A wildland fire emission inventory for the western United States - uncertainty across spatial and temporal scales

Abstract

Emissions of trace gases and aerosols by biomass burning (BB) have a significant influence on the chemical composition of the atmosphere, air quality, and climate. BB emissions depend on a range of variables including burned area, fuels, meteorology, combustion completeness, and emission factors (EF). Emission algorithms provide BB emission inventories (EI) which serve as critical input for Chemical Transport Models (CTM) employed in atmospheric sciences in a wide array of studies. Many different BB EI are commonly used and agreement among these EI is often poor. In general, the sensitivity of the emission estimates to the algorithm components is not well characterized and the performance of most algorithms have not been examined across the scales they are used. Understanding the sensitivity of EI to algorithm component uncertainties is crucial for assessing their impact on CTM simulations. We examine the spatial and temporal sensitivity of BB emission estimates of CO to uncertainties in mapped fuel loading, fuel consumption, burned area and emission factors. The study focuses on wildland fire in the western United States(2003 - 2008). Two fuel loading maps and 2 fuel consumption models provided 4 fuel load consumption emission scenarios with identical burned area and meteorology. The burned area used in the study was mapped using a MODIS burn scar algorithm.

Citation

Urbanski, S. P.; Hao, W. 2010. A wildland fire emission inventory for the western United States - uncertainty across spatial and temporal scales. American Geophysical Union. Fall Meeting: Abstract #A21B-0069.