Radiative Properties of Boundary Layer Clouds: Droplet Effective Radius versus Number Concentration

The plane-parallel model for the parameterization of clouds in global climate models is examined in order to estimate the effects of the vertical profile of the microphysical parameters on radiative transfer calculations for extended boundary layer clouds. The vertically uniform model is thus compared to the adiabatic stratified one. The validation of the adiabatic model is based on simultaneous measurements of cloud microphysical parameters in situ and cloud radiative properties from above the cloud layer with a multispectral radiometer. In particular, the observations demonstrate that the dependency of cloud optical thickness on cloud geometrical thickness is larger than predicted with the vertically uniform model and that it is in agreement with the prediction of the adiabatic one. Numerical simulations of the radiative transfer have been performed to establish the equivalence between the two models in terms of the effective radius. They show that the equivalent effective radius of a vertically uniform model is between 80% and 100% of the effective radius at the top of an adiabatic stratified model. The relationship depends, in fact, upon the cloud geometrical thickness and droplet concentration. Remote sensing measurements of cloud radiances in the visible and near infrared are then examined at the scale of a cloud system for a marine case and the most polluted case sampled during the second Aerosol Characterization Experiment. The distributions of the measured values are significantly different between the two cases. This constitutes observational evidence of the aerosol indirect effect at the scale of a cloud system. Finally, the adiabatic stratified model is used to develop a procedure for the retrieval of cloud geometrical thickness and cloud droplet number concentration from the measurements of cloud radiances. It is applied to the marine and to the polluted cases. The retrieved values of droplet concentration are significantly underestimated with respect to the values measured in situ. Despite this discrepancy the procedure is efficient at distinguishing the difference between the two cases.