The impact of urban areas on regional aerosol concentrations and visibility is an important air quality issue in the western United States. To examine the relationship of urban emissions and regional air quality in the Pacific Northwest, we completed an analysis of potential source regions which may impact the Columbia River Gorge National Scenic Area. The assessment involved the application of a footprint modeling system to a selection of ten sample days from 1997-98 in which high aerosol loadings were observed at two IMPROVE sites. In addition, the Community Multi-Scale Air Quality (CMAQ) model was used to examine aerosol concentration patterns within the region and at the two sites for a two-day period in July 1998.

The footprint assessment involved the MM5/CALMET/CALPUFF modeling system applied in backward trajectory mode to develop an upwind probability source distribution. The probability source distribution was then overlaid with gridded emission inventory data to create a source footprint which defines the 24-hr fractional source contribution area on pollutant concentrations for the two IMPROVE sites. Two footprints were defined for each site on each modeling day, where one footprint is that of an inert pollutant and the other is for a pollutant undergoing first order transformation. Emission inventories of CO and SO2 were used as surrogates for the inert and chemically reactive cases, respectively. It was shown that chemistry effects play an important role in defining the footprint and results in increasing the importance of emissions further from the receptor. It was also shown that the Portland region can be a significant source of pollutants into the Columbia River Gorge, but that the Puget Sound, Yakima, and Tri Cities areas are either significantly far away or are too small of a source area to contribute to the pollutant concentrations at the IMPROVE sites.

CMAQ model results show that for both the fine mass budget and visibility impairment budget, organic carbon is the greatest contributor and sulfate is the next largest contributor. Spatially, the highest PM2.5 concentrations and visibility impairment occurred in Portland, along the west slope of the Cascades, and within the Seattle region. At both sites, the PM2.5 temporal pattern closely matches the variations in organic carbon concentrations, which increase during the night and decrease during the day. Statistical data show that CMAQ performed best for predicting PM2.5 and the extinction coefficient, but that the model over-predicted organic carbon and under-predicted sulfate and nitrate.