Emergent constraints (ECs) have the potential to remove the cloud of uncertainty surrounding anthropogenic warming, but only if they impose real constraints on the climate system. As a result, verifying the credibility of proposed ECs is critically important. An essential ingredient in this verification is the presence of a clear physical explanation for a given EC's predictive power. These explanations are hard to test, but a new methodology for decomposing correlation between an EC and equilibrium climate sensitivity into terms related to physical processes and geographical regions makes this task easier. Of the 19 previously-published constraints evaluated in this study, only 4 are found to be consistent with their proposed explanation and robust to changes in ensemble. Six ECs were discredited by this evaluation and 6 more lacked a testable physical explanation. Two ECs were found to be repeats of other constraints in the ensemble. This low acceptance rate suggests that more care is needed in evaluating potential constraints. A framework for combining the predictions of the 4 seemingly-credible ECs is introduced. This framework implies stronger climate sensitivity than predicted by the CMIP5 models as a whole.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. All authors are supported by DOE’s Regional and Global Climate Modeling Program under the project “Identifying Robust Cloud Feedbacks in Observations and Models”. IM Release LLNL-ABS-754906