Sergey Yuryevich Rozov

Approximately one third of the scientific publications dedicated to digital soil mapping includes morphometric variables as predictors for soil cover models. For each specific territory with its own set of specific landforms, a unique set of significant morphometric variables is needed to describe the features of such a complex system as soil cover, selected directly with the research tasks. Thus, the aim of this research is to find relationships between soils taxa as vell as its morphological features determining soil diagnostic at different taxonomic levels, and morphometric variables in the case study for the farmlands soil cover on the left bank of the Don River in the Voronezh region, Liskinsky district. Soil cover consists not only of bleached, leached, and typical chernozems, but also chernozem-like meadow soil. Methods of parametric and nonparametric statistics confirm that the most important morphometric variables influencing all analyzed soil properties with diagnostic importance for determining soil allocations are  catchment area, vertical curvature and minimum curvature. The largest number of correlations between morphometric variables and soil properties occurs at maximum level of model generalization in the range of 17 m, 51 m, 102 m, 306 m.