Abstract: The development of airborne LiDAR technology provides a new solution for geological hazard investigation in densely vegetated mountainous areas.This technology highly relies on the density of ground points, which is closely related to the laser penetration rate and the original point cloud density.It is necessary to evaluate the optimal number of point clouds obtained under different vegetation density conditions in order to meet the DEM (Digital Elevation Model) interpolation requirements at the survey scale.Taking the mountainous area with complex terrain around Huangshan City, Anhui Province as the study area, the relationship between the penetration rate and the ground point density under different forest canopy density and terrain slope, as well as the relationship between the original point cloud density and the ground point density were studied, and digital terrain products with different precision were obtained according to the requirements of the survey scale.Then, the DEM constructed by the density of each point was quantitatively evaluated to inversely calculate the recommended value of airborne LiDAR optimal cloud acquisition density for geological disaster investigation in mountainous areas of Anhui Province.The results indicated that there was a negative correlation between canopy density and slope on ground point density, and canopy density had a greater impact.After the original point cloud density exceeded a certain threshold, the identification of geological hazards cannot be significantly optimized.Based on the experimental results, the recommended values for the optimal cloud collection density of airborne LiDAR for geological hazards in mountainou areas of Anhui Province at a scale of 1:500 were established, when the forest canopy density was 0.7, 1.0), 0.2, 0.7), 0, 0.2), the corresponding recommended values for point cloud density were 65, 90), 45, 65), 16, 45 pts/m².