Effectiveness of a soil mapping geomatic approach to predict the spatial distribution of soil types and their properties

A soil map (1:50,000 scale) was recently produced in Sardinia (Italy) using a cost-effective GIS approach. In this study we aimed to verify, two pilot areas and by means of statistical analysis, the effectiveness adopted methodology representing predicting spatial distribution types properties. We focused on evaluation 1) influence landforms parent materials (WRB Reference Soil Groups) selected properties 2) suitability for calibrating model predict land unit composition terms different types. Leptosols, Regosols Cambisols were prevalent slopes, with Leptosols being more frequent convex slopes concave slopes. flat areas, mainly depended type age material, prevailing Holocene deposits highly developed soils (mainly Luvisols) largely Pleistocene deposits. On hard rock, very terrigenous metamorphic rock granite. Besides occurred frequently than both materials. Landforms strongly influenced depth available water capacity. Soils plains deeper those where forms had shallower forms. similar trend applied The material significant effect topsoil (thickness, texture, pH organic carbon content) belonging same WRB Group (analysis done most relevant Groups, i.e. Cambisols). calibrated tested stepwise multiple linear regressions (MLR) general models (GLM) units Groups. gave results, distinct patterns that coherent relationships observed between groups specific combination terrain attributes Results showed reliable absence rather presence given type.