THE GPM CORE OBSERVATORY CARRIES A DUAL-FREQUENCY PRECIPITATION RADAR (DPR) AND A MULTI-FREQUENCY MICROWAVE IMAGER (GMI); THE LATTER MEASURES MICROWAVE RADIATION AT 13 CHANNELS WITH FREQUENCIES RANGING FROM 10 TO 183 GHZ. IN PARTICULAR THE MILLIMETER-WAVE CHANNELS (166 AND 183 GHZ) ARE SPECIFICALLY DESIGNED TO DETECT AND RETRIEVE LIGHT RAIN AND SNOW PRECIPITATION. RESPONDING TO THE NASA PMM SCIENCE TEAM PROPOSAL CALL S TOPIC ON EXTRACT INFORMATION FROM THE FULL RANGE OF RADAR AND/OR PASSIVE MICROWAVE FREQUENCIES THAT IMPROVE RETRIEVALS OF LIGHT RAIN AND FALLING SNOW WE PROPOSE TO CONDUCT RESEARCH ON BUILDING ALGORITHM COMPONENTS FOR GPM SNOWFALL RETRIEVALS. BUILT UPON OUR PREVIOUS RESEARCH WE PROPOSE TO CONDUCT: (1) BASIC STUDIES FOR PRECIPITATION ALGORITHM DEVELOPMENT. THE MAIN FOCI ARE: (I) SNOW-RAIN SEPARATION. SINCE EITHER RADIOMETER OR RADAR CANNOT SEPARATE BETWEEN SNOW AND RAIN PIXELS IT IS NECESSARY TO USE AUXILIARY DATA TO SEPARATE SNOWING FROM RAINING PIXELS BEFORE ACTUAL RETRIEVALS ARE TO BE PERFORMED. WE HAVE BEEN USING LONG-TERM SURFACE STATION DATA TO DEVELOP AND REFINE A SEPARATION ALGORITHM; THE CURRENT VERSION HAS BEEN IMPLEMENTED IN GPM FACILITY ALGORITHMS. FURTHER REFINEMENT OF THE ALGORITHM IS UNDERWAY. (II) SCATTERING OF NONSPHERICAL ICE/SNOW PARTICLES. IN DEVELOPING PHYSICALLY-BASED PRECIPITATION RETRIEVAL ALGORITHMS FORWARD SIMULATION OF RADAR BACKSCATTERING AND RADIOMETER BRIGHTNESS TEMPERATURE IS A NECESSARY STEP. WE HAVE BEEN BUILDING A NONSPHERICAL ICE SCATTERING DATABASE FOR THESE SIMULATORS. NEXT WE WILL FOCUS ON LARGE AGGREGATE PARTICLES OF WHICH CURRENT DATABASE IS STILL LACKING. ADDITIONALLY WE PROPOSE TO EXAMINE ICE PARTICLE TYPE INFORMATION FROM TRIPLE-FREQUENCY RADAR MEASUREMENTS WHICH WILL LEAD TO A BETTER UNDERSTANDING OF THE PREVALENT ICE PARTICLE SHAPES UNDER DIFFERENT PRECIPITATION TYPES (CONVECTIVE STRATIFORM ETC.) THUS IS USEFUL FOR MAKING ASSUMPTIONS IN RETRIEVAL ALGORITHMS. (III) CHARACTERIZATION OF ICE AND LIQUID WATER IN SNOWING CLOUDS BASED ON OBSERVATIONAL DATA. THE RESULTS OF THIS CHARACTERIZATION CAN BE USED TO VALIDATE THE DATABASE USED IN SNOWFALL RETRIEVAL ALGORITHMS. (2) EXAMINATION OF SNOWFALL DETECTION AND RETRIEVAL METHODS OVER LAND. DURING THE PREVIOUS FUNDING CYCLE WE HAVE BEEN DEVELOPING A CLOUDSAT CPR PLUS DPR-TRAINED HIGH-FREQUENCY MICROWAVE RADIOMETER ALGORITHM FOR SNOWFALL DETECTION AND RETRIEVAL OVER LAND. WE HAVE MADE SEVERAL INTERESTING FINDINGS I.E. HIGH-FREQUENCY CHANNELS (89 166 AND 183 GHZ) ARE NECESSARY AND POSSIBLY SUFFICIENT FOR SNOWFALL RETRIEVALS; DETECTING SNOWFALL OCCURRENCE AND RETRIEVING SNOWFALL RATE HAVE SIMILAR SKILLS ETC. A VERSION OF THE GMI ALGORITHM HAS BEEN DEVELOPED AND PRODUCED REASONABLE RESULTS IN TERMS OF GLOBAL SNOWFALL MAPS COMPARED TO THOSE OF CLOUDSAT AND MHS. THE TECHNIQUE WILL BE FURTHER EXAMINED VALIDATED AND IMPROVED BY SURFACE-BASED MEASUREMENTS - RADAR NETWORKS DATA FROM U.S. JAPAN AND S. KOREA AS WELL AS NASA-FUNDED FUTURE FIELD EXPERIMENTS ETC. (3) DEVELOPMENT OF SNOWFALL VS. BRIGHTNESS TEMPERATURE DATABASE OVER OCEAN. WE HAVE BEEN DEVELOPING THIS DATABASE BASED ON COINCIDENT CLOUDSAT AND GPM DATA BY SEEKING THE CONSISTENCY BETWEEN BRIGHTNESS TEMPERATURES AND HYDROMETEOR PROFILES USING RADIATIVE TRANSFER MODELING AND 1D-VAR OPTIMIZATION. THE DATABASE WAS EXAMINED FOR SNOWFALL RETRIEVALS UNDER A BAYESIAN FRAMEWORK. SINCE A LARGE PORTION OF SNOWFALL EVENTS ARE MISSED BY DPR (WITH A MINIMUM DETECTION OF 13 DBZ) CLOUDSAT DATA ARE USED TO FILL THE MISSING PORTION. THE DATABASE CONTAINS BRIGHTNESS TEMPERATURES FROM GMI SNOWFALL RATE AT SURFACE SNOW WATER CONTENT PROFILES LIQUID WATER CONTENT ATMOSPHERIC VARIABLES AND SURFACE EMISSIVITY. THIS DATABASE ITSELF AND THE LESSONS LEARNED FROM PERFORMING THIS DATABASE BUILDING WILL HELP IMPROVE GPM FACILITY PRECIPITATION ALGORITHM.