Classification of marine environment samples based on chromatographic analysis of hydrocarbons and principal component analysis

Abstract

The concentration of polycyclic aromatic hydrocarbons (phenanthrene, anthracene, fluoranthene, pyrene, benz(a)anthracene/chrysene, benzo(a)pyrene, benzo(e)pyrene/perylene, benzo(k)fluoranthene and benzo(ghi)perylene) and n-alkanes ranging from n-hexadecane (C-16) to n-dotriacontane (C-32) in sixty sediment and water samples from the southern Baltic Sea were determined. Since hydrocarbon concentration itself cannot be used for estimation of marine environment pollution, multivariate classification and comparison using principal component analysis (PCA) has been made on the data matrices. The connection of the principal component score values with the geography of the sampling area was investigated. Some suggestions on the usefullness of the aforementioned parameters for the estimation of marine environment contamination are given.