A University of Florida research team spent two weeks in June and July 2026 collecting paired field measurements and ground-based lidar data across recently burned landscapes and a range of forest conditions in Fish Lake National Forest, Utah, and North Kaibab National Forest, Arizona. The campaign combined terrestrial and mobile laser scanning with conventional stand inventory and clip-plot fuel sampling in order to calibrate models of surface, ladder, and canopy fuels. The resulting datasets feed EMS4D, a Joint Fire Science Program project developing an open-source system for mapping fuel structure and fuel consumption across the western United States.
Fuel maps constitute a foundational input to wildfire management. Fire behavior modeling, the design of fuel treatment prescriptions, and the allocation of thinning and suppression resources all depend on estimates of fuel loading at a given location and of how that fuel is distributed vertically through the stand. Such estimates have conventionally been derived by extrapolating measurements from a limited network of field plots across large and heterogeneous landscapes, an approach that introduces substantial uncertainty and produces map products that become outdated as vegetation growth and disturbance alter the underlying conditions.
A recent field campaign from researchers at the University of Florida’s School of Forest, Fisheries, and Geomatics Sciences is aimed squarely at that problem. The team led by the Forest Biometrics, Remote Sensing and Artificial Intelligence Laboratory, known as the SilvaLab, collected field and lidar data from June 17 to July 2 across Fish Lake National Forest in Utah and North Kaibab National Forest in Arizona.
Lidar x Clip Plots
The project paired conventional forest inventory with ground-based lidar at every plot. Crews recorded tree species, diameter at breast height, total height, crown characteristics, and crown base height, alongside surface, ladder, and canopy fuel measurements, including 1×1 meter surface fuel “clip plots”. In parallel, they acquired high-resolution terrestrial laser scanning (TLS) and mobile laser scanning (MLS) datasets to capture the three-dimensional structure of post-fire forest ecosystems.
Standard inventory metrics fail to accurately estimate crown base height and canopy bulk density, even though these two fundamental fuel variables are the primary drivers of crown fire initiation and spread. In contrast, lidar point clouds can map these attributes directly. The 1×1 clip plots provide the reference measurements. Crews cut and remove all vegetation within each square and weigh it, producing a direct physical measurement of fuel mass. The lidar sensors record the three-dimensional geometry of the same locations, and the models are trained on the correspondence between the two sets of observations.
The team’s stated purpose for the data is to quantify fuel consumption following prescribed fire and wildfire, improve characterization of fuel dynamics, and develop and validate next-generation AI models for scalable mapping of forest structure, surface fuels, ladder fuels, canopy fuels, fuel consumption, and fire behavior across the western United States.
Ecosystem Monitoring System for 4D Fuel Mapping and Decision Support (EMS4D)
The Ecosystem Monitoring System for 4D Fuel Mapping and Decision Support (EMS4D) is led by Carlos Alberto Silva of University of Florida and Andrew Hudak of the USDA Forest Service Rocky Mountain Research Station.
The project’s ambition is an open-source monitoring system that fuses field observations, terrestrial and mobile laser scanning, airborne and satellite lidar, artificial intelligence, and cloud computing into dynamic maps of vegetation structure and forest fuels. The purpose of the “4D” classification is to provide fuel layers that monitor structural alterations over time, allowing managers to assess if a treatment successfully achieved its objectives. The 2026 datasets will contribute directly to the EMS4D platform, adding capabilities for mapping fuels, estimating consumption, and monitoring ecosystem change over time.
The UF crew included Silva as principal investigator, postdoctoral researchers Inacio Bueno (field crew lead), Cesar Alvites, and Nadeem Fareed, lab manager Ana Terra, master’s student Alex Gaskins, undergraduate Simon Caldwell, and research assistant Durga Siva Deepak. Hudak joined as co-principal investigator, along with collaborators Mickey Campbell and his master’s student Jake Howell of the University of Utah, Manuel Gomez Roux of the University of Valladolid, and Kathleen Clough of the Desert Research Institute.
Source material
- EMS4D publication: “Comparing Terrestrial and Mobile Laser Scanning Approaches for Multi-Layer Fuel Load Prediction in the Western United States,” Remote Sensing 17(16), 2757. https://www.mdpi.com/2072-4292/17/16/2757
- Companion perspective paper: Leite et al., “Leveraging the next generation of spaceborne Earth observations for fuel monitoring and wildland fire management,” Remote Sensing in Ecology and Conservation. https://zslpublications.onlinelibrary.wiley.com/doi/10.1002/rse2.416. Co-authored by Silva and Hudak, covering the satellite tier that EMS4D scales its ground-based lidar into.
- Adjacent federal work: USDA Forest Service RMRS on lidar for fuel and consumption measurement across scales. https://research.fs.usda.gov/rmrs/articles/lidar-data-capture-fuel-and-consumption-across-scales
