Case study: Phase-component amplitude variation with angle

In detectable and isolated thin layers below seismic resolution, phase decomposition can theoretically be used to discriminate relatively high-impedance thin-layer responses from low-impedance reservoir responses. Phase decomposition can be used to isolate seismic amplitudes with a particular phase response or to decompose the seismic trace into symmetrical and antisymmetrical phase components. These components sum to form the original trace. Assuming zero-phase seismic data and normal American polarity, seismically thin layers that are high impedance relative to overlying and underlying half-spaces are seen on the +90° phase component, whereas a relatively low-impedance thin layer will appear on the −90° phase component. When such phase decomposition is applied to prestack attributes on a 2D line across a thin, 8 m thick, gas-saturated reservoir in the Western Canadian Sedimentary Basin of Alberta, Canada, amplitude-variation-with-angle is magnified on the −90° phase component. The −90° far-offset component allows the lateral extent of the reservoir to be better delineated. This amplification is also seen on the −90° phase component of the gradient attribute. These results are corroborated by seismic modeling that indicates the same phase-component relationships for near- and far-angle stacks as are observed on the real data. Fluid substitution and seismic modeling indicate that, relative to full-phase data, the mixed-phase response observed in this study exhibits variations in fluid effects that are magnified and better observed at far angles on the −90° phase component.