Use of hourly Geostationary Operational Environmental Satellite (GOES) fire emissions in a Community Multiscale Air Quality (CMAQ) model for improving surface particulate matter predictions

[1] Large changes in surface‐level PM2.5 concentrations and columnar aerosol optical thickness (AOT) were observed downwind of fires that originated in Georgia and Florida during the April–May 2007 period. In order to quantify the impacts of these wildfires on particulate matter air quality, Community Multiscale Air Quality (CMAQ) simulations were conducted by adding hourly fire emissions derived from the Geostationary Operational Environmental Satellite (GOES) imagers. The simulations include ambient aerosols by accounting for background emissions using the Sparse Matrix Operator Kernel Emissions (SMOKE) model. The impacts of fire emissions are obtained by comparing the CMAQ simulations with and without fire emissions. Overall, the CMAQ‐derived PM2.5 reproduces the major smoke transport features of the Moderate Resolution Imaging Spectrometer (MODIS) AOT, but is systematically lower than the ground‐based observations of PM2.5 mass concentrations during the fires. An increase of the satellite‐ derived fire emissions improves the simulated magnitude of PM2.5 concentrations. We also show that the disagreement between the CMAQ predictions and ground‐based observations during the high PM2.5 episodes occurs when the predicted position of fire plume is not accurately located. The smoke position error grows more rapidly due to drift behavior of model wind error, so the position error dominates the accuracy of site‐specific CMAQ PM2.5 predictions.