Combining risk programming with biophysical simulation offers potential benefits for helping farmers in developing countries choose cropping and irrigation strategies or for the study of farmer behaviour. Risk can have a significant impact on the way resources are allocated and should therefore be considered in empirical studies. This study uses risk programming and biophysical simulation models to find the expected utilityMmaximizing irrigation strategy and crop choice for southwestern Kansas farmers. Biophysical simulation models allow the researcher to obtain yield data for a longer time period than is typically available from agronomic studies, and to study risk on a very localized level. Direct expected utility maximization is used to determine the optimal strategies. Results from the study suggest that biophysical simulation models offer a promising avenue to further understanding of the impacts of risk on farm management decisions. Because biophysical simulation models are transferable to different regions of the world, biophysical simulation can be an attractive alternative to conducting risk research in developing countries.