Using the chemical mass balance model toestimate pollution source contributions fromcorrelated air quality observations

摘要

In the environmental sciences, receptor models are used to evaluate the contributionrnof various pollution sources to the air composition at a location. Using pollutionrnsource profles, profle uncertainties, and measurement error variances, the chemicalrnmass balance (CMB) model can be ft in order to partition the ambient pollutantsrnmeasured at the receptor into a collection of source contributions. We discuss thernuse of the CMB model for the analysis of a multivariate time series of air qualityrnmeasurements, and we consider estimation and inference procedures which accountrnfor the multiple sources of correlation in the data. Using computer simulation,rnapproaches are compared under various scenarios in which standard model assump-rntions are violated.