Retrieval of Wind Velocity Vectors Over the Ocean Surface with Spaceborne Synthetic Aperture Radar: Implementation and Quality Assessment

Rapport, 2011

Wind velocity measurements at 10 meters over the ocean surface represent an important parameter to study the ocean behavior. Wind vector fields are available from different sources, such as satellite derived measurements and weather forecast models. Nevertheless, the limited availability in time, low spatial resolution and poor coverage close to the coasts motivate research on the use of Synthetic Aperture Radar (SAR) for wind velocity retrieval. This report presents an description and assessment of a wind velocity algorithm implemented for C-band (frequency 4–8 GHz) SAR data. Results obtained with the algorithm are compared and validated with wind velocity estimates from the Global Forecast System (GFS) model and the Advanced SCATterometer (ASCAT) sensor on-board the MetOp-A satellite. The statistical results show that the wind velocity retrievals are in very good agreement with the validation data when the wind field is slowly varying, leading to a bias in the wind speed smaller than 0.5 ms−1 and a root mean square error (RMSE) smaller than 2 m/s, and a RMSE about 20◦ in the wind direction. For winds with high variability, the algorithm agrees better with ASCAT measurements than with the GFS model, provided that sufficient amount of samples are available to perform the comparison.