Retrieval of Ocean Surface Wind Speed using Sentinel-1 HH Polarization Data

Transformation of normalized radar cross section (NRCS) in HH-polarization to that in VV-polarization NRCS is essential step to retrieve sea surface wind speed from SAR data when the Geophysical Model Function (GMF) is applied. In this paper, four different Polarization Ratio (PR) models including Mouche-PR1 model, Mouche-PR2 model, Zhang-PR model, and Liu-PR model are combined with the GMF CMOD5.N to retrieve sea surface wind speed from 130 HH-polarized Sentinel-1 SAR images. The results suggest that wind speeds retrieved using the Zhang-PR model agree better with the National Data Buoy Center (NDBC) buoy measurements and ERA-Interim reanalysis model dataset of the European Centre for Medium-Range Weather Forecasting (ECMWF) than the other three models. Compared to the buoy measurements, the root mean square error (RMSE) and scatter index (SI) of wind speed retrieved by the Zhang-PR model are 1.60m/s and 22.40% , respectively, while compared to the ERA-Interim model results the RMSE and SI are 2.30m/s and 30.03% , respectively. It is suggested that the Zhang-PR model combined with the CMOD5.N function has a potential application in retrieving sea surface wind speed from Sentinel-1 images acquired at HH-polarization.