Multivariate statistical analyses of air pollutants and meteorology in Chicago during summers 2010-2012

Abstract

Aerosol, trace gas, and meteorological data were collected in Chicago, Illinois during 2010–2012 summer air studies. Ozone, nitrogen oxides, acetate, formate, chloride, nitrate, sulfate, and oxalate concentrations as well as temperature, wind speed, wind direction, and humidity data were explored by both principal component analysis (PCA) and canonical correlation analysis (CCA). Multivariate statistical techniques were applied to uncover existing relationships between meteorology and air pollutant concentrations and also reduce data dimensions. In PCA, principal components (PCs) revealed a relationship of ozone and nitrate concentrations with respect to temperature and humidity, coupled with transport of species from the south in relation to the sampling site (PC1). PC2 was a measure of secondary aerosols but also suggested acetate and formate presence was a result of primary emissions or transport. Both PC3 and PC4 contained residual information with the former representing days of lower pollution and the latter representing northerly wind transport of chloride, nitrate, and ozone to the sampling site. In CCA, three canonical functions were statistically significant. The first indicated high temperature and low wind speed had a strong linear relationship ozone, oxalate, and nitrogen oxide concentrations whereas the second function showed a strong influence of wind direction on acetate, formate, and chloride concentrations. Residuals of temperature, wind speed, trace gases, and oxalate also were in the second function. The only new information in the third function was humidity. Overall, PCA and CCA bring forth multivariable relationships, not represented in descriptive statistics, useful in understanding pollution variability.

from # All Medicine by Alexandros G. Sfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2wvdmKF
via IFTTT