UCLA

Abstract Wildfire emission inventories exhibit large variability that complicates assessments of smoke impacts. Here we compare fuel consumed (in mass per burned area units) from multiple burn area‐based and energy‐based approaches for fires in the western US during 2020. Average fuel consumed can vary by up to factors of 2–16 between approaches across burn severity classes and fuel types. Fuel consumed estimates typically increase with burn severity, except for the energy‐based approaches for forest land cover, where it decreases for high burn severity. Also, in contrast to other approaches, energy‐based estimates decrease for tree cover greater than 40% regardless of burn severity class. This implies that corrections to the energy‐based approach are likely needed across burn severity categories to account for canopy and smoke shading. The methodological recommendations provided would likely result in greater consistency between wildfire emission estimates and highlight the need to better constrain fuel loading and consumption. Plain Language Summary Estimates of smoke emitted into the atmosphere can be highly variable, which can result in uncertainties when estimating smoke impacts. Fuel consumption is one of the primary drivers for this variability. Here, we compare different methods to estimate fuel consumption for fires in the western US in 2020. We find that the different methods result in large variability in fuel consumed estimates, with a factor of 10 or larger differences commonly found. For grasslands, shrublands and savannas, all estimation methods show an increase in fuel consumed as the fire burn severity increases, which is consistent with the expectation that the more severe the fire, the more fuel is consumed. However, approaches based on observing the energy radiated from the fire and relating that to fuel consumed shows the opposite tendency over forests, especially for areas with dense tree cover, which is against expectations. This likely indicates that corrections to account for the energy that cannot penetrate the canopy or that gets absorbed by the combustion gases need to be implemented. We provide recommendations to improve estimates of fuel consumption that we expect will reduce the uncertainty in smoke emission estimates. Key Points Large differences (factors 2–16 on the mean) in fuel consumed for emissions estimates across burn severity classes are explained Fuel consumed increases with burn severity across all approaches except for the energy‐based approach for forest cover Fire radiative power corrections that account for canopy and smoke shading during extreme fire behavior likely need to be developed