Cloud feedbacks are the primary uncertainty in overall climate feedbacks and climate sensitivity. In recent years, a growing number of modeling groups have been developing and running global convection-resolving models that simulate the global atmosphere at kilometer-scale with fewer subgrid parameterizations. The US Department of Energy’s (DOE’s) Simple Cloud Resolving E3SM Atmosphere Model (SCREAM) is one such model capable of running multi-year simulations on DOE’s exascale computers. An outstanding question is whether these simulations will produce feedbacks that are qualitatively different from those produced by 100-km scale climate models.

Building on recent results that show that one-year long Cess-type simulations can produce global cloud feedbacks very similar to those from multi-decade fully-coupled abrupt-4xCO2 simulations, we will present recent analyses of the feedbacks from a pair of 13-month simulations forced with present-day sea-surface temperatures (SSTs) and with those SSTs perturbed uniformly by +4K. We will discuss our results in relation to feedbacks from CMIP-class climate models and evaluate the impact of modeled cloud biases on the simulated feedbacks.

This work was conducted under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-852353.