Validation of a SAR-only wind-vector retrieval against shipborne in situ wind observations in the European Arctic

Abstract

Space-borne synthetic aperture radar (SAR) observations provide broad coverage of high-resolution snapshots of the sea surface conditions in polar regions. However, their potential has not yet been fully harnessed for meteorological applications. For instance, standard methods for SAR wind-vector retrieval rely on wind direction inputs from numerical weather prediction models, which hampers the high-resolution capabilities of SAR wind retrievals and the use of these in data assimilation. A recently proposed SAR-only wind-vector retrieval method, that uses SAR information more exhaustively than standard methods do, is compared to in situ ship observations and is found to perform similarly to a standard method under average wind conditions at open sea. However, in coastal regions, at high wind speeds, and in complex meteorological conditions this new application outperforms the standard method. It is concluded here that wind fields obtained from the SAR-only wind-vector retrieval are suitable for data assimilation in high-resolution weather prediction models, since they can provide model-independent, high-quality, and high-resolution observational wind information. In addition, a simple interpolation technique is introduced to substitute land in the calibration procedure of the Doppler centroid anomaly for open-ocean SAR scenes.