OPTIMAL SURVEY DESIGN

Abstract

Methods of analyzing and optimizing a seismic survey design are described. Specifically, the sampling quality is analyzed as opposed to the overall quality of the whole survey. This allows for analysis of the impact of the offsets, obstacles, and other aspects of the survey on the sampling quality, which will improve the ability to compress the resulting data and minimize acquisition footprints.

Claims

1. A system comprising: a plurality of seismic receivers disposed in a survey area at a plurality of receiver locations; and a plurality of seismic sources disposed in the survey area at a plurality of source locations, the plurality of receiver locations and the plurality of source locations specified by a seismic survey design minimizing any artifacts identified in a filtered spectrum obtained by applying a frequency-wavenumber filter to a central midpoint space array.

2. The system of claim 1, wherein the artifacts are iteratively corrected in optimizing the seismic survey design.

3. The system of claim 1, wherein the plurality of receiver locations and the plurality of source locations are determined based on a comparison of the filtered spectrum to a second filtered spectrum, the second filtered spectrum being for a second central midpoint space array.

4. The system of claim 1, wherein the seismic survey design is selected from a plurality of seismic survey designs.

5. The system of claim 1, wherein the seismic survey design corresponds to at least a portion of the survey area.

6. The system of claim 1, wherein the central midpoint space array is generated based on responses, offsets, and azimuth relationships summed in a central midpoint space for the plurality of source locations and the plurality of receiver locations

7. A system comprising: a plurality of seismic receivers disposed at a plurality of receiver locations of a seismic survey; and a plurality of seismic sources disposed at a plurality of source locations of the seismic survey, the plurality of receiver locations and the plurality of source locations being determined by: generating a source array based on the plurality of source locations of a seismic survey design for the seismic survey; generating a receiver array based on the plurality of receiver locations of the seismic survey design; generating a filtered source array by applying a frequency-wavenumber filter to the source array; generating a filtered receiver array by applying a frequency-wavenumber filter to the receiver array; generating a combined filtered array by combining the filtered receiver array and the filtered source array; and identifying any artifacts in the combined filtered array, the seismic survey design minimizing the artifacts.

8. The system of claim 7, wherein the artifacts are iteratively corrected.

9. The system of claim 7, wherein the seismic survey design is selected from a plurality of seismic survey designs.

10. The system of claim 7, wherein the seismic survey design corresponds to at least a portion of the survey area.

11. A method comprising: obtaining a seismic survey design for a survey area; generating a central midpoint space array for the seismic survey design; obtaining a filtered spectrum by applying a frequency-wavenumber filter to the central midpoint space array; identifying any artifacts in the filtered spectrum; and optimizing the seismic survey design by determining at least one of one or more receiver locations of one more seismic receivers or one or more source locations of one or more seismic sources, the optimized seismic survey design minimizing the identified artifacts.

12. The method of claim 11, wherein the identified artifacts are iteratively corrected in optimizing the seismic survey.

13. The method of claim 11, further comprising: obtaining a second filtered spectrum for a second central midpoint space array, the at least one of the one or more receiver locations or the one or more source locations being optimized based on a comparison of the filtered spectrum to the second filtered spectrum.

14. The method of claim 11, further comprising: obtaining a plurality of seismic survey designs, the plurality of seismic survey designs including the seismic survey design, the seismic survey design being optimized based on a selection of the seismic survey design from the plurality of seismic survey designs, the at least one of the one or more receiver locations or the one or more source locations being optimized based on the selection.

15. The method of claim 11, wherein the central midpoint space array is generated based on at least one of responses, offsets, or azimuth relationships summed in the central midpoint space for the one or more source locations and the one or more receiver locations.

16. The method of 11, wherein the frequency-wavenumber filter includes a two-dimensional Fourier transform.

17. The method of claim 11, wherein the one or more receiver locations are at least one of perpendicular, non-orthogonal, or parallel relative to each other.

18. The method of claim 11, wherein the one or more source locations are at least one of perpendicular, non-orthogonal, or parallel relative to each other.

