Evaluation and improvement of ammonia emissions inventories

Two case studies are performed to improve ammonia emissions inputs used to model fine particulate matter (PM2.5 is the portion of particulate matter smaller than 2.5 mm aerodynamic diameter) formation of ammonium sulfate and ammonium nitrate. Ammonia emissions are analyzed in detail for North Carolina and the San Joaquin Valley (SJV) of California, with a focus on the Charlotte, NC, and Fresno, California metropolitan areas. A new gridded ammonia emissions inventories suitable for atmospheric modeling for the two case study cities was also developed. Agricultural sources accounted for the bulk of ammonia emissions in both case studies. Livestock waste contributed about 80% in North Carolina and 64% in the SJV, while fertilizer application contributed about 6–7% in both domains. Forests and non-agricultural vegetation contributed 5% in North Carolina and 12% in the SJV. Motor vehicles accounted for about 6% of ammonia emissions in North Carolina and 14% in the SJV. In the Charlotte and Fresno urban areas, the distribution of emissions is less heavily weighted toward agricultural sources and more heavily weighted toward highway vehicles (highway vehicles account for an estimated 64% of emissions in Charlotte and 51% of emissions in Fresno). The emissions estimates for agricultural sources (livestock and fertilizer application) decline to approximately 14% in the winter for both the Charlotte and Fresno urban areas. Emissions estimates for soils and vegetation also decline to approximately 0 during the winter for both the Fresno and Charlotte area. As a result, motor vehicles account for a larger fraction (approximately 73% and 70% for Charlotte and Fresno, respectively) of winter ammonia emissions, particularly in the Charlotte urban area. r 2003 Elsevier Ltd. All rights reserved.