Simulations of SAR imagery of the ocean compared with data

Spectra of SAR images of the ocean have been simulated using quasilinear, velocity-bunching, and full time-dependent modelling. These simulations have been compared with X, C, and L-band imagery collected during the SAXON-FPN experiment using parameter values measured from the research tower, FPN, during the experiment. The results indicate that in most cases the three models agree well with each other and with actual imagery within the domain of applicability of each model. This is to be expected since the various models are simply different approximations to the full SAR imaging theory. The authors show that a coherence time approximation is not adequate to explain the azimuthal roll-off observed in SAR image spectra at all altitudes. They also find shifts of the spectral peak of azimuthally-travelling waves in both the simulations and images compared to measured buoy spectra; these can be understood in terms of the imaging mechanism. Finally, they find that the phase of the MTF measured on the tower cannot explain the observed artificial tilt of spectra of range-travelling waves, in agreement with the results of Bruning et al. 1993