Theoretical And Application Study Of Remoting Sensing Ocean Wave By Interferometric Synthetic Aperture Radar

Interferometric synthetic aperture radar (InSAR) is new microwave imaging radar that can obtain information about three-dimensional earth surface and movement of sea surface scattering. The source of information is the phase extracted from the complex data of synthetic aperture radar. InSAR system acquires a pair of complex images of earth or sea surface by using two antennas to observe at same time (single pass mode) or at two parallel paths (repeat-pass mode). Since 1990s, studying on land and ocean with InSAR has become a hot spot of microwave remote sensing. InSAR has widely used in surface deformation monitoring, measurement of Antarctic ice streams, detection of slowly moving target on ground or sea surface.Recently, along-track or across-track interferometric synthetic aperture radar has been gradually used to measure sea surface current velocity and study imaging mechanism of sea surface waves. InSAR has two advantages to measure ocean surface waves over the conventional SAR: (1) The phase of the complex InSAR image is approximately proportional to the radial velocity of the sea surface, and thus the InSAR inherent imaging mechanism offers the opportunity to measure the dynamic motions of the sea surface directly. (2) The real aperture radar modulation transfer function (RAR-MTF) has practically no effect on the InSAR phase image, but seriously affect the conventional SAR image.Based on the advantages of InSAR measurement of sea surface waves, we have done some research work about theoretical and applicaitonal study of remoting sensing ocean waves by interferometric synthetic aperture radar. The main content can be summarized as following:(1) Based on the new nonlinear integral transform between along-track interferometric synthetic aperture radar (ATI-SAR) phase spectra and ocean wave spectra, ATI-SAR phase spectra are calculated for various sea states and radar configurations. The numerical simulations show that the slant range to velocity ratio, the radar incident angle, the baseline, the significant wave height to wavelength ratio affect ATI-SAR imaging ocean waves. The ATI-SAR imaging theory is validated by means of X-band, HH-polarized ATI-SAR phase images and C-band, HH-polarized ATI-SAR phase image spectra of ocean waves. It is shown that ATI-SAR phase image spectra calculated by forward mapping are well in agreement with available ATI-SAR observations. ATI-SAR spectral correlation coefficients between observed and simulated are greater than 0.6 and are not sensitive to the degree of nonlinearity.(2) We establish an interferometric phase model of swell for the across-track interferometric synthetic aperture radar (XTI-SAR) with ocean surface elevation and velocity bunching. An analytical presentation of swell imaging by XTI-SAR is derived. The imaging mechanism of swell propagating in azimuth direction is further investigated. The ratio of the amplitude of the second harmonic and that of the foundamental wave is used to represent the nonlinearity of the imaging. By comparing the second order harmonic components of XTI-SAR phase and that of ATI-SAR phase, analyzing numerical simulations for different sea states and typical interferometric SAR parameters, it is found that XTI-SAR phase suffers stronger nonlinear distortion than ATI-SAR phase when velocity bunching is weak, so ATI-SAR is more appropriate than XTI-SAR to measure ocean wave. Otherwise, when velocity bunching is strong, ATI-SAR phase suffers stronger nonlinear distortion than XTI-SAR phase, XTI-SAR is more suitable than ATI-SAR to measure ocean wave.(3) We present a new nonlinear integral transform relating the ocean wave spectra to the XTI-SAR phase image spectra. Firstly, a phase model is proposed for the XTI-SAR phase includes ocean surface elevation and velocity bunching. Then, a new nonlinear integral transform is derived based on the phase model and the characteristic method function. The new transform differs from the one previously derived by Bao (1999) by an additional term containing the derivative of the radial component of the orbital velocity associated with the long ocean waves. The numerical simulations show that, in general, the additional term can not be neglected. Furthermore, XTI-SAR phase image spectra are estimated for different sea states and radar configurations. We demonstrate that the slant range to velocity ratio and significant wave height to ocean wavelength ratio are crecial factors to affect the XTI-SAR imaging ocean wave.(4) Based on the new nonlinear mapping between ATI-SAR phase spectra and ocean wave spectra, we develop a parametric algorithm to retrieve ocean wave spectra by means of ATI-SAR phase image. Furthermore, the wavelength and wave direction, as well as significant wave height are derived from the parametric algorithm. These retrieval parameters are well in agreement with the in situ buoy observations. The advantages of the parametric algorithm are given as following: (1) it does not require any additional information about first-guess ocean wave spectra from ocean wave model and wind velocity from scatterometer. (2) ATI-SAR phase image does not need to radiometric calibration, significant wave height can be directly calculated from the retrieval ocean wave spectra. (3) Wind speed of local imaging area is also derived when the retrieval process finished. Therefore, the parametric retrieval algorithm is able to achieve the joint retrieval of wind and wave information.