Polycyclic aromatic hydrocarbons (PAHs) are among the most widespread and potentially toxic contaminants in Great Lakes (USA/Canada) tributaries. The sources of PAHs are numerous and diverse, and identifying the primary source(s) can be difficult. The present study used multiple lines of evidence to determine the likely sources of PAHs to surficial streambed sediments at 71 locations across 26 Great Lakes Basin watersheds. Profile correlations, principal component analysis, positive matrix factorization source-receptor modeling, and mass fractions analysis were used to identify potential PAH sources, and land-use analysis was used to relate streambed sediment PAH concentrations to different land uses. Based on the common conclusion of these analyses, coal-tar-sealed pavement was the most likely source of PAHs to the majority of the locations sampled. The potential PAH-related toxicity of streambed sediments to aquatic organisms was assessed by comparison of concentrations with sediment quality guidelines. The sum concentration of 16 US Environmental Protection Agency priority pollutant PAHs was 7.4-196 000 µg/kg, and the median was 2600 µg/kg. The threshold effect concentration was exceeded at 62% of sampling locations, and the probable effect concentration or the equilibrium partitioning sediment benchmark was exceeded at 41% of sampling locations. These results have important implications for watershed managers tasked with protecting and remediating aquatic habitats in the Great Lakes Basin. Environ Toxicol Chem 2020;39:1392-1408. © 2020 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Our understanding of atmospheric oxidation chemistry has improved significantly in recent years, greatly facilitated by developments in mass spectrometry. The generated mass spectra typically contain vast amounts of information on atmospheric sources and processes, but the identification and quantification of these is hampered by the wealth of data to analyze. The implementation of factor analysis techniques have greatly facilitated this analysis, yet many atmospheric processes still remain poorly understood. Here, we present new insights into highly oxygenated products from monoterpene oxidation, measured by chemical ionization mass spectrometry, at a boreal forest site in Finland in autumn 2016. Our primary focus was on the formation of accretion products, i.e., dimers. We identified the formation of daytime dimers, with a diurnal peak at noontime, despite high nitric oxide (NO) concentrations typically expected to inhibit dimer formation. These dimers may play an important role in new particle formation events that are often observed in the forest. In addition, dimers identified as combined products of NO3 and O3 oxidation of monoterpenes were also found to be a large source of low-volatility vapors at night. This highlights the complexity of atmospheric oxidation chemistry and the need for future laboratory studies on multi-oxidant systems. These two processes could not have been separated without the new analysis approach deployed in our study, where we applied binned positive matrix factorization (binPMF) on subranges of the mass spectra rather than the traditional approach where the entire mass spectrum is included for PMF analysis. In addition to the main findings listed above, several other benefits compared to traditional methods were found.

Recent advancements in atmospheric mass spectrometry provide huge amounts of new information but at the same time present considerable challenges for the data analysts. High-resolution (HR) peak identification and separation can be effort- and time-consuming yet still tricky and inaccurate due to the complexity of overlapping peaks, especially at larger mass-to-charge ratios. This study presents a simple and novel method, mass spectral binning combined with positive matrix factorization (binPMF), to address these problems. Different from unit mass resolution (UMR) analysis or HR peak fitting, which represent the routine data analysis approaches for mass spectrometry datasets, binPMF divides the mass spectra into small bins and takes advantage of the positive matrix factorization's (PMF) strength in separating different sources or processes based on different temporal patterns. In this study, we applied the novel approach to both ambient and synthetic datasets to evaluate its performance. It not only succeeded in separating overlapping ions but was found to be sensitive to subtle variations as well. Being fast and reliable, binPMF has no requirement for a priori peak information and can save much time and effort from conventional HR peak fitting, while still utilizing nearly the full potential of HR mass spectra. In addition, we identify several future improvements and applications for binPMF and believe it will become a powerful approach in the data analysis of mass spectra.

