Advancing National Air Quality Forecasts through Emission Data Assimilation (Invited Presentation)

The World Health Organization has estimated that outdoor exposure to air pollution caused 4 million premature deaths per year, making it the single largest environmental risk today. In the United States, over one-third of the population lives in areas that do not meet the health-based National Ambient Air Quality Standards for ozone or fine particles (PM_2.5). To mitigate the widespread health risks, regional and national air quality forecasting systems, including the National Air Quality Forecast Capability (NAQFC) operated by US National Weather Service, were established to provide communities with forecast pollution levels up to 48 to 72 hours lead time. These forecasts provide critical information so that health authorities can take actions to protect sensitive groups with early warnings and other mitigation measures. We present here recent progress in using satellite data to improve the NAQFC forecasts, with a focus on reducing emission uncertainties through a new technique called emission data assimilation (EDA). Emissions data is a key input to and also a large source of uncertainty in air quality forecasting. EDA assimilates satellite and ground observations into emission modeling processes to reduce emission uncertainties. A suite of EDA algorithms have been developed, including 1) using OMI NO₂ to update anthropogenic NO_x emission inventories; 2) using OMPS/OMI SO₂ retrievals to constrain volcanic emissions; 3) using MODIS and VIIRS fire observations to estimate fire emissions, and 4) using MODIS black-sky albedo to improve windblown dust emissions. In this work, we demonstrate how these EDA algorithms affect air quality model performance through evaluation of model outputs with satellite, aircraft, and ground observations.