Source apportionment of organic carbon in Centreville, AL using organosulfates in organic tracer-based positive matrix factorization

Organic tracer-based positive matrix factorization (PMF) was used to apportion fine particulate (PM2.5) organic carbon (OC) to its sources in Centreville, AL, USA, a rural forested site influenced by anthropogenic emissions, during the Southern Oxidant and Aerosol Study (SOAS) in the summer of 2013. Model inputs included organosulfates, a group of organic compounds that are tracers of anthropogenically-influenced biogenic secondary organic aerosols (SOA), as well as, OC, elemental carbon, water-soluble organic carbon, and other organic tracers for primary and secondary sources measured during day and night. The organic tracer-based PMF resolved eight factors that were identified as biomass burning (11%, average contribution to PM2.5 OC), vehicle emissions (8%), isoprene SOC formed under low-NOx conditions (13%), isoprene SOC formed under high-NOx conditions (11%), SOC formed by photochemical reactions (9%), oxidatively aged biogenic SOC (6%), sulfuric acid-influenced SOC (21%, that also includes isoprene and monoterpene SOC), and monoterpene SOC formed under high-NOx conditions (21%). These results indicate that OC in Centreville during summer is mainly secondary in origin (81%). Fossil fuel combustion is the major source of NOx, ozone, and sulfuric acid that play a key role in SOA formation in the southeastern US. Fossil fuel was found to influence 61-76% of OC through vehicle emissions and SOA formation. Together with prescribed burns, which were the major type of biomass burning during this study, the OC influenced by anthropogenic activities reached 87%. The organic tracer-based PMF results were further compared with two complementary source apportionment techniques: PMF factors resolved for submicron organic aerosols measured using aerosol mass spectrometry (AMS) by Xu et al. (2015a) in Centreville during SOAS; biomass burning organic aerosols (BBOA, 11% of OC), isoprene-derived organic aerosols (isoprene-OA, 20% of OC), more-oxidized oxygenated organic aerosols (MO-OOA, 34% of OC), and less-oxidized oxygenated organic aerosols (LO-OOA, 35% of OC); and PM2.5 OC apportioned by chemical-mass balance model (CMB), considering the same chemical species as this study, save for organosulfates; biomass burning (5%), diesel engines (2%), gasoline smokers (3%), vegetative detritus (1%), isoprene SOC (23%) and monoterpene SOC (34%), and other (likely biogenic secondary) sources (33%). Overall, this study indicates the primary and secondary sources resolved by the organic tracer-based PMF are in good agreement with CMB and AMS-PMF results, while the organic tracer-based PMF provides additional insight to the SOC formation pathways through the inclusion of organosulfates and other organic tracers measured during day and night. © 2018 Elsevier Ltd