Forest Fuel Carbon Maps for CONUS in 2020 and 2022

FuelMap 2020 and 2022 provide imputed 30x30 meter raster maps of carbon stored in forest fuel strata across the conterminous United States. The maps, created by Scott N. Zimmer, model litter, duff, fine woody debris, and coarse woody debris loadings derived from USDA Forest Service FIA plot data using a random forests algorithm. Data for seven distinct carbon pools are presented in GeoTIFF format.

Use Cases

• Model wildland fire behavior and fuel consumption using the litter, duff, and fine woody debris size class loadings.
• Analyze spatial patterns of total carbon in downed woody material across CONUS forests.
• Calibrate regional carbon stock estimates by comparing coarse woody debris (1000-hr size class) loadings with other biomass datasets.
• Study forest disturbance impacts by correlating fuel loadings from the 1-hr, 10-hr, and 100-hr fine woody debris classes with LANDFIRE disturbance history data.

Strengths

• Data is derived from authoritative USDA Forest Service Forest Inventory and Analysis (FIA) program plot measurements.
• Provides 30x30 meter spatial resolution for detailed landscape-level analysis across the entire conterminous United States.
• Includes seven distinct modeled carbon pools: litter, duff, three fine woody debris size classes, coarse woody debris, and a calculated total.
• Uses a random forests machine-learning algorithm for imputation, leveraging a suite of biophysical and topographic predictor variables.

Limitations

• Loadings are imputed values, not direct measurements, introducing model uncertainty, especially where FIA plots lacked downed woody material measurements.
• The dataset's spatial scope is limited to the conterminous United States (CONUS), excluding Alaska, Hawaii, and U.S. territories.
• Data represents snapshots circa 2020 and 2022; it is not a continuously updated time series.

Provenance

Source

USDA Forest Service Forest Inventory and Analysis (FIA) program plot data, imputed using LANDFIRE and Daymet rasters.

Collection Method

Random forests machine-learning algorithm used to impute FIA plots with downed woody material measurements to a 30x30m grid based on forest structure, topography, location, climate normals, and disturbance history.

Time Range

Circa 2020 and 2022.

Freshness

Snapshots circa 2020 and 2022; metadata last updated March 2026.

Geography

Conterminous United States (CONUS).