Retrieval of microphysical, geometrical, and radiative properties of marine stratocumulus from remote sensing

Abstract

Radiative transfer calculations in boundary layer stratocumulus are performed and used to retrieve cloud properties from remotely measured reflected radiances in the visible and near‐infrared. Calculations based on the vertically uniform plane‐parallel cloud model have been currently used for the retrieval of cloud optical thickness and droplet effective radius. They were extended in a previous work to the retrieval of cloud droplet number concentration and cloud geometrical thickness, by assuming a vertically stratified cloud model, with prescribed droplet spectrum. The retrieved values of cloud droplet number concentration though were substantially underestimated. Improvements are presented here with a more realistic parameterization of the droplet size distribution, based on the theory of droplet growth by vapor diffusion. The retrieval technique is applied to eight case studies of the second Aerosol Characterization Experiment and the retrieved values are compared to estimates derived from in situ cloud microphysical measurements. The technique provides accurate retrievals of cloud droplet number concentration, with no bias. The retrieval of cloud geometrical thickness though is systematically overestimated compared to that directly measured. This bias partly reflects the difference of cloud sampling between the two strategies. Remote sensing measurements, which are continuous, are capable of detecting the thickest cells of the cloud layer within the field of view of the radiometer, while discontinuous in situ vertical profiles provide a statistical average of the geometrical thickness. Part of the bias could also reflect three‐dimensional effects of the radiative transfer in spatially heterogeneous clouds, which are not accounted for by the plane‐parallel models. The data are then processed for deriving statistics of the retrieved values of cloud droplet number concentration, optical thickness, and liquid water path, for validation of large‐scale parameterizations of the aerosol indirect effect.