The global soil carbon reservoir, which is dynamic and sensitive to climate and human disturbance, exceeds the amount of carbon stored in biosphere and the atmosphere. Anthropogenic and climatic factors may convert a land surface into a source or sink of atmospheric CO2. In this presentation I will provide several examples of using environmental variables and geospatial modeling to predict the land use and climate change impacts on soil organic carbon (SOC) at regional scales. The first study uses spatially varying estimates of SOC stocks to modify the IPCC carbon inventory approach to predict the impact of land use change on SOC pools. This approach showed the spatial distribution of SOC sequestration rates for the croplands of 7 states of the Midwest US. The second study combines process-based and geospatial modeling to predict the rainfed biomass productivity across the croplands of US. We predicted that cultivating miscanthus would result in a SOC sequestration at the rate of 0.16–0.82 Mg C ha-1 yr -1 across the US croplands due to cessation of tillage and increased biomass carbon input into the soil system. The final study predicts spatially resolved SOC stocks from surface to bedrock, distinguishing active-layer and permafrost-layer SOC stocks, across the State of Alaska. Assuming temperature changes from the A1B Intergovernmental Panel on Climate Change emissions scenario, our geospatial model indicated that the average 2100 Alaska active-layer thickness could deepen by 11 cm, resulting in a thawing of 13 Pg C currently in the permafrost. The equilibrium SOC loss associated with this warming would be highest under continuous permafrost, followed by discontinuous, isolated, and sporadic permafrost areas. Our high-resolution predictions of SOC sequestration rates, biomass productivity, and SOC stocks demonstrate the impacts of land use and climate change on soil carbon pool at management scales, which is critical for policy implications.