Deriving World Reference Base Reference Soil Groups from the prospective Global Soil Map product — A case study on major soil types of Africa

摘要: Abstract The main objectives of the Global Soil Map (GSM) project are development high resolution maps for defined soil properties and new digital mapping tools. Although many current land use, capability models related applications use type information, a global map is not among GSM project. In this presented research legacy data was used to derive centroid values World Reference Base Groups based on in specified depth intervals test applicability numerical approaches classifying soils, these specifications. variables were strictly following definitions such as: organic carbon content, pH, electric conductivity, sand, silt, clay gravel bulk density effective cation exchange capacity, all calculated (0–5, 5–15, 15–30, 30–60, 60–100, 100–200 cm). No environmental covariates study. A taxonomic distance Random Forest algorithm tested classes with training validation profiles accuracy assessment. Results studied overall each Group. Reasons misclassifications identified, along properties, which could increase methods. study did class map, promising results show possibility from product as long reliable archive dataset available population. forest performed better 68% classification success compared 47% total number profiles. Classification differed by Group, those strict, well easily measureable and/or observable criteria certain property (like Arenosols) classified both methods other groups broader Cambisols). also unfolded completeness problems datasets, varying revealed narrow, precise may need future system performance