Comparison of models used for the calculation of national ammonia emission inventories from agriculture in Europe

Precipitation aids the natural removal of atmospheric pollutants through the process of wet deposition. This process affects the ambient pollutant concentrations along with the amount and chemical composition of ground-level precipitation. Because of the direct linkage between meteorological conditions and types and amounts of atmospheric constituents in precipitation, an accurate representation of meteorological fields is vital for accurate simulations of wet deposition of chemical species by air quality models (AQMs). In this work, performance evaluation is conducted for both meteorological and chemical predictions for August and December 2002 using the Pennsylvania State University (PSU)/National Center for Atmospheric Research (NCAR) Mesoscale Model Generation 5 (MM5)-the U.S. EPA Community Multi-scale Air Quality (CMAQ) modeling system. To understand the uncertainties in simulating precipitation and wet deposition, the evaluation focuses on several most influential parameters including cloud fractions, precipitation, and mass concentrations and wet deposition amounts of sulfate, nitrate and ammonium. Results from the preliminary evaluation of the August baseline simulation show overpredictions of the precipitation, cloud fraction, and wet deposition amounts and underpredictions of the PM concentrations. The December baseline simulation results show overprediction of PM mass concentrations and wet deposition of NH4+ and NO3-, and both over and underprediction of SO42- depending on the observations used in the model evaluation. Both precipitation and cloud fraction are underpredicted by MM5 in December. Sensitivity simulations of MM5 using an alternate cloud microphysics scheme are also performed. The preliminary evaluation of the sensitivity simulation for August 2002 shows an improved model performance for precipitation. Further analysis and additional sensitivity simulations are being conducted to identify major uncertainties in simulating precipitation with MM5 and wet deposition with CMAQ