Multivariate analysis of particulate sulfate and other air quality variables by principal components-Part I. Annual data from Los Angeles and New York

摘要: A multivariate statistical analysis of particulate sulfate, meteorological and air quality data is presented for the Los Angeles New York City areas. The are from Community Health Air Monitoring Program (CHAMP). Other aerometric parameters gathered at nearby sites were assembled with to produce a comprehensive set. highly intercorrelated nature many variables, oxidant maximum temperature, example, can make results ordinary regression techniques uncertain. application Principal Component Analysis (PCA) set (not including sulfate) produces statistically independent linear combinations original variables. sulfate values then regressed on these parameters. It found that only three components necessary account 50% variability all These be interpreted physically as indicative balance influences photochemical other chemical processes, sulfur dioxide sources atmospheric dispersion transport. In southern California, PCA indicates ambient SO2 factors unimportant in explaining variability. Instead, associated activity moisture content more than half variation. This agrees previous analyses data. similar component found. However, concentration also important method derived which relates coefficients apparent, first order transformation rates sulfate. Maximum apparent range 4–10% h, but about 10 times smaller York.