Estimation of Northern Hardwood Forest Inventory Attributes Using UAV Laser Scanning (ULS): Transferability of Laser Scanning Methods and Comparison of Automated Approaches at the Tree- and Stand-Level

UAV laser scanning (ULS) has the potential to support forest operations since it provides high-density data with flexible operational conditions. This study examined use of ULS systems estimate several tree attributes from an uneven-aged northern hardwood stand. We investigated: (1) transferability raster-based and bottom-up point cloud-based individual detection (ITD) algorithms data; (2) automated approaches retrieval tree-level (i.e., height, crown diameter (CD), DBH) stand-level count, basal area (BA), DBH-distribution) inventory attributes. These objectives were studied under leaf-on leaf-off canopy Results achieved cross-compared ALS TLS better understand challenges faced by different methodological in environments. The best results that characterized trees conditions using a ITD. latter outperformed ITD, improving accuracy (from 50% 71%), delineation R2 = 0.29 0.61), prediction DBH 0.36 0.67), when compared values estimated reference data. Major improvements observed for lower layer 9% ITD 51% ITD) intermediate 24% 59% ITD). Under conditions, LiDAR aerial include substantial signal occlusion incurred upper canopy. these was unable detect low-level 5% 15% detected layers, respectively), resulting rate about 40% both cylinder-fitting method used did not meet standards DBH, RMSE 7.4 cm, Bias 3.1 0.75. Yet, yielded more accurate estimates BA (+3.5%) DBH-distribution stand than allometric models ?12.9%), situ field measurements. suggest on leads promising estimating attributes, which opens up new possibilities supporting inventories operations.