Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02325874 2000-07-28
WO 99/41623 PC"T/US99/02711
-1-
~~,1 tiu~ r oR NHAN IN l~~rrr n s T ~
This invention relates generally to the field of seismic prospecting and, more
particularly, to seismic data processing and interpretation. Specifically, the
invention
is a method for enhancing seismic data so that subtle geologic features are
easier to
identify and interpret.
BACKGROUND OF TH 1NVFNT1(1N
In the oil and gas industry, seismic prospecting techniques are commonly used
to aid in the search for and evaluation of subterranean hydrocarbon deposits.
In
seismic prospecting, a seismic source is used to generate a physical impulse
known as
a "seismic signal" that propagates into the earth and is at least partially
reflected by
subsurface seismic reflectors (i.e., interfaces between underground formations
having
different acoustic impedances). The reflected signals (known as "seismic
reflections")
are detected and recorded by seismic receivers located at or near the surface
of the
earth, in an overlying body of water, or at known depths in boreholes, and the
resulting seismic data may be processed to yield information relating to the
subsurface
formations.
Seismic prospecting consists of three separate stages: data acquisition, data
processing, and data interpretation. The success of a seismic prospecting
operation
depends on satisfactory completion of all three stages.
The seismic energy recorded by each seismic receiver during the data
acquisition stage is known as a "seismic data trace." During the data
processing stage,
the raw seismic data traces are refined and enhanced so as to facilitate the
data
interpretation stage. For example, one common method for enhancing seismic
data
traces is through the common-midpoint (CMP) stacking process. As will be well
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
-2-
known to persons skilled in the art, the "midpoint" for a seismic data trace
is the point
midway between the source location and the receiver location for that trace.
According to the CMP method, the recorded seismic data traces are sorted into
common-midpoint gathers each of which contains a number of different seismic
data
traces having the same midpoint but different source-to-receiver offset
distances. The
seismic data traces within each CMP gather are corrected for statics (i.e.,
the effects of
variations in elevation, weathered layer thickness and/or velocity, and
reference
datum) and normal moveout (i.e., the variation of traveltime with respect to
source-to-receiver offset) and are then summed or "stacked" to yield a stacked
data
trace which is a composite of the individual seismic data traces in the CMP
gather.
Typically, the stacked data trace has a significantly improved signal-to-noise
ratio
compared to that of the unstacked seismic data traces in the CMP gather.
Stacked data traces for a series of CMP locations falling along a particular
survey line may be displayed side-by-side to form a stacked seismic section
which
simulates a zero-offset seismic section (i.e., a seismic section where every
trace is the
result of a coincident source and receiver). Thus, a stacked seismic section
is a
representation, in two-way seismic signal traveltime, of a vertical cross-
section of the
earth below the survey line in question. Stacked seismic sections are used in
the data
interpretation stage to predict subsurface structure and stratigraphy.
Typically, the seismic data traces recorded during the data acquisition stage
are "minimum-phase," or nearly so. In other words, at the instant that a
seismic signal
reaches a subsurface reflector, a reflected signal begins to form. As the
downgoing
seismic signal rises in strength, the upgoing reflected signal also rises in
strength.
Similarly, as the downgoing seismic signal begins to decline in strength, the
upgoing
reflected signal also begins to decline. The result of this process is that in
a
conventional stacked seismic section, each subsurface reflector is marked by
the
leading edge of a seismic pulse or "wavelet."
Because a seismic data trace represents a convolution of many overlapping
reflections, it is often difficult to clearly identify the leading edge of a
seismic
wavelet. It would facilitate interpretation of seismic data if the subsurface
reflectors
CA 02325874 2000-07-28
WO 99!41623 PCT/US99/02711
-3-
were marked by peaks or troughs in the data rather than by a rising or falling
edge of a
seismic wavelet because peaks and troughs are easier to identify. A procedure
known
as "zero-phase processing" is commonly used in the industry to accomplish this
result.
