RAINFALL OVER THE OCEAN PRODUCES SMALL PATCHES OF FRESHWATER AT THE SURFACE. THESE FRESHWATER LENSES EVOLVE RAPIDLY YET ARE VERY IMPORTANT FOR MEDIATING AIR-SEA INTERACTIONS. CAPTURING OBSERVATIONS OVER THE WIDE RANGE OF POSSIBLE OCEANIC CONDITIONS AND EVOLUTION SCENARIOS IS DIFFICULT BECAUSE OF THE STRONG RAIN EVENTS ARE RELATIVELY RARE AND OFTEN IN REGIONS DIFFICULT TO ACCESS. ACCURATE RAIN RATES CAN NOW BE ESTIMATED FROM AUTONOMOUS PLATFORMS USING PASSIVE ACOUSTICS. WHEN THESE SENSORS ARE INTEGRATED ONTO SEAGLIDERS SIMULTANEOUS MEASUREMENT OF THE VERTICAL STRUCTURE OF THE WATER COLUMN AND OF THE TIME SERIES OF RAIN RATE IS THEREFORE POSSIBLE. THIS PROPOSAL AIMS TO EXPLORE DATASETS ALREADY COLLECTED FROM GLIDERS AND PROPOSE TO IMPLEMENT A SYSTEM THAT NOT ONLY CAN SAMPLE WITH A FINER TIME RESOLUTION DURING THE INTENSE RAIN EVENTS BUT ALSO USES REMOTE SENSING FROM SATELLITES AND OTHER REAL-TIME INSTRUMENTS TO INFORM THE SAMPLING OF THE GLIDERS. MULTI-MONTHS TIME SERIES OF RAIN RATES FROM PASSIVE ACOUSTICS ARE NOW AVAILABLE FOR SEVERAL SEAGLIDER MISSIONS AROUND THE PACIFIC OCEAN. IN ALL CASES PROFILES OF TEMPERATURE AND SALINITY WERE ALSO COLLECTED. WE PROPOSE TO EXPLORE THE SENSITIVITY OF THE FRESHWATER PUDDLES TO RAIN RATES AND WIND SPEEDS IN SEVERAL DIFFERENT OCEANOGRAPHIC ENVIRONMENTS. SINCE RAIN IS EXTREMELY PATCHY IN SPACE AND TIME KNOWLEDGE OF LOCAL AND INSTANTANEOUS TIME SERIES OF RAIN RATE AS THE GLIDER IS PROFILING AND SAMPLING TEMPERATURE AND SALINITY IS KEY TO UNDERSTANDING THE OBSERVED SURFACE STRUCTURE. RAIN RATE ESTIMATES FROM ACOUSTICS WILL BE COMPARED TO THAT OF SATELLITE ESTIMATES OF PRECIPITATION LOCAL WEATHER RAIN RADAR AND WHEN AVAILABLE NEARBY DIRECT RAIN GAUGE ESTIMATES. REAL-TIME ON-BOARD PROCESSING OF THE ACOUSTIC RAIN GAUGE DATA MAKES POSSIBLE TO ADAPTIVELY CHANGE THEIR SAMPLE INTERVAL BETWEEN ESTIMATES BASED ON THE DETECTED PRESENCE (OR ABSENCE) OF RAIN. IN ADDITION WE PROPOSE TO BUILD A SYSTEM THAT INCORPORATES INFORMATION FROM SATELLITE REMOTE SENSING (GLOBAL PRECIPITATION RATE ESTIMATES) TO PRODUCE A `RAIN PROBABILITY PARAMETER WHICH WILL CONTROL THE INTERVAL AT WHICH THE INSTRUMENT CHECKS FOR RAIN. THIS WILL ALLOW FOR A BETTER ALLOCATION OF THE (FINITE) NUMBER OF SAMPLES DURING LONG MISSIONS WITH FREQUENT RAIN RATE ESTIMATES DURING PERIODS OF ACTIVE ATMOSPHERIC CONVECTION WHERE MEASURING BOTH THE PRESENCE AND ABSENCE OF RAIN IS IMPORTANT FOR UNDERSTANDING AIR-SEA INTERACTIONS. FINALLY WE PROPOSE TO CONDUCT A GLIDER MISSION DEDICATED TO STUDYING THE IMPACT OF HEAVY RAIN ON THE UPPER OCEAN. IN ADDITION TO TESTING THE ADAPTIVE RAIN SAMPLING (INCLUDING REMOTE SENSING AND ENVIRONMENTAL DATA) THE DIVE PATTERN OF THE GLIDER WILL ALSO BE MODIFIED IN RESPONSE TO RAIN EVENTS. THIS PROJECT RESPONDS TO THE A.9 CALL FOR THE OCEAN SALINITY SCIENCE TEAM AND AIMS TO ADVANCE UNDERSTANDING OF NEAR-SURFACE SALINITY INTEGRATING INNOVATIVE USES OF IN SITU OBSERVATIONS AND REMOTE SENSING. IN PARTICULAR WE PROPOSE TO USE NEAR-REAL TIME NASA RAIN AND SSS SATELLITE MEASUREMENTS TO BETTER GUIDE GLIDER SAMPLING. THE PROPOSED ADAPTIVE SAMPLING TECHNIQUES AND ANALYSIS OF EXISTING DATA CAN BE USED TO BETTER UNDERSTAND NEAR-SURFACE SALINITY STRATIFICATION IN THE UPPER FEW METERS AND THE UNDERLYING PHYSICAL PROCESSES A MAJOR OBJECTIVE OF THE OSST. THE ADAPTIVE METHOD ADDRESSES THE NEED FOR PERSISTENT MONITORING OF SPARSE EVENTS USING AUTONOMOUS INSTRUMENTS EVEN IN REGION WHERE ACCESS IS EXTREMELY LIMITED.