Comparison of Automated Landform Classification and Soil Mapping Units at a Farm Level

Abstract

More than twenty morphometric attributes derived from digital elevation models (DEMs) were correlated with soil mapping units available from conventional soil maps at a scale of 1 : 2,000. The following topographic attributes were involved into the analysis: slope steepness, slope aspect, elevation, horizontal curvature, vertical curvature, mean curvature, surface insolation, maximal curvature, minimal curvature, catchment area, depression depths, dispersal area, hill heights, gradient factor, difference curvature, horizontal and vertical excess curvatures, rotor, unsphericity, total Gaussian curvature, total accumulation curvature, total ring curvature, and topographic index. A DEM with a grid size of 30 m was best suited to describe the relationships between soil units and terrain variables. Soil typological units and their classes correlated stronger with the landform types derived using the Shary landform classification (the contingency coefficients were over 0.6) rather than the Troeh and Gauss landform classifications. Mean and horizontal curvatures, elevation, slope, gradient factor, and topographic index were best correlated with the soil typological units of the soil classification system of Lithuania, which is based on the international soil classification.