Fuel2Fire, Enhancing Wildland Fire Emissions Estimates: The devil is in the detail and defined by the purpose of the emission estimates

Fire plays a significant role on both national and global scales, profoundly impacting landscapes shaped by human activity as well as those left wild. Even though fire can be devastating, wildland fire is a natural and integral force on our landscapes. Fires can also serve to reduce fuels to mitigate wildfire risk and maintain healthy ecosystem functions. However, the smoke produced by fires, regardless of their size or purpose, can pose adverse effects on human health when inhaled.

The focus of this work is to investigate the parameters used to estimate wildland fire emissions and to enhance these estimates. We consider three areas for consideration: estimating fuels (ground to satellite); quantifying small fires from space; and the diurnal fire cycle. We developed a comprehensive national- and global-fuel consumption database that connects distinct fuel estimates to fire danger. This database was developed using Fuel Characteristic Classification System (FCCS) biomass parameters (nationally at 30 meter) and using the Pettinari et al. (2015) FCCS-based Global Fuelbed database (300 meter). Consumption was estimated for five Fire Danger categories (Low, Moderate, High, Very High, Extreme), for both ‘new’ and ‘residual’ burning scenarios. The ability of satellite data (MODIS, VIIRS, GOES) to define small fires was statistically defined for each instrument in areas of the southeastern U.S. Then, the diurnal fire, hence emissions, cycle was developed using GOES data, and these data will be shown for individual fires. We implemented the fuel and diurnal cycle approaches during the 2019 Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) campaign to estimate daily carbon emissions. Our emissions estimates were rigorously compared against in-situ measurements of CO2, CO, and black carbon aerosols, revealing a robust agreement between the two datasets.

Our data and results are presented as separate suggested enhancements, with the understanding that depending on the purpose of the wildland fire emissions model, these enhancements may or may not be applicable or feasible. Our estimates are not as accurate as ground-based estimates at a small scale, although they compare well. At the other end of the spectrum, global large-scale 1-degree models require emission overestimates, due to horizontal and vertical dispersion, to maintain the reality of smoke transport. Some of the disparity between emissions models has more to do with the intended purpose of a model, than accuracy and precision, while other models require both accuracy and precision.