Mapping soil fertility properties in central Ethiopia at 100 m spatial resolution

• Published in: Geoderma Regional Vol. 41; p. e00952
• Main Authors: Redi, Musefa A., Heuvelink, Gerard B.M., Leenaars, Johan G.B.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.06.2025
• Subjects: Central Ethiopia; data collection; Digital soil mapping; DSM comparison; Ethiopia; exchangeable potassium; fertilizers; nitrogen; nutrient management; organic carbon; phosphorus; prediction; Prediction uncertainty; regression analysis; soil fertility; Soil fertility properties; soil profiles; soil properties; subsoil; topsoil; Topsoil data integration; uncertainty; Ethiopia; Central Ethiopia; Soil fertility properties; Digital soil mapping; Prediction uncertainty; Topsoil data integration; DSM comparison
• DOI: 10.1016/j.geodrs.2025.e00952
• ISSN: 2352-0094

Accurate soil property maps are essential for effective soil nutrient management. In Ethiopia, fertilizer applications often ignored spatial variability in soil properties, leading to inefficiencies. This study employed digital soil mapping to generate three-dimensional (3D) maps of five key soil fertility properties—total nitrogen (TotalN), extractable phosphorus (OlsenP), exchangeable potassium (ExchK), pH-H2O (pH), and organic carbon (OC)—at 100 m resolution across central Ethiopia. The objectives were to (1) develop maps at six depth intervals while assessing prediction uncertainty; (2) evaluate the integration of topsoil data with soil profile data for model training; and (3) compare the maps with Africa-SoilGrids, SoilGrids, and iSDAsoil maps. We used two datasets: soil profile (1,379 profiles with 4,179 layers) and topsoil (13,724 locations), harmonized with transfer functions. Quantile regression forest was used to generate maps with 90 % prediction intervals. Models were calibrated with 80 % of the dataset and 194 covariates, including depth, and evaluated with the remaining 20 %. Integrating topsoil data with the soil profile dataset improved prediction accuracy for the five soil fertility properties in the topsoil (0–20 cm), demonstrating near-zero bias, reduced root mean squared error, and a higher model efficiency coefficient (MEC), compared to only using the soil profile dataset. It also enhanced uncertainty quantification for pH, OlsenP, TotalN, and ExchK in the topsoil. However, these benefits diminished with depth, with slight improvements in the subsoil (20–50 cm) but none in the deeper layers (50–200 cm) where pH and OC predictions were even slightly biased. Among the combined dataset models, the highest performance was for pH (MEC = 0.80), while the lowest was for OlsenP (MEC = 0.13). The maps generated from the combined dataset models showed MEC improvements of 27 % to over 1,000 % compared to SoilGrids, Africa-SoilGrids, and iSDAsoil. Additionally, the prediction intervals were also a realistic representation of the prediction uncertainty, with prediction interval coverage probability (PICP) values close to their ideal value. This markedly outperformed SoilGrids and iSDAsoil, which had unrealistically low PICP values. We conclude that the 100 m resolution 3D maps from this study offer satisfactory accuracy and realistic uncertainty quantification, making them the currently best available resource for developing location-specific fertilizer recommendations in central Ethiopia.