Sediment-quality assessment by intelligent data analysis

Abstract

The present study deals with the assessment of the quality of sediments using chemometric interpretation of monitoring data from physicochemical parameters and toxicity tests. The dataset comprised results from lake-sediment samples. All 260 samples were analyzed for 44 physicochemical parameters, as well as for acute and chronic toxicity by respective biological tests. The monitoring results were considered both separately (chemical and toxicity data) and simultaneously in order to identify pollution sources in the region and to detect correlation between chemical and toxicity tests. The chemometric methods applied were cluster analysis and principal components analysis. Our effort to establish some simple relation between the two different types of parameters measured (chemical and ecotoxicity) did not give any hint of serious correlation between them. Obviously, sediment assessment needs to be done in two separate analytical tracks - chemical (as until now) and ecotoxicological (which uses biological tests to obtain independent information about acute or chronic toxicity). © 2007 Elsevier Ltd. All rights reserved.