Automated seismic fault detection and picking

• US 6,018,498 A
• Filed: 09/02/1998
• Issued: 01/25/2000
• Est. Priority Date: 09/02/1998
• Status: Expired due to Term

• Current Assignee: Conocophillips Company (ConocoPhillips)
• Original Assignee: Phillips Petroleum Company (ConocoPhillips)
• Inventors: Lucas, William Allen, Grismore, John Richard, Neff, Dennis B.
• Primary Examiner(s): Oda, Christine K.
• Assistant Examiner(s): JOLLY, ANTHONY S
• Application Number:
  US09/145,811
• Time in Patent Office: 510 Days
• Field of Search: 367/15, 367/16, 367/106, 367/72, 367/38, 367/73, 367/68, 340/15.5
• US Class Current: 367/72
• CPC Class Codes: G01V 1/288 Event detection in seismic signals, e.g. microseismics G01V1/36 takes precedence
• CPC Class Codes: G01V 2210/65 Source localisation, e.g. faults, hypocenters or reservoirs

Abstract

A computer implemented method and apparatus is disclosed for automatically picking faults in a recorded 3D seismic trace data volume. The method employs test planes, which are mathematically inserted into the seismic data volume to approximate dip and azimuth of a potential fault plane surface. A large number of data points, which are selected points on the seismic traces, are defined within the seismic volume, such that each test plane positioned in the seismic volume contains data points corresponding to at least a significant portion of a trace. The method then determines a factor for each data point which is representative of the probability that the data point resides on a fault plane. This probability is based on planar discontinuity and average amplitude difference between corresponding traces in adjacent parallel test planes, and the method selects locations, in an x, y grid, of a strip of locations having high probability of residing on a fault surface. The strip of the selected locations is smoothed to a line and used to construct fault lineaments displays in seismic sections or time slices. The fault lineaments are stored in a computer data file, and conventional, stratigraphically enhanced, or other seismic data enhanced for seismic attributes is merged with the fault lineament files to create consolidated displays to aid interpretation of the data volume.

Claims

1. A computer implemented method for automatically determining a fault lineament detectable in a three-dimensional (3D) seismic data volume containing a plurality of seismic traces arranged as a plurality of data points having x, y and z coordinates defining a physical location in said seismic data volume, wherein said data points are selected individual points on said seismic traces, said method comprising:

a) converting said 3D seismic data volume to a fault plane (FP) data volume containing a plurality of probability factors, referred to herein as FP values, wherein an FP value and coordinates for said physical location for said FP values are associated with each of said plurality of data points in said 3D seismic data volume;

b) determining at least one strip of said physical locations in said FP data volume having a high probability of residing on a subterranean fault plane detectable in said seismic data volume;

c) converting said at least one strip of physical locations having a high probability to a line; and

d) displaying said line as a fault lineament.

2. A method in accordance with claim 1, wherein said step of converting said seismic data volume to said FP data volume comprises:

a) defining a plurality of parallel test planes of selectable area positionable throughout said 3D seismic data volume, wherein each of said plurality of test planes contains a number of said data points corresponding to a significant portion of at least one of said plurality of seismic trace;

b) determining an FP value and an optimal azimuth for at least one of said plurality of test planes, herein designated as an evaluation plane, wherein said FP value is based on cross correlation and average amplitude difference between corresponding traces in at least two of said parallel test planes;

c) wherein said evaluation plane is associated with one of said data points located at the center of said evaluation plane, herein designated as an evaluation point;

d) assigning said FP value and said azimuth value for said evaluation plane to said evaluation data point; and

e) repeating steps a), b), c) and d) for each of said plurality of data points in said seismic data volume, thereby forming said fault plane data volume.

3. A method in accordance with claim 2, wherein each of said FP values is determined with said plurality of test planes oriented in dip and azimuth substantially matching the dip and azimuth of said subterranean fault plane detectable in said seismic data volume.

4. A method in accordance with claim 3, wherein said substantially matching azimuth of said plurality of test planes, herein designated as an optimal azimuth, is determined according to the following steps:

orienting said plurality of test planes vertically in a plurality of different azimuths ranging from seismic in-line to cross-line positions;

determining said FP value of said evaluation plane at each of said plurality of different azimuths; and

selecting said optimal azimuth as the azimuth orientation producing the maximum FP value determined for said plurality of different azimuths.

5. A method in accordance with claim 3, wherein said substantially matching dip of said plurality of test planes is determined according to the following steps:

orienting said plurality of test planes at said optimal azimuth in a plurality of different dips ranging from about +45 degrees to about -45 degrees;

determining said FP value of said evaluation point at each of said plurality of different dips; and

selecting said substantially matching dip as the dip orientation producing the maximum FP value for said plurality of different dips.

6. A method in accordance with claim 2, wherein said step of determining said FP value comprises:

computing a crosscorrelation value between corresponding traces in at least two adjacent parallel test planes;

computing an average amplitude difference value between corresponding traces in said at least two adjacent parallel test planes; and

averaging said crosscorrelation value and said average amplitude difference value to determine said FP value.

7. A method in accordance with claim 1, wherein said step of determining at least one strip of locations in said FP data volume having a high probability of residing on said subterranean fault plane, herein referred to as a probability strip, comprises:

selecting FP values in said FP data volume that are above a threshold value to define a plurality of selected FP values having a high probability of residing on said subterranean fault plane;

storing said plurality of selected FP values and associated physical locations in a threshold data file;

sorting said threshold data file into a plurality of time or depth slices, wherein each of said time or depth slices contains at least two of said selected FP values having a high probability of residing on said subterranean fault plane;

building one or more of said probability strips in each of said plurality of time slices, wherein each of said probability strips is a generally linear series of locations of said selected FP values mapped onto an x, y grid; and

wherein said probability strips are constrained by criteria for minimum strip length, maximum strip length and a correlation coefficient.

