부산대학교 도서관

In this study, we model electromagnetic (EM) scattering from a realistic ocean surface to assess through simulation the effect of varying key slick properties on backscatter at microwave frequencies of L-, C-, and X-bands for both thin and emulsified mineral oil. An ocean surface model is implemented by generating randomly rough ocean surface instances from ocean wave spectra corresponding to a variety of slick properties and different wind speeds. The finite difference time domain (FDTD) method, based on Maxwell’s equations, is used to calculate the normalized radar cross section (NRCS) from the ocean surfaces, which we validate with radar observations. Results show that the effect on the NRCS does not scale linearly with the spectral damping caused by the oil layer. By changing various layer properties, we determine that the surface elasticity and oil kinematic viscosity most strongly impact the NRCS. The model is run with different oil layer thicknesses to evaluate the capability of synthetic aperture radar (SAR) to determine absolute or relative slick thickness. We find that the thickness cannot be accurately determined from SAR backscatter alone in the absence of information about the key slick properties or calibration against known thicknesses in the given environmental conditions. The simulations indicate that ocean wave spectral components outside the expected Bragg scattering regime contribute significantly to the backscatter in some cases; furthermore, the presence of an emulsion layer under certain conditions and for certain radar frequencies creates constructive interference that causes the NRCS to be enhanced rather than reduced when the layer thickness increases.