Non-unique seismic lithologic inversion for subterranean modeling

• US 5,706,194 A
• Filed: 06/01/1995
• Issued: 01/06/1998
• Est. Priority Date: 06/01/1995
• Status: Expired due to Term

• Current Assignee: Conocophillips Company (ConocoPhillips)
• Original Assignee: Phillips Petroleum Company (ConocoPhillips)
• Inventors: Neff, Dennis B., Runnestrand, Scott A., Butler, Edgar L.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/457,927
• Time in Patent Office: 950 Days
• Field of Search: 364/421, 364/422, 395/929, 367/71, 367/73, 367/70
• US Class Current: 702/14
• CPC Class Codes: G01V 1/282 Application of seismic models, synthetic seismograms
• CPC Class Codes: G01V 11/00 Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
• CPC Class Codes: G01V 2210/614 Synthetically generated data
• CPC Class Codes: G01V 2210/66 Subsurface modeling

Abstract

A method of modeling subsurface reservoir conditions deals with non-uniqueness of seismic reflection signals with respect to reservoir conditions by proposing and then comparing both seismic and lithologic data for forward model sites to select the most likely model for each site. Forward modeling techniques are used to build a synthetic seismic trace catalog, which includes a range of corresponding pseudo-logs for reservoir characteristics that could reasonably exist within a defined region. Then each synthetic seismic trace in the catalog is compared against every real seismic trace, and a number (e.g., 10 to 50) of synthetic traces is selected for association with each real trace that best match the real trace. Finally, the pseudo-logs corresponding to the number of selected synthetic seismic traces for each forward model site are compared to the pseudo-logs selected for adjacent sites to select the most lithologically coherent pseudo-log at each site for use in generating a model display of reservoir characteristics.

Claims

1. A computer implemented method for converting real seismic traces from a seismic survey composed of reflection signals which typify subterranean discontinuities, and lithologic well log data representative of the structure and texture of subterranean layers, into a model image of subterranean characteristics and structure to be displayed on a display means, said method comprising:

(a) determining an initial model based on said lithologic well log data and said real seismic traces, which are representative of known petrophysical parameters at an initial site;

(b) perturbing said initial model to propose a plurality of new petrophysical parameters that bracket said known petrophysical parameters at said initial site for a plurality of forward model sites, wherein said forward model sites are spaced apart in an area included in said seismic survey, and wherein each forward model site is associated with a single one of said real seismic traces, which is thus designated as a characteristic seismic trace for the corresponding forward model site;

(c) building a catalog of synthetic seismic model traces and corresponding pseudo-logs, and storing said catalog in a memory of said computer, said catalog containing a synthetic seismic model trace and a corresponding pseudo-log for each one of said plurality of new petrophysical parameters;

(d) selecting a set of said synthetic seismic model traces having a corresponding set of said pseudo-logs from said catalog for association with each one of said characteristic seismic traces;

(e) designating a selected one pseudo-log from said set of pseudo-logs associated with each of said characteristic traces, as a prime pseudo-log for association with each one of said characteristic seismic traces selected in step (d) to provide a plurality of prime pseudo-logs, whereby each one of said plurality of prime pseudo-logs is associated with a corresponding one of said forward model sites; and

(f) using said plurality of prime pseudo-logs selected in step (e) for generating said model image of subterranean characteristics.

2. A method in accordance with claim 1, wherein said step (d) of selecting said set of synthetic seismic model traces for association with each site of said characteristic seismic traces comprises:

(a) comparing each synthetic model trace in said catalog to a first characteristic seismic trace corresponding to a first forward model site;

(b) selecting said set of synthetic model traces for said first forward model site as the set that best match said first characteristic seismic trace corresponding to said first forward model site; and

(c) repeating steps (a) and (b) for each remaining characteristic seismic trace.

3. A method in accordance with claim 2, wherein said step of comparing each synthetic seismic model trace in said catalog includes:

establishing a numerical measure of the similarity between said synthetic seismic model traces and said real seismic traces, wherein said numerical measure includes values for the following factors;

i. a crosscorrelation coefficient;

ii. a crosscorrelation lag;

iii. an absolute average difference; and

iv. an RB factor.

4. A method in accordance with claim 1, wherein said step (e) for designating said one pseudo-log as said prime pseudo-log for each site of said plurality of forward model sites comprises:

(a) comparing each pseudo-log in said set of pseudo-logs corresponding to said set of synthetic seismic model traces selected for association with said characteristic trace for a first forward model site, to a plurality of like pseudo-logs selected for association with at least a second forward model site, wherein said second forward model site is adjacent said first said forward model site;

(b) using the comparisons made in step (a) for determining a numerical measure for the similarity between each pseudo-log associated with said first forward model site and said plurality of like pseudo-logs associated with at least said second forward model site;

(c) selecting said prime pseudo-log for describing said first forward model site as the pseudo-log having the highest similarity to said plurality of like pseudo-logs of at least said second forward model site; and

(d) repeating steps (a), (b) and (c) for each remaining forward model site of said plurality of forward model sites.

