The use of biomass for energy production has increasingly been encouraged in the United States, in part motivated by the potential to reduce greenhouse gas (GHG) emissions relative to fossil fuels. However, the GHG-intensity of biomass-derived energy is highly dependent on how the biomass is obtained and used. We explore scenario uncertainty in GHG estimates in the Calculating Uncertainty in Biomass Emissions (CUBE) model and find that direct land-use change emissions that result during the biomass production often dominate the total "farm-to-hopper" GHGs. CUBE represents each land-use change decision as a conversion of land from one of four specified baseline ecosystem to produce one of seven feedstock crops, both distinct by geographic region, and then determines the implied changes in soil organic carbon, root carbon, and above-ground biomass. CUBE therefore synthesizes and organizes the existing literature to represent direct land-use change emissions in a way that can be more readily incorporated into life cycle assessment. Our approach to representing direct land-use change literature has been applied to a specific set of data and offers immediate implications for decisionmakers, but it can also be generalized and replicated in the future, making use of improved scientific data on the magnitude and rates of direct land-use change emissions as it becomes available. (C) 2012 Elsevier Ltd. All rights reserved.