Skip to main content
ARS Home » Plains Area » Temple, Texas » Grassland Soil and Water Research Laboratory » Research » Publications at this Location » Publication #416716
Research Project: Development of Enhanced Tools and Management Strategies to Support Sustainable Agricultural Systems and Water Quality

Location: Grassland Soil and Water Research Laboratory

Title: Digital mapping of agricultural soils texture of the Brazilian Cerrado biome

Author
item PELEGRINO, MARCELO - Federal University Of Lavras
item GUILHERME, LUIZ - Federal University Of Lavras
item Adhikari, Kabindra
item POPPIEL, RAUL - Federal University Of Sao Paulo
item DEMATTE, JOSE ALEXANDRE - Federal University Of Sao Paulo
item CURI, NILTON - Federal University Of Lavras
item DE MENEZESA, MICHELE - Federal University Of Lavras

Submitted to: Geoderma Regional
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/23/2025
Publication Date: 1/27/2025
Citation: Pelegrino, M.H., Guilherme, L.R., Adhikari, K., Poppiel, R.R., Dematte, J.M., Curi, N., De Menezesa, M.D. 2025. Digital mapping of agricultural soils texture of the Brazilian Cerrado biome. Geoderma Regional. https://doi.org/10.1016/j.geodrs.2025.e00922.
DOI: https://doi.org/10.1016/j.geodrs.2025.e00922

Interpretive Summary: Information about the distribution of soil texture in a landscape is crucial for assessing soil quality, crop suitability, and land management. However, mapping soil texture precisely at a large scale is challenging. This study used >32,280 soil observations from the Brazilian Cerrado biome and utilized advanced geospatial modeling tools to predict and map soil texture with high prediction accuracy. It improved the understanding of model interpretability and provided high-resolution, accurate soil texture maps, aiding users and public policies.

Technical Abstract: Soil texture is crucial for assessing soil quality, crop suitability, and land management. However, precise large-scale soil texture mapping remains challenging. This study integrated a Synthetic Soil Image (SySI) with standard environmental covariates in a digital soil mapping framework to map soil particle size distribution in Brazil's Cerrado biome. Four Random Forest (RF) model arrangements were explored for soil texture modeling. Using an extensive legacy dataset of Cerrado topsoil (0-20 cm) on Google Earth Engine, the models explained approximately 83% of clay, 86% of sand, and 74% of silt variance, with RMSE values of 89 g kg-1 (clay), 102 g kg-1 (sand), and 53 g kg-1 (silt). The findings highlighted SySI's strong predictive capacity, especially when climate data were available. Elevation was the only relevant terrain derivative for predicting soil texture. This approach improved model interpretability and provided high-resolution, accurate soil texture maps, aiding users and public policies. Given that 65% of Cerrado soil classes (Ferralsols and Arenosols) do not significantly increase clay content with depth, this work adds value to agricultural soil mapping.