UAV LiDAR point cloud and ground measurement dataset for assessing the impact of flight parameters over a poplar...

This data article presents a UAV LiDAR dataset acquired over a 0.23-ha managed poplar (Populus spp.) plantation located in southwestern France. The dataset was designed to assess the influence of UAV flight parameters and acquisition context on the extraction of tree-level structural metrics, including canopy height and trunk circumference. A total of 22 UAV LiDAR flights were conducted between January 2024 and April 2025 using a DJI Matrice 300 platform equipped with a Zenmuse L1 sensor. Acquisition parameters were systematically varied, including flight altitude (35-120 m above ground level), flight speed (1.7-6 m·s⁻¹), return mode (single, dual, triple), flight plan orientation relative to plantation rows (0°, 45°, 90°), overlap margins, and seasonal conditions (leaf-off and leaf-on). Resulting point densities range from 447 to 12,990 points·m⁻². Raw LiDAR data were processed into georeferenced LAS point clouds (Lambert-93, EPSG:2154; EGM96 geoid). The dataset includes several complementary data products supporting the evaluation of UAV LiDAR-derived structural metrics. These comprise: (i) canopy height estimates and trunk circumference measurements derived from the UAV LiDAR point clouds using three methods (mean circle fitting, maximum inscribed circle fitting, and spline fitting); (ii) in situ trunk circumference measurements collected during three field campaigns (December 2024, April 2025, October 2025) and (iii) tree height and trunk circumference measurements derived from terrestrial LiDAR acquisitions (TLS) using a backpack-mounted system (Viametris MS-96).