Interpretation of Heavy Metal Concentrations in the Surficial Sediment from the Makirina Bay Using Aluminium as Reference Element

Abstract

Heavy metal concentrations were studied in surficial sediment from Makirina Bay to determine whether natural or anthropogenic enrichment of the metals is occurring. Metal concentrations of A1, Co, Cu, Ni, Pb and Zn were determined using inductively coupled plasma emission spectrometry. The coastal sediment of the Makirina Bay in the central Adriatic Sea represents an important source of healing marine mud which could be used for medical treatment. The sediment constitutes mostly of clay minerals, quartz, dolomite and calcite. Halite, gypsum, aragonite and pyrite are also present. Particulate metals from natural and anthropogenic sources accumulate together in the sediment. In order to evaluate the level of metal pollution in the area the proportions originating from the two sources should be determined (Din, 1992). The problem is that metals occur naturally in sediments and that their concentrations vary with mineral composition and grain size. To overcome this problem, normalization to a reference element is a useful tool (Loring, 1991, Schropp et al. , 1990). With such a geochemical method it is possible to compensate for mineralogical as well as the natural granulometric variability of metal concentrations in the sediment. With such normalization method it is also possible to detect the anthropogenic metal contribution to the sediment (Din, 1992). Heavy metals in coastal sediments are usually associated with fine-grained fraction mainly constituted of clay minerals (Loring, 1991). One of the main components of these minerals is aluminium which is generally not influenced by anthropogenic sources (Din, 1992) and because of this it is recommended as a reference element (Din, 1992, Schropp et al. , 1990). Regression analysis for log-transformed data was nm for Co, Cu, Ni, Pb and Zn using A1 as the independent variable. Then heavy metal:aluminium ratios were calculated. Data points on the regression line with higher or lower ratios than the 95% prediction limit were removed. High ratios probably indicate anthropogenic metal enrichment (Schropp e t al. , 1990). In our case three data points in the vicinity of the Kapela-Tisno road with high heavy metal:aluminium ratios show an anthropogenic enrichment in the surficial sediment of Makirina Bay. Those data points were removed. All five heavy metals exhibit a highly significant positive linear correlation with aluminium (Table 1) with correlation coefficient 0.92 < r < 0.98. After Loring (1990) the strength of the correlation varies from element to element and from area to area because of variation in mineralogical composition related to origin rather than to particle size. Scatterplots for Co, Cu, Ni, Pb and Zn vs A1 are shown on Fig. 1. It is interesting that the regression lines of Ni, Pb and Zn have very similar slopes (Fig. 1, Table 1). The slopes of regression lines of Co and Cu are slightly gentle (Fig. 1, Table 1). The similar slopes of regression lines and high positive correlation