Different global climate models give different answers about future climate. The range of answers across the different models—what climate scientists refer to as the spread—can be very large. One approach to constraining, or narrowing the range of, answers across climate models is the use of emergent constraints.
The basic concept underlying the emergent constraint approach is that when you analyze large ensembles of global climate models, a clear relationship may emerge between a variable X in models’ simulations of the current climate and a different variable Y in models’ projections of future climate. When this relationship is identified and tested in various ways, you can then plug in actual observations of variable X to narrow the range of answers for variable Y.
The field of emergent constraints began to develop when our group pioneered the use of observations of the seasonal snowmelt cycle to constrain global climate model projections of snow albedo feedback. Since then, the list of relationships being investigated by the climate science community as potential emergent constraints has grown to more than more than 30. We remain active in identifying and vetting potential emergent constraints and applying confirmed ones to reduce uncertainty in global climate model projections.