This sections examines the amplitude behavior of AMO through numerical experiments on a dipping bed model. In each AMO reconstruction, the output amplitudes are compared to the theoretical amplitudes predicted by synthetic modeling. The three experiments consisted of applying AMO for azimuth rotation, absolute-offset continuation and vector-offset transformation. The input data simulate a constant-offset section recorded at an offset of 1600 meters and azimuth of 5 degrees, measured from the dip direction (the in-line direction). Figure comp-amo compares the AMO reconstructed amplitudes to the theoretical amplitudes for each experiment.
Figure comp-amoa shows the results of applying AMO to rotate the geometry by 40 degrees while keeping the offset constant. The peak amplitudes extracted from the AMO result at the new azimuth of 45 degrees are almost identical to the theoretical amplitudes predicted by modeling. Similar to comp-amoa, Figure comp-amob compares the results of AMO applied for an offset continuation of 800 meters. The transformation is a 2D operation since the azimuth is held constant. The peak amplitudes from the 800 meter offset-section follow the theoretical amplitudes very closely with an error of less than a few percent. Finally, figure comp-amoc shows the amplitude preservation by AMO when applied as a full 3D operator to rotate the azimuth by 40 degrees and change the offset by 800 meters. The AMO transformation again correctly reconstructed the peak amplitudes at the new geometry. The amplitude curves match very closely and the differences are mainly attributed to cumulative errors in the processing sequence surrounding AMO, which includes spherical divergence and NMO corrections prior to AMO and inverse NMO after AMO.