Determination of the Radiative Properties of Stratiform Clouds from a Nadir-Looking 95-GHz Radar

Abstract

Several space agencies are presently considering missions with active instruments (radar, lidar), which are able to document cloud stratification and cloud microphysical properties on the global scale. The objective of this paper is to develop an algorithm to derive as much information as possible from a single-frequency, nadir-looking cloud radar operating from an airborne or spaceborne platform. It is impossible to derive all parameters of interest in the radiative budget of a cloud from only the radar reflectivity profile, unless some a priori knowledge of cloud processes is introduced in the formulation of the algorithm itself. The a priori knowledge considered here (total concentration of particles invariant with altitude, adiabatic liquid water content) does not apply to all cloud types but only to warm stratiform clouds where entrainment is weak. The algorithm concept and inversion procedure, including a stable scheme for correcting the radar reflectivity for attenuation, are first described. A test of the algorithm is then performed using numerical simulations in order to investigate the sensitivity of the retrieval to measurement noise, degradation of the range resolution, shape of the cloud droplet distribution, and presence of entrainment. In the realistic conditions of an airborne experiment, the retrieval of cloud base hb, total number concentration of particles NT, profiles of the liquid water content, and effective radius re can be performed with good accuracy (provided the entrainment coefficient is below 1 km?1). With the sampling characteristics of a spaceborne radar, retrievals of the cloud base and liquid water content remain reasonably accurate, but the estimates of NT and re are degraded to a level where they become meaningless. A test of the algorithm is performed using a dataset from the zenith-pointing ground-based 94-GHz radar of The Pennsylvania State University, obtained during the Continental Stratus Experiment. The algorithm is found to be successful in 43% of cases. An attempt of evaluation of the retrieval is made by comparison with ceilometer data. Most failure cases are probably due to the presence of drizzle.