Identification of sources affecting fog formation using receptor modeling approaches and inventory estimates of sectoral emissions

Positive matrix factorization (PMF) was used to identify factors affecting fog formation in Kanpur during the ISRO-GBP land campaign-II (LC-II) in December 2004. PMF predicted factors were validated by contrasting the emission strength of sources in the foggy and clear periods, using a combination of potential source contribution function (PSCF) analysis and quantitative emission inventory information. A time series aerosol chemical data set of 29 days and 12 species was decomposed to identify 4-factors: Secondary species, Biomass burning, Dust and Sea salt. PMF predicted particle mass with a satisfactory goodness-of-fit (slope of 0.83 +/- 0.17 and R-2 of 0.8), and strong species within 11-12% relative standard deviation. Mean contributions of anthropogenic factors were significantly higher during the foggy period for secondary species (2.9 +/- 0.3) and biomass burning (1.2 +/- 0.09) compared to the clear period. Local Sources contributing to aerosols that mediated fog events at Kanpur, based on emissions in a 200 km x 200 km area around Kanpur City Were thermal power plants and transportation (SO2) and biofuel combustion (BC and OM). Regional scale sources influencing emissions during the foggy period, in probable source regions identified by PSCF included thermal power plants, transportation, brick kilns and biofuel combustion. While biofuel combustion and transportation are distributed area sources, individual point sources include coal-fired thermal power plants located in Aligarh, Delhi, Ghaziabad, Jhansi, Kanpur, Rae Bareli and Rupnagar and brick kilns located in Allahabad, Agra, Farrukhabad, Ghaziabad, Kanpur, Ludhiana, Lucknow and Rae Bareli. Additionally, in the foggy period, large areas of probable source regions lay Outside India, implying the significance of aerosol incursion from Outside India. (C) 2008 Elsevier Ltd. All rights reserved.