TEMPL: a Multi-Platform LiDAR Dataset of a Temperate Forest in the Eastern United States

Abstract. Laser scanning from multiple Light Detection and Ranging (LiDAR) platforms enables the acquisition of 3D point clouds from complementary sensing perspectives and at varying resolutions, supporting cross-platform forest measurement, point-cloud enhancement, tree species classification, and ecological analyses. However, publicly available multi-platform LiDAR datasets with corresponding individual-tree information remain scarce, particularly for temperate hardwood forests in the Central Hardwood Forest (CHF) region of the eastern United States. To address this gap, we collected data using an airborne LiDAR system (ALS), an uncrewed aerial vehicle LiDAR system (ULS), and two backpack LiDAR systems (BLS) in a temperate forest in the CHF of the eastern United States. The TEMPL dataset provides spatially well-aligned and georeferenced point clouds from multiple LiDAR platforms for nine continuous forest inventory (CFI) plots. A key feature of the dataset is the availability of corresponding individual-tree point clouds across multiple platforms, including 838 individual tree point clouds representing 231 trees from 16 tree species and a separate snag class. Associated tree attributes are also provided, including field-measured metrics and point-cloud-derived metrics estimated using quantitative structure models (QSMs). Together, these data provide a resource for cross-platform algorithm development and ecological analyses of tree structure within a forest region that is underrepresented in existing individual-tree LiDAR datasets. This paper describes the data acquisition, processing workflow, dataset contents, and potential applications. The entire dataset is available at https://doi.org/10.4231/5DGX-DP81.