19. The method of claim 11, wherein the artifacts are associated with one or more of an edge of the survey area or obstacles.

20. The method of claim 11, wherein the survey area corresponds to one or more oil and gas wells.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0060] FIG. 1. Schematic of method according to the CMP array embodiment.

[0061] FIG. 2. Schematic of method according to the Total Survey array embodiment.

[0062] FIG. 3. Exemplary graphic of step one of the Total Survey array embodiment displaying synthetic data and a grid of shots and receivers before any artifact correction or optimization. The F-K transformed and filtered data is shown on the left in a plan view (top) and cross section view (bottom). The actual layout is shown in the upper right.

[0063] FIG. 4A. Display of the impact of the Total Survey array on the sources only at step 1 before any artifact correction or optimization.

[0064] FIG. 4B. Display of the sources in FIG. 4A after the first pass of cleanup of artifacts.

[0065] FIG. 4C. Display of the sources in FIG. 4B after a second cleanup using the F-K transformation to optimize the spectrum. Compare the impact of rounding the edges of the lake in the center of the survey to the to FIG. 4A in left FK filtered displays.

[0066] FIG. 5A. Display of the receivers at step 1 before any optimization or artifact correction. Note the variability in the FK plan view display in the upper left panel.

[0067] FIG. 5B. Display of the sources in FIG. 5A after the first round of cleanup of artifacts.

[0068] FIG. 5C. Display of the receivers in FIG. 5A after a second (final) round of cleanup using the F-K transformation. Compare the impact of rounding the edges of the lake in the center of the survey to FIG. 5A in left FK filtered displays.

[0069] FIG. 6. Exemplary graphic of the grid of shots and receivers in FIG. 3 after undergoing optimization by the Total Survey method.

DESCRIPTION OF EMBODIMENTS OF THE DISCLOSURE

[0070] The disclosure provides a novel method of analyzing a 3D seismic survey and predicting quality of survey and, optionally, means of improving the quality by adjusting survey design parameters.

[0071] The present methods include any of the following embodiments in any combination(s) of one or more thereof:

[0072] A method of evaluating or optimizing a seismic survey design comprising, determining the location of a plurality of seismic sources and a plurality of receivers geographically in a seismic survey design; summing the responses, offsets, and azimuth relationships for the locations determined in the first step in the central midpoint space (CMP); compiling said summed responses, offsets and azimuth relationships into a CMP array; applying an F-K transform to said CMP array; applying a frequency-wavenumber filter to said transformed CMP array; evaluating the filtered array for artifacts; modifying said survey design to correct said artifacts; and repeating steps a-f until an optimal survey is produced, and applying said optimal seismic survey design to a reservoir.

[0073] A method of creating or optimizing a seismic survey design for a hydrocarbon-containing reservoir, comprising: determining the location of a plurality of seismic sources and a plurality of receivers in one or more proposed seismic survey designs for a reservoir being developed; summing the responses, offsets, and azimuth relationships for the locations determined in step a in the central midpoint space (CMP) for each proposed seismic survey design; compiling said summed responses, offsets and azimuth relationships into a CMP array for proposed seismic survey design; applying a frequency-wavenumber filter to said CMP array for each proposed seismic survey design; comparing the filtered array for artifacts in each proposed seismic survey design; selecting the proposed seismic survey design with the minimal artifacts; and applying said selected seismic survey design to said reservoir.

[0074] A method of evaluating a seismic survey design comprising, determining the location of a plurality of seismic sources and a plurality of receivers geographically in a seismic survey design; inputting the complete set of sources into an array design software to form a sources array; inputting the complete set of receivers into said array design software to form a receivers array; applying an F-K transform to said sources array and said receivers array; applying interactive frequency-wavenumber filters to said sources array and said receivers array; combining filtered sources array and receivers array; evaluating the source array, receiver array and the combined filtered array for artifacts; modifying said survey design to correct said artifacts and repeating steps a-h until an optimal survey is produced; and applying said optimal seismic survey design to a reservoir.

