A promising option to sequester carbon in agricultural soils is the inclusion of cover crops in cropping systems. The advantage of cover crops as compared to other management practices that increase soil organic carbon (SOC) is that they neither cause a decline in yields, like extensification, nor carbon losses in other systems, like organic manure applications may do. However, the effect of cover crop green manuring on SOC stocks is widely overlooked. We therefore conducted a meta-analysis to derive a carbon response function describing SOC stock changes as a function of time. Data from 139 plots at 37 different sites were compiled. In total, the cover crop treatments had a significantly higher SOC stock than the reference croplands. The time since introduction of cover crops in crop rotations was linearly correlated with SOC stock change (R2=0.19) with an annual change rate of 0.32±0.08Mgha-1yr-1 in a mean soil depth of 22cm and during the observed period of up to 54 years. Elevation above sea level of the plot and sampling depth could be used as explanatory variables to improve the model fit. Assuming that the observed linear SOC accumulation would not proceed indefinitely, we modeled the average SOC stock change with the carbon turnover model RothC. The predicted new steady state was reached after 155 years of cover crop cultivation with a total mean SOC stock accumulation of 16.7±1.5Mgha-1 for a soil depth of 22cm. Thus, the C input driven SOC sequestration with the introduction of cover crops proved to be highly efficient. We estimated a potential global SOC sequestration of 0.12±0.03PgCyr-1, which would compensate for 8% of the direct annual greenhouse gas emissions from agriculture. However, altered N2O emissions and albedo due to cover crop cultivation have not been taken into account here. Data on those processes, which are most likely species-specific, would be needed for reliable greenhouse gas budgets.
