Strong External Radiative Forcing Abnormal Climate Sensitivity And Feedback Process

A version 1.4 of the National Centre for Atmospheric Research (NCAR) coupled climate model (CSM1.4) is integrated to investigate the nature and behaviour of the climate feedback/sensitivity of the model. The model is integrated for 50 years, after increasing the solar constant by 2.5, 10 and 25% of its control value. The main contents and results are below.(1) If the system is to evolve toward a new equilibrium, feedbacks processes in the system must be negative and of sufficient strength to balance the positive forcing. Feedback processes are analyzed both locally and globally in terms of longwave and shortwave, clear-sky/surface and cloud forcing components. The overall feedback does become more negative as temperature increases, but the rate of change is not linear, especially for strong warmings.(2) Longwave feedback processes become increasingly negative with temperature in a reasonably linear form. This is not the case for shortwave feedback processes. Shortwave clear-sky/surface feedback (i.e. ice/snow albedo feedback) becomes increasingly positive with temperature. At smaller warmings, the ice/snow albedo positive feedback is partially counteracted by solar cloud feedback, which becomes increasingly negative with temperature. However, solar cloud feedback ceases to decrease beyond a critical temperature and begins to increase as temperatures increase. For the 25% forcing case the net solar feedback ultimately becomes positive as temperatures warm.(3) The conditions under which a runaway warming might occur have previously been investigated using simple models. Under some circumstances, the effect of the forcing and the greenhouse effect of increasing water vapour in a warmer atmosphere can overwhelm the negative feedback of the longwave cooling to space as temperature increases. This is not, however, the reason for the runaway warming experienced by the GCM for the case of a 25% (or larger) increase in solar constant. Instead, the model experiences a "runaway cloud feedback" warming whereby cloudiness begins to decreases as temperature increases beyond some critical value. The resulting decrease in planetary albedo and increase in solar radiation absorbed by the system overwhelms negative feedbacks and a runaway warming results.(4) The decrease in cloudiness and planetary albedo is associated with an increase in precipitation and regions of both increasing and decreasing relative humidity. Thus cloudiness decreases even as the hydrological cycle accelerates and precipitation and evaporation increase. While the decrease in cloudiness may imply that the empirical cloud parameterization used in the model is not applicable for largeforcings and warmings, but there is no obvious evidence that this is the case. Adecrease in cloudiness and planetary albedo implies a strong positive feedback with increasing amounts of solar radiation absorbed by the system. Nevertheless, if this behaviour is atypical or if it represents a physically reasonable response to strong forcing in the climate system is required for further investigation with other models and cloud parameterizations.