CMIP3 Climate Models May Overestimate Global Warming by Underestimating Future Low Cloudiness
Small changes in low cloudiness have a major impact on Earth’s climate, but simulating low clouds is a major struggle for global climate models (GCMs). Specialized local-area models are capable of accurately simulating these clouds but their regional scope precludes prediction of large-scale environmental changes which are also critical to low cloud formation. This study explores a novel framework combining the large-scale information from GCMs with the accuracy of a regional model. This is done by using large-scale information for current-climate and future-climate conditions from ten GCMs participating in phase 3 of the Coupled Model Intercomparison Project (CMIP3) to drive an atmospheric Mixed Layer Model (MLM).
The MLM is much better at reproducing observed cloud responses to current-climate forcing changes, lending credibility to our model results. In contrast to most GCMs, the MLM predicts low clouds to increase in the future in response to robust increases in GCM-predicted boundary-layer inversion strength. Because low clouds act to cool the planet, MLM low cloud increases imply that global warming will be slightly weaker than predicted by GCMs. Using the MLM as the cloud scheme for all ten GCMs does not reduce inter-model spread, however, because MLM sensitivity to large differences in GCM-predicted inversion strength effectively disperses the MLM predictions. This suggests that improved treatment of low clouds is a necessary but not sufficient condition for more accurate climate forecasts.
Thanks goes to modeling centers around the world for sharing their data, which was hosted by the Program for Climate Model Diagnosis and Intercomparison (PCMDI) under the DOE Office of Science funding, with oversight from the World Climate Research Program’s Working Group on Coupled Modeling. This work was performed under the auspices of DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.