ABSTRACT
Assessment of landslide probability usually derived from three major approaches - statistical approach, mechanical approach, semi-mechanical approach. Landslide history or geotechnical parameters are required time and budget to complete, which is very difficult to finish. In order to search location of landslide with basin scale, a remote sensing approach with radar signal processing is adopted in this research. With processing of radar signal data, which is called Differential Interferometry of Synthetic Aperture Radar (DInSAR), can generate ground deformation fringes. The fringes represent ground deformation thus potential landslide deformation can be detected. Landslide inventory maps are produced from temporal optical remote sensing data for comparison. Landslide probability maps are generated from DInSAR analysis from 2007 to 2010 and landslide inventory maps in the same period are used for verification. The results show that landslide probability map generated from DInSAR analysis can represent most landslides except some creeping or not sliding slope cases.
