Increase in Precipitation Efficiency With Surface Warming in Radiative-Convective Equilibrium

The precipitation efficiency of convection (epsilon) plays an important role in simple models of the tropical atmosphere as well as in global climate models' projections of future climate changes, but remains poorly understood and poorly constrained. A particularly urgent question is how epsilon will change in warmer climates. To address these issues, this study investigates the precipitation efficiency in simulations of radiative-convective equilibrium with a cloud-resolving model forced by a wide range of sea surface temperatures (SSTs). Two different domains are considered: a small, doubly periodic domain, and a 2-D (x-z) mock-Walker domain with a sinusoidal SST profile that resembles the equatorial Pacific, and the sensitivities of the results to the microphysical scheme and to the horizontal resolution are also explored. It is found that epsilon generally increases with warming in the small domain simulations because of increases in the efficiency with which cloud condensate is converted into precipitation, with changes in the re-evaporation of falling precipitation playing a secondary role. This picture is complicated in the 2-D simulations by substantial changes in the degree of convective organization as the underlying SSTs are varied. epsilon is found to decrease as convection becomes more organized, because convective organization results in relatively more low clouds, which have small (<= 0.1) precipitation efficiencies, and relatively less high clouds, which have larger (similar to 0.4) precipitation efficiencies. Plain Language Summary The precipitation efficiency of convection (epsilon) quantifies the fraction of water that condenses in a cloud that reaches the surface as precipitation. Recent work has shown that changes in epsilon can play an important role in determining the warming of climate models in response to increases in atmospheric carbon dioxide concentrations, and epsilon is also a key factor in theories for the dynamics of the tropical atmosphere. Despite this importance, however, epsilon is poorly understood and poorly constrained. In this study, we take a first step to addressing this issue by investigating how precipitation efficiency behaves in idealized simulations of the tropical atmosphere, in which the underlying sea surface temperature is varied across a wide range of values. We find that epsilon generally increases with warming because clouds become denser and so form precipitation more easily, though in some cases epsilon decreases because of changes in the large-scale flow of the tropical atmosphere.