[0075] A non-transitory machine-readable storage medium, which when executed by at least one processor of a computer, performs the steps of any method herein described.

[0076] Any method as herein described, further comprising the step of changing one or more locations of one or more seismic sources or receivers or both to minimize artifacts.

[0077] Any method as herein described, further comprising comparing artifacts for two or more survey designs.

[0078] Any method as herein described, wherein said plurality of seismic sources or said plurality of receivers or both are about perpendicular, or about parallel, or both, e.g., orthogonal, but they can also be non-orthogonal.

[0079] One embodiment of the present disclosure is exemplified with respect to the description below and FIG. 1. However, this is exemplary only of the “CMP method”. The following is intended to be illustrative only, and not unduly limit the scope of the appended claims.

[0080] A schematic of the basic steps taken in the described CMP method is shown in FIG. 1. First, test seismic data 101 is collected for a proposed seismic survey design. The source and receiver locations are combined with the acquisition template to determine the CMP's for each bin 102 and then used to determine the responses, offsets, and azimuth relationships in the test data. These relationships are then summed 103 in the CMP space to form a CMP array 104.

[0081] The CMP array then undergoes transformation using a F-K filter algorithm 105 and the responses summed. The summed responses are then interactively filtered and analyzed as if it were a geophone array using geophone array design software to bring out artifacts and other sampling issues in the data 106. The F-K domain will show the artifacts clearly whereas in the spatial or geographic domain it is more difficult to spot by eye.

[0082] The user can then clean up the artifacts by moving the locations of the source and or receivers geographically to new or better points and thus, improve the quality of the data. Regions containing artifacts are commonly associated with survey edges, obstacles like railroads, lakes and no permit regions or similar real world encumbrances that naturally degrade the preferred sampling of the survey.

[0083] In addition to analyzing a single survey for artifacts, two or more survey designs can be compared to analyze the quality of the different designs. Aspects from each design can then be implemented into the final design. This correction process and then re-collection of the CMP array and retransforming with analysis can be repeated until the survey is optimized. Once the final optimized design is created, data can be collected according to known methods in the art.

[0084] In a second embodiment, shown in FIG. 2, the “Total Survey” method, the approach is similar. Test data is collected and the source and receiver locations for the whole survey are input into a geophone array design software package 202. The survey sources are then F-K transformed and analyzed in F-K space for patterns and sampling artifacts 203. The same approach is used on the receivers 204. Once each subgroup is handled, the two F-K spectrums are combined in the geophone array analysis software 205 and the combined spectrums are analyzed again for anomalies in sampling and inconsistencies in the whole survey 206. These are iteratively corrected and then the process repeated until the whole survey is optimized 207.

[0085] Results from using the Total Survey method are shown in FIGS. 3-6 using exemplary data representing an obstacle encountered during a Barnett 3D seismic survey near Denton, Tex. in 2012. The obstacle was a no permit zone next to a lake and this example recreates how the problem was addressed, while not using the actual survey data. The example data was loaded into the Omni 3D seismic survey design package for this example.

[0086] FIG. 3 displays an example graphic of source and receivers positions, per step 202 of FIG. 2. This organization of display is exemplarily only and a user will be able to modify it for his needs. In this particular layout, the figures are, starting from upper left and moving clockwise, the power spectrum or F-K spectrum in plan view, the sources and receivers locations and weights with the average actual geographic location removed, the derived exemplary wavelet from the convolution of the sources and receivers (middle right), the element weights of the array along the line of analysis (bottom right), the decibel (dB) power spectrum for the composited array (bottom left) and a cross section view of the F-K spectrum (middle left).

[0087] In FIG. 3, we have mapped a grid of shots and receivers laid out with some duplication and some gaps that are caused by no permit region and a lake of the target area in center of the upper right corner of the display. The plot in the upper left corner is the combined signature of source and receiver data. The roughness in this plot is clearly visible. There is a strong grain in both the north-south and east west direction, but that is due to the grid nature. There are also wings and the 45° diagonals caused by the sharp corners caused by the gaps in permit regions.

