Wind Speed Retrieval Of Ocean Surface Using RADARSAT-2 Radar Data

With the development of the traditional energy consumption increasing, solar energy, wind energy and other renewable resources have great application prospects. Offshore wind energy resource which will become China’s resources transformation direction in our country is particularly abundant.Using scatterometers and radiometers can obtain large area of ocean surface wind field data.With the wide application of the synthetic aperture radar satellite, it can not only get the high resolution ocean surface wind field data, but also can obtain the ocean surface wind field data in the near coastal area. It can provide long time series, high accuracy and high resolution ocean surface wind field data for offshore wind resource development and utilization.In the paper, based on the principle of the ocean surface wind field retrieved by the high spatial resolution synthetic aperture radar, the RADARSAT-2 satellite data with the characteristics of multi polarization modes, multi imaging modes and high spatial resolution, can be used to retrieve the ocean surface wind speed in the east ocean of China using GMF models and C-2PO model.In this paper, the ERA-Interim reanalysis wind field data is compared with the measured data from the buoys which are provided by NOAA. The result shows that the RMSEs of the ocean surface wind speed and direction of ERA-Interim data are 0.98m/s and 12.79° respectively, which show a high accuracy, and which are determined to be credible data. In this paper, we choose ERA-Interim wind field data as the initial wind direction data and wind speed validation data of ocean surface wind speed retrieved by RADARSAT-2 satellite images.Based on the characteristics of multi spatial resolution and multi polarization modes of RADARSAT-2 radar data, the co-polarization and cross-polarization data are both used to retrieve ocean surface wind speed in different models. We use three GMF models (CMOD4, CMOD5 and CMOD-IRF2) to retrieve ocean surface wind speed for W polarization of RADARSAT-2 radar data. The results show that we can get the highest accuracy ocean surface wind speed with the root mean square error is less than 1.5m/s when using the CMOD4 model to retrieve ocean surface wind speed. For the HH polarization of RADARSAT-2 radar data, we first use three different polarization ratio models (Bragg, Thompson and Kirchhoff models) to transform respectively, and then use three GMF models (CMOD4, CMOD5 and CMOD-IRP2) to retrieve ocean surface wind speed. When Kirchhoff models are applied to transform, all the three GMF models can get high accuracy ocean surface wind speed which have no obvious difference between them with the RMSEs are less than 2m/s. C-2PO model performs well in retrieving ocean surface speed for VH and HV polarization of fine quad RADARSAT-2 radar data and there is no obvious difference between them. Co-polarization data show higher accuracy than cross-polarization data in retrieving ocean surface wind speed. When we apply C-2PO model to retrieve ocean surface wind speed for ScanSAR cross-polarization radar data, the result can’t perform well.Overall, using RADARSAT-2 dats can retrieve high accuracy ocean surface wind speed. Especially the cross-polarization radar data retrieved result is better than the co-polarization data, without requiring the radar incident angle and wind direction. The cross-polarization of the fine quad mode RADARSAT-2 radar data show a significant advantage over the traditional method, which will become the development direction of ocean surface wind speed retrieved in the future.