Contrasting approaches for estimating soil carbon changes in Amazon and Cerrado biomes

The Brazilian Amazon and Cerrado are among the principal agricultural frontiers of the world, and have attracted global attention because of their potential to be a source or sink of atmospheric CO2. However, accurately quantifying alterations in soil organic carbon (SOC) due to land-use changes remains to be a major challenge. Using a meta-analysis, our objective was to evaluate tillage effects on SOC stocks in Brazil's Amazon and Cerrado biomes. Two approaches: (1) the classical SOC annual change rate which provides only an arithmetic average of SOC changes, and (2) the rate of annual change adjusted from management factors derived according to the Intergovernmental Panel on Climate Change (IPCC) SOC methodology, which allows a temporal dataset adjustment, were used. Regardless of approach, conservation technologies such as no-tillage (NT) and integrated crop-livestock (ICL) systems were identified as viable options for SOC sequestration, increasing it between 9 and 16%. One exception was adoption of NT in the Amazon forest, where SOC decreased by 9% over 20 years. Comparing results of the two approaches showed substantial differences between them. For conversions from NT to ICL or from native vegetation to full tillage (FT), the IPCC-based approach produced SOC stock estimates that were 65 and 47% lower, respectively, than with the classical approach. The IPCC approach also identified modeling covariate effects and was influenced less by outliers, thus reducing the risk of under or overestimating soil management effects on SOC