Near Sea Surface Wind Estimations Using Airborne GPS Systems

Studies have shown that airborne GPS remote sensing techniques can be used to retrieve near sea surface wind speeds. Methods for determining directions of the winds using the GPS techniques, however, have not been fully developed. This study tries to use the GPS signals reflected from the sea surface to estimate not only near sea surface wind speeds but also directions. The method in estimating sea surface winds utilizes aircraft-received multiple path length GPS signals reflected from the sea surface. The intensity and waveform of these sea surface reflected GPS signals are dependent on the roughness (or slope) and spatial anisotropy of the rough sea surfaces caused by near sea surface winds. A theoretical model based on Cox and Munk's sea surface slope probability and the GPS signal transfer and reflection process is used to simulate airborne GPS signals. The modeled GPS waveforms are generally a function of the wind speed and direction, aircraft altitude, and GPS satellite elevation angle. Using a matched filter, these simulated GPS signal waveforms are compared with actual airborne data that is simultaneously measured from two GPS satellites at different elevation and azimuth angles. The best match of the simulated GPS signals and the measurements from the GPS satellites provides estimated values for near sea surface wind speed and direction. The GPS estimated results have been compared to buoy recordings and TOPEX satellite measurements. The retrieved wind speeds are generally within ±2 m/s of buoy recordings and other satellite measurements. The wind direction retrievals deviate from about 0° to 35° away from corresponding buoy data. 180° ambiguity was found in the directional retrievals. To overcome this ambiguity, a third set of simultaneous GPS satellite measurements may be needed, which is left for future studies.