Soil aggregate stability within morphologically diverse areas

Knowledge of spatial distribution of soil aggregate stability as an indicator of soil degradation vulnerability and its possible prediction are required for many scientific and practical environmental studies. The goal of our study was to provide a model for predicting soil aggregate stability within morphologically diverse areas, where soil properties have been affected by soil material redistribution due to erosion. The study was performed on a study site (6 ha area) in the loess region of Southern Moravia, Czech Republic. Haplic Chernozem, which is an original dominant soil unit, has been transformed into different soil units (eroded phases of Chernozem, Regosol, colluvial Chernozem and Colluvial soil). 36 sampling spots were selected in order to represent diverse soils. The following soil properties were measured: oxidable organic carbon content (C-ox), CaCO3 content, pH(H2O), PHKCl, soil particle density (rho(s)), bulk density (rho(d)), porosity (P), actual field soil-water content (theta(field)), content of iron and manganese (in ammonium oxalate extract, Fe-o and Mn-o, and dithionite-citrate extract, Fe-d and Mn-d) and mass specific magnetic susceptibility (chi(lf) and chi(hf)). The aggregate stability was assessed using various tests to study different disruption mechanisms. Terrain attributes were derived from a digital elevation model. In general, the lowest soil aggregate stability was observed on steep slopes, which were highly impacted by soil erosion. The highest aggregate stability was measured on soils sampled at relatively flat upper parts, which were less influenced by erosion processes. Higher stability was also obtained on toe slopes, where the sedimentation of previously eroded soil material occurred. The simple correlations revealed that characteristics resulting from the tests studying aggregate slaking due to the compression of the entrapped air (Water Stable Aggregate index and coefficient of vulnerability from fast wetting test) were positively impacted by the C-ox, P, Fe-o, Mn-o, Fe-d, Mn-d, chi(lf) and chi(hf) values, and negatively by the rho(d) value. The soil aggregate stability was also negatively influenced by the plan and total terrain curvatures, i.e. larger aggregate stability was measured at concave parts in comparison with that at convex parts. Almost no statistically significant relationships were found in the case of the tests evaluating either aggregate disintegration caused by the micro-cracking due to the different swelling, or by the physico-chemical dispersion due to the osmotic stress or the mechanical aggregate breakdown. The multiple linear regressions resulted in the model for estimating the WSA index using the C-ox content, total terrain curvature and actual field soil-water content (theta(field)). In this model the C-ox content positively and the total terrain curvature and theta(field) value negatively influenced the value of the WSA index. Since C-ox was positively related with iron content and thus also with the magnetic susceptibility, the alternative model was proposed for less costly and time consuming WSA estimation. The WSA index may be predicted by combining the mass specific magnetic susceptibility (chi(lf) and chi(hf)), total terrain curvature and actual field soil-water content (theta(field)). (C) 2015 Elsevier B.V. All rights reserved.