Advanced multivariate techniques for the classification and pollution of marine sediments due to aquaculture.

Aquaculture production has globally increased and its environmental impact is not well understood and assessed yet. Therefore, in this work nine metals and metalloids (Cu, Cd, Pb, Hg, Ni, Fe, Mn, Zn and As) and three nutrients (P, N and C) that seem to accumulate in marine sediments, were determined under the fish cages (zero distance) and about 50 and 100 m away from them, in three aquacultures in Greece. The analysis of these data is crucial due to the negative impact of the intensive aquaculture activities on fish population, human health and marine environment. This study investigated the environmental impact associated with aquaculture cages on marine sediments, using Supervised Artificial Neural Networks (ANNs) in parallel with Classification Trees (CTs). Optimised models were constructed in order to detect the significance of each variable, predict the origin of the sediment samples and successfully visualise their results. Three popular ANN architectures, as multi-layer perceptrons (MLPs), radial basis function (RBF) and counter propagation artificial neural networks (CP-ANNs) were used to assess the impact of the intensive aquaculture activities on marine sediments. In addition, more traditional multivariate chemometric techniques like CTs were applied to the same data set for comparison purposes. The modelling study showed that P, N, Cu, Cd were the most critical (and polluting) factors of those metals studied. Moreover, single-element models achieved elevated predictive percentages. The results were justified due to the usual practices used for fish feeding or cages maintenance.

Environmental impact of intensive aquaculture: investigation on the accumulation of metals and nutrients in marine sediments of Greece.

The impact of intensive aquaculture activities on marine sediments along three coastal areas in Greece was studied. The content of nine metals/metalloids (Cu, Cd, Pb, Hg, Ni, Fe, Mn, Zn, As), and three nutrients (P, N and C), that seem to accumulate in marine sediments, was determined under the fish cages (zero distance) and away (50 or 100 m) from them. Elevated concentrations for phosphorus, nitrogen, copper, zinc and cadmium were recorded in the areas where farming establishments are moored. In parallel, the intrinsic differences between the aquaculture facilities and their seasonal variations were investigated. The individual characteristics of each farm (local water currents, facilities' capacity, transferring mechanisms or the geological background) were the determinant factors. On the contrary, significant seasonal differences were not recorded. Statistical techniques, as the non-parametric Mann-Whitney U and Kruskal-Wallis tests and principal components analysis (PCA), factor analysis (FA) and discriminant analysis (DA), were used for the evaluation of the results. These chemometric tools succeeded to discriminate the sampling points according to their distance from the cages or the origin of the sample. Variables' significance, correlations and potential accumulation sources were also investigated.

A comparative chemometric study for water quality expertise of the Athenian water reservoirs.

The aim of the present study is to compare the application of unsupervised and supervised pattern recognition techniques for the quality assessment and classification of the reservoirs used as the source for the domestic and industrial water supply of the city of Athens, Greece. A new optimization strategy for sampling, monitoring, and water management is proposed. During the period of October 2006 to April 2007, 89 samples were collected from the three water reservoirs (Iliki, Mornos, and Marathon), and 13 parameters (metals and metalloids) were analytically determined. Generally, all the elements were found to fluctuate at very low levels, especially for Mornos that comprises the main water reservoir of Athens. Iliki and Marathon showed relatively elevated values, compared to Mornos, but below the legislative limits. Multivariate unsupervised statistical techniques, such as factor analysis/principal components analysis, and cluster analysis and supervised ones, like discriminant analysis and classification trees, were applied to the data set, and their classification abilities were compared. All the chemometric techniques successfully revealed the critical variables and described the similarities and dissimilarities among the sampling points, emphasizing the individual characteristics in every sample and revealing the sources of elements in the region. New data from posterior samplings (November and December 2007) were used for the validation of the supervised techniques. Finally, water management strategies were proposed concerning the sampling points and representative parameters.