Source apportionment of submicron organic aerosol collected from Atlanta, Georgia, during 2014–2015 using the aerosol chemical speciation monitor (ACSM)

Abstract

The Aerodyne Aerosol Chemical Speciation Monitor (ACSM) was redeployed at the Jefferson Street (JST) site in downtown Atlanta, Georgia (GA) for 1 year (March 20, 2014–February 08, 2015) to chemically characterize non-refractory submicron particulate matter (NR-PM1) in near real-time and to assess whether organic aerosol (OA) types and amounts change from year-to-year. Submicron organic aerosol (OA) mass spectra were analyzed by season using multilinear engine (ME-2) to apportion OA subtypes to potential sources and chemical processes. A suite of real-time collocated measurements from the Southeastern Aerosol Research and Characterization (SEARCH) network was compared with ME-2 factor solutions to aid in the interpretation of OA subtypes during each season. OA tracers measured from high-volume filter samples using gas chromatography interfaced with electron ionization-mass spectrometry (GC/EI-MS) also aided in identifying OA sources. The initial application of ME-2 to the yearlong ACSM dataset revealed that OA source apportionment by season was required to better resolve sporadic OA types. Spring and fall OA mass spectral datasets were separated into finer periods to capture potential OA sources resulting from non-homogeneous emissions during transitioning periods. NR-PM1 was highest in summer (16.7 ± 8.4 μg m−3) and lowest in winter (8.0 ± 5.7 μg m−3), consistent with prior studies. OA dominated NR-PM1 mass (56–74% on average) in all seasons. Hydrocarbon-like OA (HOA) from primary emissions was observed in all seasons, averaging 5–22% of total OA mass. Strong correlations of HOA with carbon monoxide (CO) (R = 0.71–0.88) and oxides of nitrogen (NOx) (R = 0.55–0.79) indicated that vehicular traffic was the likely source. Biomass burning OA (BBOA) was observed in all seasons, with lower contributions (2%) in summer and higher in colder seasons (averaging 8–20% of total OA mass). BBOA correlated strongly with levoglucosan (R = 0.78–0.95) during colder seasons, which supports that BBOA is likely derived from fresh biomass/residential burning. However, weaker correlation with levoglucosan (R = 0.38) in summer suggested a more aged aerosol. During warmer seasons, OA from the reactive uptake of isoprene epoxydiols (IEPOX) onto acidic sulfate aerosol was resolved by ME-2 (denoted as IEPOX-OA), averaging 25–29% of the total OA mass. Temporal variation of IEPOX-OA was nearly coincident with that of 91Fac OA (a factor dominated by a distinct ion at m/z 91). The largest contribution of IEPOX-OA to total OA (29%) was found in summer, whereas the largest contribution of 91Fac to total OA (24%) occurred in early fall. Moderate negative correlation between IEPOX-OA and aerosol acidity was observed during late spring (−0.67) and summer (−0.42), consistent with laboratory studies showing that IEPOX-OA is enhanced in the presence of acidic aerosols. Finally, the largest OA mass in all seasons (46–70% of total OA) was derived from oxygenated OA denoted as low-volatility oxygenated OA (LV-OOA) and semi-volatile oxygenated OA (SV-OOA).