Assimilation of VIIRS Aod To Improve Smoke Forecasts Over The Western United States

Abstract

According to US Bureau of Land Management wildfires cost taxpayers millions of dollars that are spent on extinguishing fires, repairing damages to property and ecosystem, and healthcare due to air and water pollution. Prediction of smoke dispersion helps mitigating effects of poor air quality on populations. In this work we examine quality of simulations of smoke using WRF-Chem over the Western U.S. during 2016 and 2017 fire seasons. For a comparison two simple aerosol parameterizations, GOCART and Smoke, are employed. Next, 3D-Var assimilation of MODIS and VIIRS products, AOD at 550 nm and smoke masks, is introduced. Verification statistics are presented to assess performance of the model and the impact of the assimilation.