C. Bruning

Estimation of the ocean wave–radar modulation transfer function from synthetic aperture radar imagery

Estimates of the ocean wave-radar modulation transfer function (MTF) are derived from synthetic aperture radar (SAR) imagery acquired by the American Naval Air Development Center airborne three-frequency SAR over the North Sea during the U.S./German SAR and X Band Ocean Nonlinearities-Forschungsplattform Nordsee experiment in November 1990. This is achieved by comparing measured and simulated SAR image spectra. The simulated SAR image spectra are computed from ocean wave height spectra measured by a pitch and roll buoy and by applying the generalized velocity-bunching model. First, SAR simulations are carried out by using the theoretical MTF which contains the tilt and range-bunching MTFs as well as the hydrodynamic MTF calculated from the relaxation time model. Second, SAR simulations are carried out by taking the modulus and phase of the MTF as free parameters. For waves traveling toward the radar antenna, best agreement is achieved when using the following values for the modulus |M0| and phase η of the nondimensional MTF defined by (5) and (10): |M0| = 8 - 13 for VV polarization and |M0| = 12 - 15 for HH polarization; η = 60° - 90° past the long-wave crest when the wind is blowing downwave, and η = 0° - 60° past the long-wave crest when the wind is blowing upwave. The values derived for the modulus of the MTF are in good agreement with values obtained from tower-based radar backscatter measurements. However, for the downwave case the phase of the MTF disagrees with the phase obtained from tower-based measurements, where usually values between 20° and 60° are found.

Imaging of ocean waves by SIR-C/X-SAR over the North Sea and North Atlantic

First results of the imaging of ocean waves by SIR-C/X-SAR over the North Sea and North Atlantic are presented. It is shown that the SAR imaging of ocean waves depends only weakly on the radar frequency and polarization, but depends strongly on wind speed and direction. Wind streaks are imaged best at high radar frequencies and VV polarization.

Simulation of interferometric synthetic aperture radar imaging of two-dimensional ocean surface wave fields

An interferometric synthetic aperture radar (INSAR) is capable of measuring ocean surface waves and surface currents. An INSAR velocity bunching model is derived which includes the radar cross section modulation and the velocity bunching modulation, as well as the velocity spread within the SAR resolution cell (parameterized by a scene coherence time). The authors simulate INSAR images of two-dimensional ocean wave fields by using the Monte-Carlo method, from which the INSAR amplitude and phase image spectra are calculated. It is shown that the INSAR phase image spectrum is almost independent of the radar cross section modulation, while the conventional SAR image spectra are more strongly affected by the interference of the radar cross section and the velocity bunching modulations. The authors find that the scene coherence time and the distance between the two antennas are the limiting parameters for imaging of ocean waves by INSAR.<<ETX>>

Comparison of ERS-1 and ALMAZ-1 SAR ocean wave imaging

The performance of the synthetic aperture radar (SAR) aboard the Russian ALMAZ-1 and the European ERS-1 satellites for imaging ocean surface waves is investigated. Collocated ALMAZ-1 and ERS-1 SAR images were acquired quasi-simultaneously on Oct. 6 and Oct. 8, 1992, over the North Atlantic in the vicinity of the island of Iceland. The SAR image spectra calculated from the ALMAZ-1 and ERS-1 images are compared with simulated SAR image spectra which are obtained from a wave spectrum hindcast by the WAM wave prediction model.<<ETX>>

Global statistics of integrated ocean wave parameters extracted from ERS-1 SAR wave mode image spectra

