A review of the main strategies used in the interpretation of similar chemical profiles yielded by receptor models in the source apportionment of particulate matter

Receptor models have been widely used for the source apportionment of airborne particulate matter. However, in the last 10 years, the use of factor analysis-based models, such as PMF and UNMIX, has increased significantly. The results yielded by these models must be interpreted by users who must know all variables influencing the modeling, and without this knowledge, the probability of incorrect interpretation of the source profiles may increase, especially when two or more sources have similar chemical profiles. Concerning the quality of data, this work shows that a broad characterization of PM composition, including inorganic, organic, and mineralogical species can improve this process, avoiding misinterpretation and the attribution of mixed or unidentified sources. This work aims to provide readers with some answers for a question often risen during source apportionment studies: Which source markers should be used for better separation and interpretation of source profiles? This review shows there is no right answer for this because different strategies can be used for this purpose. Therefore, this review aims to compile and highlight qualitatively the key strategies already used by several experienced receptor models users, combining the use of inorganic, organic, and mineralogical markers of PM for better separation and interpretation of the profiles yielded by receptor models. Also, this work presents a compilation in tables of the main chemical species reported in the literature as markers for interpreting the source profiles. (C) 2020 Elsevier Ltd. All rights reserved.

Automated summary

This article discusses the increasing use of factor analysis-based models in source apportionment of airborne particulate matter and the importance of considering all variables influencing the modeling process. It emphasizes the role of data quality in improving the interpretation of results and presents key strategies for better separation and interpretation of source profiles, highlighting the flexibility of different approaches.