Apportionment of Ambient Primary and Secondary Fine Particulate Matter at the Pittsburgh National Energy Laboratory Particulate Matter Characterization Site Using Positive Matrix Factorization and a Potential Source Contributions Function Analysis

Abstract

Fine particulate matter (PM2.5) concentrations associated with 202 24-hr samples collected at the National Energy Technology Laboratory (NETL) particulate matter (PM) characterization site in south Pittsburgh from October 1999 through September 2001 were used to apportion PM2.5 into primary and secondary contributions using Positive Matrix Factorization (PMF2). Input included the concentrations of PM2.5 mass determined with a Federal Reference Method (FRM) sampler, semi-volatile PM2.5 organic material, elemental carbon (EC), and trace element components of PM2.5. A total of 11 factors were identified. The results of potential source contributions function (PSCF) analysis using PMF2 factors and HYSPLIT-calculated back-trajectories were used to identify those factors associated with specific meteorological transport conditions. The 11 factors were identified as being associated with emissions from various specific regions and facilities including crustal material, gasoline combustion, diesel combustion, and three nearby sources high in trace metals. Three sources associated with transport from coal-fired power plants to the southeast, a combination of point sources to the northwest, and a steel mill and associated sources to the west were identified. In addition, two secondary-material-dominated sources were identified, one was associated with secondary products of local emissions and one was dominated by secondary ammonium sulfate transported to the NETL site from the west and southwest. Of these 11 factors, the four largest contributors to PM2.5 were the secondary transported material (dominated by ammonium sulfate) (47%), local secondary material (19%), diesel combustion emissions (10%), and gasoline combustion emissions (8%). The other seven factors accounted for the remaining 16% of the PM2.5 mass. The findings are consistent with the major source of PM2.5 in the Pittsburgh area being dominated by ammonium sulfate from distant transport and so decoupled from local activity emitting organic pollutants in the metropolitan area. In contrast, the major local secondary sources are dominated by organic material.

Identifying the sources driving observed PM<sub>2.5</sub> temporal variability over Halifax, Nova Scotia, during BORTAS-B

“…The PMF method has an extensive heritage, having been applied to many PM 2.5 source apportionment studies (Paatero, 1997;Paatero and Trapper, 1994;Martello et al, 2008;Jeong et al, 2011;Brown et al, 2007;Bari et al, 2009). Details of the PMF model are provided in Hopke (1991).…”

“…Ship emissions also contain large quantities of BC particulate (Lack and Corbett, 2012). Unambiguous markers of fugitive surficial dust include Fe, Al, Ca and Si (Jeong et al, 2011;Martello et al, 2008;Gugamsetty et al, 2012). Primary sea salt markers include Na + , Cl − , Mg 2+ and Ca 2+ (Gibson et al, 2009), and Na + , Mg 2+ , Ca 2+ and NO − 3 for aged marine secondary aerosol (Jeong et al, 2011;Gibson et al, 2009).…”

Review of Recent Advances in Detection of Organic Markers in Fine Particulate Matter and Their Use for Source Apportionment

“…The PMF analysis of PM 2.5 reported in this study was comparable to results obtained in a PMF analysis of independent data from a study in Pittsburgh at the Pittsburgh Energy Technology Center conducted over the same time period. 215 Finally, this group has published the comparison of PMF solutions using different combinations of 54 organic marker compounds. 216 The PMF analysis was done with and without added elemental data.…”

“…that are indicators of atmospheric conversion processes. 139,210,215 The factors identified by PMF must be interpreted as to their probable origin in the atmosphere. Usually this is done by relating the profile of the factor and, if the data are obtained in a short enough time frame, 139,221 on the diurnal patterns of the material associated with the factor.…”

Chemical Characterization and Preliminary Source Contribution of Fine Particulate Matter in the Mexicali/Imperial Valley Border Area

“…No seasonal variation was assigned to this ratio in the absence of additional information, although it is probable that the final conclusions are unaffected by this drawback. 29 For example, if the total PM 2.5 mass was reconstructed using an OM/OC ratio that provided the best fit (in a least-squares sense) between the reconstructed and observed PM 2.5 (estimating the rest of the classes as indicated above), a value for the OM/OC ratio of 1.75 for the fall data and 1.2 for the spring data would be obtained.…”