Source apportionment using reconstructed mass calculations

Abstract

A long-term study was undertaken to investigate the air quality of the Islamabad/Rawalpindi area. In this regard fine and coarse particulate matter were collected from 4 sites in the Islamabad/Rawalpindi region from 1998 to 2010 using Gent samplers and polycarbonate filters and analyzed for their elemental composition using the techniques of Neutron Activation Analysis (NAA), Proton Induced X-ray Emission/Proton Induced Gamma-ray Emission (PIXE/PIGE) and X-ray Fluorescence (XRF) Spectroscopy. The elemental data along with the gravimetric measurements and black carbon (BC) results obtained by reflectance measurement were used to approximate or reconstruct the particulate mass (RCM) by estimation of pseudo sources such as soil, smoke, sea salt, sulfate and black carbon or soot. This simple analysis shows that if the analytical technique used does not measure important major elements then the data will not be representative of the sample composition and cannot be further utilized for source apportionment studies or to perform transboundary analysis. In this regard PIXE/PIGE and XRF techniques that can provide elemental compositional data for most of the major environmentally important elements appear to be more useful as compared to NAA. Therefore %RCM calculations for such datasets can be used as a quality assurance (QA) measure to treat data prior to application of chemometrical tools such as factor analysis (FA) or cluster analysis (CA).