APPLICATION OF OPTICAL MEASURE THEORY FOR REMOTE TEMPERATURE SENSING

REDUCTION OF SATELLITE RADIANCE DATA TO TEMPERATURE PROFILES INVOLVES BOTH DIFFICULT MATHEMATICAL INVERSION ANALYSIS AS WELL AS COMPLICATED DATA PROCESSING. RECENTLY, THE APPLICATION OF THE OPTICAL MEASURE THEORY REPRESENTS A NOVEL APPROACH WHICH PROVIDE A SOLUTION FREE FROM THE NEED OF EXTERNALLY-IMPOSED CONSTRAINTS (KING, 1987). THE OBJECTIVE OF THIS PROPOSAL IS TO DEVELOP A SYSTEMATIC TECHNICAL APPROACH FOR DETERMINATION OF THE SENSITIVITY OF THE ALGORITHM TO CERTAIN INSTRUMENT DESIGN PARAMETERS; AMONG THEM ARE THE SPECTRAL RESOLUTION OF THE SENSOR, THE SHARPNESS OF THE WEIGHT FUNCTIONS, THE MEASUREMENT NOISE, AND THE NUMBER OF CHANNELS IN THE MEASUREMENT. THE ALGORITHM WILL BE FURTHER STUDIED BY USING SOME SYNTHESIZED RADIANCE DATA, WHICH WILL BE GENERATED IN THE COURSE OF THE STUDY, AND MEASURED SATELLITE RADIANCE DATA FROM NOAA'S TOVS SYSTEMS. THE DATA ANALYSIS WILL INCLUDE THE APPLICATION OF THE ALGORITHM'S PERFORMANCE AGAINST GROUND TRUTH DATA. THE INITIAL STUDY OF APPLYING ADAPTIVE SYSTEMS TECHNIQUES TO THIS RETRIEVAL ALGORITHM WILL ALSO BE GIVEN IN THE PHASE I PERIOD.