Sensitivity of simulated global climate to perturbations in low-cloud microphysical properties. Part I: Globally uniform perturbations

Abstract

The sensitivity of the global climate to perturbations in the microphysical properties of low clouds is investigated using a general circulation model coupled to a static mixed layer ocean with fixed cloud distributions and incorporating a new broadband parameterization for cloud radiative properties. A series of GCM experiments involving globally uniform perturbations in cloud liquid water path or effective radius (albedo perturbations), along with one for a doubling of carbon dioxide (greenhouse perturbation), lead to the following results: 1) The model's climate sensitivity (ratio of global-mean surface temperature response to the global-mean radiative forcing) is virtually independent (to ∼10%) of the sign, magnitude, and the spatial pattern of the forcings considered, thus revealing a linear and invariant nature of the model's global-mean response. 2) Although the total climate feedback is very similar in all the experiments, the strengths of the individual feedback mechanisms (e.g., water vapor, albedo) are different for positive and negative forcings. 3) Changes in moisture, tropospheric static stability, and sea ice extent govern the vertical and zonal patterns of the temperature response, with the spatial distribution of the response being quite different from that of the radiative forcing. 4) The zonal surface temperature response pattern, normalized with respect to the global mean, is different for experiments with positive and negative forcings, particularly in the polar regions of both hemispheres, due to differing changes in sea ice. 5) The change in the surface radiative fluxes is different for the carbon dioxide doubling and cloud liquid water path decrease experiments, even though both cases have the same radiative forcing and a similar global-mean surface temperature response; this leads to differences in the vigor of the hydrologic cycle (evaporation and precipitation rates) in these two experiments.