In zero-phase processing, the minimum-phase seismic wavelet embedded in the
seismic data is converted to a zero-phase wavelet. Zero-phase wavelets are
symmetrical, and the time scale is shifted so that the center of the wavelet
indicates
the arrival time. In other words, the center of a zero-phase wavelet coincides
with the
subsurface seismic reflector that caused the reflection. The conversion to
zero-phase
is preferably performed on the individual seismic data traces within a CMP
gather
prior to stacking; however, the conversion may also be performed after
stacking has
occurred. See, e.g., Sheriff, R. E. and Geldart, L. P., Exploration
Seismology,
Volume l: History, theory, 8~ data acquisition and Volume 2: Data processing
and
interpretation, sections 4.3.4, 8.1.4, and 10.6.6d, Cambridge University
Press, 1982.
The result of this process is a zero-phase seismic section in which the
subsurface
reflectors generally are marked by peaks and/or troughs in the stacked zero-
phase data
traces.
Another advantage of zero-phase processing is that the resulting zero-phase
seismic data traces typically have better seismic resolution (i.e., the
ability to
distinguish two reflectors which are close together) than the seismic data
traces
recorded during the data acquisition stage. See Schoenberger, M., "Resolution
comparison of minimum-phase and zero-phase signals," Geophysics, Vol. 39, No.
6,
pp. 826-833, December 1974. Accordingly, converting the recorded seismic data
traces to zero-phase data traces permits identification and interpretation of
shorter
geologic intervals than is possible with conventional seismic data processing.
As is well known in the art, the seismic data resulting from zero-phase
processing may deviate from true zero-phase by as much as 30 degrees.
Nevertheless,
this "near-zero-phase" seismic data is generally considered to be
substantially
equivalent to true zero-phase seismic data. Accordingly, as used herein and in
the
claims, "zero-phase" will be deemed to include both true zero-phase and
near-zero-phase seismic data.
CA 02325874 2000-07-28
WO 99/41623 PCTNS99/02711
Some seismic data processing operations utilize another type of data trace
known as a "quadrature" trace. For example, quadrature traces are used to
determine
instantaneous seismic attributes for use in seismic attribute analysis. As is
well
known in the art, a quadrature trace is a 90-degree phase-shifted version of
the
recorded minimum-phase seismic data trace. It is obtained by taking the
Hilbert
transform of the recorded trace.
The "quadrature" concept may be extended to other types of processed data
traces. For example, a 90-degree phase shift may be applied to a zero-phase
(or
near-zero-phase) data trace to yield a "zero-phase quadrature" trace.
Techniques for
IO applying the 90-degree phase shift to a zero-phase data trace are well
known to
persons skilled in the art.
The zero-phase, quadrature, and zero-phase quadrature transformations
described above may be used to facilitate many seismic data processing
operations.
However, these transformations do not resolve all seismic data processing and
interpretation problems. For example, seismic resolution may be a problem for
thin
geologic features. Many subsurface geologic features of interest to the
petroleum
industry are from about five to about 50 feet in thickness. The cycle of a
seismic
pulse is typically sinusoidal and from about 80 to about 800 feet in length.
Because a
cycle consists of both a positive phase and a negative phase, the approximate
resolution of a typical seismic pulse is from about 40 to about 400 feet. A
seismic
reflection is generated each time the seismic pulse encounters an impedance
boundary. When the impedance boundaries are closer together than the
resolution of
the seismic pulse, the seismic reflections overlap, as noted above. Thus, the
presence
of an impedance boundary of interest may appear as only a small anomaly on the
seismic data trace, such as a subdued peak or a departure from sinusoidal
(i.e., a bend
or kink in the data). Failure to identify and interpret these anomalies can
result in
erroneous conclusions regarding the subsurface stratigraphy.
Thus, there is a need for a method for enhancing seismic data to make subtle
geologic features more easily identifiable. Such a method should be applicable
to all
types of seismic data traces, including but not limited to minimum-phase, zero-
phase,
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/OZ711
-5-
quadrature, and zero-phase quadrature data traces. Such a method also should
permit
the identification and interpretation of geological features marked only by an
anomaly
in the data. The present invention satisfies this need.
