An UAV-LiDAR ultra-high-density dataset collected in leaf-off and partial-leaf-cover forests in Guangxi, China

The area falls within the eastern subtropical mountainous humid climate zone, with an average annual temperature of 13.7°C and an average annual rainfall of 1,657 mm. There are two forest types within the study site, separated by a road: to the west lies a planted Betula alnoides Buch. -Ham. ex D. Don forests, planted along contour lines with an initial planting density of 2 × 4 m; to the east lies a mixed-planted forest of Betula alnoides and Chinese fir (Cunninghamia lanceolata (Lamb.) Hook.), with Betula alnoides representing approximately two-thirds and Chinese fir one-third.

The tree distributions show no obvious pattern.2. Field dataField measurements were conducted on Jan. 28, 2026. One unbounded plot was established in the planted Betula alnoides forests and another in the mixed Betula alnoides-Chinese fir forests.

A diameter tape was used to measure the DBH (1.3 m above ground level) of sample trees with a DBH greater than 5.0 cm, and the tree species were recorded. A network real-time kinematic global navigation satellite system (RTK-GNSS) was used to position the sample trees. A total of 167 trees were measured, comprising 83 trees in the Betula alnoides forest and 84 trees in the mixed forest.

Plot 1 contained only one Chinese fir; the rest were all Betula alnoides. To simplify the analysis, all trees in this plot were treated as Betula alnoides.Due to errors in RTK measurements, tree position adjustments were required. To this end, in the ArcGIS software environment, the field-survey tree map was overlaid with the trunk point clouds (height range: 0.2–2.4 m), and the tree positions were carefully moved manually to the centers of adjacent point-cloud clusters, with the expected positioning error of the trees not exceeding 5 cm.

The final sample tree positions were taken as the initial centers of the sample trees, which could be regarded as the centers of the cross-sections at a height of 1.3 m above the ground.The field measurement and recording data for the two plots were contained in the Tree_location.shp file, including plot ID, tree ID, DBH, and tree species. The projection system for the files was WGS_1984_World_Mercator.3. LiDAR dataThe Fly2Map Done SF-1200 UAS integrated with an AcuteLas SA130 laser scanner (South Surveying & Mapping, Guangzhou, China).

The URL for the equipment presentation is:southinstrument.com/product/details/pro_tid/1/id/211.html;https://www.southinstrument.com/product/details/pro_tid/1/id/270.html.To acquire ultra‑high‑density point cloud data with complete and uniform coverage around individual tree trunks, this study adopted a dual‑grid crossing flight plan with dual 45° oblique intersections. The workflow involved (1) constructing a primary base grid (0° and 90° orthogonal) by flying the UAV in a lawnmower‑style zigzag pattern along tree rows at 6–10 m strip spacing, followed by a second pass with a 90° heading rotation at the same spacing.

(2) Generating a secondary crossing grid (45° and 135° oblique) by rotating the entire route matrix 45° about the base grid centroid and repeating the reciprocating flight logic. The staggered cross‑coverage of these two independent grids enabled comprehensive, blind‑spot‑free scanning of the target plot. Upon mission completion, the UAV exited along a pre‑defined egress route, forming a fully closed operational loop.ULS datasets were acquired from two campaigns on Mar. 05, 2026, and Mar. 25, 2026, using the following flight configurations.

Campaign one: flight altitude 90 m above ground level (AGL); flight path spacing 6 m; and flight speed 6 m/s. Campaign two: flight altitude: 70 m AGL; flight path spacing: 10 m; and flight speed: 3 m/s. As a result, this study obtained two point cloud datasets.The raw point cloud data were first preprocessed using the software provided by the sensor manufacturer.

Subsequently, the point clouds were denoised and classified into ground and non-ground points using the improved triangulated irregular network (TIN) densification (IPTD) filtering algorithm in the LiDAR360 software (v 5.0) (Green Valley Inc., Beijing, China). The point cloud elevations were then normalized using the ground points. The four LAS files:l Plot1_0305.las;l Plot1_0325.las;l Plot2_0305.las;l Plot2_0325.las.The projection system for all of the point cloud files was WGS_1984_World_Mercator.