Mapping Wildfire Fuels with Lidar: LANDFIRE's Canopy Upgrades

By Birgit E. Peterson and Kurtis J. Nelson. USGS Fact Sheet 2015–3086, September 2016.

Why lidar

The Wildfire Sciences Team at the USGS Earth Resources Observation and Science (EROS) Center produces national maps of vegetation type, structure and fuel, mostly through the LANDFIRE program (Landscape Fire and Resource Management Planning Tools). LANDFIRE is kept current by periodic updating and remapping, each a chance to add better data. Lidar (light detection and ranging) measures vegetation in three dimensions, making it ideal for mapping forest structure.

A new canopy-height map for Alaska

A prototype begun in 2010 tested lidar for mapping canopy structure and fuels in the Yukon Flats Ecoregion of interior Alaska:

• Airborne lidar covered part of the ecoregion, with field plots to train and check the results.
• Satellite lidar: to go beyond the airborne survey, samples from GLAS (the Geoscience Laser Altimeter System), a space-borne lidar, trained regression tree models to predict height.
• The result: all of Alaska remapped. The forest-height legend grew from two classes (10 meters or less, and over 10) to five, matching the classes used for the conterminous U.S. A canopy cover remap followed.

LANDFIRE's 2012 Alaska forest canopy height, with the revised five-class legend. Gray marks shrubland, ice, snow, water and other non-forest. Map: USGS.

CHISLIC: lidar for local managers

The Creating Hybrid Structure from LANDFIRE/Lidar Combinations (CHISLIC) tool lets people without lidar experience turn lidar point clouds into useful products and update their local fuel data. It was built with the U.S. Forest Service, the University of Montana and NASA, using algorithms from published research tested in several forest types.

• More detail: compared with LANDFIRE products trained on Landsat imagery, CHISLIC's lidar outputs capture far more of the landscape's structural variety.
• Lots of data to use: about 1,663,000 square kilometers of public airborne lidar had been collected in the U.S.
• Filling gaps: CHISLIC can also bring in GLAS data beyond airborne coverage.

Canopy height in Grand County, Colorado, from LANDFIRE (top) and from the CHISLIC lidar tool (bottom). Maps: USGS.

What's next

3D fuels for fire models: new physics-based fire simulation models need detailed three-dimensional fuel maps. A prototype uses lidar collected in 2012 over parts of Grand Canyon National Park and Kaibab National Forest:

1. 1-meter canopy-height grids derived from the lidar;
2. individual tree stems located within them;
3. canopy height, crown diameter and trunk diameter inferred for each tree;
4. the result built into 3D fuel profiles to test in the models.

A 1-meter canopy-height grid for Grand Canyon National Park and Kaibab National Forest; dots are lidar-inferred tree stems. Map: USGS.

Canopy bulk density: CHISLIC is also being extended to map canopy bulk density — the biomass in a canopy, a key input for modeling crown fires.

Sources

Based on Enhanced Canopy Fuel Mapping by Integrating Lidar Data, by Birgit E. Peterson and Kurtis J. Nelson, USGS Fact Sheet 2015–3086, U.S. Geological Survey; a work of the United States government in the public domain. The three maps are taken from the fact sheet's PDF.