Ocean wave height spectra are extracted routinely from ERS-1 synthetic aperture radar (SAR) wave mode image spectra since July 1, 1992 by applying a SAR inversion scheme based on Hasselmanns closed nonlinear spectral integral transform. The scheme requires a first-guess wave height spectrum as regularization term obtained from the WAM wave prediction model. Global monthly statistics of integrated spectral wave parameters, such as significant wave height, mean wavenumber, mean wave direction and directional spread, are calculated from wave model and SAR extracted wave spectra. The general agreement between the mean spectral wave parameters is quite good. They have a correlation between 0.83 and 0.94. However, the SAR extracted mean wave height is 10-15% larger and the mean wavenumber is about 10% smaller than the wave model values. The results demonstrate the good performance and high accuracy of the SAR ocean wave inversion scheme.<<ETX>>

Ocean wave-radar modulation transfer function inferred from synthetic aperture radar imagery

Estimations of the ocean wave-radar modulation transfer function (MTF) at X-, C-, and L-band are obtained from synthetic aperture radar (SAR) imagery acquired by the American NADC airborne 3-frequency SAR over the North Sea during the US/German SAXON-FPN experiment in November 1990. This is achieved by comparing measured SAR image spectra with simulated ones which are computed from measured ocean waveheight spectra. SAR simulations are carried out by taking the modulus and phase of the MTF as free parameters. Best agreement is achieved when using the following values for the modulus |M/sup RAR/|, and the phase, /spl etasup RAR/, of the nondimensional MTF: |M/sup RAR/|=8-15 (VV and HH polarization), /spl etasup RAR/=60/spl deg/-90/spl deg/ for the case that the wind is blowing in the direction of the long ocean waves, and /spl etasup RAR/=0/spl deg/-60/spl deg/ for the case that the wind is blowing against this direction.<<ETX>>

On the nonlinear imaging of two-dimensional ocean surface wave fields by interferometric SAR

The nonlinear INSAR imaging mechanism of ocean waves is studied by applying an INSAR velocity bunching model and by using Monte-Carlo simulation techniques. Furthermore, a simplified INSAR phase imaging model is presented, which includes velocity bunching. A nonlinear integral ocean wave-INSAR phase spectral transform is derived. It is shown that the modulation by velocity bunching also enters into the INSAR imaging mechanism of ocean waves and gives rise to nonlinearity effects in the INSAR imaging mechanism. In some cases the INSAR phase image spectrum is split into two parts due to the interference between the velocity bunching term and the velocity term.

A comparison of ocean wave-radar modulation transfer functions at different radar frequencies and polarizations determined from tower and aircraft measurements

For the inversion of SAR image spectra into ocean wave spectra the ocean wave-radar modulation transfer function (MTF) plays an important role for waves traveling in or near the range direction. MTFs can be measured from sea-based platforms or inferred from the comparison of measured and simulated SAR image spectra. The MTFs determined by the two experimental methods have comparable moduli, but different phases. Also they do not agree with the values derived from various modulation theories. A short summary of the present status is given and possible reasons for the observed discrepancies are discussed.

Ocean wave-radar modulation transfer functions inferred from airborne synthetic aperture radar imagery

Estimates of the ocean wave-radar modulation transfer function (MTF) at X-, C-, and L-band are obtained from synthetic aperture radar (SAR) imagery acquired by the American NADC airborne 3-frequency SAR over the North Sea during the US/German SAXON-FPN experiment in November 1990. This is achieved by comparing measured SAR image spectra with simulated ones which are computed from measured ocean waveheight spectra. SAR simulations are carried out by taking the nondimensional modulus, |M/sup RAR/|, and phase, /spl etasup RAR/, of the MTF as free parameters. The phase, /spl etasup RAR/, is defined in such a way that it is positive in front of the wave crest and negative behind the wave crest in the reference system of the long ocean wave. Best agreement is achieved when using the following values for |M/sup RAR/| and /spl etasup RAR/: |M/sup RAR/|=8-15 (VV and HH polarization), /spl etasup RAR/=60/spl deg/-90/spl deg/ for the case that the wind is blowing in the direction of the long ocean waves, and /spl etasup RAR/=0/spl deg/-60/spl deg/ for the case that the wind is blowing against this direction.<<ETX>>