Influence of seismic diffraction for high-resolution imaging: applications in offshore Malaysia

Abstract

Small-scale geological discontinuities are not easy to detect and image in seismic data, as these features represent themselves as diffracted rather than reflected waves. However, the combined reflected and diffracted image contains full wave information and is of great value to an interpreter, for instance enabling the identification of faults, fractures, and surfaces in built-up carbonate. Although diffraction imaging has a resolution below the typical seismic wavelength, if the wavelength is much smaller than the width of the discontinuity then interference effects can be ignored, as they would not play a role in generating the seismic diffractions. In this paper, by means of synthetic examples and real data, the potential of diffraction separation for high-resolution seismic imaging is revealed and choosing the best method for preserving diffraction are discussed. We illustrate the accuracy of separating diffractions using the plane-wave destruction (PWD) and dip frequency filtering (DFF) techniques on data from the Sarawak Basin, a carbonate field. PWD is able to preserve the diffraction more intelligently than DFF, which is proven in the results by the model and real data. The final results illustrate the effectiveness of diffraction separation and possible imaging for high-resolution seismic data of small but significant geological features.

Appendix

$$ \left( {\frac{\partial \tau }{\partial x}} \right)^{2} + \left( {\frac{{{\text{d}}\tau }}{{{\text{d}}x}}} \right)^{2} = \frac{1}{{v^{2} \left( {x,z} \right)}}, $$

(9)

$$ \sin \emptyset = v\frac{{{\text{d}}\tau }}{{{\text{d}}x}} , $$

(10)

$$ \frac{\partial \sigma }{\partial x}\frac{\partial \tau }{\partial x} + \frac{\partial \sigma }{\partial z}\frac{\partial \tau }{\partial z} = 1, $$

(11)

$$ \frac{\partial \beta }{\partial x}\frac{\partial \tau }{\partial x} + \frac{\partial \beta }{\partial z}\frac{\partial \tau }{\partial z} = 1, $$

(12)

$$ \frac{\partial }{\partial z}\left( {\frac{\partial \beta }{\partial x}} \right) + \frac{\partial }{\partial x}\left[ {\mu (x,z)\frac{\partial \beta }{\partial x}} \right] = 0, $$

(13)

where v(x, z) is the velocity and

$$ \mu (x,z) = \frac{\partial \tau }{\partial x}\left[ {\frac{\partial \tau }{\partial z}} \right]^{ - 1} . $$

Equation (9) is the eikonal equation, Eqs. (11) and (12) are derived from Pusey and Vidale (1991) and Eq. (13) follows from Eq. (11).