Abstract:
Concentrations of heavy metals in the soils of contaminated river floodplains can vary greatly over short distances, making it difficult for spatial modeling with the usual approaches. Linking the trace metal contents to the morphography of the terrain using modern geostatistical methods, applied in digital soil mapping, would allow the creation of a predictive model for spatial distribution of pollutants. The research in the project „Relationship of the spatial distribution of heavy metals in the soil with the morphology of polluted floodplains (TOPOMET)“ aims to perform predictive modeling of the spatial distribution of certain heavy metals in polluted alluvial deposits and floodplain soils in selected sections of the Danube river valleys – Archaro-Orsoyska lowland, Ogosta and Lom rivers catchments, using remote sensing methods, spatial data analysis and geostatistics. They will be used for GISbased analysis of a set of morphometric parameters of the topography of the valleys. The information on the parameters and contents of heavy metals in nearly 400 soil samples will be processed using Regression-Kriging Techniques and Generalized Linear Models (GLM) to reveal the dependencies between them and to fit regression models. With the help of the latter, an attempt will be made to predict the spatial distribution of selected heavy metals in the soils of the studied areas. As a result of the project, it is expected to establish morphometric parameters that the best describes spatial variation of heavy metal contents in soils and can be used in different predictive models. An attempt will be made to develop a technology for the spatial modeling of heavy metal content in the soils of floodplain terraces. This will make it possible to identify and detail the extent of pollution in river valleys as a result of mining extraction and metallurgical production and to take measures to minimize the health risk for the local population.