Ocean Surface Wind Speed Estimation Combining Radiometric and Geometric Characteristics of Ocean Waves From ICESat-2 Photon-Counting Lidar

The ICESat-2 photon-counting lidar can simultaneously observe high-resolution radiometric and geometric characteristics of ocean waves. In this study, combining the wave radiometric and geometric characteristics, ICESat-2 can provide robust and accurate wind speeds over oceans that cannot be achieved by using a single characteristic. Specifically, with an along-track segment of 700 m, the sea surface photons and surface profile are detected and generated from ICESat-2 ATL03 geolocated photons by which the received energy and wave height are obtained. The unknown atmospheric transmittance and swell wave height have significant impacts on deriving the water surface reflectance (radiometric characteristic) and the wind-driven water height (geometric characteristic), respectively. We use all available ICESat-2 data in 2018 and 2019 to determine the unknown parameters and further separately estimate wind speeds. The radiation-based method performs better at low wind speeds, while the geometry-based method performs better at high wind speeds, and we combine each specific advantage to obtain more reliable wind speeds. In the Pacific Ocean, Atlantic Ocean, and Indian Ocean, thousands of ICESat-2 ground tracks from 2020 are used to estimate wind speeds, and the performance is validated against a trustworthy data source of ERA5 with a root-mean-square error (RMSE) of similar to 0.3 m for wave heights and an RMSE of similar to 1.5 m/s for wind speeds. ICESat-2 has shown great ability to provide independent wind speeds over oceans with high accuracy and resolution, which may fill the gap of wind speeds on a small scale as the large-scale results are well observed by spaceborne microwave devices.