Retrieval of Wind Velocity Vectors Over the Ocean Surface with Spaceborne Synthetic Aperture Radar

Sammanfattning: Wind over the ocean (normally defined at 10 meter height) is one of the main parameters affecting sea state, generating waves and driving currents. Wind field maps with large area coverage and high resolution therefore represent valuable information for a wide range of applications such as information for shipping, which carries 80% of the world’s trade, oil platforms and other maritime infrastructures, modeling, analysis and validation of weather models, improving performance of sensors onboard maritime infrastructures, etc. Spaceborne Synthetic Aperture Radar (SAR) is a promising sensor for providing wind information with global coverage and high resolution irrespective of weather conditions. This thesis presents research carried out on the implementation and evaluation of an algorithm for retrieval of wind parameters (wind speed and direction) over the ocean surface with SAR data. The implemented algorithm uses C-band (frequency 4-8GHz) SAR data to retrieve wind speed and direction information at 10 meters above the ocean surface. The wind direction retrieval is based on the local gradient method that extracts streak features in the SAR image aligned with the wind. The wind speed is inverted by providing the extracted wind direction to the CMOD-IFR2 geophysical model function. A quality assessment of the wind retrievals by comparison with data from a numerical weather model (NWM) and a spaceborne scatterometer has been performed for various SAR images taken at different locations showing a diversity of features. The wind retrievals agree very well with the reference (wind speed bias smaller than 0.5 m/s and root-mean-square difference smaller than 2 m/s for the wind speed and about 20º for the wind direction) when the wind presents a smooth behavior. It was found that for data with high spatial variability the SAR retrievals agree better with the scatterometer retrievals than the NWM data. Future work requires further quality assessment of the wind retrievals against in-situ measurements and scatterometer data.