Source Identifications of Airborne Fine Particles Using Positive Matrix Factorization and U.S. Environmental Protection Agency Positive Matrix Factorization

Abstract The widely used source apportionment model, positive matrix factorization (PMF2), has been applied to various air pollution data. Recently, U.S. Environmental Protection Agency (EPA) developed EPA positive matrix factorization (PMF), a version of PMF that will be freely distributed by EPA. The objectives of this study were to conduct source apportionment studies for particulate matter less than 2.5 μm in aerodynamic diameter (PM2.5) speciation data using PMF2 and EPA PMF (version 1.1) and to compare identified sources between the two models. In the present study, ambient PM2.5compositional datasets of 24-hr integrated samples collected at EPA Speciation Trends Network monitoring sites in Chicago, IL, and Portland, OR, were analyzed. Both PMF2 and EPA PMF extracted eight sources for the Chicago data and 10 sources for the Portland data. The model-resolved source profiles were similar between two models for both datasets. However, in several sources, the average contributions did not agree well and the time series contributions were not highly correlated. The differences between PMF2 and EPA PMF solutions were caused by the different least-square algorithm and the different nonnegativity constraints. Most of the average source contributions resolved by both models were within 5–95% uncertainty provided by EPA PMF, indicating that the sources resolved by both models were reproducible.

[1]  Charles L. Lawson,et al.  Solving least squares problems , 1976, Classics in applied mathematics.

[2]  W. J. Mitchell,et al.  East versus West in the US: Chemical Characteristics of PM2.5 during the Winter of 1999 , 2001 .

[3]  Investigation of sulfate and nitrate formation on mineral dust particles by receptor modeling , 2005 .

[4]  Philip K. Hopke,et al.  Incorporation of parametric factors into multilinear receptor model studies of Atlanta aerosol , 2003 .

[5]  Philip K Hopke,et al.  Source Apportionment of Fine Particles in Washington, DC, Utilizing Temperature-Resolved Carbon Fractions , 2004, Journal of the Air & Waste Management Association.

[6]  J. Schwartz,et al.  Hierarchical bivariate time series models: a combined analysis of the effects of particulate matter on morbidity and mortality. , 2004, Biostatistics.

[7]  Philip K. Hopke,et al.  Sources of fine particle composition in the northeastern US , 2001 .

[8]  J. Brook,et al.  Identification of the major sources contributing to PM2.5 observed in Toronto. , 2003, Environmental science & technology.

[9]  R. Henry.  Multivariate receptor modeling by N-dimensional edge detection , 2003 .

[10]  P. Paatero.  The Multilinear Engine—A Table-Driven, Least Squares Program for Solving Multilinear Problems, Including the n-Way Parallel Factor Analysis Model , 1999 .

[11]  P. Hopke,et al.  Analysis of ambient particle size distributions using Unmix and positive matrix factorization. , 2004, Environmental science & technology.

[12]  R. Cary,et al.  Elemental Carbon-Based Method for Monitoring Occupational Exposures to Particulate Diesel Exhaust , 1996 .

[13]  P. Paatero.  Least squares formulation of robust non-negative factor analysis , 1997 .

[14]  P. Hopke,et al.  Estimation of Organic Carbon Blank Values and Error Structures of the Speciation Trends Network Data for Source Apportionment , 2005, Journal of the Air & Waste Management Association.

[15]  Suilou Huang,et al.  Testing and optimizing two factor-analysis techniques on aerosol at Narragansett, Rhode Island , 1999 .

[16]  Sheldon Landsberger,et al.  Atmospheric aerosol over Finnish Arctic: source analysis by the multilinear engine and the potential source contribution function , 2003 .

[17]  P. Hopke,et al.  Source Identification of Atlanta Aerosol by Positive Matrix Factorization , 2003, Journal of the Air & Waste Management Association.

[18]  Philip K. Hopke,et al.  Discarding or downweighting high-noise variables in factor analytic models , 2003 .

[19]  P. Hopke,et al.  Identification of Fine Particle Sources in Mid-Atlantic US Area , 2005 .

[20]  Ronald C. Henry,et al.  Current factor analysis receptor models are ill-posed , 1987 .

[21]  J. Lewtas,et al.  Source apportionment of PM2.5 at an urban IMPROVE site in Seattle, Washington. , 2003, Environmental science & technology.

[22]  Philip K. Hopke,et al.  Identification of Sources of Phoenix Aerosol by Positive Matrix Factorization , 2000, Journal of the Air & Waste Management Association.

[23]  Antonella Zanobetti,et al.  The concentration-response relation between PM(2.5) and daily deaths. , 2002, Environmental health perspectives.

[24]  P. Hopke,et al.  Comparison of Positive Matrix Factorization and Multilinear Engine for the source apportionment of particulate pollutants , 2003 .

[25]  Philip K. Hopke,et al.  Identification of Source Nature and Seasonal Variations of Arctic Aerosol byPositive Matrix Factorization , 1999 .

[26]  Philip K. Hopke,et al.  Factor Analysis of Seattle Fine Particles , 2004 .

[27]  G A Norris,et al.  Associations between air pollution and mortality in Phoenix, 1995-1997. , 2000, Environmental health perspectives.

[28]  P. Paatero,et al.  Investigation of sources of atmospheric aerosol at urban and suburban residential areas in Thailand by positive matrix factorization , 2000 .

[29]  Philip K. Hopke,et al.  The use of constrained least-squares to solve the chemical mass balance problem , 1989 .

[30]  J. Sarnat,et al.  Assessing the Relationship between Personal Particulate and Gaseous Exposures of Senior Citizens Living in Baltimore, MD , 2000, Journal of the Air & Waste Management Association.

[31]  P. Paatero,et al.  Application of positive matrix factorization in source apportionment of particulate pollutants in Hong Kong , 1999 .

[32]  J. Schwartz,et al.  Association of fine particulate matter from different sources with daily mortality in six U.S. cities. , 2000, Environmental health perspectives.

[33]  P. Hopke,et al.  Atmospheric aerosol over Vermont: chemical composition and sources. , 2001, Environmental science & technology.

[34]  Kehinde O. K. Oduyemi,et al.  Comparative testing of PMF and CFA models , 2002 .

[35]  Philip K. Hopke,et al.  Utilizing wind direction and wind speed as independent variables in multilinear receptor modeling studies , 2002 .