CFOSAT Microwave Scatterometer Calibration And Wind Field Reprocessing Researc

The China-France Ocean Satellite(CFOSAT)microwave scatterometer(CSCAT)has been providing sea surface wind field products to global users for four consecutive years since its launch in 2018,during which the wind field inversion method has been continuously optimized and the wind field product versions have been continuously developed,but there are still some differences in the wind field data characteristics of different versions of the scatterometer.In addition,recent monitoring of the CSCAT wind field quality has shown that the CSCAT wind field errors are significantly large and the quality is unstable during certain time periods.Therefore,it is necessary to re-visit the CSCAT wind field inversion process,analyze the reasons for the fluctuation of the scatterometer wind field quality,and re-invert the scatterometer wind field for the whole period to obtain a CSCAT reprocessed wind field product with consistent temporal characteristics.In this paper,the calibration coefficients of CSCAT in 2021 are first obtained based on weekly time windows using the ocean calibration method,which is a function of the incident angle and the antenna azimuth of the scatterometer.By analyzing the temporal stability of the calibration coefficients,it is found that the HH polarization calibration coefficients vary widely in the 25th-32nd week of 2021,which is consistent with the time when the scatterometer wind field shows large deviations in 2021.The measurement state of the scatterometer under low signal-to-noise ratio conditions such as large incidence angle and low wind speed was then investigated.And the calculation problem of noise subtraction might be the key to cause the unstable quality of the CSCAT near-real-time wind field by analyzing the temporal variation of the negativeσ~0 data percentage under low signal-to-noise ratio.The obtained weekly calibration coefficients,as well as weekly MLE lookup tables and quality control threshold lookup tables,are then applied to the wind field inversion process,while iterative calculations will be performed in the wind field inversion process until the calibration coefficients converge.In the final obtained CSCAT reprocessed sea surface wind field,the bias of sea surface wind speed from the original 1 m/s to less than 0.2 m/s for 25 weeks-32weeks,and the wind field inversion residuals and wind vector cell rejection rate are also significantly reduced,thus more wind data is tained in the scatterometer reprocessed wind field.Based on the above results,the quality of the reprocessed wind field under different quality control methods is investigated by using the same quality control rejection data ratio,and it is found that the mean maximum likelihood estimate(MLE_m)is more suitable for the quality control in the inner swath,while the singularity exponent(SE)performs better on the outer swath.The spatio-temporal matching of the scatterometer wind data with rainfall also reveals that the MLEm and SE values also identify more precipitation in the inner and outer swath,respectively.The wind speed bias at different wind speeds show that the SE-based quality control method has lower wind speed deviations at low and high wind speeds,while the MLE_m is more advantageous at moderate wind speeds.Finally,this paper establishes a new ratio of wind vector cells rejected by quality control for reprocessed wind fields,preserving more scatterometer wind speed data.