NUMERICAL WEATHER PREDICTION (NWP) CENTERS SYSTEMATICALLY EXCLUDE CLOUD OBSERVATIONS FROM THE ASSIMILATION PROCESS AND ONLY ASSIMILATE CLEAR-SKY RADIANCES. THIS PROCESS NOT ONLY EXCLUDES A LARGE PORTION OF THE DATA BUT ALSO CAUSES LACK OF OBSERVATIONS IN REGIONS WHERE CLOUDS ARE DOMINANT E.G. IN THE RAINBANDS OF TROPICAL CYCLONES (TC). INFRARED OBSERVATIONS ARE VERY SENSITIVE TO CLOUDS AND ARE SATURATED IN THE PRESENCE OF CLOUDS THUS DO NOT PROVIDE MUCH INFORMATION ON THE STRUCTURE OF TCS. HOWEVER MICROWAVE OBSERVATIONS ARE LESS SENSITIVE TO CLOUDS AND PROVIDE VALUABLE INFORMATION ON STATE OF THE ATMOSPHERE AS WELL AS THE HORIZONTAL AND THE VERTICAL STRUCTURE OF CLOUDS EVEN IN THE PRESENCE OF TCS.HOWEVER ASSIMILATION OF MICROWAVE CLOUDY RADIANCES IS FUNDAMENTALLY DIFFERENT FROM ASSIMILATING CLEAR SKY SATELLITE RADIANCES. THE EFFORTS SO FAR HAVE FOCUSED ON THE ASSIMILATION OF CLOUDY RADIANCES USING VARIATIONAL TECHNIQUES I.E. MINIMIZING THE DIFFERENCE BETWEEN THE BEST ESTIMATE OF THE ATMOSPHERE NORMALLY PROVIDED BY A NWP MODEL (ALSO KNOWN AS FIRST GUESS OR BACKGROUND) AND OBSERVATIONS. SINCE THE FIRST GUESS AND THE SATELLITE OBSERVATIONS CANNOT BE DIRECTLY COMPARED OBSERVATIONS ARE SIMULATED USING THE FIRST GUESS PROFILES USING A RADIATIVE TRANSFER MODEL ALSO KNOWN AS FORWARD OPERATOR. SEVERAL FACTORS ADVERSELY IMPACTS THE ACCURACY OF THE FORWARD CALCULATIONS FOR CLOUDY RADIANCES INCLUDING: (I) THE NWP MODEL FIELDS DO NOT PROVIDE A CLOSE ESTIMATE OF THE ATMOSPHERE FOR INSTANCE IN MANY CASES THE OBSERVED CLOUDS ARE DISPLACED OR COMPLETELY MISSING IN THE FIRST GUESS (II) THE NWP MODELS DO NOT PROVIDE ALL THE INPUTS AND PARAMETERS REQUIRED BY THE FORWARD OPERATOR (III) MANY OF THE PARAMETRIZATIONS IN THE FORWARD MODEL (E.G. THE SIZE AND SHAPE OF HYDROMETERS) CANNOT BE DIRECTLY MEASURED THUS ARE ESTIMATED FROM OTHER MEASUREMENTS WITH VERY LIMITED ACCURACY AND (IV) THE VARIATIONAL TECHNIQUES ASSUME THAT THE FORWARD OPERATOR IS LINEAR AND THE FIRST GUESS DEPARTURES (I.E. THE DIFFERENCES BETWEEN OBSERVATION AND FIRST GUESS) HAVE GAUSSIAN DISTRIBUTIONS BUT IN THE CASE OF CLOUDY RADIANCES ESPECIALLY IN THE CASE OF OPTICALLY THICK CLOUDS E.G. RAINBANDS OF TCS THE FORWARD OPERATOR IS HIGHLY NON-LINEAR AND THE DEPARTURES DO NOT HAVE GAUSSIAN DISTRIBUTIONS.WE PROPOSE A NOVEL BAYESIAN MONTE CARLO INTEGRATION (BMCI) TECHNIQUE TO RETRIEVE THE PROFILES OF TEMPERATURE WATER VAPOR AND CLOUD LIQUID/ICE WATER CONTENT FROM MICROWAVE CLOUDY MEASUREMENTS IN THE PRESENCE OF TCS. THESE RETRIEVALS THEN CAN EITHER BE DIRECTLY USED BY METEOROLOGISTS TO ANALYZE THE STRUCTURE OF TCS OR BE ASSIMILATED TO PROVIDE ACCURATE INITIAL CONDITIONS FOR THE NWP MODELS. WE PROPOSE TO UTILIZE DATA FROM THE ADVANCED TECHNOLOGY MICROWAVE SOUNDER (ATMS) ONBOARD SUOMI NATIONAL POLARORBITING PARTNERSHIP (NPP) AND GLOBAL PRECIPITATION MEASUREMENT (GPM) MICROWAVE IMAGER (GMI). WE PROPOSE TO ASSIMILATE THE BMCI RETRIEVALS INTO THE HURRICANE WEATHER RESEARCH AND FORECASTING (HWRF) USING THE GRIDPOINT STATISTICAL INTERPOLATION (GSI) DATA ASSIMILATION SYSTEM. WE PROPOSE TO RUN A SET OF DATA ASSIMILATION EXPERIMENTS FOR SOME SELECTED HURRICANES FOR THE PERIOD 2014-2015. THE TCVITAL DATA WILL BE USED TO VALIDATE THE TRACK AND INTENSITY OF TCS SIMULATED BY HWRF. IN ADDITION THE COLLOCATED CLOUDSAT AND GMP DUAL-FREQUENCY PRECIPITATION RADAR (DPR) WILL BE SUED TO ANALYZE THE STRUCTURE OF THE SIMULATED TCS.