Improvement of summertime surface ozone prediction by assimilating Geostationary Operational Environmental Satellite cloud observations

Clouds play an important role in the Earth's climate system since they can affect various physical and chemical processes within the atmosphere. Misplacement of clouds is a major source of error in the numerical weather prediction (NWP) models, and it also impacts the accuracy of air quality simulations since the meteorology and air quality are directly coupled. In this study, a cloud assimilation technique was utilized to improve cloud placement within the Weather Research and Forecasting (WRF) model by assimilating Geostationary Operational Environmental Satellite (GOES)-derived cloud products. Meteorological outputs from the WRF model were then used as inputs for the Community Multiscale Air Quality (CMAQ) model. The impact of cloud assimilation on air quality was tested over the June-September 2016 period. The results indicated that, by modifying model clouds, cloud assimilation corrected surface solar radiation and photochemical reaction rates, altered light sensitive biogenic emissions, adjusted horizontal transport and vertical mixing, and finally improved the prediction of surface ozone concentration. Cloud assimilation improved daytime surface ozone prediction over most of the U.S. domain, with exceptions in California. On average, cloud assimilation improved the prediction of daytime peak ozone and reduced bias by 47% (similar to 1.5 ppb). The largest improvement was seen over the southeast U.S. region (similar to 2.6 ppb reduction in daytime peak ozone), where convective clouds are more frequent and transient and biogenic volatile organic compound (VOC) emissions are more intense than elsewhere.