5. A method in accordance with claim 4, wherein each pseudo-log in said set of pseudo-logs associated with said first forward model site is compared to a plurality of like pseudo-logs for a plurality of adjacent forward model sites, and wherein said numerical measure comprises a numerical difference calculation, said method includes:

selecting said prime pseudo-log from said set of pseudo-logs for as the pseudo-log having the lowest composite numerical difference compared to said plurality of like pseudo-logs of said plurality of adjacent forward model sites for use in generating said model image display.

6. A method in accordance with claim 1, wherein said petrophysical parameters are selected from the group of parameters consisting of:

layer thickness, shaliness, water or hydrocarbon saturation, porosity, pay zone, sonic and density, any derivative of one or more of the aforementioned parameters, and combinations thereof.

7. A method in accordance with claim 1, wherein said lithologic well log data is expressible as a curve representing the data as a function of time or depth.

8. A method in accordance with claim 1, wherein said catalog of synthetic seismic model traces contains from about 1,000 traces to about 30,000 traces.

9. A method in accordance with claim 1, wherein said set of synthetic seismic model traces for each of said characteristic traces includes from about 10 synthetic seismic model traces to about 50 such traces.

10. A method in accordance with claim 3, wherein said crosscorrelation coefficient is determined over a time window of about 30 milliseconds to about 100 milliseconds.

11. A method in accordance with claim 1 additionally comprising:

(a) using a seismic source to inject acoustic signals into the earth;

(b) detecting reflections from said acoustic signals transmitted back to the surface of the earth by subterranean discontinuities;

(c) recording said reflection signals; and

(d) processing said reflection signals to produce said real seismic traces in said seismic survey.

12. Apparatus for convening real seismic traces from a seismic survey composed of reflection signals which typify subterranean discontinuities, and lithologic well log data representative of the structure and texture of subterranean layers, into a model image of subterranean characteristics and structure to be displayed on a display means, said apparatus comprising:

a computer programmed to perform method steps comprising;

(a) determining an initial model based on said lithologic well log data and said real seismic traces, which are representative of known petrophysical parameters at an initial site;

(b) perturbing said initial model to propose a plurality of new petrophysical parameters that bracket said known petrophysical parameters at said initial site for a plurality of forward model sites, wherein said model sites are spaced apart in an area included in said seismic survey, and wherein each forward model site is associated with a single one of said real seismic trace, which is thus designated as a characteristic seismic trace for the corresponding forward model site;

(c) building a catalog of synthetic seismic model traces and corresponding pseudo-logs, and storing said catalog in a memory of said computer, said catalog containing a synthetic seismic model trace and a corresponding pseudo-log for each one of said plurality of new petrophysical parameters;

(d) selecting a set of said synthetic seismic model traces having a corresponding set of said pseudo-logs from said catalog for association with each one of said characteristic seismic traces;

(e) designating a selected one pseudo-log from said set of pseudo-logs associated with each of said characteristic traces, as a prime pseudo-log for association with each one of said characteristic seismic traces selected in step (d) to provide a plurality of prime pseudo-logs, whereby each one of said plurality of prime pseudo-logs is associated with a corresponding one of said forward model sites; and

(f) using said plurality of prime pseudo-logs selected in step (e) for generating said model image of subterranean characteristics.

13. Apparatus in accordance with claim 12, wherein said computer programmed method step (d) of selecting said set of synthetic model traces for association with each of said characteristic seismic traces comprises:

(a) comparing each synthetic model trace in said catalog to a first characteristic seismic trace corresponding to a first forward model site;

(b) selecting said set of synthetic model traces for said first forward model site as the set that best match said first characteristic seismic trace corresponding to said first forward model site; and

(c) repeating steps (a) and (b) for each remaining characteristic seismic trace.

14. Apparatus in accordance with claim 13, wherein said step (a) of comparing each synthetic seismic model trace in said catalog includes:

establishing a numerical measure of the similarity between said synthetic seismic model traces and said real seismic traces, wherein said numerical measure includes values for the following factors;

i. a crosscorrelation coefficient;

ii. a crosscorrelation lag;

iii. an absolute average difference; and

iv. an RB factor.

15. Apparatus in accordance with claim 12, wherein said computer programmed method step (e) for designating said one pseudo-log as said prime pseudo-log for each site of said plurality of forward model sites comprises:

(a) comparing each pseudo-log in said set of pseudo-logs corresponding to said set of synthetic seismic model traces selected for association with said characteristic trace for a first forward model site, to a plurality of like pseudo-logs selected for association with at least a second forward model site, wherein said second forward model site is adjacent said first said forward model site;

(b) using the comparisons made in step (a) for determining a numerical measure for the similarity between each pseudo-log associated with said first forward model site and said plurality of like pseudo-logs associated with at least said second forward model site;

(c) selecting said prime pseudo-log for describing said first forward model site as the pseudo-log having the highest similarity to said plurality of like pseudo-logs of at least said second forward model site; and

(d) repeating steps (a), (b) and (c) for each remaining forward model site of said plurality of forward model sites.

