Objectives Unmanned aerial vehicle (UAV)-borne laser detecting and ranging (LiDAR), which is relatively insensitive to environmental conditions and offers a distinctive viewing geometry with the capability to penetrate vegetation, has become an effective tool for geohazard investigations in densely vegetated mountainous regions. However, point-density requirements tailors to geohazard identification remain underdeveloped. Current practice largely follows general surveying specifications, which often fail to support the fine-scale depiction of micro-topographic features (e.g., scarps and gullies) needed for early-stage detection.
Methods Focusing on Guangxi, China, where geohazards are typically small in scale and vegetation cover is dense, we examined the effects of varying point-cloud densities on DEM (digital elevation model) quality using four relatively large map scales. Local terrain complexity was introduced as a quantitative indicator to derive the optimal ground-point density. LiDAR penetration rates under different canopy-closure conditions were further incorporated to back-calculate recommended acquisition point densities.
Results Local terrain complexity effectively reflected the completeness of micro-topographic preservation in DEM. Multi-group comparisons across point-cloud densities determined the optimal ground-point density and the corresponding acquisition point density, and produced canopy-closure–adaptive recommended acquisition densities.
Conclusions We establishe a point-density reference standard for unmanned aerial vehicle-borne LiDAR–based geohazard surveys in densely vegetated mountainous areas of Guangxi, enabling more reliable DEM representation of micro-topography to support early geohazard identification.
DownLoad: