Landslide Detection And Monitoring Over Southwestern Mountainous Area With InSAR

China is a country of geological disaster frequently happened, especially landslide disaster. Landslide is one of the natural disasters that caused huge economic losses besides earthquake of the world. This event will cause huge economic losses and human casualties. Landslide detection and monitoring should be strengthened. But conventional landslide detection and monitoring often need a lot of manpower and material resources, and uaually restricted by external factors.InSAR technique has been widely applied in the study of landslide detection and monitoring as its ability to acquire accurate deformation and its advantages of day and night, all-weather conditions, etc. However the southwestern mountainous area has complex mountainous terrain, which is covered by dense vegetation. It makes many layovers and shadows in the SAR image and a low coherence of the interferograms. Combined with the serious affection of atmospheric, there still exist many problems require in-depth study in the wide range of landslide detection and high-precision monitoring.To solve the above problems, the paper studies landslide detection and monitoring of southwestern mountainous area based on InSAR technology, and then summarize InSAR techniques’ applicable characteristics and the processing of different applications. Different areas are chosen to carry out experimental study. Accordingly, the main research contents and achievements in the paper are as follows:(1) Different kinds of InSAR techniques, including Stacking, PS-InSAR, SBAS and multi-source SAR data integration are expounded, and their advantages and applicability of landslide detection and monitoring are analyzed. The landslide monitoring procedues of different techniques are given, especially in southwestern mountainous area.(2) ALOS/PALSAR and Envisat/ASAR data are involved to detect the potential landslides over Wudongde Hydropower Station Section, Jinsha River. As for Stacking interferogram method, four criteria are set to delimitate potential landslides, that is, intensity, coherence, deformation rate and DEM slope, accordingly ten known and unknown landslides are successfully detected. ALOS/PALSAR data are applied to detect the potential landslides over the mountainous area of Wulong County, Chongqing City based on Interferometric Point Target Analysis(IPTA) method and seven potential landslides are successfully detected. The results are validated by field investigation.(3) As for InSAR time series monitoring, we Jinpingzi landslide is chosen as a typical case. Small Baseline Subset(SBAS) technique is applied to calculate the time-series deformation by using ALOS/PALSAR images. The deformation characteristics in the space and time domain are analyzed. Then, in-situ georobot measurements are used for point-wise comparison. The precision in landslide direction is around 1.8 cm by comparing with in-situ georobot measurements. Erpingcun landslide located in Wudongde reservoir area is selected to calculate the vertical and horizontal deformation components by fusing the ascending and descending line-of-sight results.(4) ALOS/PALSAR images are involved for Guanling landslide monitoring. The time-series deformation before the landslide is acquired through PS-InSAR technique. The mechanism of GuanLing landslide and the deformation characteristics both in space and time domain have been analyzed.