Spatial modeling applied to environmental monitoring: identifying sources of potentially toxic metals in aquatic system.

Aquatic contamination by potentially toxic metals is a problem that has been aggravated, especially due to the quantity and the diversity of sources. Locating these sources is not always an easy task, especially because of the wide variety of possibilities. In this context, the application of geostatistical methods may represent an excellent tool to find out sources of metal contaminants in aquatic systems. Thus, the objective of this work was to elaborate an approach to identify sources of potentially toxic metals (Zn, Ba, Pb, Cr, Mn and Fe), by relating their spatial-temporal variations with the local land use patterns, along a longitudinal profile of the Pirapora River, located in the State of Sao Paulo, Brazil. For this purpose, water samples were collected at different points, taking into consideration each specific land use pattern and quantifying the metals contents by microwave plasma atomic emission spectrometry. In this work, thirteen land use patterns have been identified: mining, forestry, abandoned pasture, water, urban area, human occupation, floodplain, bare soil, temporary crop, roads, forest, streets and pasture. The results revealed temporal variations for the metals Ba, Cr, Fe, and Pb and spatial for Zn and Mn, making possible to correlate the presence of these two latter metals with mining and forestry, the most proeminent activities in the region. Overall, this work proposes a model which brings together geoprocessing and analytical methods, in order to correlate spatial-temporal variations of potentially toxic metals with specific land use patterns of a determined region, aiming the environmental monitoring.