Annals of “dunarea De Jos” University of Galati – Fascicle

Multivariate statistical methods such as Cluster Analysis (CA) and Principal Component Analysis (PCA) were used for the study of distribution and sources of 16 polycyclic aromatic hydrocarbons (PAHs) in topsoils from seventeen locations in SE Romania, including industrial, agricultural and natural soils. The investigated dataset consists of concentrations of individual PAHs listed by the US Environmental Protection Agency (EPA) and the European Community as priority pollutants – naphthalene (NAP), acenaphthylene (ACY), acenaphthene (ACE), fluorene (FLU), phenanthrene (PHE), anthracene (ANT), fluoranthene (FLN), pyrene (PYR), chrysene (CHR), benz[a]anthracene (BaA), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), dibenzo[a,h]anthracene (DahA), benzo[g,h,i]perylene (BghiP) and indeno[1,2,3-cd]pyrene (InP) and the total concentration of PAHs (TPAHs) in Romanian surface soils, which have been previously determined by GC-MS technique. Clustering of the variables highlights that NAP is the only variable which differs drastically from all others. It may mean that this parameter comes from a different source of pollution as compared to the other variables. The rest of variables form a homogeneous cluster with the exception of FLU and ACE, but they should not be excluded from the group of the other chemicals, since they fall into significance level of clustering (33.3% of squared Euclidian distance Dmax). The slight difference between FLU and ACE on one hand, and the rest of PAHs, on the other hand, could be due to different sources of emission, different absorption on the soil surface, soil properties etc. Clustering of cases (sampling sites) also points out the similarity between all sampling sites. The only drastic exception is a cattle farm site (Galati County), being very different from the other 16 sites, as mentioned by the input data the most polluted site with the highest concentrations of organics. The source of PAHs in this location could be the systematic incineration of agricultural residues and the using of vehicles for various farm practices or generators. Using PCA factor loadings, only one latent factor was identified, explaining almost 90% of the total variance. It means that one major source of pollution (combustion) is available in the region of study, being the source of all chemicals except for NAP, as indicated also by cluster analysis. The fact that NAP is not related to the other chemicals (even it shows a negative correlation to the others) speaks for the possible existence of another pollution source specific to NAP.