A one-dimensional model is developed to simulate azimuthal synthetic aperture radar (SAR) imaging of the ocean surface. The model developed here is general and can admit both the ''distributed surface'' and the ''velocity bunching'' approaches. Computer simulations are performed to test the validity of this model. The simulations show that the time-dependent modulation patterns due to the radar cross section variation and the velocity bunching effects give optimum focusing around half the phase velocity of the long wave. It is shown that the focus dependence is due to an incoherent integration process and not due to motion induced phase errors. It is argued that the radar reflectivity is correlated over distances small in comparison with the SAR resolution. It is argued that the radar reflectivity is correlated over distances small in comparison with the SAR resolution. In the simulations the radar reflectivity is assumed to be spatially decorrelated. Also given are the relative importance of the various effects that degrade the azimuthal resolution. It is shown that the limited radar temporal coherence due to the velocity spread of short waves degrades the azimuthal resolution of the SAR. This coherence time depends on sea state and radar wavelength. It is shown that the orbital acceleration effects are small in comparison with the smearing due to the velocity spread. It is found that the degraded azimuthal resolution of the radar minimizes the nonlinear effects of velocity bunching. The simulations show that in the Tower Ocean Wave and Radar Dependence Experiment (TOWARD), SAR imaging at L band is reasonably linear. ¿ American Geophysical Union 1988