Source apportionment of 1 h semi-continuous data during the 2005 Study of Organic Aerosols in Riverside (SOAR) using positive matrix factorization

Positive matrix factorization (PMF2) was used to elucidate sources of fine particulate material (PM2.5) for a study conducted during July and August 2005, in Riverside, CA. One-hour averaged semi-continuous measurements were made with a suite of instruments to provide PM2.5 mass and chemical composition data. Total PM2.5 mass concentrations (non-volatile plus semi-volatile) were measured with an R&P filter dynamic measurement system (FDMS TEOM) and a conventional TEOM monitor was used to measure non-volatile mass concentrations. PM2.5 chemical species monitors included a dual-oven Sunset monitor to measure both non-volatile and semi-volatile carbonaceous material, an ion chromatographic-based monitor to measure sulfate and nitrate and an Anderson Aethalometer to measure black carbon (BC). Gas phase data including CO, NO2, NOx and O3 were also collected during the sampling period. In addition, single-particle measurements were made using aerosol time-of-flight mass spectrometry (ATOFMS). Twenty different single-particle types consistent with those observed in previous ATOFMS studies in Riverside were identified for the PMF2 analysis. Finally, time-of-flight aerosol mass spectrometry (ToF-AMS) provided data on markers of primary and secondary organic aerosol. Two distinct PMF2 analyses were performed. In analysis 1, all the data except for the ATOFMS and ToF-AMS data were used in an initial evaluation of sources at Riverside during the study. PMF2 was able to identify six factors from the data set corresponding to both primary and secondary sources, primarily from automobile emissions, diesel emissions, secondary nitrate formation, a secondary photochemical associated source, organic emissions and Basin transported pollutants. In analysis 2, the ATOFMS and ToF-AMS data were included in the analysis. In the second analysis, PMF2 was able to identify 16 factors with a variety of both primary and secondary factors being identified, corresponding to both primary and secondary material from both anthropogenic and natural sources. Based on relationships with Basin meteorology, the PMF identified source profiles and diurnal patterns in the source concentrations, sources were identified as being of local origin or resulting from transport of pollutants across the Basin due to onshore flow. Good agreement was observed between the PMF2 predicted mass and the FDMS measured mass for both analyses.