Publication details

Predictive spatial modelling of POPs background concentrations in the area of Czech Republic

Authors

KUBOŠOVÁ Klára KOMPRDA Jiří JARKOVSKÝ Jiří HÁJEK Ondřej DUŠEK Ladislav SÁŇKA Milan HOLOUBEK Ivan KLÁNOVÁ Jana

Year of publication 2008
Type Article in Proceedings
Conference Abstract book
MU Faculty or unit

Faculty of Science

Citation
Field Soil contamination adn decontamination incl. pesticides
Keywords predictive modelling; POPs; soil; spatial; CART
Description The goal of this study was to use the modelling tools for prediction of background environmental concentrations of selected in soils of the Czech Republic and to produce maps showing spatial distribution of selected POPs in the Czech Republic. A number of available data from several projects conducted by RECETOX and CISTA was collected for this purpose. Database of concentrations from the measurements mentioned above represents 534 samples distributed over the all area of the Czech Republic. The predictors used for the distribution modelling of the selected POPs were environmental variables such as a type of soil, organic carbon content, distance from industry, distance from populated areas, elevation and land use. Total area of the Czech Republic was divided into grid of 80 033 squares of 1 km2, using the ArcGIS 9.2. Our model was based on regression trees. Regression trees are well interpretable according to expectable behaviour of POPs in the environment and also models explain most of data variability.
Related projects: