Study On Wind Retrieval Simulation And Precision Improvement For FY-3Dual-Frequency WFR

Tropical cyclone is one of the primary disastrous synoptic systems in China.With thecontinuous observation, global coverage and the ability to penetrate through precipitationlayer, microwave sensors on polar orbit satellites can provide more precise observations of thetropical cyclone location and intensity for marine extreme weather forecasting, which willcompensate for the shortage of conventional observations. In the condition of mid-and-lowwind speed and non-precipitation, scatterometer is an effective instrument of measuring seasurface wind vector, but is hlghly affected by high wind and heavy rain. Compared withsingle-frequency scatterometer, dual-frequency scatterometer has advantages in higher spatialresolution and better response to winds in extreme weather conditions, where high winds areusually associated with high rain rates. The FY-3satellite microwave scatterometer, namedWind Field Radar (WFR), which will be launched in2016is designed to use thedual-frequency and dual-polarization time-sharing observation pattern.is the only way tomeasure ocean vector winds (OVW).The purpose of this study is to investigate the potential of the WFR proposed to flyaboard FY-3satellite to measure OVW in rain fall areas of hurricane Ivan. A theoretical modelbased on radiation transfer equation including rain attenuation and scattering, has beendeveloped to quantify the modification by rain of the measured backscatter. Based on thesimulated normalize radar cross section (NRCS) dataset generated by forward model, theimpact on the retrieved winds has been analyzed, and the dual-frequency retrieval algorithmhas been firstly studied.Analysis shows that changes of contaminated NRCS briefly depend on relative size ofvolume scattering item and attenuation item. Precipitation and hydrometeor in cloud oftropical cyclone can then significantly degrade the OVW accuracy. It’s indicated that Ku-bandscatterometer underestimate the wind speed by up to18m·s-1for wind speed condition of30-40m·s-1and rain rate condition of30mm·hr-1for Ku bands. Wind vectors measured atKu-band can be more severely altered by rain, cloud and gas in the atmospheric layer then atC-band. Analysis of random rough surface scattering physical mechanism shows Ku-band andC-band have different sensitivity intervals. Therefore, dual-frequency measuring can expandthe dynamic range of retrieved wind speed. Retrieval results show that new methods combined Ku-band and C-band (dual-frequencyMLE and partitioned wind retrieval technique) have higher spatial resolution then C-bandretrieval, and better performance in rain fall region then Ku-band retrieval. Especially,dual-frenquency MLE with rain adjustment can significantly reduce the rainfall error, is aneffective way to improve the wind retrieval accuracy in tropical cyclone.