Originally aimed at studying cloud geometrical and microphysical vertical properties, the 94-GHz Cloud Profiling Radar (CPR) on board of CloudSat is also well-suited to the problem of determining the distribution of light precipitation including drizzle and light rainfall, which proved difficult to detect using current observing systems. CloudSat is part of the A-Train constellation that gathers a wide variety of sensors (both passive and active) which is well suited for comprehensive cloud and aerosol studies at many spatial-scales. This data set can lead to a better understanding and characterization of our climate system and have a positive role in improving current numerical weather prediction (NWP) models.

Present work focuses on developing and implementing the CloudSat's light rainfall algorithm in the NRL's near real-time processing system (NRTPS). Since the CPR can only provide one measurement at each gate, the forward model description used in the retrieval algorithm must overcome this shortage of information. This proves to be a challenging problem, given the well known variety and complexity of clouds. Thus, various assumptions about the drop size distribution (DSD) and cloud phase are being tested and implemented. In addition, melting phase processes responsible for the bright band (BB) in radar images must also be modeled. Moreover, given the large field of view of the CPR, multiple scatter radiation must also be properly accounted for. Since in-situ precipitation measurements are taken at the surface, information about surface radar return under given physical conditions can also be exploited. Thus, whenever possible, the path-integrated attenuation (PIA) is used as an additional constraint in the light rainfall retrieval algorithm. In the end, retrieval of the cloud vertical structure, including its light precipitation aspects, is possible under a given range of conditions.