Improved identification of pollution source attribution by using PAH ratios combined with multivariate statistics

Abstract

Polycyclic aromatic hydrocarbons (PAHs) are contaminants introduced by different pathways in the marine ecosystem, affecting both aquatic and sediment bodies. Identification of their sources is of vital importance for protecting the marine ecosystem. The attribution of the pollution sources is usually made by using diagnostic molecular ratios of PAHs isomers. The reliability of this approach diminishes when PAHs contents are measured far from their original source, for example in water bodies or in bottom sediments. Conventionally the source attribution is based on time consuming univariate methods. In the present work coupling of molecular ratios with advanced supervised statistical techniques was used to increase the accuracy of the PAH source attribution in bottom sediments. Data on PAHs distribution within 5 port areas, with known pattern port activity, were collected. Evaluation of multiple PAHs ratios at once by means of supervised OPLS-DA technique was performed. A robust descriptive and predictive model was set up and successfully validated. The proposed methodology helps identify PAH transport pathways, highlighting interactions between pollution patterns, port activities and coastal land-use supporting decision makers in defining monitoring and mitigation procedures.