Spaceborne altimeters are an important data source for obtaining global sea surface wind speeds (U10). Although many altimeter U10 algorithms have been proposed and they perform well, there is still room for improvement. In this study, the data from ten altimeters were collocated with buoys to investigate the error of the altimeter U10 retrievals. The U10 residuals were found to be significantly dependent on many oceanic and atmospheric parameters. Because these oceanic and atmospheric parameters are intercorrelated, an asymptotic strategy was used to isolate the impact of different parameters and establish a neural-network-based correction model of altimeter U10. The results indicated that significant wave heights and mean wave periods are effective in correcting U10 retrievals, probably due to the tilting modulation of long-waves on the sea surface. After the wave correction, the root-mean-square error between the U10 from altimeters and buoys was reduced from 1.45 to 1.24 m/s and the impacts of thermodynamic parameters, such as sea surface (air) temperate, became negligible. The U10 residuals after correction showed that the atmospheric instability can lead to errors on extrapolated buoy U10. The buoy measurements with large air-sea temperature differences need to be excluded in the Cal/Val of remotely sensed U10.

中文翻译：

---

使用海浪参数改进高度计风速反演


星载高度计是获取全球海面风速（U10）的重要数据源。尽管已经提出了许多高度计 U10 算法并且表现良好，但仍有改进的空间。在本研究中，来自 10 个高度计的数据与浮标搭配，以调查高度计 U10 反演的误差。研究发现 U10 残差显着依赖于许多海洋和大气参数。由于这些海洋和大气参数是相互关联的，因此采用渐近策略来隔离不同参数的影响，并建立基于神经网络的高度计U10校正模型。结果表明，显着波高和平均波周期对U10反演的校正是有效的，这可能是由于海面长波的倾斜调制所致。经过波浪校正后，高度计和浮标U10之间的均方根误差从1.45减少到1.24 m/s，并且海面（空气）温度等热力学参数的影响变得可以忽略不计。校正后的 U10 残差表明，大气不稳定会导致外推浮标 U10 出现误差。遥感U10的Cal/Val需要排除海气温差较大的浮标测量值。