.............................................................................................................. 3

The new version of EPA's positive matrix factorization (EPA PMF) software, 5.0, includes three error estimation (EE) methods for analyzing factor analytic solutions: classical bootstrap (BS), displacement of factor elements (DISP), and bootstrap enhanced by displacement (BS-DISP). These methods capture the uncertainty of PMF analyses due to random errors and rotational ambiguity. To demonstrate the utility of the EE methods, results are presented for three data sets: (1) speciated PM2.5 data from a chemical speciation network (CSN) site in Sacramento, California (2003–2009); (2) trace metal, ammonia, and other species in water quality samples taken at an inline storage system (ISS) in Milwaukee, Wisconsin (2006); and (3) an organic aerosol data set from high-resolution aerosol mass spectrometer (HR-AMS) measurements in Las Vegas, Nevada (January 2008). We present an interpretation of EE diagnostics for these data sets, results from sensitivity tests of EE diagnostics using additional and fewer factors, and recommendations for reporting PMF results. BS-DISP and BS are found useful in understanding the uncertainty of factor profiles; they also suggest if the data are over-fitted by specifying too many factors. DISP diagnostics were consistently robust, indicating its use for understanding rotational uncertainty and as a first step in assessing a solution's viability. The uncertainty of each factor's identifying species is shown to be a useful gauge for evaluating multiple solutions, e.g., with a different number of factors.

The performance and the uncertainty of receptor models (RMs) were assessed in intercomparison exercises employing real-world and synthetic input datasets. To that end, the results obtained by different practitioners using ten different RMs were compared with a reference. In order to explain the differences in the performances and uncertainties of the different approaches, the apportioned mass, the number of sources, the chemical profiles, the contribution-to-species and the time trends of the sources were all evaluated using the methodology described in Bells et al. (2015).In this study, 87% of the 344 source contribution estimates (SCEs) reported by participants in 47 different source apportionment model results met the 50% standard uncertainty quality objective established for the performance test. In addition, 68% of the SCE uncertainties reported in the results were coherent with the analytical uncertainties in the input data.The most used models, EPA-PMF v.3, PMF2 and EPA-CMB 8.2, presented quite satisfactory performances in the estimation of SCEs while unconstrained models, that do not account for the uncertainty in the input data (e.g. APCS and FA-MLRA), showed below average performance. Sources with well-defined chemical profiles and seasonal time trends, that make appreciable contributions (>10%), were those better quantified by the models while those with contributions to the PM mass close to 1% represented a challenge.The results of the assessment indicate that RMs are capable of estimating the contribution of the major pollution source categories over a given time window with a level of accuracy that is in line with the needs of air quality management. (C) 2015 The Authors. Published by Elsevier Ltd.

The EPA PMF (Environmental Protection Agency positive matrix factorization) version 5.0 and the underlying multilinear engine-executable ME-2 contain three methods for estimating uncertainty in factor analytic models: classical bootstrap (BS), displacement of factor elements (DISP), and bootstrap enhanced by displacement of factor elements (BS-DISP). The goal of these methods is to capture the uncertainty of PMF analyses due to random errors and rotational ambiguity. It is shown that the three methods complement each other: depending on characteristics of the data set, one method may provide better results than the other two. Results are presented using synthetic data sets, including interpretation of diagnostics, and recommendations are given for parameters to report when documenting uncertainty estimates from EPA PMF or ME-2 applications.

This report contains a guide and a European harmonised protocol for the identification of air pollution sources using receptor models. The document aims at disseminating and promoting the best available methodologies for source identification and at harmonising their application across Europe. It was developed by a committee of leading experts within the framework of the JRC initiative for the harmonisation of source apportionment that has been launched in collaboration with the European networks in the field of air quality modelling (FAIRMODE) and measurements (AQUILA). The protocol has been conceived as a reference document that includes tutorials, technical recommendations and check lists connected to the most up –to-date and rigorous scientific standards. As a guide, it is structured in sections with increasing levels of complexity that make it accessible to readers with different degrees of familiarity with this topic, from air quality managers to air pollution experts and modellers.

F. Karagulian, C.A. Belis, F. Amato, D.C.S. Beddows, V. Bernardoni, S. Carbone, D. Cesari, E. Cuccia, D. Contini, O. Favez, I. El Haddad, R.M. Harrison, T. Kammermeier, M.Karl, F. Lucarelli, S.Nava, J. K. Nojgaard, M. Pandolfi, M.G. Perrone, J.E. Petit, A. Pietrodangelo, P. Prati, A.S.H. Prevot, U. Quass, X. Querol, D. Saraga, J. Sciare, A. Sfetsos, G. Valli, R. Vecchi, M. Vestenius, J.J. Schauer, J.R. Turner, P. Paatero, P.K. Hopke

