Back-trajectory-based source apportionment of airborne sulfur and nitrogen concentrations at Rocky Mountain National Park, Colorado, USA

The Rocky Mountain Atmospheric Nitrogen and Sulfur Study (RoMANS), conducted during the spring and summer of 2006, was designed to assess the sources of nitrogen and sulfur species that contribute to wet and dry deposition and visibility impairment at Rocky Mountain National Park (RMNP), Colorado. Several source apportionment methods were utilized for RoMANS, including the Trajectory Mass Balance (TrMB) Model, a receptor-based method in which the hourly measured concentrations are the dependent variables and the residence times of back trajectories in several source regions are the independent variables. The regression coefficients are estimates of the mean emissions, dispersion, chemical transformation, and deposition between the source areas and the receptors. For RoMANS, a new ensemble technique was employed in which input parameters were varied to explore the range, variability, and model sensitivity of source attribution results and statistical measures of model fit over thousands of trials for each set of concentration measurements. s showed that carefully chosen source regions dramatically improved the ability of TrMB to reproduce temporal patterns in the measured concentrations, and source attribution results were also very sensitive to source region choices. Conversely, attributions were relatively insensitive to trajectory start height, trajectory length, minimum endpoints per source area, and maximum endpoint height, as long as the trajectories were long enough to reach contributing source areas and were not overly restricted in height or horizontal location. Source attribution results estimated that more than half the ammonia and 30–45% of sulfur dioxide and other nitrogen-containing species at the RoMANS core site were from sources within the state of Colorado. Approximately a quarter to a third of the sulfate was from within Colorado.