Romps, Climate sensitivity and the direct effect of carbon dioxide in a limited-area cloud-resolving model, J Climate, 2020





Much has been learned about the impacts of global warming (for example, with regards to changes in storm intensity and the distribution of tropical water vapor and clouds) by studying "global" warming in small-domain cloud-resolving simulations. These domains are much too small (e.g., 100-km wide) to resolve large-scale atmospheric circulations, but, by virtue of explicitly resolving the convection (i.e., having grid cells that are small enough to resolve rain clouds), they faithfully reflect some of the most basic aspects of a convecting atmosphere.

Unfortunately, it takes a large number of computer hours to run even just a few weeks of model time in these cloud-resolving simulations. Considering that it can take many years of model time to equilibrate these simulations to an altered CO2 concentration, it is no wonder that cloud-resolving models have been used only a handful of times to study "global" warming.

Here, a technique is presented that accelerates the equilibration of these cloud-resolving simulations by a factor of 30 or more. The approach is to fix the sea-surface temperature (SST) and equilibrate the CO2 concentration. With this technique, it is possible to run many dozens of equilibrated "global" warming simulations in a cloud-resolving model. Here, 36 different simulations are equilibrated over SSTs ranging from 285 to 320 K. These and other simulations are used to quantify the equilibrium climate sensitivity (ECS) and the direct effect of CO2 on clouds. One of the most interesting results is the SST-dependence of the ECS: the equilibrium climate sensitivity is found to increase with warming, with a pronounced maximum at around 310 K.

Equilibrium climate sensitivity (ECS) in cloud-resolving simulations of tropical RCE at 1-K sea-surface-temperature (SST) intervals.