Accession Number : ADA303138

Title :   Global Evaluation of Special Sensor Microwave/Imager Ocean Surface Wind Speed Retrieval Algorithms for the Period September 1991-April 1992.

Descriptive Note : Master's thesis,

Corporate Author : NAVAL POSTGRADUATE SCHOOL MONTEREY CA

Personal Author(s) : Hesser, William A.

PDF Url : ADA303138

Report Date : JUN 1995

Pagination or Media Count : 110

Abstract : The Fleet Numerical Meteorology and Oceanography Center (FNMOC) has the charter to provide Special Sensor Microwave/Imager (SSMI) data to the DOD and the NOAA. This has led FNMOC to examine new methods for processing SSM/I data to generate SSM/I products. Of particular interest is the ability to use the SSM/I to remotely sense ocean surface winds. For this study four candidate wind retrieval algorithms initially proposed at the SSM/I Algorithm Symposium held in June, 1993 are examined for potential implementation at FNMOC. Previous calibrarion/validarion studies of the efficacy of wiod speed algorithms focused on regional (mid-latitude or tropical) data sets prompting the requirement to develop a more encompassing, global data set on which to evaluate the proposed algorithms. Comparisons of SSM/I wind retrieval methods reveal that the current FNMOC operational algorithm overestimates wind speeds when atruospheric water vapor content exceeds 5O kg/sq-m2. Adjustments made to this algorithm effectively mitigate the high wind speed bias, but at the cost of eliminating a significant amount of data. Neural network algorithms display high wind speed bias for winds above 11 m/s and low wind speed bias for winds below 4 m/s. The performance of neural network algorithms is largely independent of atmospheric moisture content. A new, global training data set is necessary to enable neural network algorithms to perform properly over the full range of global wind speeds. The use of brightness temperature based rain flags are recommended for use in all wind speed retrieval methods.

Descriptors :   *OCEANOGRAPHIC DATA, *WIND VELOCITY, *MARINE METEOROLOGY, DATA BASES, ALGORITHMS, GLOBAL, NEURAL NETS, OCEAN SURFACE, WIND, NUMERICAL ANALYSIS, THESES, MICROWAVE EQUIPMENT, RADAR IMAGES, METEOROLOGICAL SATELLITES, MOISTURE CONTENT, INFORMATION RETRIEVAL, BIAS, REMOTE DETECTION, HUMIDITY, WATER VAPOR.

Subject Categories : Meteorology
      Physical and Dynamic Oceanography

Distribution Statement : APPROVED FOR PUBLIC RELEASE