U.S. flag

An official website of the United States government


Main content area

Environmetrics to evaluate marine environment quality

Spanos, Thomas, Simeonov, Vasil, Simeonova, Pavlina, Apostolidou, Eleni, Stratis, John
Environmental monitoring and assessment 2008 v.143 no.1-3 pp. 215-225
heavy metals, lead, marine environment, cluster analysis, carbon, Mollusca, sediments, linear models, data collection, least squares, cadmium, coasts, cobalt, copper, Polychaeta, benthic organisms, principal component analysis, chromium, Crustacea, zinc, nickel, Bulgaria, Black Sea
The environmetric data analysis of analytical datasets from sediment and benthic organisms samples collected from different sampling sites along the coast of Black Sea near to City of Varna, Bulgaria has given some important indications about the bioindication properties of both type of samples. Various multivariate statistical methods like cluster analysis, principal components analysis, source apportioning modeling and partial least square (PLS) modeling were used in order to classify and interpret the parameters describing the chemical content of the coastal sediments (major components, heavy metals and total organic carbon) and benthic organisms (heavy metals). It has been shown that seriously polluted coastal zones are indicated in the same way by all benthic species, although some specificity could be detected for moderate polluted regions' e.g. polychaeta accumulated preferably Co, Cr, Cu, and Pb; crustacea - As, Cd, and Ni; mollusca - Zn. The identified latent factors responsible for the dataset structure are clearly indicated and apportioned with respect to their contribution to the total mass or total concentration of the species in the samples. The linear regression and PLS models indicated that a reliable forecast about the relation between naturally occurring chemical components and polluting species accumulated in the benthic organisms is possible.