Thin-layer Reservoir Property Interpretation Using Seismic Odd and Even Peak Amplitude Attributes

Interpretation of seismic amplitude attributes for thin layers often suffers from the limited seismic resolution and related wavelet tuning. The tuning amplitude response from thin layers is especially complex when layer top and base reflections have unequal magnitude and variable sign. The complex tuning behavior can be simplified using the fact that any arbitrary seismic signal can be uniquely decomposed into odd and even components that have distinct sensitivities to variation in thin-layer and overburden properties. Extraction of the two components can be achieved using the Fourier transform. Spectral division applied to each component is effective in removing the tuning effect from peak-amplitude attributes. Numerical analysis and a synthetic seismic example show that peak amplitude of the seismic data odd part is more sensitive to thin-reservoir property change compared to that of conventional peak amplitude (total-waveform) and its even part. When applied in analyzing real seismic data in a tight-dolomite reservoir in Sichuan basin, China, conventional seismic amplitude was not obviously correlated to porosity variations. However, the odd-part peak-amplitude attribute achieved acceptable match with porosity measured in four wells. These results suggest that, for thin layers, amplitude-based interpretation and inversion may benefit from isolation of even and odd amplitude attributes.