A Study Of HY-2A Satellite Microwave Scatterometer Sea Surface High-wind Field Retrieval Model

The sea surface wind field is an important physical parameter for studying meteorology,climate,and the exchange of matter and energy between the ocean and the atmosphere.Spaceborne scatterometer is a kind of radar instrument which can be used to observe the sea surface wind field.Compared with the traditional sea surface wind observation technology,scatterometer has the advantage of large-scale real-time access to a large number of sea surface wind data.In the low wind speed without rainfall conditions,the precision of scatterometer wind measurement is reliable.However,under high wind speed conditions,the accuracy of the scatterometer observes the sea surface wind field is decrease,and the high wind area is usually accompanied by rainfall,making the scatterometer subject to the additional impact.The HY-2A satellite launched in August 2011 is a remote sensing satellite for monitoring the marine environment.HY-2A loaded the first China operational operation of the scatterometer HY2-Scat,and the working band of HY2-Scat is Ku-band.HY-2A satellites in the orbit period obtained a large number of scatterometer to ground observation data.In this paper,based on the L2 A data of HY-2A scatterometer and SSM / I data and NCEP data,a high-wind geophysical model function(HW-GMF)is established by neural network training,in order to improve the accuracy of scatterometer inversion of wind vector under high wind speed.The results show that the error of the backscattering coefficient and the backscattering coefficient measured by HY-2A using the HW-GMF simulation is less than 0.75 dB.The geophysical model function,which is established by the neural network training,has high reliability.The wind accuracy of the geophysical model NSCAT-2 currently used by HY-2A scatterometer and HW-GMF is compared by the method of wind vector inversion.The results show that the error of HW-GMF reverses wind speed and SSM/I wind speed is less than 2m/s,which indicates that HW-GMF improves the accuracy of scatterometer in high wind speed.Using the HY-2A scatterometer inversion of the wind field L2 B standard product data from 19 July to 19 August 2012,NCEP reanalyzes the sea surface wind field data,the SSM/I radiometer measures the rainfall rate and the buoy observation data,the wind speed and wind direction error of L2 B data in the presence or absence of rainfall were statistically tested,and the influence of rainfall on the wind measuring of the scatterometer was verified.In order to improve the accuracy of HY-2A scatterometer under rainfall conditions,a neural network model which can be used to correct the influence of rainfall on scatterometer is established by using L2 B and NCEP pairing data,and the corrected results are tested.Compared with the error statistical results,under the rainfall condition,the error between the NCEP wind speed and the L2 B wind speed corrected by the neural network is less than 0.1m/s,and which is equivalent to the wind speed error without rainfall.It is shown that the neural network is suitable for correcting the influence of rainfall on the wind measuring of the scatterometer,which can effectively improve the accuracy of the wind speed in the scatterometer retrieval wind.