Synthetic aperture radar satellite data for offshore wind assessment: A strategic sampling approach

Many studies have shown that the synthetic aperture radar (SAR) approach is an adequate tool for wind assessment and mapping in coastal and offshore areas. However, a large SAR scene sample is required to precisely calculate wind statistics. The proposed sampling algorithm, called strategic sampling, is based on a long term data series (e.g. QuikSCAT) from a nearby location to accurately evaluate the wind statistics with a reasonable number of wind observations (<45). Strategic sampling methodology relies on the complementarity of two databases: the SAR satellite RADARSAT-1 and the QuikSCAT scatterometer were chosen. Random sampling, on the other hand, requires 75 and 225 wind observations to estimate, respectively, the average wind speed and mean power output of a modern wind turbine, with an error less than 10% for a 90% confidence level. This represents a difference of at least two to seven times more SAR scenes for random sampling than for strategic sampling. The SAR sample selected via strategic sampling at a specific offshore location can then be used for wind predictions in neighboring regions. Finally, the strategic sampling algorithm is used to estimate the minimum number of SAR images for a coastal region located in eastern Canada.