Preliminary assessment of the anthropogenic and biogenic contributions to secondary organic aerosols at two industrial cities in the upper Midwest

Abstract

The contributions of anthropogenic and biogenic secondary organic carbon (SOC) to total PM2.5 mass are of interest to air quality management agencies required to demonstrate maintenance of the PM2.5 NAAQS. Reductions of SOC can be used in conjunction with the mitigation of other PM2.5 constituents to maintain PM2.5 concentrations below the regulatory limit. Currently, quantitative tools to understand the SOC source contributions to PM2.5 mass are not well developed, and the spatial variation of different types of SOC is not known.In this study concentrations of anthropogenic and biogenic SOC mass were determined using PM2.5 measurements made in Cleveland, OH and Mingo Junction, OH. Twenty-four hour averaged samples were collected on the EPA 1-in-6 day schedule over the course of one year between June 2007 and May of 2008. Organic molecular markers for anthropogenic and biogenic SOC were extracted from the PM2.5, silylated, and then analyzed by GC–MS. Source apportionment calculations were conducted using the EPA CMB (v.8.2) software and organic molecular markers as source tracers.SOC concentrations calculated from SOC tracers measurements followed the expected seasonal patterns with maximum contributions during the summer and minimum contributions during the winter. Anthropogenic SOC constituted approximately 37% to the apportioned SOC and 6% to the measured OC, on average across both sites. Biogenic SOC contributed the 42% to the apportioned SOC, and 4% to the measured OC. Anthropogenic SOC contributed strongly to organic PM2.5 meaning that SOC may by partially controllable by reductions in VOC emissions from anthropogenic sources.Similarities in the month-to-month patterns in α-pinene markers were observed between Cleveland and Mingo Junction, suggesting a regional character to this type of SOC. However, such patterns were not readily apparent in the isoprene markers.Limitations were found in the current version of the model. Approximately half of the water soluble organic carbon unrelated to biomass burning (NB-WSOC) during spring, summer and early fall could not be apportioned by the CMB model with the SOC markers available during this study. This suggested that additional sources not included in the CMB model used in this study contributed to SOC, or that models using markers measured in chamber oxidations are not entirely representative of the study sites. The unapportioned OC did not correlate particularly well with any of the known OC sources. While performance of the model is limited due to uncertainties in the source profiles, the apportionments calculated still give a preliminary insight into the relative contributions to SOC from anthropogenic and biogenic emissions.