Abstract: In order to improve the accuracy of sea surface wind fields retrieved by the spaceborne fully polarimetric microwave radiometer, a wind field correction model based on statistical algorithm was constructed. Firstly, the WindSat sea surface wind fields and rainfall products coupled with buoy data from the American national data buoy center ranging from 2003 to 2020 are evaluated with obtained 18 996 coupled samples. Then, the wind field correction model is trained to correct the WindSat wind field product. The results show that the wind speed under rainfall in WindSat product was severely overestimated. The wind direction error increases with the increase of rain rate. After applying the wind field correction model, the error of low wind speed decreases obviously. The corrected wind speed accuracy is much higher than that of the WindSat product, regardless of the rain rate. The wind speed error reduces from 2.9 m/s to 2.1 m/s, and the wind direction error reduces from 26.9° to 26.3°.