The activity is focused on the assessment of soil load, the spatial distribution of pollutants, definition of stress and especially vulnerable areas, mapping local variability of soil properties at the level of the land in relation to the behaviour of substances in the soil, and on mathematical modeling transport and behaviour of substances, in order to provide modern means, IE. the validated models and data bases for the effective evaluation of the transport of substances in the environment. One of the input data for these models obtained in the framework of a project is a detailed analysis of the spatial layout of the soil (soil structure, the distribution of the pores, the layout of the roots), among others. using the methods of computer tomography.
This is the crucial information that contribute significantly to improve the predictive ability of the models, which will then be able to better predict the possible contamination of ground water, plants, agricultural products, etc.
Mapping and spatial estimation are focused in two directions:
1) Reviews the basic layout of the soil properties important for the conduct of controlled substances in the soil (e.g., pH, organic matter content, grain size, etc.) as inputs into simulation models. The result of the conjunction of the spatial estimation and simulation models are the prediction of the behaviour of substances in soil and their variability in space.
This direction is focused more on the local level land surround the estimate, and as the input values for the 3-d models of the estimate will be used. the yield maps or soil sensors (especially soil spectroscopy).
2) load evaluation of soil and spatial distribution of pollutants, definition of the traffic, and especially vulnerable areas.
It is a spatial estimates of greater range, i.e. regional to nationwide. For the production of maps of the spatial distribution of soil properties and the content of pollutants are used the methods of pedometriky and digital mapping of soils, such as geostatistika, regression trees and random forests, artificial neural networks, etc.
