The retrieval of dry and wet snow distributions from SSM/I measurements and MM5 forecast results

Three dimensional distributions of precipitation in a North Atlantic cyclone were inferred from microwave radiances measured by the Special Sensor Microwave/Imager, SSM/I, on the Defense Meteorological Satellite Program, DMSP, satellite, using a radiative transfer hydrometeor retrieval model. Microwave brightness temperatures were not unique functions of the physical characteristics of the suspended hydrometeors; the SSM/I data were therefore analyzed in combination with estimates of snow, rain and liquid cloud water distributions obtained from the Fifth-Generation Pennsylvania State University-NCAR Mesoscale Model, MM5, to remove that ambiguity. The present snow model required the density to decrease with increasing particle size to significantly reduce the scattering effect at 85.5 GHz. As the snow particles melted, and their liquid fraction increased, the emission efficiency increased strongly. Both effects improved the fit of the theoretical brightness temperatures to the observations. Snow size distributions inferred from the modified microphysics retrieval agreed with in situ airborne measurements