Method for optimal wave field separation

• US 7,676,327 B2
• Filed: 04/26/2007
• Issued: 03/09/2010
• Est. Priority Date: 04/26/2007
• Status: Expired due to Fees

• Current Assignee: WesternGeco LLC
• Original Assignee: WesternGeco LLC
• Inventors: Ozbek, Ali, Ozdemir, Ahmet Kemal
• Primary Examiner(s): Cosimano; Edward R
• Assistant Examiner(s): Washburn; Douglas N
• Application Number:
  US11/740,641
• Publication Number:
  US 20080275649A1
• Time in Patent Office: 1,048 Days
• Field of Search: 367/21, 367/14, 367/15, 367/24, 367/124, 702/13, 702/14, 702/17
• US Class Current: 702/17
• CPC Class Codes: G01V 1/28 Processing seismic data, e.g. for interpretation or for event detection G01V1/48 takes precedence
• CPC Class Codes: G01V 1/364 Seismic filtering G01V1/37 takes precedence
• CPC Class Codes: G01V 2210/34 Noise estimation quality control G01V2200/14
• CPC Class Codes: G01V 2210/56 De-ghosting; Reverberation compensation

Abstract

A technique for decomposing a recorded wave field represented in a set of multicomponent, marine seismic data centers around a computer-implemented method including: estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in a seismic data set acquired in a marine survey; modeling the physical propagation of a recorded wave field represented in the seismic data set from the estimated statistics; and estimating a directional component of the recorded wave field from the physical propagation model that minimizes error relative to the seismic measurements.

Claims

1. A computer-implemented method, comprising:

estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in a seismic data set acquired in a marine survey;

modeling the physical propagation of a recorded wave field represented in the seismic data set from the estimated statistics; and

estimating a directional component of the recorded wave field from the physical propagation model that minimizes error relative to the seismic measurements.

2. The computer-implemented method of claim 1, wherein the directional component is an upgoing component.

3. The computer-implemented method of claim 1, wherein the directional component is a downgoing component.

4. The computer-implemented method of claim 1, wherein the estimated statistics include estimated correlations.

5. The computer-implemented method of claim 1, wherein the seismic measurements include pressure and particle displacement.

6. The computer-implemented method of claim 1, wherein the seismic measurements include pressure measurements at at least two different depths.

7. The computer-implemented method of claim 1, wherein estimating the statistics of the noise terms includes estimating second order statistics of noise terms on different measurements.

8. The computer-implemented method of claim 7, wherein the second order statistics include correlations and covariances.

9. The computer-implemented method of claim 7, wherein the second order statistics comprise a rough estimate of the statistics of the noise terms.

10. The computer-implemented method of claim 1, wherein the contribution of each measurement to the model is a function of correlations between different measurements and the physical propagation model.

11. The computer-implemented method of claim 1, further comprising estimating a further wave field quantity.

12. The computer-implemented method of claim 11, wherein the further wave field quantity comprises a propagation characteristic.

13. The computer-implemented method of claim 11, wherein the further wave field quantity comprises down-going pressure, up-going particle displacement and down-going particle displacement.

14. The computer-implemented method of claim 1, wherein the group of seismic measurements comprises a subset of a larger group of seismic measurements.

15. The computer-implemented method of claim 14, wherein estimating the directional component includes approximating a full three-dimensional solution with two-dimensional data.

16. The computer-implemented method of claim 15, wherein estimating the directional component includes performing a zero crossline wavenumber approximation.

17. The computer-implemented method of claim 16, wherein the two-dimensional data comprises frequency and wavenumber.

18. The computer-implemented method of claim 14, wherein estimating the directional component includes approximating a full three-dimensional solution with one-dimensional data.

