[Objective] To clarify the spatial distribution characteristics and laws of geological disasters in Huangshui Basin, and to provide data support and scientific basis for disaster prevention and reduction. [Methods] Geological hazard sites were identified by Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR). Combining topographic factors, geological factors, environmental factors, meteorological factors and human activities, the spatial distribution law of disasters was analyzed, and a Logistic Regression (LR) -frequency Response (FR) model was established and tested. The return probability value was used to evaluate the susceptibility. [Results] The main disasters in the Huangshui watershed of Qinghai Province include landslide, collapse and unstable slope. Landslides and unstable slopes usually develop on gently sloping slopes, posing a threat to the safety of traffic and residents below the mountain. They may also block rivers and form barrier lakes, further aggravating disaster risks. The collapse mostly develops in the steep mountain walls with relatively loose rock structure or severe weathering, posing a threat to the residential areas and traffic lines below. (1) The distribution of geological hazards in the area of 2425-3650m elevation is more, and the northeast direction is the slope direction of geological hazards. Normalized Digital Vegetation Index (NDVI) decreased with the increase of geological hazards, increased with the increase of slope, relief and daily precipitation, and decreased with the increase of distance from the fault. (2) High-prone and slightly high-prone areas in Huangshui Basin, covering an area of about 5937.60km2, accounting for about 38.78% of the total area of the study area, mainly concentrated in the south and north border areas of Huangshui Basin, Huangzhong, Datong, Haiyan junction, and the surrounding slope of the construction area. (3) The evaluation results were tested. The Area Under Curve (AUC) of Receiver Operating Characteristic Curve (ROC curve) of model prediction performance was 0.787. The FR in the prone area decreases step by step from high to low, which is in good agreement with the distribution of actual disaster points. (4) Fault core density is the main control factor of geological hazard development in Huangshui Basin, followed by slope direction, relief degree and road core density, and section curvature has the least influence on geological hazard development. [Conclusion](1) SBAS-InSAR technology can effectively identify potential geological hazards in Huangshui basin, and the susceptibility evaluation results obtained by LR-FR model are reliable. (2) The distribution of geological disaster prone areas in the Huangshui Basin has obvious spatial differences, mainly distributed in the areas with higher elevation, lower vegetation coverage rate, larger rainfall and close to the fault, and the fault core density is the main control factor of geological disaster susceptibility. (3) The geological disasters in the Huangshui Basin are characterized by multiple, sudden occurrence and high risk, which have brought serious impact on local people's life, regional economic development and ecological environment. Therefore, the monitoring, early warning and prevention of these disasters are particularly important.