WE PROPOSE TO 1) ADAPT THE CLOUD FEEDBACK MODEL INTERCOMPARISON PROJECT (CFMIP) OBSERVATION SIMULATOR PACKAGE (COSP) TO TROPICAL RAINFALL MEASURING MISSION (TRMM) PRECIPITATION RADAR (PR) AND GLOBAL PRECIPITATION MEASUREMENT (GPM) DUAL-FREQUENCY RADAR (DPR) OBSERVATIONS AND IMPROVE ITS SUB-GRID SCALE ALGORITHM TO BETTER ALIGN WITH OBSERVATIONS AND 2) COMPREHENSIVELY EVALUATE COUPLED MODEL INTERCOMPARISON PROJECT PHASE 6 (CMIP6) CLIMATE MODEL PRECIPITATION PROCESSES BY COMPARING COSP SIMULATED REFLECTIVITY TO TRMM PR AND GPM DPR OBSERVATIONS. PARTICULAR EMPHASIS WILL BE GIVEN TO EVALUATING NASA GODDARD INSTITUTE FOR SPACE STUDIES (GISS) MODEL-E. OUR PRIMARY GOAL IS TO DETERMINE CLIMATE MODEL FIDELITY AT REPRESENTING CONVECTIVE AND STRATIFORM PRECIPITATION PROCESSES AND PINPOINT POSSIBLE CAUSES FOR MODEL-OBSERVATION DISCREPANCIES. THIS WORK IS MOTIVATED BY THE DIFFICULTIES ASSOCIATED WITH MODEL-OBSERVATION COMPARISONS BECAUSE THEY OFTEN DO NOT MEASURE OR SIMULATE SIMILAR VARIABLES. PREVIOUS WORK HAS HAD SUCCESS AT IDENTIFYING CLIMATE MODEL DEFICIENCIES BY SIMULATING PASSIVE SATELLITES AND CLOUD RADARS. HOWEVER FULL EVALUATION OF MODEL PRECIPITATION PROCESSES (I.E. ABOVE THE LIGHT-TO-MODERATE RAIN SENSITIVITY OF A CLOUD RADAR) HAS BEEN STYMIED BY THE ABSENCE OF A PRECIPITATION RADAR SIMULATOR WITHIN COSP. EVALUATION OF CONVECTIVE VS. STRATIFORM PRECIPITATION PROCESSES IS IMPORTANT AS THE TWO RAIN TYPES HAVE VERY DIFFERENT LATENT HEATING PROFILES WITH SIGNIFICANTLY DIFFERENT IMPACT ON THE LARGE-SCALE CIRCULATION.AS A FIRST STEP WE WILL ADAPT COSP TO ACCOUNT FOR THE SCANNING STRATEGY OF THE TRMM PR AND GPM DPR SINCE THE CURRENT COSP RADAR SIMULATOR IS ONLY CONFIGURED FOR NADIR POINTING RADARS. NEXT WE WILL IMPROVE VARIOUS SUB-GRID SCALE ASSUMPTIONS USED WITHIN COSP TO DESIGNATE CLOUD AND PRECIPITATION AMONG MODEL GRID BOX SUBCOLUMNS. THESE ASSUMPTIONS ARE OFTEN INCONGRUOUS WITH OBSERVATIONS. FOR EXAMPLE THE ASSUMPTION IS MADE THAT ALL CLOUDY SUBCOLUMNS RAIN WHEREAS CLOUDSAT OBSERVATIONS SHOW THAT IS NOT TRUE. LASTLY EVALUATION OF MODEL PRECIPITATION PROCESSES WILL BE DONE PRIMARILY BY COMPARING MODEL-SIMULATED VS. OBSERVED DISTRIBUTIONS OF RADAR REFLECTIVITY WITH HEIGHT AND GEOGRAPHICAL MAPS OF PRECIPITATION PROPERTIES TO HIGHLIGHT HEIGHTS AND REGIONS OF MODEL DEFICIENCIES.