Concentração de metais em um córrego urbano no município de Alfenas-MG: caracterização e modelagem bayesiana

Abstract

Launching of domestic and industrial effluents, treated or untreated, directly into water bodies is the major cause of depreciation of water quality. Insertion of contaminants such as trace metals, through point and diffuse sources of pollution, can hugely alter natural biogeochemical cycles of these metals, can be toxic for living organisms and cause harm for biota and human health. Thus, it is essential to monitor environmental samples contaminated by metals through effective and precise models. This study aimed to characterize and establish bayesian models for water and sediment quality from different points of an affluent of Reservatório de Furnas, Alfenas-MG, which receives discharges of industrial and domestic sewage. Water and sediment samples were collected monthly in five points of the stream, during 2010/10-2011/07. Trace metals lead, cadmium and zinc were quantified and were fitted bayesian simple linear regression models for cadmium from water and sediment. The stream has metal concentrations and physicochemical parameters in disagreement with CONAMA 357/2005, Class II. Lead concentrations in nascent stream are above the limits established by decree nº 2914 of Ministério da Saúde, representing risk for population. During the study period was observed transfer of cadmium in the water column to sediment. Bayesian simple linear regression models presented as a precise tool for assessing and monitoring environmental samples contaminated by trace metals, which have linear behavior.