Effects of Cloud Vertical Structure on Atmospheric Circulation in the GISS GCM

Abstract

Thirteen experiments have been performed using the Goddard Institute for Space Studies General Circulation Model (GISS GCM) to investigate the response of the large-scale circulation to different macroscale cloud vertical structures (CVS). The overall effect of clouds, the role of their geographic variations, and difference between the transient and equilibrium responses of the atmospheric circulation are also studied. Clouds act to suppress the Hadley circulation in the transient response, but intensify it in the equilibrium state. Changing CVS affects the atmospheric circulation directly by modifying the radiative cooling profile and atmospheric static stability, but the effect is opposed, on average, by an indirect effect on the latent heating profile produced by deep (moist) convection. Different interactions of radiation and convection with land and ocean surfaces mean that this cancellation of CVS effects on radiative and latent heating is not the same at all locations. All three parameters of the CVS seem equally important: the cloud-top height of the uppermost cloud layer, the presence of multiple layers, and the separation distance between two consecutive layers in a multilayered cloud system. In experiments with a globally uniform, single-layered cloud, the one with the cloud located somewhere at middle levels (720?550 mb in this model) results in the strongest Hadley circulation; with a single-layered cloud located above or below this level, both the circulation intensity and its vertical extent decrease. Inserting another cloud layer below a cloud in the upper troposphere also intensifies the Hadley circulation, the effect increasing with decreasing separation distance. Separately, vertical gradients in the cloud distribution appear to be more important to the circulation strength than horizontal gradients, but horizontal variations in the CVS are needed to explain the strength of the mean circulation in the model atmosphere. The results also suggest that explicitly resolving cloud-top radiative cooling and base warming for each cloud layer is important to modeling the Hadley circulation.