The land CO2 flux plays a key role in the global carbon cycle. The projected climate change is likely to alter CO2 flux at the surface by increasing both biomass productivity (photosynthesis) and soil respiration. The magnitude and sign of the net CO2 flux are reflected by changes in size of soil carbon pools. In this study, we use the DayCENT model, the CENTURY model with a daily time step, driven by past climate and IPCC AR4 scenario climates to project the trends of soil organic carbon (SOC) pools for the Midwest cropland under various management strategies. The B1, A1b, and A2 scenario climates from three GCMs representing respectively cold/wet, normal, and warm/dry models were selected for three time windows: 1961-2000, 2046-2065, and 2081-2100. Despite differences among the models, the three GCMs projected growing season (May-September) warming by 1.5-1.8oC around mid-21st century under the “greenest” B1 scenario. The regionally diverse A2 scenario produced an additional 1oC warming than the B1 scenario. The balanced A1b scenario predicted a climate that is about 0.2oC colder than the A2 scenario. Future precipitation amounts during the growing season increase by 10-30% depending on the model and scenario.

Under all these scenario climates, DayCENT simulations indicate a 10-30% decline in SOC within the top 20 cm of soil profile by mid-century using current management options. The simulated SOC differences among the scenarios vary within 8% of each other, smaller than the inter-model range of 28%, underlining the uncertainty attributable to individual GCM models. Further diagnoses indicate that the SOC decline is more (less) due to heterotrophic soil respiration (NPP), suggesting the dominance of warming effects on soil microbial decomposition. No-till practice combined with appropriate crop rotations and harvest methods can increase SOC by up to 20%, thereby reversing its trend of decline.