Using unmanned aerial vehicles (UAV) as platforms for light detection and ranging (LiDAR) sensors offers the efficient operation and advantages of active remote sensing; hence, UAV-LiDAR plays an important role in forest resource investigations. However, high-precision individual tree segmentation, in which the most appropriate individual tree segmentation method and the optimal algorithm parameter settings must be determined, remains highly challenging when applied to multiple forest types. This article compared the applicability of methods based on a canopy height model (CHM) and a normalized point cloud (NPC) obtained from UAV-LiDAR point cloud data. The watershed algorithm, local maximum method, point cloud-based cluster segmentation, and layer stacking were used to segment individual trees and extract the tree height parameters from nine plots of three forest types. The individual tree segmentation results were evaluated based on experimental field data, and the sensitivity of the parameter settings in the segmentation methods was analyzed. Among all plots, the overall accuracy F of individual tree segmentation was between 0.621 and 1, the average RMSE of tree height extraction was 1.175 m, and the RMSE% was 12.54%. The results indicated that compared with the CHM-based methods, the NPC-based methods exhibited better performance in individual tree segmentation; additionally, the type and complexity of a forest influence the accuracy of individual tree segmentation, and point cloud-based cluster segmentation is the preferred scheme for individual tree segmentation, while layer stacking should be used as a supplement in multilayer forests and extremely complex heterogeneous forests. This research provides important guidance for the use of UAV-LiDAR to accurately obtain forest structure parameters and perform forest resource investigations. In addition, the methods compared in this paper can be employed to extract vegetation indices, such as the canopy height, leaf area index, and vegetation coverage.

Direct adoption of nationwide generalized biomass models often causes systematic prediction errors when applied to forest biomass estimation in Hunan Province. This study developed localized multi–component compatible biomass model systems for Chinese fir, poplar and oak based on 461 felled–tree specimens sampled from 14 administrative cities of Hunan Province during 2022–2023. Individual allometric equations were first fitted via weighted least squares, and hierarchical compatible model structures were further established by algebraic component summation and proportional scaling under simultaneous equation constraints. The final total–biomass compatible equations were: Chinese fir: $B_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}=0.065662D^{1.750420}{H}^{0.780381}+0.024760{(D}^2{H)}^{0.774387}$; Poplar: $B_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}=0.046471D^{2.171796}{H}^{0.462424}+0.013116·{(D}^2{H)}^{0.822310}$; Oak: $B_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}=0.0927573{\left(D^{2}H\right)}^{0.89013}+0.018969{\left(D^{2}H\right)}^{0.89162}$. R² values for total and aboveground biomass exceeded 0.92 and prediction accuracy exceeded 96%, with TRE within ±3% for all species. Root biomass models reached prediction accuracy higher than 78% with R² > 0.78. Stem, wood and bark sub–models achieved distinctly better fitting performance than branch and foliage sub–models, and strict hierarchical compatibility was guaranteed among total and component biomass predictions. Calibrated exclusively with local field data, these models effectively eliminate regional systematic estimation bias and only require readily measurable DBH and tree height as input variables, serving as reliable tools for regional forest biomass assessment and routine forest resource monitoring across Hunan Province.