| Synthetic Aperture Radar(SAR)is an active microwave sensor that can capture large-scale images with high spatial resolution under all-day and all-weather conditions.Currently,the orbiting satellite SAR systems have a wealth of polarization information that allows for acceptable inversion of wind fields in complex sea states.The C-band satellite SAR,particularly,has a well-established Geophysical Model Function(GMF)in the co-polarization(VV,HH)backscattering signal,making it well-suited for quantitatively inverting the sea surface wind fields with high accuracy at low and medium wind speeds.The cross-polarization(VH,HV)backscattering coefficient shows a general upward tendency at C-band without any saturation following the increasing wind speed.Neither the wind direction nor radar incidence angle exerts a visible impact on the backscattering coefficient in this case.Therefore,the wind field inversion method using cross-polarized data shows better performance for inverting wind speed at the high sea state,e.g.,typhoons,than those based on the co-polarized radar data.In this paper,we discuss the best inversion models and potential application models of different polarization data in wind speed inversion based on physical models,and the main contents and conclusions are as follows:(1)This paper aims to determine how the wind fields affect the roughness of the sea surface and,further,how the sea surface roughness influences the corresponding radar backscattering coefficient by examining the physical mechanism of microwave scattering from random rough sea surfaces.To this end,the correlation function of the superimposed rough sea surfaces at different scales is analyzed,and an exponential function describes the associated Fourier transform.Subsequently,the backscattering coefficient could be yielded by the AIEM model together with the correlation length and root mean square height under different wind fields.On this basis,the relationship between the wind field and the backscattering coefficient of the sea surface is established.Afterward,this study probes the sensitivity of backscattering coefficients to wind fields(wind speed and direction)and radar incidence angles under multipolarizations(cross and co-polarizations).Thereupon,it compared the advantages and disadvantages of the empirical models(e.g.,the co-polarized CMOD series functions,linear cross-polarized models,etc.),following which the advantages and applicability of those models are clarified.(2)Regarding the wind field inversion under typhoon weather,the dual-polarization Sentinel-1 SAR images in the Extra Wide(EW)mode,together with various inversion models,are adopted to invert the wind speed of Typhoons “Mindulle”,“Malou”,and “Hinnamnoe” in 2021.The results clearly state that the co-polarized data can obtain high accuracy in the wind field inversion under low and medium wind speeds.By contrast,the cross-polarization is more suitable for inverting the wind speed under high sea conditions(>20m/s),with a maximum inversion result of around 45m/s.However,cross-polarization still faces constraints concerning inversion accuracy at low and medium wind speeds.Therefore,it is preferable to apply the inversion model based on the cross-polarization method,combined with the CMOD function inversion model based on co-polarization,to improve the inversion accuracy for the wind field of the sea surface.The wind speed deviation obtained from the multi-polarization joint inversion model has the previous wind speed underestimation negative value of 1.55m/s to1.12m/s,the root mean square difference is reduced by 2.03m/s,and the correlation coefficient is improved from 0.755 to 0.766.This paper discusses the optimal inversion model and potential applications of different polarization data in the wind speed inversion according to the physical models,which is vital for typhoon forecasting and disaster prevention and reduction.Also,this study provides a reference basis and theoretical support for the related studies. |
PDF Full Text Request