Abstract:Reservoir landslide is a common geological disaster in hydropower project construction, when the landslide slides into the reservoir area may cause surge, river blockage or even dam failure, thus causing huge economic losses and casualties, so how to efficiently and timely carry out the early identification and monitoring of the reservoir bank landslides is of great significance. This paper utilizes short baseline subsets interferometric synthetic aperture radar (SBAS-InSAR) to detect and identify active landslides in the Maoergai reservoir area, and on this basis, we com-bined high-resolution optical remote sensing images, UAV aerial photogrammetry and on-site investigation to make a preliminary analysis of the relationship between the deformation characteristics of typical active landslides in the reservoir area and the reservoir water level changes response. Through the study, it is concluded that: (1) there are 65 active landslides in the study area, of which 36 are resurrection-type ancient landslide accumulations and 29 are new breeding-type landslides; (2) in combination with the two types of typical landslides, there is a positive cor-relation between the deformation of the slopes and the rapid decline of the reservoir level, and when the average daily rate of decline of the reservoir water level is greater than 0.67 m/d, the accelerated deformation occurs in both the InSAR time-series curves; (3) for large deformation landslides occurring within a short period of time, it is difficult for SBAS-InSAR to effectively capture the real acceleration and deformation information of landslides before destabilization, which leads to the inability to identify and monitor the landslides in advance, therefore, the combination of multi-temporal high-resolution optical remote sensing imagery can play an effective complemen-tary role. The results of the study can provide a reference for the early identification and monitoring of reservoir bank landslides.