Method for rapid fault interpretation of fault surfaces generated to fit three-dimensional seismic discontinuity data

• US 20070078604A1
• Filed: 09/17/2003
• Published: 04/05/2007
• Est. Priority Date: 10/18/2002
• Status: Active Grant

• Current Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Original Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Inventors: Dobin, Mark, Hauge, Thomas, Gillard, Dominique, Kim, Chung-Sung, Holl, James
• Granted Patent: US 7,330,791 B2
• Time in Patent Office: Days
• US Class Current: 702/16
• CPC Class Codes: G01V 1/301 for determining seismic cross-sections or geostructures
• CPC Class Codes: G01V 1/32 Transforming one recording into another or one representation into another
• CPC Class Codes: G01V 2210/641 Continuity of geobodies

Abstract

A method to extract fault surfaces from seismic data (1,3) is disclosed. The method comprises: (a) generating at least two fault sticks (5) from the same fault from at least two slices of the seismic data wherein each slice comprises at least one fault stick from the same fault, (b) constructing an initial three-dimensional fault surface (7) containing the fault sticks, and (c) reconstructing the initial fault surface (9) using a deformable surface model to fit discontinuity or coherency information in the seismic data in an iterative process. Techniques are disclosed for constructing the initial fault surface from interpreter-provided fault nodes, and for performing the deformable surface iteration by defining an energy function for the fault surface and then minimizing the surface energy.

Claims

1. A method to create a fault surface from a three-dimensional seismic data volume, comprising:

constructing an initial fault surface in three-dimensions, said surface containing at least two fault sticks, the fault sticks being from the same fault but from different slices of the seismic data volume, each fault stick being defined by at least two fault nodes; and

reconstructing the initial fault surface to fit the three-dimensional seismic data, said reconstruction using an iterative evolution of a deformable surface model of the fault surface, said evolution being based on smoothness of the fault surface and a fault-indicating parameter of each location on the fault surface.

2. The method of claim 1, wherein the fault nodes are obtained by manual interpretation of the data volume.

3. The method of claim 1, wherein the data volume slices comprise a vertical slice and an horizontal slice.

4. The method of claim 1, wherein the seismic data comprise seismic amplitude discontinuity data, and the fault-indicating parameter is the seismic amplitude discontinuity.

5. The method of claim 4, wherein the iteration of the deformable surface model uses local minimization of an energy function of such surface, said function having as variables the curvature of the fault surface and the discontinuity in seismic amplitudes, each variable being itself a function of spatial location on the fault surface, and wherein the iteration is terminated when the change in the surface energy relative to the preceding iteration is less than a pre-determined value.

6. The method of claim 5, wherein the energy function is a weighted sum of an internal force and an external force, said internal force being a function of curvature of the fault surface and said external force being a function of seismic amplitude discontinuity on the fault surface.

7. The method of claim 1, wherein the seismic data comprise seismic amplitude coherency data, and the fault-indicating parameter is the seismic amplitude coherency.

8. The method of claim 7, wherein the iteration of the deformable surface model uses local minimization of an energy function of such surface, said function having as variables the curvature of the fault surface and the coherency in seismic amplitudes, each variable being itself a function of spatial location on the fault surface, and wherein the iteration is terminated when the change in the surface energy relative to the preceding iteration is less than a pre-determined value.

9. The method of claim 8, wherein the energy function is a weighted sum of an internal force and an external force, said internal force being a function of curvature of the fault surface and said external force being a function of seismic amplitude coherency on the fault surface.

10. The method of claim 6 or claim 9, wherein the weighted sum uses a weighting factor λ

for the internal force and a weighting factor (1-λ

) for the external force, λ

being selected to provide substantially equal contributions to the energy function from the internal and external forces, except for noisy data where λ

is chosen substantially equal to 1.

11. The method of claim 1, wherein a fault surface is assumed to be a single-valued function of two orthogonal geographic coordinates, and all voxels in the fault surface are allowed to move only in the third orthogonal direction in the course of iterative deforming of the surface model, said single-valued function being in the form of x=f(y,z) if the angle between a fault surface and the y-z plane is less than 45 degrees, and y=f(x,z) if the angle between a fault surface and the y-z plane is greater than 45 degrees, the coordinate z being in the vertical direction.

12. The method of claim 1, wherein constructing an initial fault surface comprises the steps of:

(a) generating a pre-selected density of fault points along each fault stick by performing linear interpolation between fault nodes;

(b) identifying the extent of the boundaries of the fault surface by using the interpolated fault points from step (a); and

(c) constructing an initial fault surface within said boundaries by using bilinear interpolation of the interpolated fault points from step (a).

13. The method of claim 1, wherein the reconstructed fault surface is defined by a discrete set of fault sticks.

14. The method of claim 1, wherein the seismic data comprise seismic amplitude discontinuity and coherency data, and the fault-indicating parameter is a combination of the seismic amplitude discontinuity and the seismic amplitude coherency.