The impact of changing nitrogen oxide emissions on wet and dry nitrogen deposition in the northeastern USA

This study is an attempt to quantify the relation between changes in NOx emissions and nitric acid (HNO3) in the northeastern USA. From this relation, and previous work relating NOx emission changes and wet NO3- deposition, we can estimate how changing NOx emissions may impact total (wet + dry) measured nitrogen (N) deposition. Electric utility emissions account for 1/4, and vehicle emissions account for over 1/2 of the total NOx emissions in the eastern USA. Canadian NOx emissions from the seven easternmost provinces (Manitoba and east) represent less than 10% (1.2 teragrams (Tg) NOx) of the NOx emissions compared with those from the eastern USA. Emissions from eastern Canada are dominated by vehicle NOx emissions, which account for 2/3 of the total NOx emissions from eastern Canada. Data from the EPA National Emissions Inventory show, for the period 1991-2001, that nitrogen oxide (NOx) emissions in the eastern USA have declined by 17% to from 16.1 to 13.1 Tg. Large declines in vehicle emissions in 2001 may be questionable. If 2001 data are excluded the decline in total NOx is only 7%. A recent assessment of EPA's emissions estimates suggest that vehicle NOx emissions may be underestimated, and total NOx emissions reductions may be less than what is reported by the EPA. The CASTNet (Clean Air Status and Trends Network) measurements of N dry deposition include HNO3, particulate NO3- and NH4+. The dominant N dry deposition product measured is HNO3, which represents 80% of measured N dry deposition for the sites used in this study. Amounts of NH3, NO2, organic nitrate and PAN dry deposition are not measured by CASTNet. The NH3 and NO2 deposition are probably significant, and may be major N dry deposition components in some areas. Random coefficient models with total NOx emissions as the independent variable, and HNO3 concentrations as the dependent variable, show that reducing total NOx emissions by 50% should reduce HNO3 concentrations by 36%. The average efficiency (the ratio of % change in HNO3 to % change in NOx emissions) is 72%. Random coefficient models with non-vehicle NOx emissions as the independent variable, and HNO3 concentrations as the dependent variable, show a 50% decline in non-vehicle NOx emissions (which is a 23% decline in total NOx emissions) should reduce HNO3 by 17-20%. The average efficiency in this case is 81 %. Because non-vehicle NOx emissions data are more reliable than vehicle NOx emissions, non-vehicle NOx models are likely more accurate than the total NOx models.