Scientists develop computer models of real, complex systems to increase understanding of their behaviour and make predictions. A prime example is the Earth's climate. Complex climate models are used to compute the climate change in response to expected changes in the composition of the atmosphere due to man-made emissions. Years of research have improved the ability to simulate the climate of the recent past but these models are still far from perfect. The model projections of the globally averaged temperature increase by the end of this century differ by as much as a factor of two, and differ completely in regard to projections for specific regions of the globe.Current practice commonly averages the predictions of the separate models. Our proposed approach is instead to form a consensus by combining the models into one super model. The super model has learned from past observations how to optimally exchange information among individual models at every moment in time. Results in nonlinear dynamics suggest that the models can be made to synchronize with each other even if only a small amount of information is exchanged, forming a consensus that best represents reality. This innovative approach to reduce uncertainty might be compared to a group of scientists resolving their differences through dialogue, rather than simply voting or averaging their opinions.Experts from non-linear dynamics, machine-learning and climate science are brought together within SUMO to produce a climate change simulation with a super model combining state-of-the-art climate models. The super-modelling concept has the potential to provide improved estimates of global and regional climate change, so as to motivate and inform policy decisions. The approach is applicable in other situations where a small number of alternative models exist of the same real-world complex system, as in economy, ecology or biology.