The variability of clouds and water vapor in CMIP6 models

The simulation of clouds and precipitation with global climate models has proven notoriously difficult as a multitude of nonlinear processes spanning a large range of spatial and temporal scales are involved. Biases in the simulation of present-day clouds do not only affect confidence in the capability of the models to correctly reproduce present-day climate but also raise concerns about the accurate representation of cloud feedback processes in climate change projections. In this study we investigate the performance of state-of-the-art global coupled GCMs from the sixth phase of the Coupled Model Intercomparison Project (CMIP6) by comparing simulated cloud properties with observations. A main focus is the variability of cloud properties and water vapor on seasonal and interannual time scales. The analysis is performed with the Earth System Model Evaluation Tool (ESMValTool), a community diagnostic and performance metrics tool for the evaluation of Earth System Models. The ESMValTool is applied to the ensemble of models participating in CMIP6 to evaluate the new models with satellite observations and for comparison with results from the previous generations of global coupled models (CMIP3 and CMIP5) to monitor the progress made over the recent years. The present-day performance of the models is then put into perspective with the simulated changes in cloud forcing in the abrupt 4xCO2 runs and with the equilibrium climate sensitivity of the models