[0088] In the Total Survey Method, step 203 of FIG. 2, requires a user to first transform the source data and analyze transform space for artifacts. FIG. 4A-C displays the source data before (4A), after a first transformation (4B) and a final, cleaned up source display (4C). In this example (and in the real project) we did not actually move points from the pre-plotted position. What was done instead, was determine through the inventive method, which positions were critical to obtain and we then worked with the land-owner and seismic crew to obtain access to these positions and actually acquire some data in the lake during a dry period when access became available.

[0089] FIG. 4A displays the original source data before any processing. The light colored t-shape in the upper right spectrum is due to an area without sources because of e.g. lack of permits, rivers, lakes, etc. The sharp inner edges are problematic because they act as diffractors of the signal. Thus, the optimization of the design will focus on smoothing these corners. The smoother the corners, the less disruption in signaling and the more cost effectiveness of the acquisition. While most who are skilled at the art recognize that smoother boundaries are probably a better approach then sharp corners, there has not been any easy way prior to the inventive method to parameterize or quantify the improvements or impact of changes in the survey design.

[0090] After processing with the first F-K transformation, the FK spectrum in the upper left corner has significant changes. FIG. 4B appears to be the best we can do with the sources and the area we can scan. Sometimes a user just cannot get the permit for the entire plot of land (or in this case the lake was too deep to source) so there is a hole in the upper right plot. By rounding the edges off of the hole, but not running another transform, we were able to clean up the dark vertical lines internally in the upper left plot to achieve the display in FIG. 4C.

[0091] The next step, step 204, is to analyze just the receivers. The display for the receivers is shown in FIG. 5A-C. In the upper left spectrum of FIG. 5A, the dark lines intersecting in the middle of the spectrum are from the sharp corners caused by the no permit zone.

[0092] After the first round of transformation, shown in FIG. 5B, the intensity of the dark lines have been reduced. FIG. 5C shows the results after a second transformation where the artifacts were further attenuated by working with the seismic crew to obtain some receiver locations in the lake area are effectively rounding the edges of the hole. FIG. 5C is the final receiver cleanup. Again, sometimes a no permit zone or lake or river cannot be fixed and a hole in the upper right spectrum remains. This second transformation made the spectrum as clear as possible and rounded the sharp edges.

[0093] FIG. 6 demonstrates the final combined results from the fixed source and receiver data. There is a strong grain in both the north-south and east west direction, but that is due to the grid nature of the sources and receivers. If we had not shot on cardinal orientation (NS-EW) the grain would be oriented in a different direction. The Total survey method did address the wings and the 45° diagonals caused by the sharp corners shown in FIG. 3 and was able to smooth them out and reduce the disruption to signal.

[0094] This process can be applied over and over to improve and clean up the overall FK spectrum in the upper left corner of FIG. 3-5, until the survey is optimally designed. The final layout of the sources and receivers can the be performed in the field according to the optimized survey design. Because the no permit areas and means to reduce the artifacts are known in advance, both time and money can be saved using the optimized survey design.

[0095] Hardware for implementing the inventive methods may preferably include massively parallel and distributed Linux clusters, which utilize both CPU and GPU architectures. Alternatively, the hardware may use a LINUX OS, XML universal interface run with supercomputing facilities provided by Linux Networx, including the next-generation Clusterworx Advanced cluster management system.

[0096] Another system is the Microsoft Windows 7 Enterprise or Ultimate Edition (64-bit, SP1) with Dual quad-core or hex-core processor, 64 GB RAM memory with Fast rotational speed hard disk (10,000-15,000 rpm) or solid state drive (300 GB) with NVIDIA Quadro K5000 graphics card and multiple high resolution monitors.

[0097] Slower systems could be used but are less preferred since seismic data processing may already compute intensive.

[0098] The results may be displayed in any suitable manner, including printouts, holographic projections, display on a monitor and the like. Alternatively, the results may be recorded to memory for use with other programs, e.g., reservoir modeling, and the like.

[0099] The following references are incorporated by reference in their entirety: U.S. Pat. No. 7,660,674.