Retrieving Ocean Surface Wind Speeds in Real Time on Spaceborne GNSS-R Receivers: Algorithm and Validation

Based on delay-Doppler maps (DDMs) in raw counts generated by spaceborne global navigation satellite system reflectometry (GNSS-R) receivers, retrieving ocean surface wind speeds is feasible so that several spaceborne GNSS-R missions have been carried out. However, it is currently troubled by global data latency of several hours or even more, due to the bottleneck in the satellite downlink. Consequently, this article, for the first time, presents an algorithm for spaceborne GNSS-R receivers to conduct the DDM calibration in orbit and then to retrieve ocean surface wind speeds in real time, which contributes to not only lightening the burden on downloading a wealth of scientific data but also broadcasting real-time ocean surface wind speeds to users. Since there is a power correlation between direct and reflected signals from the same GNSS satellite with respect to the GNSS-R receiver, this algorithm calibrates direct signal power first, and then it estimates the real-time GNSS transmitter effective isotropic radiated power (EIRP) at the reflected signal according to the normalized antenna pattern of the corresponding GNSS satellite. Afterwards, DDMs in raw counts are calibrated. Finally, ocean surface wind speeds are computed using pre-trained geophysical model functions (GMFs). Exploiting the scientific data from the GNOS-II onboard the China's FY-3E satellite, this algorithm is validated carefully, and final retrieved ocean surface wind speeds against collocated ECMWF wind speeds have an overall root mean square error (RMSE) of 1.68 m/s and 1.50 m/s for GPS-R and BDS-R, respectively.