Source identification of airborne PM2.5 at the St. Louis-Midwest Supersite



[1] Daily 24-hour integrated PM2.5 (particulate matter ≤2.5 μm aerodynamic diameter) mass and species concentrations were measured at the St. Louis-Midwest Supersite in East St. Louis, Illinois, during the sampling period from June 2001 to May 2003. The PM2.5 speciation data were analyzed using a receptor model, positive matrix factorization (PMF), to identify sources contributing to the observed PM2.5 burdens. Species profiles for the identified sources and their contributions to the observed mass concentration at the receptor were derived from the PMF modeling. These source-specific contributions were then coupled with on-site wind data to identify the directionality of the identified sources which are compared to known point source locations. Overall, ten source categories were resolved (study average contribution to the PM2.5 mass in parentheses): secondary sulfate (33%), carbon-rich sulfate (20%), gasoline exhaust (16%), secondary nitrate (15%), steel processing (7%), airborne soil (4%), diesel emissions/railroad traffic (2%), zinc smelting (1.3%), lead smelting (1.3%), and copper production (0.5%). Temperature-resolved organic and elemental carbon fractions enhanced the source separation between secondary sulfate and carbon-rich sulfate and between gasoline exhaust and diesel emissions/railroad traffic. A major Saharan dust plume observed throughout the Midwestern U.S. in July 2002 was also observed in this study. Overall, about half (48%) of PM2.5 mass concentration measured at this site was distinctly apportioned to secondary sulfate and secondary nitrate. Contributions from distinct primary emissions included local industrial sources (9%), transportation (gasoline/diesel/railroad, 19%), and airborne dust (4%). The remaining 20% of the PM2.5 mass was apportioned to a carbon-rich sulfate factor which is likely an admixture of primary emissions and secondary formation. More work is needed to identify the distinct sources contributing to this factor.