Analysis of Polycyclic Aromatic Hydrocarbons (PAHs) Sources and Vertical Distribution in Soils of the Age-Diverse Brownfields of Southern Poland Using Positive Matrix Factorisation and Data Mining Model

Abstract

In this study, we aimed at understanding a relation between PAH accumulation in the soils of brownfields that differed in origin (i.e., waste after coal mining, power plant, zinc–lead- or iron-ore processing) and age (400 years old to present). PAH contents were determined after their extraction with 2-propanol using a gas chromatograph (GC) equipped with a flame ionisation detector (FID) and a FactorFour VF-5MS capillary column. Sources of PAHs were identified using positive matrix factorisation (PMF). In regard to the relevance of the variables that helped explain the PAH accumulation, a random forest (RF) method was used. PMF analysis highlighted the connection of PAH sources with the industrial activity performed in the area. PAH concentrations increased with the age of tailings, but always exceeded the threshold-effect levels, indicating their potential to cause harm to humans. The RF method showed that for 2–3-ring PAHs the most important factor was the carbon content followed by clay, site, depth, pH, and then nitrogen, while for other PAHs, only the N content and depth exceeded 50% importance. PAHs’ vertical distributions reflected the history and the time passing from the brownfield site formation. Severe contamination of the brownfield soils indicates the necessity for remediation measures, especially where there are plans to build on these areas.