Inversion-based reflector dip estimation

Patent metadata

Field	Value
Publication number	US-9671512-B2
Application number	US-201414491538-A
Country	US
Kind code	B2
Filing date	Sep 19, 2014
Priority date	Oct 29, 2013
Publication date	Jun 6, 2017
Grant date	Jun 6, 2017

Abstract

Method for estimating reflector dips in a window of post stack image traces ( 51 ) of seismic data for use in velocity tomography ( 57 ). The method iteratively ( 56 ) flattens ( 55 ) the image traces against a specified reference trace through the application of conjugate-gradient least-squares inversion ( 53 ). Different from other dip estimation methods which emphasize on strong-amplitude reflectors, the inventive method automatically inverts for the reflector dip for every grid point in the image window.

First claim

The invention claimed is:

1. An automated method for estimating reflector dips in an image window of post-stack image traces of seismic data, comprising:
flattening the post-stack image traces against a reference trace in the post-stack image traces using reflector dips to adjust depths of image points for the post-stack image traces, wherein the reflector dips are obtained from horizontal and vertical gradients of the image points in the image window of the post-stack image made up of the post-stack image traces through computer-implemented iterative inversion that inverts for the reflector dips of every grid point in the image window;
constructing a model of seismic wave velocity in a subsurface region from the reflector dips;
migrating the seismic data with the model of seismic wave velocity; and
producing an image from the migrated seismic data showing structural features of the subsurface region and prospecting for hydrocarbons according to the structural features of the subsurface region.

2. The method of claim 1 , wherein the inversion is a conjugate-gradient least-squares process.

3. The method of claim 2 , wherein the conjugate-gradient least-squares process comprises:
(a) computing the horizontal and vertical gradients within the image window, at all points on a selected computational grid; and
(b) applying iterative least-squares inversion to the horizontal and vertical gradients to compute the reflector dips.

4. The method of claim 3 , further comprising calculating at least one coherence function at each of said grid points and then using the at least one coherence function as a weighting function in a cost function selected for the conjugate-gradient least-squares process, wherein the at least one coherence function constrains the iterative inversion to make it more stable.

5. The method of claim 3 , wherein the horizontal and vertical gradients are computed using a space-domain finite-difference operator.

6. The method of claim 1 , wherein in each iteration, the reflector dips are first estimated relative to a reference trace corresponding to a selected common depth point in the image window, followed by partly flattening the image traces against the reference trace.

7. The method of claim 1 , wherein each cycle of the iterative inversion comprises flattening the image traces by rotating seismic events using the reflector dips.

8. The method of claim 1 , wherein the post-stack image traces are in depth-CDP domain and are generated by a pre-stack depth imaging scheme.

9. The method of claim 1 , wherein the optimizing comprises updating the reflector dips in each cycle of the iterative inversion.

10. The method of claim 1 , wherein the image traces are flattened by adjusting dip angles of reflectors in the image window.

11. The method of claim 1 wherein the image traces are flattened through a vertical interpolation to alleviate a problem of waveform distortion.

12. The method of claim 1 , wherein high frequency noise is removed by a band-pass filter from the image traces before the iterative inversion.

13. A computer-implemented method for transforming seismic data into a post-stack image showing structural features of a subsurface region, comprising:
building an initial seismic wave velocity model in the subsurface region;
applying a depth imaging algorithm to the seismic data to generate a post-stack image made up of image traces;
estimating reflector dips in an image window of the seismic image traces, the estimating including flattening the image traces against a reference trace in the image traces using reflector dips to adjust depths of image points for the image traces, wherein the reflector dips are obtained from horizontal and vertical gradients of the image points in the image window of the post-stack image made up of the post-stack image traces through computer-implemented iterative inversion that inverts for the reflector dips of every grid point in the image window;
constructing a model of seismic wave velocity in the subsurface region from the reflector dips;
migrating the seismic data with the model of seismic wave velocity and producing an image showing structural features of the subsurface region; and
prospecting for hydrocarbons according to the structural features of the subsurface region.

14. The method of claim 13 , further comprising:
conducting a seismic survey of the subsurface region;
designing a well for the subsurface region in accordance with the structural features of the subsurface region; and
drilling the well and producing hydrocarbons from the subsurface region.

Assignees

• Sun Hongchuan
• Deng Feng
• Marcinkovich Carey M
• Exxonmobil Upstream Res Co

Inventors

• Sun Hongchuan
• Deng Feng
• Marcinkovich Carey M

Classifications

• G01V2210/514

  Post-stack · CPC title

• G01V1/303 Primary

  for determining velocity profiles or travel times · CPC title

• G01V2210/6222

  Velocity; travel time · CPC title

• G01V2210/522

  Dip move-out [DMO] · CPC title