A statistical approach applied to trace metal data from biomonitoring studies.

This work underlines the importance of the statistical procedures necessary for the significant assessment of analytical data of heavy metals derived from marine organisms (biomonitors) collected at five stations in Ustica at the Southern Tyrrhenian Sea. The statistical method described indicates that, where the differences among the averages of Pb, Cu and Zn concentrations in the molluscs of the sites analysed tested with ANOVA are significant, it is necessary to conduct multiple comparison tests in order to determine which average is significantly different from the other. Using the method described, it is possible to indicate which site or sites is/are more contaminated in comparison with the other sites. This way it is possible to make a preliminary evaluation of the basal levels of trace metals present in the reference marine ecosystem under study. [ABSTRACT FROM AUTHOR]