F. Karagulian, C.A. Belis, F. Amato, D.C.S. Beddows, V. Bernardoni, S. Carbone, D. Cesari, E. Cuccia, D. Contini, O. Favez, I. El Haddad, R.M. Harrison, T. Kammermeier, M.Karl, F. Lucarelli, S.Nava, J. K. Nojgaard, M. Pandolfi, M.G. Perrone, J.E. Petit, A. Pietrodangelo, P. Prati, A.S.H. Prevot, U. Quass, X. Querol, D. Saraga, J. Sciare, A. Sfetsos, G. Valli, R. Vecchi, M. Vestenius, J.J. Schauer, J.R. Turner, P. Paatero, P.K. Hopke

The identification of sources is one of the prerequisites for the implementation of the Air Quality Directive (AQD). It provides scientific support to the development and periodic revision of air quality plans and short term action plans and to the quantification of categories with special status like long range transport, natural sources and winter road salting and sanding. The suitability of receptor models (RM) for the apportionment of pollutant sources in the implementa-tion of the AQD is testified by the amount of published studies in 2005 and 2010 in correspondence with the entry into force of new provisions for PM10 and PM2,5, respectively. In recent years, Member States were requested to provide official estimations of source contribution to the Commission like the quantification of natural sources in 2006 and PM10 time extension reports in 2009 (Fragkou et al., 2011). These experiences have shown that al-though this kind of methodology is used by about 60 % of the European experts involved in source apportion-ment there is a considerable variability in the methodo-logical approaches and adopted tools. Furthermore, there are critical steps that require strict quality assurance standards and skilled practitioners (e.g. identification of the appropriate number of sources). In order to foster harmonization in this field, the JRC has promoted a number of interconnected initiatives linked to FAIRMODE. One of those was to set up a group of experts with skills in different areas to assess RM methodologies and propose common criteria and procedures for source apportionment studies. The infor-mation collected was summarized in a document which is intended to serve as a basis for a common Receptor Model Technical Protocol (RMTP). The RMTP is addressed to different kinds of us-ers: Policy makers and managers interested in the output of RMs for development of mitigation measures, air quality experts and scientists unfamiliar with these tech-niques, and RM practitioners involved in the model ex-ecution and interpretation of results. In order to address such heterogeneous readership the document was designed to be accessible at different levels. The RMTP is organized in three sections : an introduction to present the methodology to the unskilled reader, illustrating its capabilities and recom-mending when and how to use it ; a core section concerning the most common RM methodologies for source apportionment with in-depth analysis boxes for more experienced readers. The section is structured in 13 chapters following, as a check list, the logical steps to accomplish a source apportionment study. The first chapters deal with preliminary activities like the evaluation of the study area, collection of exist-ing information and experimental design. In the follow-ing chapters data collection and data treatment are dis-cussed. The section includes also chapters regarding spe-cific aspects of widely used methodologies like CMB, Factor Analysis, and PMF. The evaluation of test per-formance and reporting are discussed in the last part of this section ; the third section of the document was conceived to give an insight on the capabilities and the possible future trends in RM methodology. It consists of four chapters dealing with advanced, innovative techniques for which ready- to-use tools are already available or under development : trajectory analysis combined with RM, constrained and expanded models, AMS data processing, and the aethalometer model. The document includes a number of annexes to provide additional and practical information on specific topics, and examples.

Ambient non-refractory PM1 aerosol particles were measured with an Aerodyne High Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-AMS) at an elementary school 18 m from the US 95 freeway soundwall in Las Vegas, Nevada, during January 2008. Additional collocated continuous measurements of black carbon (BC), carbon monoxide (CO), nitrogen oxides (NOx), and meteorological data were collected. The US~Environmental Protection Agency's (EPA) positive matrix factorization (PMF) data analysis tool was used to apportion organic matter (OM) as measured by HR-AMS, and rotational tools in EPA PMF were used to better characterize the solution space and pull resolved factors toward known source profiles. Three- to six-factor solutions were resolved. The four-factor solution was the most interpretable, with the typical AMS PMF factors of hydrocarbon-like organic aerosol (HOA), low-volatility oxygenated organic aerosol (LV-OOA), biomass burning organic aerosol (BBOA), and semi-volatile oxygenated organic aerosol (SV-OOA). When the measurement site was downwind of the freeway, HOA composed about half the OM, with SV-OOA and LV-OOA accounting for the rest. Attempts to pull the PMF factor profiles toward source profiles were successful but did not qualitatively change the results, indicating that these factors are very stable. Oblique edges were present in G-space plots, suggesting that the obtained rotation may not be the most plausible one. Since solutions found by pulling the profiles or using Fpeak retained these oblique edges, there appears to be little rotational freedom in the base solution. On average, HOA made up 26% of the OM, while LV-OOA was highest in the afternoon and accounted for 26% of the OM. BBOA occurred in the evening hours, was predominantly from the residential area to the north, and on average constituted 12% of the OM; SV-OOA accounted for the remaining third of the OM. Use of the pulling techniques available in EPA PMF and ME-2 suggested that the four-factor solution was very stable.

