Future food security will be challenged by the likely increase in demand, changes in consumption patterns and the effects of climate change. Framing food availability requires adequate agricultural production planning. Decision-making can benefit from improved understanding of the uncertainties involved. The aim of the study is to identify and quantify the sources of this uncertainty and explore their interactions and influence on precision and accuracy of agricultural estimates, with emphasis on modeling of wheat. Wheat is the most widely grown cereal worldwide and Europe one of its major producers. Results demonstrated the complex interaction between the level of knowledge and complexity of crop models, the availability of data and the projection targets. Interactions lead us to believe that uncertainty may be more robustly reduced by improved datasets than using more complex models. International model inter-comparison projects (as AgMIP or MACSUR) may represent an ideal framework to conduct a deeper analysis of this issue. Ignorance about future wheat varieties and phenological development features are a significant source of uncertainty in the future agricultural socio-economic context for adaptation to climate change (and a significant aspect for the design of the Representative Agricultural Pathways).