Source apportionment of time- and size-resolved particulate matter over two sites in New Delhi & NCR

Size resolved data of chemical species carries a lot of latent information about the sources and atmospheric processes which lead to their formation and growth. Source apportionment techniques on organic or inorganic aerosols provide a fair amount of information about the sources but this analysis only provides a partial picture owing to the complicated nature of the ambient aerosols which may contain both, organic as well as inorganic particulate matter. Traditionally, potential emission sources are distinguished by either the organic or inorganic tracers present in ambient aerosol, but recently several studies have performed PMF on both the species (Sun et al, 2012). However, it tells more about the final transformed products which could be formed from different pathways but not much about the transformation pathways. Insights about the source and the atmospheric processes involved can be derived from the analysis of size-resolved data of the ambient aerosol. PMF on Size-resolved information helps us to narrow down the possible pathways of the transformed products.

However, there is very limited literature available to help us understand more about size-resolved bulk particulate matter. In this manuscript, a novel approach to perform Positive Matrix Factorization (PMF) on real-time size-resolved Unit Mass Resolution (UMR) data from Aerosol Mass Spectrometer (AMS) is presented. Both size- and time-resolved PMF is performed on non-refractory particle composition (organic & inorganic) on the UMR PTOF data of two sites in one of the most polluted cities in the world. The sampling through Long Time of flight mass spectrometer (LToF-AMS) was carried out at Indian Institute of Technology, Delhi which is located in Hauz Khaz area, at the heart of Delhi NCR, whereas parallel sampling through High-resolution Time of flight aerosol mass spectrometer (HR-ToF-AMS) was carried out at Manav Rachna University which is located in Faridabad within Delhi NCR at a downwind location. PMF was performed on the data by using Multi-linear Engine (ME-2) on PMF model by SoFi (Source Finder) tool. A seven-factor solution was chosen based on the factor profiles, time series, diurnals and correlation with the external factors obtained by supplementary instruments. The size-resolved spectra of the species at an individual site was studied and the difference between the sites was compared.