Discrepancies in Soil Organic Carbon Maps: Insights for Pará State

Soil organic carbon (SOC) accounts for roughly two-thirds of terrestrial carbon and is critical for understanding carbon dynamics and emissions from land-use changes. Several remote sensing SOC products vary in methodology, resolution, and spatial coverage. Assessing map discrepancies is key to evaluating SOC emission uncertainties from land-use changes in Brazil. This study compares SOC stock maps for Pará state at two depths (0-30 cm and 0-100 cm) to identify differences and assess implications. Four SOC stock products are analyzed: MapBiomas Solos, SoilGrids, Embrapa Solos, and Bernoux et al. (2002), with [15,729] original data points for MapBiomas Solos, [150,000] for SoilGrids, [1,712] for Embrapa Solos, and [17,918] for Bernoux et al. (2002). For the 0-100 cm depths, only SoilGrids and Embrapa Solos product were available. SOC stocks were analyzed using descriptive statistics, regridding all maps to the highest resolution available (250m, from SoilGrids) using bilinear interpolation. The reference year for comparison was 2020, except for Bernoux et al. (2002), which reflects historical vegetation cover. Differences in SOC stock estimates were observed across products. MapBiomas Solos, SoilGrids, and Embrapa Solos provided different SOC values, with discrepancies mainly due to variations in data sources, modeling techniques, and sampling densities. The Embrapa product shows moderate variation with lower stocks in southeastern Pará. The MapBiomas product estimates higher SOC concentrations near rivers. The Bernoux product presents well-distributed SOC classes, while the SoilGrids product demonstrates greater heterogeneity influenced by elevation. SoilGrids and Embrapa Solos exhibited higher SOC stock estimates for the 0-100 cm depth. Differences in sampling density, particularly in tropical regions, contributed to variations in SOC estimates. The analysis reveals significant SOC map variability, underscoring the need for standardization. Emission estimates vary by dataset, impacting carbon accounting. This study enhances SOC assessment by advocating for high-resolution, region-specific data. The findings align with Sustainable Development Goal (SDG) 13 (Climate Action) by improving carbon accounting accuracy and supporting informed environmental decision-making.