Development of Advanced Factor Analysis Methods for Carbonaceous PM Source Identification and ApportionmentEPA Grant Number: R831078
Title: Development of Advanced Factor Analysis Methods for Carbonaceous PM Source Identification and Apportionment
Investigators: Hopke, Philip K. , Henry, Ronald C. , Spiegelman, C. , Paatero, P.
Institution: Clarkson University
Current Institution: Clarkson University , Texas A & M University , University of Helsinki , University of Southern California
EPA Project Officer: Chung, Serena
Project Period: October 1, 2003 through September 30, 2006 (Extended to December 31, 2006)
Project Amount: $450,000
RFA: Measurement, Modeling, and Analysis Methods for Airborne Carbonaceous Fine Particulate Matter (PM2.5) (2003) RFA Text | Recipients Lists
Research Category: Air , Air Quality and Air Toxics , Particulate Matter
Because controls on precursor gases have reduced sulfate and nitrate formation, carbonaceous particles are becoming a larger fraction of the fine particle mass. Accurate source identification and apportionment will be important for developing effective control strategies for areas found to be out of attainment of the PM2.5 standard. In addition, there is increasing interest in epidemiological studies to relate adverse health effects to apportioned source contributions. Thus, the objective of this project is to combine the best features of the two advanced factor analysis models, UNMIX and Positive Matrix Factorization (PMF), and to test the effectiveness of this improved factor analysis methodology by analysis of the data developed at the various supersites with an emphasis on data from the New York City supersite and other data from New York State.
PMF provides an optimized explicit least squares framework on which to build models. As such it provides for individual data point weights to provide optimum scaling. UNMIX includes an edge-finding routine to define relative elemental concentrations in the source profiles as well as a bootstrapping approach for estimating uncertainties in the resulting source profiles and contributions. By working together, we can combine these approaches into an improved factor analysis approach. One effort will be developing models that can make maximal use of data collected on different time scales (i.e., hourly data from semicontinuous systems and 24-hour integrated filter samples). Particularly with the carbonaceous aerosol where there are important diurnal and weekday/weekend patterns of mobile source traffic, averaging the hourly OC/EC values loses much of the information content of the data. Thus, new data analysis approaches that can fully utilize the data from instruments operating on different time scales are critical to extracting maximal information from the data. Appropriate models will be developed and tested. The rotational ambiguity in the PMF results will be reduced by including the edge information into the analysis. An error analysis that combines the residual rotational ambiguity with the measurement uncertainties for a better estimate of the overall uncertainties in the results. These methods will be exercised using supersite data with an emphasis on the New York City data, but taking advantage of all of the available data.
This project will provide substantially improved factor analysis methodology. In addition the increased time resolution of new analytical methods being employed in the supersite program will provide data that is better suited for factor analysis. Thus, we expect to provide a substantially improved data analysis approach that will be able to take full advantage of the information content of any data set.