Based on the thermodynamic energy balance between radiation and vertical plus horizontal dynamic transports, a two-dimensional radiation-turbulence climate model is developed. This model consists of a broadband solar and IR radiation transfer scheme previously presented by the authors and vertical and horizontal dynamic eddy transports utilizing the elementary turbulent theory. In the model, three kinds of feedback mechanisms are considered: the humidity feedback via the constant relative humidity assumption, the ice-albedo feedback via a preliminary correlation between the surface albedo and the surface temperature and the dynamic transport feedback through the parameterization of eddy transports and the prescribed mean wind field. A standard temperature field, which differs from the climatological data by no more than 0.1°C, is first obtained by solving the coupled thermodynamic and surface flux budget equations using climatological distributions of H2O, CO2, O3, surface albedo and cloud properties. The model-derived atmospheric radiation budget, surface energy balance and horizontal transport patterns compare reasonably well with available observational data. Further validation of the model includes sensitivity studies on the effects of doubling of CO2 and a 2% increase in the solar constant. The temperature changes relative to the standard field on these experiments agree closely with those presented by Manabe and Wetherald utilizing a general circulation model. To investigate the two-dimensional cirrus–radiation interaction, a relationship between the cirrus IR emissivity and solar reflectance (and transmittance) is established based on the parameterization equations. On the basis of a number of experiments involving various couplings and feedbacks, we find that 1) the humidity and albedo feedbacks are most active in the tropics and arctic area, respectively, 2) the dynamic transport is a negative feedback in the equatorial and polar regions but a positive one in midlatitudes and 3) the relative importance of each feedback depends only slightly on the radiative properties of cirrus. Finally, we demonstrate that slight variations in the cirrus IR emissivity lead to significant temperature perturbations in the arctic surface and tropical troposphere.
Read full abstract