Canopy height model and NAIP imagery pairs across CONUS

Authors: ALLRED, BRADY - University Of Montana; McCord, Sarah; MORFORD, SCOTT - University Of Montana

Submitted to: Scientific Data
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/14/2025
Publication Date: 2/22/2025
Citation: Allred, B.W., McCord, S.E., Morford, S.L. 2025. Canopy height model and NAIP imagery pairs across CONUS. Scientific Data. 12. Article 322. https://doi.org/10.1038/s41597-025-04655-z.
DOI: https://doi.org/10.1038/s41597-025-04655-z

Interpretive Summary: Canopy height models (CHM) are spatially explicit representations of the vertical structure of an environment, measured relative to the ground surface. These models provide detailed information about the structure, arrangement, and organization of vegetation and the built up environment; they are used in numerous applications, including land management and conservation, carbon and climate change modeling, landscape and habitat monitoring, disaster risk assessment and management, and geospatial analysis and modeling. The paucity of high resolution, aerial derived CHMs presents difficulties when training models for broader scale application. In the United States, the United States Geological Survey (USGS) works with partners to collect aerial lidar for the 3D Elevation Program (3DEP). Although data are collected in different regions, at different times, by different contractors, and ultimately processed into various derived products (primarily digital elevation models), lidar data are publicly available for independent CHM production. The overhead of retrieving and processing these data, however, can be challenging.  We used the USGS 3DEP lidar collection to produce a geographically large, but spatially disparate, CHM dataset. We focused our efforts on United States rangelands, but ensured that other dominant land covers are included. Our dataset comprises 22,796,764 CHM images, each spatially paired with a USDA NAIP image.

Technical Abstract: Canopy height models (CHM) provide detailed environmental vertical structure information and are an important indicator and input for ecological and geospatial applications. These models are often spatiotemporally inconsistent, necessitating additional modeling to scale them in space and time. Yet, such scaling is hindered by a lack of spatially diverse data. To address this, we use United States Geological Survey 3D Elevation Program lidar data to produce 22,796,764 one meter resolution CHM chips, stratified across the dominant land covers of the conterminous United States. For each CHM, we pair a matching time-aligned aerial image from the United States Department of Agriculture National Agriculture Imagery Program. This dataset can be used to train models for large scale CHM production.