Chapter 5 - Multivariate Receptor Models

All multivariate receptor models have two fundamental difficulties that must be appreciated if they are to be applied in an intelligent manner. First, the problem of deducing the source compositions and contributions using only the data and natural, physical constraints can be shown to be mathematically indeterminate—that is, there is no unique solution. Secondly, observed correlations between species, which are assumed to bear source composition information may instead the result of the mutual effects of meteorology and coincident source location. This chapter reviews the major multivariate methods, which have been applied to receptor modeling. The Source Apportionment by Factors with Explicit Restrictions (SAFER) model is discussed in the chapter in greater detail than other models because all the others suffer from the fundamental mathematical indeterminacy. The other models continue to be important as semi-quantitative methods to estimate composition of sources from the data alone. All the multivariate methods can be extremely valuable in identifying the existence of unsuspected sources. The chapter introduces the concepts basic to multivariate receptor models and provides a detailed description of the SAFER model and its application to Los Angeles PM-10 data.

[1]  Kazuhiko Ito,et al.  The effects of number of samples and random error on the factor analysis/multiple regression (FA/MR) receptor modeling technique , 1986 .

[2]  Michael T. Kleinman,et al.  Identifying and estimating the relative importance of sources of airborne particulates , 1980 .

[3]  G. M. Hidy,et al.  Observations of aerosols over Southern California coastal waters , 1974 .

[4]  Douglas H. Lowenthal,et al.  Application of the factor-analysis receptor model to simulated urban- and regional-scale data sets , 1987 .

[5]  R. Henry,et al.  Extension of self-modeling curve resolution to mixtures of more than three components: Part 1. Finding the basic feasible region , 1990 .

[6]  John D. Spengler,et al.  A QUANTITATIVE ASSESSMENT OF SOURCE CONTRIBUTIONS TO INHALABLE PARTICULATE MATTER POLLUTION IN METROPOLITAN BOSTON , 1985 .

[7]  K. Rahn,et al.  Silicon and aluminum in atmospheric aerosols: Crust-air fractionation? , 1976 .

[8]  Ronald C. Henry,et al.  Multivariate analysis of particulate sulfate and other air quality variables by principal components—II. Salt Lake City, Utah and St. Louis, Missouri , 1982 .

[9]  D. J. Alpert,et al.  A quantitative determination of sources in the Boston urban aerosol. , 1980, Atmospheric environment.

[10]  G E Gordon,et al.  The use of multivariate analysis to identify sources of selected elements in the Boston urban aerosol. , 1976, Atmospheric environment.

[11]  Ronald C. Henry,et al.  Extension of self-modeling curve resolution to mixtures of more than three components: Part 3. Atmospheric aerosol data simulation studies☆ , 1990 .

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

[13]  Philip K. Hopke,et al.  Multielemental characterization of urban roadway dust , 1980 .

[14]  Irving H. Blifford,et al.  A factor analysis model of large scale pollution , 1967 .

[15]  Sam P. Perone,et al.  Application of pattern recognition and factor analysis for characterization of atmospheric particulate composition in southwest desert atmosphere , 1977 .

[16]  Paul J. Lioy,et al.  Development of a modified factor analysis/multiple regression model to apportion suspended particulate matter in a complex urban airshed , 1987 .

[17]  L. L. Spiller,et al.  Sulfuric acid aerosols and HCl release in coastal atmospheres: Evidence of rapid formation of sulfuric acid particulates , 1980 .

[18]  G. M. Hidy,et al.  Multivariate analysis of particulate sulfate and other air quality variables by principal components-Part I: Annual data from Los Angeles and New York , 1979 .

[19]  Theo. J. Kneip,et al.  The impact of changing air quality on multiple regression models for coarse and fine particle fractions , 1983 .

[20]  Philip K. Hopke,et al.  Target transformation factor analysis as an aerosol mass apportionment method: A review and sensitivity study , 1988 .

[21]  G R Cass,et al.  Source-receptor reconciliation of routine air monitoring data for trace metals: an emission inventory assisted approach. , 1983, Environmental science & technology.

[22]  Donald F. Gatz,et al.  Identification of Aerosol Sources in the St. Louis Area Using Factor Analysis , 1978 .

[23]  Kristian Keiding,et al.  A receptor model for urban aerosols based on oblique factor analysis , 1987 .

[24]  Michael T. Kleinman,et al.  The use of tracers to identify sources of airborne particles , 1980 .

[25]  Edmund R. Malinowski,et al.  Factor analysis of solvent shifts in proton magnetic resonance , 1970 .

[26]  Daniel J. Alpert,et al.  A determination of the sources of airborne particles collected during the regional air pollution study , 1981 .

[27]  P. Solomon,et al.  Chemical Characteristics of PM10 Aerosols Collected in the Los Angeles Area , 1989 .

[28]  Niels Z. Heidam,et al.  The components of the Arctic aerosol , 1984 .

[29]  H. J. Williamson,et al.  Review of receptor model fundamentals , 1984 .

[30]  Kristian Keiding,et al.  Absolute Modelling of Urban Aerosol Elemental Composition by Factor Analysis , 1986 .