Numerical Weather Prediction Ocean Calibration for the Chinese-French Oceanography Satellite Wind Scatterometer and Wind Retrieval Evaluation

The rotating fan-beam architecture wind scatterometer Chinese-French Oceanography
Satellite (CSCAT) onboard the Chinese-French Oceanography Satellite leads to varying geometries and
varying wind retrieval performance across the swath. To assess this, the wind vector cells (WVCs) are
classified into three groups: outer, sweet, and nadir. The sweet swath WVCs contain the most diverse
geometries, which give the best wind retrieval performance, whereas on the opposite the outer swath
WVCs have the least diverse geometries, which makes the wind retrieval and ambiguity removal more
challenging. NWP Ocean Calibration (NOC) is used to calibrate the radar backscatter prior to the wind
inversion. It computes the difference between measured backscatter and simulated backscatter from
collocated NWP winds through the geophysical model function. The difference is applied as a correction.
An improved NOC method, NOC as a function of incidence angle and azimuth angle (NOCant) is
introduced, which takes the orientation of the antenna into account. This method is compared with
the widely used NOC as a function of incidence angle only (NOCinc). It shows a better fit with the
geophysical model function, except for the outer swath WVCs which have limited geometry diversity.
It also corrects the asymmetric wind direction bias distribution across the swath and reduced the nadir
swath relative wind direction bias which is caused by the poor azimuth diversity. In conclusion, the
rotating fan-beam architecture of CSCAT leads to unique and varying data characteristics across the
swath. Overall, the performance of the proposed NOCant correction is better than NOCinc and improves
the wind statistics.