Rotational ambiguity is a major problem in the application of factor analysis to a variety of multivariate mixture resolution problems and particularly important in the analysis of environmental data. As part of the development of an implementation of positive matrix factorization for the U.S. Environmental Protection Agency to distribute in support of airborne particle source identification and apportionment, several tools have been developed to control the rotations and explore the extent of rotational solution space. This discussion of the nature of the rotational tools provides an understanding of these rotational tools as well as the basis for their incorporation into other implementations of least‐squares based solutions to the factor analysis problem. Several example algorithms are presented. Copyright © 2008 John Wiley & Sons, Ltd.

Current approaches and recent developments in methods and software associated with multivariate factor analysis and related methods in the analysis of environmental data for the identification, resolution and apportionment of contamination sources are discussed and compared. The chapter first focuses on techniques to be applied in the analysis of the various factors contributing to contamination of the environment, among which we list Principal Component Analysis, and alternative methods and tools such as: Unmix, Positive Matrix Factorization (PMF) and the Multilinear Engine (ME), and Multivariate Curve Resolution Alternating Least Squares (MCR-ALS). In cases where uncertainties were not experimentally available, the use of jackknife uncertainty estimations methods for receptor modelling is described and recommended. Time series extension of multivariate receptor modelling has been developed to account for temporal dependence in air pollution data into estimation of source compositions and uncertainty estimations. In the case of air pollution it is also of interest to estimate the average concentration of a given pollutant at the monitoring site after the air masses have travelled over a certain point (source) on the map. The use of non-parametric regression (kernel smoothing) methods proves to be useful. A method is given for source apportionment of local sources of air pollution by non-parametric regression of the concentration of a pollutant on wind speed and direction. Non-parametric methods have shown that nearby sources (such as freeways) are not always important contributors to high pollutant concentrations. Finally, a number of applications of receptor modelling techniques are presented, including source identification by the new CATT tool (Combined Aerosol Trajectory Tools). Ensemble backward trajectory techniques have been employed to identify regional origins of air pollutants subject to synoptic-scale atmospheric transport. In another application, one detailed study is reported with the analysis of the results obtained from the application of PMF and from PCA-MLRA (Principal Component Analysis with Multilinear Regression Receptor Modelling) to one dataset containing compositional PM10 data at an industrial site in Northern Spain. Even though similar results were obtained with both models, PMF achieved a higher level of detail in the apportionment of sources than PCA-MLRA. However, it was also noted that the application of PMF is more time consuming (at least 50% more time) than PCA-MLRA.

Ambient particulate chemical composition data acquired from samples collected using a three-stage Davis Rotating-drum Universal-size-cut Monitoring (DRUM) impactor in Detroit, MI, between February and April 2002 were analyzed through the application of a three-way factor analysis model. PM2.5 (particulate matter ⩽2.5μm in aerodynamic diameter) was collected by a DRUM impactor with 3-h time resolution and three size modes (2.5μm>Dp>1.15μm, 1.15μm>Dp>0.34μm and 0.34μm>Dp>0.1μm). A novel three-way factor analysis model was applied to these data where the source profiles are a three-way array of size, composition and source while the contributions are a matrix of sample by source. Nine factors were identified: road salt, industrial (Fe+Zn), cloud processed sulfate, two types of metal works, road dust, local sulfate source, sulfur with dust, and homogeneously formed sulfate. Road salt had high concentrations of Na and Cl. Mixed industrial emissions are characterized by Fe and Zn. The cloud processed sulfate had a high concentration of S in the intermediate size mode. The first metal works represented by Fe in all three size modes and by Zn, Ti, Cu, and Mn. The second included a high concentration of small size particle sulfur with intermediate size Fe, Zn, Al, Si, and Ca. Road dust contained Na, Al, Si, S, K, and Fe in the large size mode. The local and homogeneous sulfate factors show high concentrations of S in the smallest size mode, but different time series behavior in their contributions. Sulfur with dust is characterized by S and a mix of Na, Mg, Al, Si, K, Ca, Ti, and Fe from the medium and large size modes. This study shows that the utilization of time and size resolved DRUM data can assist in the identification of sources and atmospheric processes leading to the observed ambient concentrations.