Subdaily Variation in Aerosol Observations and Models, and Impacts of Geostationary Aerosol Data Assimilation

As the spatial resolution of global weather models has increased, variation on sub-daily time scales has become a larger factor in model error. Sub-daily variation can be very large for many processes important to forecasting of aerosol-related impacts on visibility, atmospheric forcing, and air quality. Many aerosol sources including open burning, wind-driven dust, and anthropogenic aerosols have strong diurnal cycles, and aerosol transport in many regions has been demonstrated to vary systematically with time of day.

Taking advantage of recently available aerosol optical depth retrievals from geostationary satellites, we can examine the impact of these phenomena on the observed variation in both satellite data and model forecasts. Based on this analysis of variability, we can construct hypotheses about the impact of geostationary aerosol data assimilation on forecasting of aerosol optical depth. Those hypotheses can then be tested through data assimilation experiments using the Navy Aerosol Analysis and Prediction System (NAAPS).

Our analysis focuses on sub-daily variation of aerosol observed during spring and summer of 2016 using geostationary AOD retrievals from the Geostationary Ocean Color Imager (GOCI) on board the Korean COMS-1 satellite as well as the Advanced Himawari Imager (AHI) on the Japanese Himawari-8 satellite. We examine conditions of greater and lesser sub-daily variation in aerosol optical depth, and compare the variation observed by the satellite to that forecast by the model. Finally, we perform experimental assimilation of geostationary AOD products and analyze the impact of this assimilation on the ability of the forecast to capture sub-daily variation in AOD.