Comparison of fixed cloud-top temperature and fixed cloud-top altitude approximations in the Manabe- Wetherald radiative-convective atmospheric model

This work demonstrates the effect of fixing the cloud-top temperature during a numerical calculation of the atmospheric temperature profile using the GMR atmospheric model. In all earlier GMR calculations of the temperature response due to a parameter change the infinitely thin high and middle cloud layers and the thick lower cloud layer were held at fixed altitudes. Since recent cloud measurements have shown it may be more realistic to hold the cloud-top temperature fixed, we have compared the temperature response of the GMR model with fixed cloud altitudes and fixed cloud-top temperatures.

If our three cloud layers are maintained so as to hold the temperature of each cloud-top fixed (and the bottom of the lower cloud adjusted to maintain the same optical thickness) the calculated temperature differences due to parameter changes are 1.43 larger than when clouds are held at fixed altitudes in the model. If on the other hand the bottom of the lower cloud is maintained at a fixed altitude while the tops of the clouds are maintained at a fixed temperature the calculated temperature is 1.56 larger. These values compare with a value of 1.46 obtained by a radiative-convective model developed by Wang et al. (1976). Feedback mechanisms between the abundances of different clouds have not been included but could be important if sufficient detailed information were known about the interaction between different clouds.