Department Seminars

General Information

When: Thursdays from 4:00–4:50 p.m.
Where: 1170 TMCB

Jay R. Turner
Washington University in St. Louis
Department of Chemical Engineering

2007-10-18

Apportioning Urban Ambient Air Particulate Matter Burdens to Emission Sources

Analytical tools are available to identify and quantify emission source contributions to measured ambient particulate matter concentrations. While these tools have been applied to data sets from numerous monitoring locations in the United States, the results and their interpretation are being more carefully scrutinized because regulatory agencies with jurisdiction over areas violating the fine particulate matter National Ambient Air Quality Standard (NAAQS) must develop emission control strategies to meet attainment deadlines. Current limitations of chemical transport models, which are the de facto tools for urban ozone air quality management, has led to a greater reliance on receptor-based models for performing ambient particulate matter source attribution. Our group has participated in source apportionment analyses for several Midwest cities including detailed work on St. Louis. Receptor modeling studies have provided substantial information on source-receptor relationships, but our current knowledge is unsatisfying from the perspective of developing rationale emission control strategies. While we continue to advance our state-of-knowledge about these complex systems, we often struggle to be responsive to the key questions asked by air quality managers. I will briefly present a conceptual model for ambient fine particulate matter over St. Louis, and subsequently show results from a series of data analysis efforts. Strengths and weaknesses of the various analyses will be highlighted, with emphasis on whether the results "pass the laugh test" and are consistent across analyses. Ideally this seminar will provide a context for identifying opportunities for statisticians to significantly contribute to this field. For example, can the strengths and limitations of existing tools be can formally stated to guide their proper use and to qualify results; can analysis methods be developed which capitalize on the structure and content of current data sets (or are these data sets simply too "flawed"); can novel methods, unconstrained by the limitations of existing data sets, be conceptualized and developed (which in turn could be used to guide future data collection activities); and can methods be developed which adapt to our evolving knowledge of the system being studied. We are particularly interested in approaches that directly inform the key questions we are asked by air quality managers.