Review of Wind Retrieval Algorithms

This report gives a brief overview of the most common wind retrieval algorithms for satellite data. Wind speed over ocean can be retrieved from several satellite instruments such as Passive Microwave radiometer, Radar Altimeter and Scatterometer. The latter instrument, which has been operational in space since the launch of ERS-1 in 1991, has been designed for ocean wind measurements. Scatterometer data provides both wind speed and direction and is currently used in operational weather forecasts. These data have all coarse resolution, tentatively 50 km, and cannot be used for wind measurements in near coastal regions. With high resolution SAR images, which have also been available from satellites since 1991, it is possible to estimate ocean wind fields in coastal regions with resolution of a few km. For wind energy estimation in these regions, SAR provide the most appropriate satellite data. However, the possibility to retrieve wind speed depends on the performance of the algorithm. Therefore several SAR algorithms for determination of wind speed and direction have been reviewed in this report. These include the SWA algorithm, the CMOD-4 wind retrieval algorithm, the CMOD-IFR2, the VIERS algorithm of University of Hamburg and the NORUT algorithm. The SAR wind algorithm review in this chapter is to a large extent based on a recent report where state-of-the-art SAR ocean imaging and interaction models were studied. Also results of the COASTWATCH experiment, where SAR wind retrieval was investigated and evaluated against in situ ship observations have been important in this review. The CMOD-4 algorithm is chosen for the use in this project. It is relatively easy to use, based on PRI images as input. It is very important, however, that correct calibration information is used so that absolute backscatter coefficients can be derived. The CMOD-4 algorithm will be used to analyse more than 50 ERS SAR images from the three test sites in Norway, Denmark and Italy. SAR derived wind speeds will then be compared with in situ wind measurements and the overall capability of SAR to estimate wind speeds in coastal regions will be assessed.