Abstract. Country-specific soil organic carbon (SOC) estimates are the baseline for the Global SOC Map of the Global Soil Partnership (GSOCmap-GSP). This endeavor is key to explaining the uncertainty of global SOC estimates but requires harmonizing heterogeneous datasets and building country-specific capacities for digital soil mapping (DSM). We identified country-specific predictors for SOC and tested the performance of five predictive algorithms for mapping SOC across Latin America. The algorithms included support vector machines (SVMs), random forest (RF), kernel-weighted nearest neighbors (KK), partial least squares regression (PL), and regression kriging based on stepwise multiple linear models (RK). Country-specific training data and SOC predictors (5 × 5 km pixel resolution) were obtained from ISRIC -World Soil Information. Temperature, soil type, vegetation indices, and topographic constraints were the best predictors for SOC, but country-specific predictors and their respective weights varied across Latin America. We compared a large diversity of country-specific datasets and models, and were able to explain SOC variability in a range between ∼ 1 and ∼ 60 %, with no universal predictive algorithm among countries. A regional (n = 11 268 SOC estimates) ensemble of these five algorithms was able to explain ∼ 39 % of SOC variability from repeated 5-fold cross-validation. We report a combined SOC stock of 77.8 ± 43.6 Pg (uncertainty represented by the full conditional response of independent model residuals) across Latin America. SOC stocks were higher in tropical forests (30 ± 16.5 Pg) and croplands (13 ± 8.1 Pg). Country-specific and regional ensembles revealed spatial discrepancies across geopolitical borders, higher elevations, and coastal plains, but provided similar regional stocks (77.8 ± 42.2 and 76.8 ± 45.1 Pg, respectively). These results are conservative compared to global estimates (e.g., SoilGrids250m 185.8 Pg, the Harmonized World Soil Database 138.4 Pg, or the GSOCmap-GSP 99.7 Pg). Countries with large area (i.e., Brazil, Bolivia, Mexico, Peru) and large spatial SOC heterogeneity had lower SOC stocks per unit area and larger uncertainty in their predictions. We highlight that expert opinion is needed to set boundary prediction limits to avoid unrealistically high modeling estimates. For maximizing explained variance while minimizing prediction bias, the selection of predictive algorithms for SOC mapping should consider density of available data and variability of country-specific environmental gradients. This study highlights the large degree of spatial uncertainty in SOC estimates across Latin America. We provide a framework for improving country-specific mapping efforts and reducing current discrepancy of global, regional, and country-specific SOC estimates.

Abstract. Country-specific soil organic carbon (SOC) maps are the baseline for the Global SOC Map of the Global Soil Partnership (GSOCmap-GSP). This endeavor requires harmonizing heterogeneous datasets and building country-specific capacities for digital soil mapping (DSM). We identified country-specific predictors for SOC and tested the performance of five predictive algorithms for mapping SOC across Latin America. The algorithms included: support vector machines, random forest, kernel weighted nearest neighbors, partial least squares regression, and regression-Kriging based on stepwise multiple linear models. Country-specific training data and SOC predictors (5 × 5 km pixel resolution) were obtained from ISRIC-World-Soil-Information-System. In general, temperature, soil type, vegetation indices and topographic constraints were the best predictors for SOC, but country-specific predictors and their respective weights varied across Latin America. We compared a large diversity of country-specific data scenarios and were able to explain ~ 53 % of SOC variability (range

No abstract

Abstract. Spatial soil databases can help model complex phenomena in which soils are decisive, for example, evaluating agricultural potential or estimating carbon storage capacity. The Soil Information System for Latin America and the Caribbean, SISLAC, is a regional initiative promoted by the FAO's South American Soil Partnership to contribute to the sustainable management of soil. SISLAC includes data coming from 49,084 soil profiles distributed unevenly across the continent, making it the region's largest soil database. However, some problems hinder its usages, such as the quality of the data and its high dimensionality. The objective of this research is twofold. First, to evaluate the quality of SISLAC and its data values and generate a new, improved version that meets the minimum quality requirements to be used by different interests or practical applications. Second, to demonstrate the potential of improved soil profile databases to generate more accurate information on soil properties, by conducting a case study to estimate the spatial variability of the percentage of soil organic carbon using 192 profiles in a 1473 km2 region located in the department of Valle del Cauca, Colombia. The findings show that 15 percent of the existing soil profiles had an inaccurate description of the diagnostic horizons. Further correction of an 4.5 additional percent of existing inconsistencies improved overall data quality. The improved database consists of 41,691 profiles and is available for public use at https://doi.org/10.5281/zenodo.6540710 (Díaz-Guadarrama, S. & Guevara, M., 2022). The updated profiles were segmented using algorithms for quantitative pedology to estimate the spatial variability. We generated segments one centimeter thick along with each soil profile data, then the values of these segments were adjusted using a spline-type function to enhance vertical continuity and reliability. Vertical variability was estimated up to 150 cm in-depth, while ordinary kriging predicts horizontal variability at three depth intervals, 0 to 5, 5 to 15, and 15 to 30 cm, at 250 m-spatial resolution, following the standards of the GlobalSoilMap project. Finally, the leave-one-out cross-validation provides information for evaluating the kriging model performance, obtaining values for the RMSE index between 1.77 % and 1.79 % and the R2 index greater than 0.5. The results show the usability of SISLAC database to generate spatial information on soil properties and suggest further efforts to collect a more significant amount of data to guide sustainable soil management.