AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Application of Advanced Factorization Techniques for Deconvolution of Cooking and Biomass Burning Source Contributions in a Polluted Megacity
SAHIL BHANDARI, Kanan Patel, Shahzad Gani, Gazala Habib, Joshua Apte, Lea Hildebrandt Ruiz, University of Texas at Austin
Abstract Number: 523 Working Group: Source Apportionment
Abstract Application of receptor modeling techniques to data from online mass spectrometers has provided insight into the nature and sources of atmospheric aerosol. Positive matrix factorization (PMF) is one such technique which deconvolutes sampled data into the summation of products of positively constrained mass spectral profiles and their corresponding time series, under the assumption that the mass spectral profiles remain constant in time. However, application of PMF to unit mass spectral (UMR) data often does not resolve primary organic aerosol (POA) to hydrocarbon-like organic aerosol (HOA), cooking organic aerosol (COA), and biomass burning organic aerosol (BBOA), even in highly polluted environments. To address this issue, we applied advanced factorization techniques including time-resolved PMF and ME-2 (a hybrid of Chemical Mass Balance (CMB) technique and PMF using a priori information of factor profiles to constrain their presence) to data collected in the Delhi Aerosol Supersite (DAS) study.
For example, applying these techniques to the evenings of winter 2017 separates HOA, oxidized organic aerosol (OOA), BBOA, and oxidized biomass burning organic aerosol (OBOA), compared to two-factor solutions in seasonal PMF runs (HOA and OOA). Also, a narrowing time window around afternoon cooking hours results in increasing fresh COA signatures in the POA profile.
Overall, these techniques enable the generation of multiple sets of factor profiles over diurnal time windows in each season. Profiles obtained using these methods, therefore, offer a more realistic picture of temporally varying sources and the evolving oxidized organic aerosol.