Cloud thermodynamic phase and particle size estimation using the 0.67 and 1.6 μm channels from meteorological satellites

Abstract. A robust method to estimate the cloud microphysical properties from visible (0.67 μm) and near infrared (1.6 μm) measurements of reflected sunlight is presented. The method does not determine cloud particle phase and size separately. Instead it assigns a cloud particle type to every pixel that is most representative for the radiation measurements. The corresponding radiative transfer model calculations will yield the most accurate values for optical thickness. Furthermore, an estimate of the particle size is obtained, which is used in estimates of liquid water path.

Radiative transfer calculations have been performed for eleven cloud particle models assuming a single, plane-parallel and homogeneous layer. Standard gamma distributions with varying effective radii have been chosen for liquid water droplet whereas imperfect hexagonal ice crystal with different aspect ratio and size were selected for ice particles. It is shown that the ratio of the visible reflectivity to the near infrared reflectivity as a function of the visible reflectivity allows a consistent classification of cloud particles with respect to size and phase over a large area. The method is tested with measurements from the Along Track Scanning Radiometer instrument (ATSR-2) on board ERS-2 for a marine stratocumulus cloud and a cirrus cloud over the North Sea. For both cases, the variation of the measured ratio as a function of the measured visible reflectivity is well simulated by liquid water droplet distribution with an effective radius between 4 and 10 micrometers for the stratocumulus and by imperfect hexagonal ice crystal with a size of 60 μm for cirrus.

The method was used in the CLIWANET-project and will be the basis to the algorithm for AVHRR and SEVIRI radiances for EUMETSAT's Sattelite Application facility on climate monitoring.