To retrieve cloud optical thickness and particle size from satellite measurements, one must first have some idea of the cloud phase, that is, whether the cloud is composed of ice or water or some combination of both. Furthermore, little work has previously been performed to infer cloud phase in operational satellite data products. This is changing with the advent of data from the MODIS (Moderate resolution Imaging Spectroradiometer) on the NASA Terra platform. An operational MODIS cloud product concerns the inference of cloud thermodynamic phase. The phase retrieval (ice, water, mixed, or uncertain) is based on analysis of measurements at 8.5 and 11 microns, and thus is applicable for both daytime and nighttime data. The problem addressed here is upon the comparison of the satellite cloud phase product with independent determinations of cloud phase measured with the University of Wisconsin High Spectral Resolution Lidar (HSRL).

The HSRL provides vertical profiles of backscatter cross section, extinction cross section, and depolarization. The HSRL divides the return signal into separate molecular and particulate scattering profiles. The Rayleigh scattering by molecules is used as a calibration target which provides robust recovery of cross section measurements. Residual depolarizations in the receiver are very small and a small receiver field of view (100 micro-radians) minimizes depolarization caused by multiple scattering. Thus, the HSRL depolarization signal allows easy distinction between scattering from spherical water droplets which induce little depolarization and ice crystals which are non-spherical and thus highly depolarizing.

This paper uses HSRL measured cloud phase and cloud altitudes acquired over Madison, Wisconsin to validate simultaneous, co-located MODIS retrievals.