Wind retrieval processing for HY-2A microwave scatterometer

The research constructed a retrieval algorithm of routine wind fields for microwave scatterometer onboard HY-2A (HY-2A/SACT). To prepare the sigma0 measurements for conducting wind vectors retrieval, the time ordered L1B product was regrouped to the sub-track aligned wind vector cells (WVCs). Besides, the authors utilized an extension land and ice mask to simplify the surface flag process. The maximum-likelihood estimator (MLE) was applied to retrieve wind vector for each WVC with the NSCAT-2 as geophysical model function (GMF). Furthermore, to greatly improve the ambiguity removal skill, the research used circular median filter to remove the ambiguities, and the NCEP wind field as a background to re-initialized wind field comprising of ambiguous wind vectors. Wind vectors observed by HY-2A/SACT were evaluated by comparing with NDBC buoy observations. The RMS difference for wind speed was less than 2 m/s or 10%; while the RMS difference for wind direction was less than 20°comparing with wind direction. This result showed that the wind products provided by HY-2A/SCAT are able to meet the requirements of science and routine meteo users.