Constrained source apportionment of coarse particulate matter and selected trace elements in three cities from the multi-ethnic study of atherosclerosis

Abstract

PM10-2.5 mass and trace element concentrations were measured in Winston-Salem, Chicago, and St. Paul at up to 60 sites per city during two different seasons in 2010. Positive Matrix Factorization (PMF) was used to explore the underlying sources of variability. Information on previously reported PM10-2.5 tire and brake wear profiles was used to constrain these features in PMF by prior specification of selected species ratios. We also modified PMF to allow for combining the measurements from all three cities into a single model while preserving city-specific soil features. Relatively minor differences were observed between model predictions with and without the prior ratio constraints, increasing confidence in our ability to identify separate brake wear and tire wear features. Brake wear, tire wear, fertilized soil, and re-suspended soil were found to be important sources of copper, zinc, phosphorus, and silicon respectively across all three urban areas.

Source and risk apportionment of selected VOCs and PM2.5 species using partially constrained receptor models with multiple time resolution data

“…In such cases, a priori information about source contributions, source profiles, or the relationships between species in a certain profile can serve as additional constraints (Norris et al, 2009;Paatero and Hopke, 2009;Paatero et al, 2002). Previous studies (Amato and Hopke, 2012;Amato et al, 2009;Sturtz et al, 2014) used measured profiles from potential source-types to constrain the PMF source factors, resulting in more physically realistic correspondence between the calculated and measured abundances. To the best of our knowledge, this concept has not been applied to data with multiple time resolutions.…”

Variations in the chemical composition of the submicron aerosol and in the sources of the organic fraction at a regional background site of the Po Valley (Italy)

“…Amato et al, 2009;Crippa et al, 2014), specific elemental ratios (e.g. Sturtz et al, 2014) or specific species contribution (e.g. Crawford et al, 2005) in a given FP.…”

Physically constrained source apportionment (PCSA) for polycyclic aromatic hydrocarbon using the Multilinear Engine 2-species ratios (ME2-SR) method

“…f pk is the factor loading element obtained from the base run in ME2. a pk is the target value and can be written as follows (Sturtz et al, 2014):…”

“…Recently, a new factor analysis model known as Multilinear Engine-2 (ME-2) was put forward and applied in many researches (Amato et al, 2009;Amato and Hopke, 2012;Sturtz et al, 2014). It was developed as a more general and flexible model to solve all types of multilinear and quasi-multilinear problems (Paatero, 2000;Ramadan et al, 2003), and permitted the imposition of the external constraints, such as pulling factor elements toward the target values or a ratio (Sturtz et al, 2014).…”

“…It was developed as a more general and flexible model to solve all types of multilinear and quasi-multilinear problems (Paatero, 2000;Ramadan et al, 2003), and permitted the imposition of the external constraints, such as pulling factor elements toward the target values or a ratio (Sturtz et al, 2014). Therefore, combing ME-2 with PAH diagnostics ratios is feasible, using the prior species ratio constraints for factor pulling.…”