Inference of Ice Cloud Particle Roughness in Optically Thin Clouds using Satellite Polarimetric Observations

Abstract:

Global statistics of cloud optical thickness and particle size from visible and near-infrared satellite observations require an appropriate model for cloud particle morphology, namely, overall shape and surface texture. This study focuses on the inference of an ice cloud particle model from satellite polarimetric observations. The Polarization and Directionality of Earth’s Reflectance (POLDER) sensor aboard Polarization and Anisotropy of Reflectances for Atmospheric Sciences coupled with Observations from a Lidar (PARASOL) satellite produced a global polarimetric reflectivity dataset covering nine years from 2004 to 2013. In the last decade, several attempts were carried out to infer particle shapes and roughness based on these data, revealing that a regular particle shape with some degree of surface roughness and a particle shape with high randomness are consistent with observations over thick ice clouds. However, the representative particle shape and optimal degree of roughness of optically thin ice clouds still remain uncertain. We demonstrate the methodology and the performance of the integrated polarimetric and non-polarimetric multispectral inference technique for the ice cloud particle shape and roughness. The technique assumes that a mixture of a few habits with multiple degrees of roughness can model cloud reflectivity to infer the best mixing ratio and degree of roughness based on the satellite dataset. The optical thickness is retrieved simultaneously by combining non-polarimetric (intensity) observations to the conventional polarization-only technique. The application of the method to one month of global data will be discussed with an emphasis on the latitudinal variation in our results.