Stream sediment pollution: a compositional baseline assessment

Abstract

A high concentration of potentially toxic elements (PTEs) can affect ecosystem health in many ways. It is therefore essential that spatial trends in pollutants are assessed and monitored. Two questions must be addressed when quantifying pollution: how to define a non-polluted sample and how to reduce the problem’s dimensionality. A geochemical dataset is a composition of variables (chemical elements), where the components represent the relative importance of each part of the whole. Therefore, to comply with the compositional constraints, a compositional approach was used. A novel compositional pollution indicator (CPI) based on compositional data (CoDa) principles such as the properties of sparsity and simplicity was computed. A dataset of 12 chemical elements in 33 stream-sediment samples were collected from depths of 0–10 cm in a grid of 1 km × 1 km and analyzed. Maximum concentrations of 3.8% Pb, 750 μg g− 1 As, and 340 μg g– 1 Hg were obtained near the mine tailings. The methodological approach involved geological background selection in terms of a trimmed subsample that could be assumed to contain only non-pollutants (Al and Fe) and the selection of a list of pollutants (As, Zn, Pb, and Hg) based on expert knowledge criteria and previous studies. Finally, a stochastic sequential Gaussian simulation of the new CPI was performed. The results of the hundred simulations performed were summarized through the mean image map and maps of the probability of exceeding a given statistical threshold, allowing the characterization of the spatial distribution and the associated variability of the CPI. A high risk of contamination along the Grândola River was observed. As the main economic activities in this area are agricultural and involve animal stocks, it is crucial to establish two lines of intervention: the installation of a surveillance network for continuous control in all areas and the definition of mitigation actions for the northern area with high levels of contamination.