A principal motivation for introducing alternative fuels is to reduce air pollution and greenhouse gas emissions. A comprehensive evaluation of the reductions must include all Life Cycle activities from the vehicle operation to the feedstock extraction. This paper focuses on the fuel upstream activities only. We compare the results and methods of the three most comprehensive existing fuel upstream models in the U.S.A. and we explore the differences and uncertainties of these types of analyses. To explicitly include the impact of uncertainties, we create a new model using the following approaches: Instead of using a single value as input, the new model deals with ranges around the most probable value. Ranges are discussed and calibrated by an expert network, in terms of their relative probability. Probabilistic function techniques are applied to study the impact of the uncertainties on the model output. The paper also presents the rationale and benefits of using each of the alternative approaches that are discussed and reviewed.