16. Apparatus in accordance with claim 15, wherein each pseudo-log in said set of pseudo-logs associated with said first forward model site is compared to a plurality of like pseudo-logs for a plurality of adjacent forward model sites, and wherein said numerical measure comprises a numerical difference calculation, said computer programmed method includes:

selecting said prime pseudo-log from said set of pseudo-logs for use in generating said model image, as the pseudo-log having the lowest composite difference compared to said plurality of like pseudo-logs of said plurality of adjacent forward sites.

17. Apparatus in accordance with claim 12, additionally comprising:

(a) a seismic source for injecting acoustic signals into the earth;

(b) means for detecting acoustic reflection signals transmitted back to the surface of the earth by subterranean discontinuities;

(c) means for recording said reflection signals; and

(d) means for processing said reflection signals to produce said real seismic traces in said seismic survey.

18. A program storage device, readable by a computer, tangibly embodying a program of instructions executable by said computer to perform method steps for converting real seismic traces from a seismic survey composed of reflection signals which typify subterranean discontinuities, and lithologic well log data representative of the structure and texture of subterranean layers, into a model image of subterranean characteristics and structure to be displayed on a display means, said method steps comprising:

(a) determining an initial model based on said lithologic well log data and said real seismic traces, which are representative of known petrophysical parameters at an initial site;

(b) perturbing said initial model to propose a plurality of new petrophysical parameters that bracket said known petrophysical parameters at said initial site for a plurality of forward model sites, wherein said forward model sites are spaced apart in an area included in said seismic survey, and wherein each forward model site is associated with a single one of said real seismic traces, which is thus designated as a characteristic seismic trace for the corresponding forward model site;

(c) building a catalog of synthetic seismic model traces and corresponding pseudo-logs, and storing said catalog in a memory of said computer, said catalog containing a synthetic seismic model trace and a corresponding pseudo-log for each one of said plurality of new petrophysical parameters;

(d) selecting a set of said synthetic seismic model traces having a corresponding set of said pseudo-logs from said catalog for association with each one of said characteristic seismic traces;

(e) designating a selected one pseudo-log from said set of pseudo-logs associated with each of said characteristic traces, as a prime pseudo-log for association with each one of said characteristic seismic traces selected in step (d) to provide a plurality of prime pseudo-logs, whereby each one of said plurality of prime pseudo-logs is associated with a corresponding one of said forward model sites; and

(f) using said plurality of prime pseudo-logs selected in step (e) for generating said model image of subterranean characteristics.

19. A program storage device in accordance with claim 18, wherein said method step (d) of selecting said set of synthetic model traces for association with each one of said characteristic seismic traces comprises:

(a) comparing each synthetic model trace in said catalog to a first characteristic seismic trace corresponding to a first forward model site;

(b) selecting said set of synthetic model traces for said first forward model site as the set that best match said first characteristic seismic trace corresponding to said first forward model site; and

(c) repeating steps (a) and (b) for each remaining characteristic seismic trace.

20. A program storage device in accordance with claim 19, wherein said step (a) of comparing each synthetic seismic model trace in said catalog includes:

establishing a numerical measure of the similarity between said synthetic seismic model traces and said real seismic traces, wherein said numerical measure includes values for the following factors;

i. a crosscorrelation coefficient;

ii. a crosscorrelation lag;

iii. an absolute average difference; and

iv. an RB factor.

21. A program storage device in accordance with claim 18, wherein said method step (e) for designating said one pseudo-log as said prime pseudo-log for each site of said plurality of forward model sites comprises:

(a) comparing each pseudo-log in said set of pseudo-logs corresponding to said set of synthetic seismic model traces selected for association with said characteristic trace for a first forward model site, to a plurality of like pseudo-logs selected for association with at least a second forward model site, wherein said second forward model site is adjacent said first said forward model site;

(b) using the comparisons made in step (a) for determining a numerical measure for the similarity between each pseudo-log associated with said first forward model site and said plurality of like pseudo-logs associated with at least said second forward model site;

(c) selecting said prime pseudo-log for describing said first forward model site as the pseudo-log having the highest similarity to said plurality of like pseudo-logs of at least said second forward model site; and

(d) repeating steps (a), (b) and (c) for each remaining forward model site of said plurality of forward model sites.

22. A program storage device in accordance with claim 21, wherein each pseudo-log in said set of pseudo-logs associated with said first forward model site is compared to a plurality of like pseudo-logs for a plurality of adjacent forward model sites, and wherein said numerical measure comprises a numerical difference calculation, said method steps embodied on said storage device includes:

selecting said prime pseudo-log from said set of pseudo-logs for use in generating said model image, as the pseudo-log having the lowest composite difference compared to said plurality of like pseudo-logs of said plurality of adjacent forward model sites.