SAR Ocean Images Research Articles

Accuracy Estimate for Determination of the Near-Surface Wind Velocity Using the SAR Ocean Image Spectrum

Numerical modeling of the SAR image of wind waves propagating azimuthally is performed for various wind velocities. It is shown that if the wind is relatively weak (up to 10 m/s), then the position of the image-spectrum maximum is a fairly sensitive parameter, such that the accuracy of the respective wind-velocity measurement technique for these wind velocities should be 1 m/s or less for a high spatial resolution. Under conditions of stronger wind, the sensitivity decreases; however, more definite results for the stronger-wind domain can be obtained only experimentally.

Wavelet analysis in SAR ocean image profiles for internal wave detection and wavelength estimation

Oceanographers and remote sensing researchers have long recognized the potential of using satellite imagery for studying oceanic internal waves. Radars are able to image internal waves because they are particularly sensitive to changes in the small-scale surface roughness (i.e. the capillary and ultragravity waves) present on the ocean surface which are altered by the velocity field associated with the internal waves. If, as seems likely, the greytone patterns of these images can be confirmed to correspond to trough and crest patterns of internal waves, then a great deal can be learnt about internal waves from satellite data. In this paper, the utility of analysis as a tool for oceanic internal wave detection and wavelength estimation is examined using both continuous and discrete versions of the transform. The theoretical background of each procedure is briefly described and applied using a specific wavelet for each case. In this first approach, the authors only consider supervised detection for the internal wave train detection problem. Normally, an unsupervised method using the two-dimensional (2D) transform is required for internal wave detection and orientation, including land-sea separation to avoid false alarms. They first present the construction of an appropriate basis, based on an oceanographic soliton internal wave analytical model, to detect and localize nonlinear wave signatures from SAR ocean image profiles. The structure of arbitrary basis derived from the compactly supported orthonormal B-splines wavelets is studied so as to obtain more optimal discrete decompositions. Comparisons are made for decompositions based on several families of compactly supported wavelets. Finally, the continuous transform is applied to estimate energies and wavelengths within soliton peaks from the detected internal wave trains. The advantages and drawbacks of the continuous and discrete transforms for the internal wave detection problem from SAR ocean image profiles are also discussed. The results from this study show that analysis is an excellent tool to detect internal waves against background noise, and to estimate, with a good degree of precision, soliton wavelengths from SAR ocean image profiles.

SAR ocean image decomposition using the Gabor expansion

Demonstrates the utility of the Gabor expansion as a new tool in geophysical research. The Gabor expansion provides good time-frequency (or space-wavenumber) localization and is ideally suited to represent nonstationary processes. The properties of this tool are demonstrated by expanding an FM-chirp waveform, and azimuth cuts taken from two different SAR ocean images. The effects of filtering in Gabor phase space are also investigated. >

SAR ocean image representation using wavelets

The utility of wavelet analysis as a tool for geophysical research is examined using both continuous and discrete versions of the wavelet transform. In both cases, waveform decomposition and reconstruction is possible using somewhat different computational procedures. The theoretical background of each procedure is briefly described and applied using a specific 'wavelet'. The wavelet used is based on a Gaussian function, and provides simultaneous time-frequency (or space-wavenumber) localization that meets the lower limit of the uncertainty principle. A representation of this type is ideally suited for the analysis of waveforms that arise from nonstationary processes. The properties of wavelet analysis are examined by expanding an FM-chirp waveform and azimuth cuts taken from two different SAR ocean images. The performance and ease of implementation are compared for the continuous and discrete formulations, and the effects of filtering in wavelet phase space using the discrete case are also examined.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Focusing simulations of synthetic aperture radar ocean images

The SAR ocean images obtained in the Tower Ocean Wave and Radar Dependence Experiment (TOWARD) are carefully analyzed at different focus settings and compared with simulated results based on various theories for imaging surface waves. The agreement between the experimental data and all of the SAR simulations except one is found to be favorable. There is also surprisingly close agreement among the different theoretical models themselves, suggesting a closer fundamental similarity among the contending theories than was previously thought. It is shown that the width of the focusing curve has an inverse dependence on the SAR integration time. For the TOWARD conditions it is found that the image modulation due to azimuth‐traveling surface waves is greater than that due to range‐traveling waves.