8. A method in accordance with claim 1, wherein said high probability is a probability greater than one-half.

9. A method in accordance with claim 7, wherein said step of converting at least one strip of said physical locations having a high probability of residing on a fault plane, to a line comprises:

relating said probability strips found in vertically adjacent time slices according to an overlap criteria, wherein said probability strips having at least a minimum overlap of locations are combined;

averaging coordinates of said physical location and azimuth of said probability strips to define a line, wherein the location of a point on said line is the average location of at least a minimum number of said coordinates of physical locations on said probability strips; and

storing the location of said lines in a line indexed file.

10. A method in accordance with claim 9, wherein a fault model file is created, said method comprising:

grouping related lines in said line indexed file into regions defined by a specified minimum fault depth, wherein groups containing a plurality of lines related in azimuth and fault depth, herein designated as line groups, are constructed;

constructing a histogram of azimuths of said line groups;

extracting lines from said line indexed file having an azimuth substantially matching a dominant azimuth in said histogram;

creating a candidate fault surface having the longest line extracted from said line indexed file in the previous step horizontally positioned at the center of said candidate surface;

determining the proximity of remaining sublines extracted from said line indexed file to said candidate surface, and rejecting sublines outside a specified maximum distance;

adjusting said candidate surface horizontally and in dip to maximize the number of sublines within said specified maximum distance to define a modified candidate surface;

filling in gap in said lines that are less than a specified lateral gap threshold; and

writing said modified candidate surface to a fault model file.

11. A method in accordance with claim 4, wherein structural dip is determined and restrained prior to executing said step of selecting said optimal azimuth.

12. Apparatus for automatically determining a fault lineament detectable in a three dimensional (3D) seismic data volume containing a plurality of seismic traces arranged as a plurality of data points having x, y and z coordinates defining a physical location in said seismic data volume, wherein said data points are selected individual points on said seismic traces, said apparatus comprising:

a computer programmed to carry out the following method steps;

a) converting said 3D seismic data volume to a fault plane (FP) data volume containing a plurality of probability factors, referred to herein as FP values, wherein an FP value and coordinates for said physical location of said FP values are associated with each of said plurality of data points in said seismic data volume;

b) determining at least one strip of said physical locations in said FP data volume having a high probability of residing on a subterranean fault plane detectable in said seismic data volume;

c) converting said at least one strip of physical locations having a high probability to a line; and

d) displaying said line as a fault lineament.

13. Apparatus in accordance with claim 12, wherein said computer is additionally programmed to carry out the following steps for converting said seismic data volume to said FP data volume, said method steps comprising:

a) defining a plurality of parallel test planes of selectable area positionable throughout said 3D seismic data volume, wherein each of said plurality of test planes contains a number of said data points corresponding to a significant portion of at least one of said plurality of seismic trace;

b) determining an FP value and an optimal azimuth for at least one of said plurality of test planes, herein designated as an evaluation plane, wherein said FP value is based on cross correlation and average amplitude difference between corresponding traces in at least two of said parallel test planes;

c) wherein said evaluation plane is associated with one of said data points located at the center of said evaluation plane, herein designated as an evaluation point;

d) assigning said FP value and said azimuth value for said evaluation plane to said evaluation data point; and

e) repeating steps a), b), c) and d) for each of said plurality of data points in said seismic data volume, thereby forming said fault plane data volume.

14. Apparatus in accordance with claim 12, wherein said computer is additionally programmed to carry out the following steps for converting at least one strip of said physical locations having a high probability of residing on a fault plane to a line, said method steps comprising:

selecting FP values in said FP data volume that are above a threshold value to define a plurality of selected FP values having a high probability of residing on said subterranean fault plane;

storing said plurality of selected FP values and associated physical locations in a threshold data file;

sorting said threshold data file into a plurality of time or depth slices, wherein each of said time or depth slices contains at least two of said selected FP values having a high probability of residing on said subterranean fault plane;

building one or more of said probability strips in each of said plurality of time slices, wherein each of said probability strips is a generally linear series of locations of said selected FP values mapped onto an x, y grid; and

wherein said probability strips are constrained by criteria for minimum strip length, maximum strip length and a correlation coefficient.

15. Apparatus in accordance with claim 12, wherein said computer is additionally programmed to carry out the following steps for converting at least one strip of said physical locations having a high probability of residing on a fault plane to a line, said method steps comprising:

relating said probability strips found in vertically adjacent time slices according to an overlap criteria, wherein said probability strips having at least a minimum overlap of locations are combined;

averaging coordinates of said physical location and azimuth of said probability strips to define a line, wherein the location of a point on said line is the average location of at least a minimum number of said coordinates of physical locations on said probability strips; and

storing the location of said lines in a line indexed file.

16. A program storage device, readable by a computer, tangibly embodying a program of instructions executable by said computer for determining a fault lineament detectable in a three-dimensional (3D) data volume, said seismic data volume containing a plurality of seismic traces arranged as a plurality of data points having x, y and z coordinates defining a physical location in said seismic data volume, wherein said data points are selected individual points on said seismic traces, said method comprising:

a) converting said 3D seismic data volume to a fault plane (FP) data volume containing a plurality of probability factors, referred to herein as FP values, wherein an FP value and coordinates for said physical location for said FP values are associated with each of said plurality of data points in said 3D seismic data volume;

b) determining at least one strip of said physical locations in said FP data volume having a high probability of residing on a subterranean fault plane detectable in said seismic data volume;

c) converting said at least one strip of physical locations having a high probability to a line; and

d) displaying said line as a fault lineament.