Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Earth science
Reexamination Certificate
2001-06-26
2003-02-04
Lefkowitz, Edward (Department: 2862)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Earth science
C702S016000
Reexamination Certificate
active
06516274
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to the field of geophysical prospecting. More particularly, the invention is a method of identifying structural and stratigraphic discontinuities in a three-dimensional (3-D) seismic data volume containing dipping reflectors.
BACKGROUND OF THE INVENTION
As part of the hydrocarbon exploration and production work process, geoscience interpreters often need to recognize and map subsurface structural features, such as faults, and stratigraphic features, such as channel- or sand-body edges, in three-dimensional seismic data. However, identifying structural and stratigraphic features in 3-D seismic data can be a time consuming, subjective, and difficult process. There is a need to generate, in a computationally efficient matter, a derivative data volume (i.e., a data volume derived from the original seismic data volume), that displays clear, sharply focussed structural and stratigraphic features that can be quickly recognized and exploited in the mapping process.
Several techniques have been used in the oil industry to enhance the interpretation of structural and stratigraphic features in 3-D seismic data. A well-known technique is to transform the original amplitude data into a coherence volume using a series of one-dimensional cross-correlation calculations. For every data sample in a volume, the cross-correlation calculation is performed using a user-defined vertical window with the equivalent portion of an adjacent trace. Typically, the vertical window is the target sample in question, plus 3-7 data samples above and below the target sample, depending upon the frequency of the data. This operation is repeated for all data samples and all traces, all in the same correlation direction. The correlation direction is generally in-line, cross-line, or either diagonal direction. The resulting coherence volume typically contains values normalized between −1 and +1. For adjacent traces that are very similar, the value of the coherence sample will be close to +1, since +1 represents high correlation. This similarity, and hence correlation, is expected for adjacent traces that do not straddle a structural or stratigraphic discontinuity. For adjacent traces that do straddle a discontinuity, lack of similarity is expected. Thus, their coherence value would be closer to 0, since 0 represents no correlation. A coherence value of −1 represents negative correlation, such as high correlation with phase reversal. Alternatively, coherence can be described with the inverse notion of discontinuity, in which low coherence equals high discontinuity and high coherence equals low discontinuity. In either case, this standard technique has limitations, though, because features perpendicular to the single correlation direction are highlighted, while features parallel to the correlation direction are poorly imaged.
Normally the cross-correlation calculations are conducted parallel to time slices. This direction of calculation can create problems if the seismic data volume contains reflections that dip significantly, because a calculation that is conducted parallel to a time slice searches across the dipping reflections. When a cross-correlation calculation searches across dipping reflections, it identifies poor correlation because it is comparing different parts of the seismic wavelet. It may then map low coherence or high discontinuity to the coherence volume, even where dipping reflections are continuous.
Dip-steering reorients the search in a cross-correlation calculation so that it is conducted parallel to dipping seismic reflections. Once dip-steering re-orients the search parallel to dipping reflections, the calculation compares the same parts of the seismic wavelet, and is able to correctly assign high correlation or low discontinuity to continuous reflections. If these reflections are cut by a discontinuity such as a fault or channel margin, this discontinuity is imaged much more clearly because of the contrast to the continuous reflections.
Prospective hydrocarbon reservoirs often have steep dips because they often are located at anticlines or inclined fault blocks. Dip-steering provides better images of discontinuities in these prospects. This enhances the ability to add reserves or make discoveries and produce complexly faulted traps. Easier and more efficient interpretation of complex fault networks should lead to cost reduction and performance improvement.
Bahorich and Farmer received U.S. Pat. No. 5,563,949, “Method of Seismic Signal Processing and Exploration”, issued Oct. 8, 1996. This patent is commonly known as the “coherence cube” patent. Bahorich and Farmer also obtained a continuation of this patent in U.S. Pat. No. 5,838,564, “Apparatus for Seismic Signal Processing and Exploration”, issued Nov. 17, 1998.
Bahorich and Farmer's '949 patent describes a method for converting a fully processed 3-D seismic data volume into a cube of coherence measurements. According to their method, the 3-D data volume is divided into a plurality of horizontal slices, and each horizontal slice is further divided into a plurality of cells, each of which contains portions of at least three seismic data traces. As described in the '949 patent, these at least three traces in each cell comprise a reference trace, an in-line trace, and a cross-line trace. The in-line trace and the cross-line trace are each compared to the reference trace in each cell using a measure of coherency. Then the in-line and cross-line coherency measures are combined to obtain a single value that is representative of the coherence of the three seismic traces for each cell. This process is repeated for every cell, using every trace in the 3-D seismic volume as a reference trace, in order to obtain a 3-D cube of coherence measurements. Bahorich and Farmer's '564 patent describes the corresponding apparatus for carrying out the process of their '949 patent.
Bahorich and Farmer's patented technique combines information from more than one correlation direction at each data sample in the 3-D seismic data volume, thereby highlighting structural and stratigraphic information along multiple azimuths. According to Bahorich and Farmer, in their invention “the concept of cross-correlation is extended to two dimensions by taking the geometric means between the classical one dimensional cross-correlations” (U.S. Pat. No. 5,563,949, column 4, lines 17-20). This technique has limitations, however. Combining information from different correlation directions may effect the image clarity of the structural and stratigraphic features. This decrease in clarity can make it more difficult to extract structural and stratigraphic information in automated mapping processes. In addition, the computational complexity of this procedure is significantly greater than the traditional method using classical one-dimensional cross-correlations. Further, Bahorich and Farmer's '949 and '564 patents do not take into account the presence of reflection dip in the seismic data.
Higgs and Luo received U.S. Pat. No. 5,724,309 “Method for Geophysical Processing and Interpretation Using Instantaneous Phase and Its Derivatives and Their Derivatives”, issued Mar. 3, 1998. Higgs and Luo's '309 patent describes a related technique for interpretation of faults and stratigraphic features. The technique uses instantaneous phase and its spatial derivatives to determine values of spatial frequency, instantaneous frequency, dip magnitude and dip azimuth. These values are plotted to produce a derivative seismic volume that highlights subsurface changes. The main advantage of this technique is its computational speed. However, the instantaneous phase and frequency images tend to be of lower resolution than traditional amplitude-derived cross-correlation images. A similar technique was also published by Hardage et al., 1998, “3-D Instantaneous Frequency used as a Coherency/Continuity Parameter to Interpret Reservoir Compartment Boundaries Across an Area o
Cheng Yao C.
Fairchild Lee H.
Farre John A.
May Steve R.
Bell Keith A.
ExxonMobil Upstream Research Company
Gutierrez Anthony
Lefkowitz Edward
LandOfFree
Method for imaging discontinuities in seismic data using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for imaging discontinuities in seismic data using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for imaging discontinuities in seismic data using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3145245