Control strategies for the reduction of airborne particulate nitrate in California's San Joaquin Valley

Abstract

Abstract The effect of NO x , volatile organic compound (VOC), and NH 3 emissions control programs on the formation of particulate ammonium nitrate in the San Joaquin Valley (SJV) was examined under the typical winter conditions that existed on 4–6 January, 1996. The UCD/CIT photochemical transport model was used for this study so that the source origin of primary particulate matter and secondary particulate matter could be identified. When averaged across the entire SJV, the model results predict that 13–18% of the reactive nitrogen (NO y =NO x +reaction products of NO x ) emitted from local sources within the SJV was converted to nitrate at the ground level. Each gram of NO x emitted locally within the SJV (expressed as NO 2 ) produced 0.23–0.31 g of particulate ammonium nitrate (NH 4 NO 3 ), which is much smaller than the maximum theoretical yield of 1.7 g of NH 4 NO 3 per gram of NO 2 . The fraction of reactive nitrogen converted to nitrate varied strongly as a function of location. Urban regions with large amounts of fresh NO emissions converted little reactive nitrogen to nitrate, while remote areas had up to 70% conversion (equivalent to approximately 1.2 g of NH 4 NO 3 per gram of NO 2 ). The use of a single spatially averaged ratio of NH 4 NO 3 /NO x as a predictor of how changes to NO x emissions would affect particulate nitrate concentrations would not be accurate at all locations in the SJV under the conditions studied. The largest local sources of particulate nitrate in the SJV were predicted to be diesel engines and catalyst equipped gasoline engines under the conditions experienced on 6 January, 1996. Together, these sources accounted for less than half of the ground-level nitrate aerosol in the SJV. The remaining fraction of the aerosol nitrate originated from reactive nitrogen originally released upwind of the SJV. The majority of this upwind reactive nitrogen was already transformed to nitrate by the time it entered the SJV. The effect of local emissions controls on this upwind material was small. A 50% reduction in NO x emissions applied to sources within the SJV reduced the predicted concentration of total nitrate by approximately 25% during the study episode. VOC emissions controls were less effective, while reasonable NH 3 emissions controls had the smallest effect on the amount of ammonium nitrate produced. A 50% reduction in VOC emissions lowered predicted concentrations of total nitrate by 17.5%, while a 50% reduction in NH 3 emissions lowered predicted concentrations of total nitrate by only 10%. This latter result is expected since the formation of ammonium nitrate aerosol is limited by the availability of gas-phase nitric acid, with large amounts of excess NH 3 available. NO x emissions controls appear to be the most efficient method to reduce the concentration of locally generated particulate nitrate in the SJV under the conditions experienced on 4–6 January, 1996.