Over the last three decades, microwave remote sensing has played a significant role in ocean surface wind measurement, and several scatterometer missions have flown in space since early 1990's. Although they have been extremely successful for measuring ocean surface winds with high accuracy for the vast majority of marine weather conditions, unfortunately, the conventional scatterometer cannot measure extreme winds condition such as hurricane. The SeaWinds scatterometer, onboard the QuikSCAT satellite is NASA's only operating scatterometer at present. Like its predecessors, it measures global ocean vector winds; however, for a number of reasons, the quality of the measurements in hurricanes are significantly degraded. The most pressing issues are associated with the presence of precipitation and Ku-band saturation effects, especially in extreme wind speed regime such as tropical cyclones (hurricanes and typhoons). Under this dissertation, an improved hurricane ocean vector wind retrieval approach, named as Q-Winds, was developed using existing SeaWinds scatterometer data. This unique data processing algorithm uses combined SeaWinds active and passive measurements to extend the use of SeaWinds for tropical cyclones up to approximately 50 m/s (Hurricane Category-3). Results show that Q-Winds wind speeds are consistently superior to the standard SeaWinds Project Level 2B wind speeds for hurricane wind speed measurement, and also Q-Winds provides more reliable rain flagging algorithm for quality assurance purposes. By comparing to H*Wind, Q-Winds achieves ~9% of error, while L2B-12.5km exhibits wind speed saturation at ~30 m/s with error of ~31% for high wind speed ( > 40 m/s).

Ocean surface wind vectors over the ocean present vital information for scientists and forecasters in their attempt to understand the Earth's global weather and climate. As the demand for global wind velocity information has increased, the number of satellite missions that carry wind-measuring sensors has also increased; however, there are still not sufficient numbers of instruments in orbit today to fulfill the need for operational meteorological and scientific wind vector data. Over the last three decades operational measurements of global ocean wind speeds have been obtained from passive microwave radiometers. Also, vector ocean surface wind data were primarily obtained from several scatterometry missions that have flown since the early 1990's. However, other than SeaSat-A in 1978, there has not been combined active and passive wind measurements on the same satellite until the launch of the second Advanced Earth Observing Satellite (ADEOS-II) in 2002. This mission has provided a unique data set of coincident measurements between the SeaWinds scatterometer and the Advanced Microwave Scanning Radiometer (AMSR). AMSR observes the vertical and horizontal brightness temperature (TB) at six frequency bands between 6.9 GHz and 89.0 GHz. Although these measurements contain some wind direction information, the overlying atmospheric influence can easily obscure this signal and make wind direction retrieval from passive microwave measurements very difficult. However, at radiometer frequencies between 10 and 37 GHz, a certain linear combination of vertical and horizontal brightness temperatures causes the atmospheric dependence to be nearly cancelled and surface parameters such as wind speed, wind direction and sea surface temperature to dominate the resulting signal. This brightness temperature combination may be expressed as ATBV-TBH, where A is a constant to be determined and the TBV and TBH are the brightness temperatures for the vertical and horizontal polarization respectively. In this dissertation, an empirical relationship between the AMSR's ATBV-TBH and SeaWinds' surface wind vector retrievals was established for three microwave frequencies: 10, 18 and 37 GHz. This newly developed model function for a passive microwave radiometer could provide the basis for wind vector retrievals either separately or in combination with scatterometer measurements.