Source Apportionment of PM10 Over Three Tropical Urban Atmospheres at Indo-Gangetic Plain of India: An Approach Using Different Receptor Models.

The present work is the ensuing part of the study on spatial and temporal variations in chemical characteristics of PM10 (particulate matter with aerodynamic diameter ≤ 10 µm) over Indo Gangetic Plain (IGP) of India. It focuses on the apportionment of PM10 sources with the application of different receptor models, i.e., principal component analysis with absolute principal component scores (PCA-APCS), UNMIX, and positive matrix factorization (PMF) on the same chemical species of PM10. The main objective of this study is to perform the comparative analysis of the models, obtained mutually validated outputs and more robust results. The average PM10 concentration during January 2011 to December 2011 at Delhi, Varanasi, and Kolkata were 202.3 ± 74.3, 206.2 ± 77.4, and 171.5 ± 38.5 µg m-3, respectively. The results provided by the three models revealed quite similar source profile for all the sampling regions, with some disaccords in number of sources as well as their percent contributions. The PMF analysis resolved seven individual sources in Delhi [soil dust (SD), vehicular emissions (VE), secondary aerosols (SA), biomass burning (BB), sodium and magnesium salt (SMS), fossil fuel combustion, and industrial emissions (IE)], Varanasi [SD, VE, SA, BB, SMS, coal combustion, and IE], and Kolkata [secondary sulfate (Ssulf), secondary nitrate, SD, VE, BB, SMS, IE]. However, PCA-APCS and UNMIX models identified less number of sources (besides mixed type sources) than PMF for all the sampling sites. All models identified that VE, SA, BB, and SD were the dominant contributors of PM10 mass concentration over the IGP region of India.

Stable carbon and nitrogen isotopic composition of PM10 over Indo-Gangetic Plains (IGP), adjoining regions and Indo-Himalayan Range (IHR) during a winter 2014 campaign.

For source identification, a field campaign involving simultaneous sampling of particulate matter (PM10) was conducted at eight sampling sites in the Indian mainland during winter 2014. The sampling sites include Delhi (upper IGP), Lucknow (middle IGP), and Kolkata (lower IGP) in the Indo-Gangetic Plains (IGP); Mohal-Kullu and Darjeeling in the Indo-Himalayan Range (IHR). In addition, Ajmer, located upwind of the IGP in NW-India and Giridih and Bhubaneswar, in the downwind to the IGP has also been chosen. To characterize the sources of the ambient PM10, stable isotope ratios of carbon (δ13CTC) and nitrogen (δ15NTN) for the total carbon (TC) and total nitrogen (TN) fractions have been considered. Ancillary chemical parameters, such as organic carbon (OC), elemental carbon (EC), and water-soluble ionic components (WSIC) mass concentrations are also presented in this paper. There was very small variation in the daily average δ13CTC ratios (- 24.8 to - 25.9‰) among the sites. Comparison with end-member stable C isotopic signatures of major typical sources suggests that the PM10 at the sites was mainly from fossil fuel and biofuel and biomass combustion. Daily average δ15NTN ratios were not observed to vary much between sites either (8.3 to 11.0‰), and the low δ15NTN levels also indicate substantial contributions from biofuel and biomass burning of primarily C3 andC4 plant matter. Graphical abstract Scatter plot of the average (± 1 standard deviation (SD)) δ13CTC (‰) compared to δ15NTN (‰) at the sampling sites.