-level precision, SAR interferometry (InSAR) has been widely recognized as an effective tool for large-area deformation mapping. However, applications of InSAR technology are limited by temporal–geometric decorrelation and atmospheric disturbance. Large-field displacement could be underestimated as a result of these limitations. A point-like target offset tracking (PTOT) method that solely uses SAR pixel amplitude information through highly accurate image matching can measure 2D displacements in the azimuth and range directions. This method is minimally affected by decorrelation and weather, and it is more suitable for mapping large displacements. In this study, landslide-prone areas were identified using high-resolution TerraSAR-X InSAR data pairs with short normal baselines and temporal baselines covering Zigui County in the Three Gorges area, and the PTOT method was applied to study potential landslides using image pairs with long temporal baselines. An estimate of the cumulative deformation was given. Time series displacements measured at corner reflectors and point-like targets were analysed. The feasibilities of landslide detection using DInSAR and PTOT methods with high-resolution SAR images were then evaluated and summarized.