Evaluating Soil Organic Carbon Stock Variability: Implications for Carbon Dynamics and Climate Action in Pará-Brazil

Maintaining and enhancing soil organic carbon (SOC) stocks are crucial for climate change mitigation. Due to the high costs of field sample analyses, modelling approaches that generate SOC stock maps based on available field data provide a valuable alternative for understanding its spatial distribution over time. This study adapts the methodology of Brazil’s Fourth Brazilian National Communication of Greenhouse Gas Emissions (4NC) to produce annual SOC maps for the Itacaiúnas River Basin, a 41,000 km² watershed within the Amazon’s deforestation arc. Since before 1985, pastures have replaced forested areas in the basin, currently occupying about half its total area. We used the reference SOC map (0–30 cm depth) from Bernoux et al. (2002) as a baseline, which provides average values based on native vegetation and soil types. Emission factors from 4NC were applied to this baseline map to assess stock changes resulting from land-use transitions from MapBiomas Collection 8 data between 1985 and 2021, assuming that SOC stocks need a 20-year period to reach a new equilibrium. The results were compared with other SOC products for the year 2020: MapBiomas Solos, SoilGrids, and Embrapa Solos. The changes in the SOC stock showed a reduction of 1.6% between 1985 and 2021, decreasing from 202 to 199 Mt-C. This slight reduction is attributed to the conservative land use change factor (0.98) recommended to Pará State, which was applied to the primary land-use change observed (forest to pasture). A Tukey test indicated statistically significant differences among all products. While the spatial distribution of SOC stocks was generally similar, discrepancies were observed for one soil type, where the reference map reported higher values than the other products. Another key difference was related to including elevation as a predictor variable in the alternative datasets, which is not considered in the baseline map. Additionally, the inventory-based methodology estimated higher SOC values for pasture areas compared to the other products. These higher SOC values obtained using the inventory-based methodology are primarily due to the emission factor applied. It is important to highlight that the study area and its surroundings have limited or low-density field data in the databases used to generate SOC products. The observed differences underscore the importance of field data for local and regional validation and the need for long-term studies to monitor SOC stock changes driven by land-use transitions.