19. The computer-implemented method of claim 18, wherein the one-dimensional data comprises frequency.

20. A computing apparatus, comprising:

a processor;

a bus system;

a storage communicating with the processor over the bus system; and

an application residing on the storage that, when executed by the processor, performs a method, comprising;

estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in a seismic data set acquired in a marine survey;

modeling the physical propagation of a recorded wave field represented in the seismic data set from the estimated statistics; and

estimating a directional component of the recorded wave field from the physical propagation model that minimizes error relative to the seismic measurements.

21. The computing apparatus of claim 20, wherein the estimated statistics include estimated correlations.

22. The computing apparatus of claim 20, wherein estimating the statistics of the noise terms in the method includes estimating second order statistics of noise terms on different measurements.

23. The computing apparatus of claim 20, wherein the contribution of each measurement to the model is a function of correlations between different measurements and the physical propagation model.

24. The computing apparatus of claim 20, wherein the group of seismic measurements comprises a subset of a larger group of seismic measurements.

25. The computing apparatus of claim 20, wherein the seismic data set resides on the storage.

26. A computer-implemented method, comprising:

modeling the physical propagation of a recorded wave field represented in a seismic data set acquired in a marine survey; and

determining an optimal estimation of a directional component of the recorded wave field from the physical propagation model.

27. The computer-implemented method of claim 26, wherein determining the optimal estimation includes:

estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in a seismic data set acquired in a marine survey;

modeling the physical propagation of a recorded wave field represented in the seismic data set from the estimated statistics; and

estimating a directional component of the recorded wave field from the physical propagation model that minimizes error relative to the seismic measurements.

28. The computer-implemented method of claim 27, wherein the estimated statistics include estimated correlations.

29. The computer-implemented method of claim 27, wherein estimating the statistics of the noise terms includes estimating second order statistics of noise terms on different measurements.

30. The computer-implemented method of claim 27, wherein the contribution of each measurement to the model is a function of correlations between different measurements and the physical propagation model.

31. The computer-implemented method of claim 27, wherein estimating the directional component includes approximating a full three-dimensional solution with two-dimensional data or with one-dimensional data.

32. A computer-implemented method, comprising:

modeling the physical propagation of a recorded wave field represented in a seismic data set acquired in a towed-array marine survey, the model being derived from each one of a predetermined group of seismic measurements in the seismic data set; and

identifying an estimation of a directional component of the recorded wave field from the physical propagation model that minimizes the fit error between the seismic measurements and the estimate derived from the seismic measurements.

33. The computer-implemented method of claim 32, wherein identifying the estimation includes:

estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in a seismic data set acquired in a marine survey;

modeling the physical propagation of a recorded wave field represented in the seismic data set from the estimated statistics; and

estimating a directional component of the recorded wave field from the physical propagation model that minimizes error relative to the seismic measurements.

34. The computer-implemented method of claim 33, wherein estimating the statistics of the noise terms includes estimating second order statistics of noise terms on different measurements.

35. The computer-implemented method of claim 33, wherein the contribution of each measurement to the model is a function of correlations between different measurements and the physical propagation model.

36. The computer-implemented method of claim 33, wherein estimating the directional component includes approximating a full three-dimensional solution with two-dimensional data or with one-dimensional data.

37. A computer-implemented method, comprising:

identifying a downgoing component of a recorded wavefield represented by a set of seismic data set acquired in a marine survey, including;

estimating the statistics of the noise terms in each one of a predetermined group of seismic measurements in the seismic data;

modeling the physical propagation of the recorded wave field from the estimated statistics; and

estimating the downgoing component from the physical propagation model that minimizes error relative to the seismic measurements; and

deghosting the seismic data using the estimated downgoing component.

38. The computer-implemented method of claim 37, wherein estimating the statistics of the noise terms includes estimating second order statistics of noise terms on different measurements.

39. The computer-implemented method of claim 37, wherein the contribution of each measurement to the model is a function of correlations between different measurements and the physical propagation model.

40. The computer-implemented method of claim 37, wherein estimating the directional component includes approximating a full three-dimensional solution with two-dimensional data or with one-dimensional data.