Modelling the Diameter Distribution of Savanna Trees with Drone-Based LiDAR

The diameter distribution of savanna tree populations is a valuable indicator health because changes in the number and size trees can signal shift from to grassland or forest. Savanna distributions have traditionally been monitored with forestry techniques, where stem at breast height (DBH) measured field within defined sub-hectare plots. However, spatial scale these plots often misaligned variability populations, there need for techniques that scale-up surveys. Dense point clouds collected uncrewed aerial vehicle laser scanners (UAV-LS), also known as drone-based LiDAR (Light Detection Ranging), be segmented into individual crowns then related application allometric scaling equations. Here, we sought test potential UAV-LS segmentation model trees. We both field-survey data five one-hectare woodland northern Australia, which were divided two calibration three validation Within plots, equations developed by linking field-surveyed DBH metrics manually delineated crowns, best performing had bias 1.8% relatively high RMSE 39.2%. A algorithm was applied segment derived clouds, level accuracy assessed against crowns. 47% accurately 68% Using site-specific allometry, modelled crown all results compared distributions. In significant differences between distributions, became similar when smaller (<10 cm DBH) excluded. Although followed overall trend surveys, non-significant result demonstrates adoption remotely detectable proxies could replace DBH, well more accurate detection methods ecosystems.