Enhancing SAR Data Analyses for Offshore Wind Energy and Coastal Applications

This Ph.D. study focuses on the use of satellite data in connection with wind energy at sea, including coastal zones. The satellite data used comes from Synthetic Aperture Radar (SAR). SAR data are known for their global coverage and independence from atmospheric conditions. SAR data is available day and night and in all weather conditions. Offshore wind energy forms a significant part in connection with the increasing demand for renewable energy. There is a need for geophysical data in relation to offshore wind farms and coastal applications. Overall, the purpose is to contribute within possibilities of how satellite SAR data from the existing archives can be used. In the Ph.D. study, SAR data is used to measure the wind over the sea and wave parameters.

In order to derive the wind from SAR optimally, there is a need to reduce the effect of non-wind-related conditions in the SAR images. Ships, wind turbines, etc. give a high signal in the form of very bright pixels. If these are not filtered out, the wind speed from SAR will be overestimated. In the Ph.D. thesis, a new implementation is introduced to automatically reduce the effect of very bright pixels, which improves the accuracy of SAR-based wind measurements.

The research includes calculation of wind speed over the sea based on SAR data over the northern European seas. A focus area is to quantify the wake effects of the larger wind farms and the coastal wind speed gradients. In coastal zones where there are significant gradients in wind speeds, the various contributions and their interaction with the wake effects of wind turbines are quantified. The research shows the potential of SAR to characterize the wake effects of offshore wind farms and their spatial extent and size.

Wave data derived from SAR can potentially be used in connection with the planning of offshore wind farms. However, validation of the SAR-based wave data in the coastal zones is crucial in order to gain confidence in this new type of information. The Ph.D. study compares SAR-based wave data with observations from buoys and a hindcast wave model in the oceans near the U.S. The SAR-based wave parameters show a high correlation in comparisons, indicating that there is potential for using SAR as an independent data source.

Furthermore, the possibility of estimating the aerodynamic roughness above the sea surface based on SAR from the satellite is investigated. Despite the SAR Sentinel-1 data's spatial resolution and high recording height, which together cause so-called azimuthal cutoff effects that limit which wavelengths can be mapped directly, a new method is being developed and tested that overcomes this limitation. The estimated aerodynamic roughness based on SAR correlates well with the roughness calculated from observations from buoys, which makes it probable that there is an opportunity to globally use SAR data for this purpose.

In summary, the Ph.D. study contributes an improved method for calculating wind speed from SAR. This is introduced by automatically reducing the effect of very bright pixels caused by conditions other than the wind over the sea. Furthermore, new aspects within the mapping of wake effects of wind turbines and quantification of the horizontal coastal wind speed gradients close to the coast based on SAR are done. Finally, results are presented that assess SAR-based wave parameters and a new method to derive the aerodynamic roughness in coastal zones based on SAR from the satellite.