Climate change science has established certain incontrovertible facts. We know that greenhouse gases are accumulating in the atmosphere as a result of human activity, and that these gases are the main drivers of the rise in global temperature. Starting form this solid scientific bedrock, the international community must take urgent steps to curb emissions of these pollutant gases. Yet current climate models are still insufficiently reliable to predict concrete impacts at a regional level, and this makes it harder to plan preventive measures to protect population centers that could be affected. In this conference, Professor Stevens will stress the urgency of developing far more accurate models, based on supercomputing, that improve our ability to predict climate behavior in a context of global warming. It is more vital than ever to improve our science, he will argue, if we wish to grapple successfully with this threat to humanity’s future.