RAIN FALLING ON THE SEA SURFACE PRODUCES BUOYANT LENSES OF RELATIVELY FRESH WATER THAT ARE SUBSEQUENTLY MIXED INTO THE WATER COLUMN. THESE LENSES ARE TYPICALLY TENS OF CENTIMETERS TO SEVERAL METERS THICK AND ARE ASSOCIATED WITH A VERTICAL SALINITY GRADIENT. A MICROWAVE RADIOMETER USED FOR SATELLITE SALINITY MEASUREMENT PROVIDES VALUES FROM THE UPPER CENTIMETER OF THE OCEAN SURFACE. THESE VERY NEAR SURFACE SALINITIES FROM THE SATELLITE ARE VALIDATED WITH IN SITU OBSERVATIONS FROM ARGO FLOATS OR OTHER IN SITU MEASUREMENTS THAT TYPICALLY SAMPLE AT A FEW METERS DEPTH. THEREFORE A NEAR-SURFACE SALINITY GRADIENT PRODUCED BY RAINFALL CAN THUS LEAD TO A MISMATCH BETWEEN SATELLITE AND IN SITU SALINITIES. THE RAIN IMPACT MODEL (RIM) WAS DEVELOPED TO ESTIMATE NEAR-SURFACE SALINITY STRATIFICATION IN THE PRESENCE OF RAIN WITH THE OBJECTIVE OF IMPROVING VALIDATION OF AQUARIUS AND SMOS SALINITIES. RIM-DERIVED ESTIMATES OF THE SURFACE FRESHENING DUE TO RAIN ARE CURRENTLY IMPLEMENTED AS A PRODUCT IN THE FINAL VERSION (V5) AQUARIUS DATASET. THE BASIC PREMISE OF RIM IS TO MODEL RAIN AS A PULSE OF FRESHWATER TO THE OCEAN SURFACE. THIS PULSE THEN MOVES DOWNWARDS THROUGH THE WATER COLUMN FOLLOWING TRANSPORT DEFINED BY A ONE-DIMENSIONAL VERTICAL DIFFUSION EQUATION. WITH RAIN GIVEN BY A SATELLITE-BASED PRECIPITATION ESTIMATE RIM ESTIMATES THE EVOLUTION OF NEAR-SURFACE SALINITY. BY USING CALCULATING THE CUMULATIVE EFFECT OF RAIN EVENTS OBSERVED OVER THE PREVIOUS 24 H THE SALINITY PROFILE AT A GIVEN TIME AND LOCATION IS PREDICTED. WHILE RIM HAS BEEN SHOWN TO WORK WELL IN MANY INSTANCES IT IS LIMITED BY THE FACT THE DIFFUSION EQUATION USED TO ESTIMATE THE SALINITY EVOLUTION ASSUMES A CONSTANT VERTICAL EDDY DIFFUSIVITY COEFFICIENT REGARDLESS OF WIND SPEED. FURTHERMORE RIM ASSUMES THAT THE FRESHWATER IS DELIVERED AS AN INSTANTANEOUS PULSE. IN FACT RECENT OBSERVATIONS AND MODELS HAVE SHOWN THAT THE FORMATION AND EVOLUTION OF RAIN LENSES ARE HIGHLY DEPENDENT ON BOTH THE WIND SPEED AND THE TEMPORAL DISTRIBUTION OF THE FRESHWATER INPUT. RIM WILL BE MODIFIED TO EXPLICITLY INCLUDE THE EFFECTS OF WIND SPEED ON EVOLUTION OF THE FRESH LENS USING TWO METHODS: 1). A PARAMETERIZATION FOR THE VERTICAL DIFFUSIVITY IN TERMS OF WIND SPEED WILL BE GENERATED USING PREVIOUSLY CONDUCTED MODEL STUDIES AND INCLUDING THIS PARAMETERIZATION WILL RESULT IN PRIM AND 2). THE CALCULATIONS USED IN RIM WILL BE IMPLEMENTED IN THE GENERALIZED OCEAN CIRCULATION MODEL (GOTM). GOTM IS A ONE-DIMENSIONAL TURBULENCE CLOSURE MODEL FOR UPPER OCEAN DYNAMICS THAT HAS USED TO MODEL RAIN LENSES. THE HYBRID GOTM-BASED RIM (GRIM) WILL HAVE THE SAME FUNCTIONALITY AS PRIM BUT IS EXPECTED TO BE MORE ACCURATE ESPECIALLY AT HIGH WIND SPEEDS. HOWEVER THIS ACCURACY WILL COME AT THE COST OF HIGHER COMPUTATIONAL OVERHEAD. GRIM WILL ALSO BE ABLE TO SIMULATE THE EFFECTS OF DIFFERENCES IN RAIN RATE ON THE BEHAVIOR OF FRESH LENSES AND THIS INFORMATION WILL BE USED TO GUIDE PRIM DEVELOPMENT TO ALSO INCLUDE THIS EFFECT. BOTH VERSIONS OF THE MODEL PRIM AND GRIM WILL BE TESTED USING IN SITU DATA AVAILABLE FROM THE RECENTLY CONCLUDED SPURS2 FIELD EXPERIMENT. ONCE EACH MODEL IS VALIDATED THEY WILL PRODUCE REGULAR ESTIMATES OF NEAR-SURFACE SALINITY STRATIFICATION FOR THE SMAP SALINITY PRODUCT AS WELL AS FOR THE ENTIRE AQUARIUS DATASET. WE WILL ALSO USE CARRY OUT STATISTICAL CHARACTERIZATIONS OF THE NEAR-SURFACE TURBULENT DISSIPATION RATES DURING DIFFERENT WIND AND RAIN CONDITIONS ALSO USING THE SPURS2 DATA SET TO PROVIDE IN SITU DATA FOR VALIDATION AND CALIBRATION.