FIRE-MAPS: SAR derived percent moisture content (PMC) for wildland fuel moisture estimation

This dataset provides per-pixel percent moisture content (PMC) derived from polarimetric radar backscatter. The data is obtained from the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) instrument between 2023-07-27 and 2023-10-10 and the Sentinel-1 satellite between 2023-07-28 and 2023-10-13 before and during a prescribed burn in Fishlake National Forest, Utah. Radiometric terrain correction (RTC) was applied to the polarimetric backscatter data to remove unreliable areas under terrain shadow. Additional urban masking was applied to the Sentinel-1 data. Field samples of PMC were collected before the prescribed burn, and used as training data with coincident UAVSAR data to train a deep neural network model to generate fuel moisture products for wildland fuels management. All files are distributed in orthorectified GeoTIFF format. For detailed methodology, see (pending) publication: An et al. (2026, August) Fuel Moisture Estimation Using Polarimetric SAR in Deep Learning Model for Fishlake National Forest, UT In IGARSS 2026-2026 IEEE International Geoscience and Remote Sensing Symposium (pp. pending). IEEE. The field samples used for training were provided by Andrew T. Hudak (U.S. Forest Service) and Carlos Alberto Silva (University of Florida) under the Joint Fire Science Program, “#22-2-02-15 EMS4D: Multi-Scale Fuel Mapping and Decision Support System for the Next Generation of Fire Management”. UAVSAR imagery was acquired as part of NASA’s FireSense campaign, courtesy of NASA/JPL-Caltech (www.uavsar.jpl.nasa.gov), and can be accessed through the Alaska Satellite Facility (ASF). The Sentinel-1 data are courtesy of ESA and the NASA JPL-Caltech ARIA/OPERA team, and can be accessed through ASF Vertex.