UAV-LiDAR point clouds from the Forest and Biodiversity Experiment 2 (2022)
Data files
Jan 09, 2025 version files 8.85 GB
-
2022-04-10_FAB2.laz
1.30 GB
-
2022-05-18_FAB2.laz
1.04 GB
-
2022-06-12_FAB2.laz
1.07 GB
-
2022-07-06_FAB2.laz
1.08 GB
-
2022-08-03_FAB2.laz
1.11 GB
-
2022-09-07_FAB2.laz
1.04 GB
-
2022-09-18_FAB2.laz
1.07 GB
-
2022-10-24_FAB2.laz
1.14 GB
-
README.md
1.29 KB
Abstract
Point clouds collected at the of the Forest and Biodiversity Experiment 2 located at the Cedar Creek Ecosystem Science Reserve (CCESR), Minnesota, USA. These point clouds were collected using a LiDAR onboard an Uncrewed Aerial Vehicle (UAV) thoughout the growing season of 2022.
README: UAV-LiDAR point clouds from the Forest and Biodiversity Experiment 2 (2022)
https://doi.org/10.5061/dryad.jdfn2z3hk
Eight point clouds were collected on the Forest and Biodiversity Experiment 2 throughout the 2022 growing season
Description of the data and file structure
We provide point clouds using the compressed format (.LAZ) clipped to the area where the experiment is located. These point clouds were not filtered by noise nor decimated by the redundancy of points. However, these were processed by stripe alignment, ground classification, and normalization by height. The height provided (i.e., Z coordinate) is in reference to the height from the ground. The X and Y columns are coordinates projected to NAD83 UMT15 (EPSG:26915). The names of the files refers to the date of data collection.
Code/Software
Point clouds can be opened using QGIS, CloudCompare, or LAStools for visualization. They can also be processed using the lidR package of R. The code used for processing and deriving forest structural metrics is available at GitHub.
Methods
We collected LiDAR data across the FAB2 experiment eight times during 2022 on the following days of the year: 100, 138, 163, 188, 215, 250, 261, and 297. Flights were conducted throughout the growing season before leaf out until after leaf senescence of the deciduous species in accordance with sufficient accumulated growing degree days for 2022. Data were collected using a Zenmuse L1 sensor onboard an Uncrewed Aerial Vehicle (UAV) DJI Matrice 300. This sensor integrates a Livox LiDAR module, an inertial measurement unit, and an RGB camera on a 3-axis stabilized gimbal. The LiDAR module has a conic footprint on ground with a field of view of 77.2° vertical and 70.4° horizontal, enabling it to capture multiple returns. All the surveys were conducted using autonomous flights programmed to take place at a speed of 8 m/s at 50 m above ground. These surveys were done in an area of ~5.8 ha with 85% overlap between sidetracks allowing capture of dense point clouds (~ 2100 point/m2). During these flights, we also collected static GNSS data using an Emlid Reach RS2+ receiver to enable kinematic corrections or post-processing corrections in case of signal loss. The data collected from the LiDAR sensor and GNSS receiver were processed in DJI Terra, delivering true color point clouds using the optimization option for manufacture strip alignment. The resulting point clouds were then processed through BayesStripAlign 2.24 to ensure a proper alignment among flight lines. The resulting point clouds were treated in a series of steps to ensure data quality. First, points were classified as ground and non-ground and then normalized by height. These procedures were done using ‘lasthin’, ‘lasground’, and ‘lasheight’ of LAStools. We provide point clouds using the compressed format (.LAZ) clipped to the area where the experiment is located. These point clouds were not filtered by noise nor decimated by the redundancy of points. The height provided (i.e., Z coordinate) is in reference to the height from the ground. The X and Y columns are coordinates projected to NAD83 UMT15 (EPSG:26915). The names of the files refers to the date of data collection.