Study of parameters affecting the behaviour of trace elements in a polluted estuary. Canonical correlation analysis as a tool in environmental impact assessment

Understanding metal behaviour in estuaries is a difficult task. It involves highly dynamic systems continuously subjected to fast changes in environmental conditions governed by alternating physico-chemical parameters and human activity. In order to distinguish the most important environmental factors that determine the behaviour of trace elements in a polluted estuary, water (superficial and deep, at low and high tides) and sediment samples were collected along the Nerbioi–Ibaizabal River estuary (Basque Country, Spain) every three months for six years. The environmental dataset consisted on the concentration of trace elements (Al, As, Cd, Co, Cu, Cr, Fe, Mg, Mn, Ni, Pb, Sn, V and Zn in sediment samples and eight of them in water samples) and physico-chemical properties (temperature, redox potential, dissolved oxygen percentage, conductivity, pH for water samples and also the content of carbonates, total organic carbon (TOC), humic acids (HA) and fulvic acids (FA) in sediment samples). The study of these datasets with canonical correlation analysis revealed the existence of a strong link between different physico-chemical properties and the concentration of certain metals in the estuary. In sediments, for example, strong correlations were found between the carbonate content and some of the studied trace elements. In the water, on the other hand, evident relationships were found between the salinity and the pollutants.

► Deep understanding of the metal behaviour in estuaries.
► CCA for combining the chemical information with environmental variables.
► Monitoring of two main matrices: water and sediments.
► Existence of a linkage between environmental properties and the metal content.