ARY OF THE 1NVFNTIfIN
In one embodiment, the present invention is a method for enhancing a seismic
data trace comprising the steps of (i) locating all peaks and troughs on at
least a
portion of the seismic data trace; and (ii) enhancing the amplitude values of
the peaks
and troughs. This enhancement can be accomplished in a variety of ways. For
example, the amplitudes of all peaks may be adjusted to equal a first
arbitrarily
selected constant amplitude, and the amplitudes of all troughs may be adjusted
to
equal a second arbitrarily selected constant, which may be equal to or
different from
the first constant. Alternatively, the amplitudes of all peaks may be made
equal to the
largest peak amplitude on the data trace, and the amplitudes of all troughs
may be
made equal to the largest trough amplitude on the data trace. In another
alternative,
minimum amplitudes are specified for the peaks and troughs, and all peaks and
troughs having amplitudes less than the specified minimum are enhanced to the
specified minimum values. The inventive method may be applied to any type of
seismic data trace including without limitation minimum-phase seismic data
traces,
zero-phase seismic data traces, quadrature traces, and zero-phase quadrature
traces.
In another embodiment of the invention, the seismic data traces are converted
to curvature traces. The curvature traces may be generated by taking the
second
derivative with respect to time of the data traces, provided that the data
traces have
been converted to continuous interpolated data traces defined at every time
point.
Alternatively, and perhaps preferably, the curvature traces can be
approximated by
calculating the negative second difference of the discretized digital data at
each
sample point and plotting the result.
CA 02325874 2000-07-28
WO 99/41623 PCTNS99/02711
-6-
BRIEF D ~.~ ~pTION OF THE nuAwlN~c
The present invention and its advantages will be better understood by
referring
to the following detailed description and the attached drawings in which:
FIG. lA illustrates a hypothetical zero-phase seismic data trace, and FIG. 1B
illustrates the same seismic data trace after application of a first
embodiment of the
present invention;
FIG. 2 is a zero-phase seismic section for a particular survey line;
FIG. 3 is the same seismic section as shown in FIG. 2 after enhancement of the
peaks and troughs according to the present invention;
FIG. 4 illustrates the negative second difference concept;
FIG S illustrates a zero-phase seismic data trace and its corresponding
curvature trace;
FIG. 6 illustrates another zero-phase seismic data trace and its corresponding
curvature trace;
FIGS. 7A and 7B illustrate a series of zero-phase stacked data traces for a
particular survey line and the corresponding curvature traces;
FIGS. 8A and 8B illustrate a zero-phase stacked seismic section and its
corresponding curvature section; and
FIG. 9 illustrates an impedance curve derived from well log data, a zero-phase
quadrature trace obtained from a location near the well, and a second
derivative of the
zero-phase quadrature trace obtained according to the present invention.
The invention will be described in connection with its preferred embodiments.
However, to the extent that the following detailed description is specific to
a particular
embodiment or a particular use of the invention, this is intended to be
illustrative only,
and is not to be construed as limiting the scope of the invention. On the
contrary, it is
intended to cover all alternatives, modifications, and equivalents which are
included
within the spirit and scope of the invention, as defined by the appended
claims.
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
_'7_
DETAILED DFS IpTION OF THE P FFF~tRFn ~~~D~.r=w~r~-~TTc
The present invention is a method for enhancing seismic data to make subtle
geologic features easier to identify and interpret. Preferably, the inventive
method is
applied to the individual stacked seismic data traces of a seismic section.
However,
the method may also be used to enhance seismic data traces prior to stacking.
As will be readily apparent to persons skilled in the art, the method of the
present invention is preferably implemented using a suitably programmed
digital
computer. Persons skilled in the art could easily develop computer software
for
practicing the inventive method based on the teachings set forth herein.
As noted above, the inventive method may be used to enhance any type of
seismic data trace. For purposes of illustrating the invention, without
thereby limiting
the scope of the invention, the following detailed description will be
directed toward
use of the method to enhance zero-phase seismic data traces. Further, the
following
description will be based on implementation of the invention in the time
domain.
However, persons skilled in the art will understand that the invention may
also be
used to enhance other types of seismic data (such as minimum-phase,
quadrature, or
zero-phase quadrature data) and may be implemented in other data domains, such
as
the frequency domain, without departing from the true spirit and scope of the
invention.
In a first embodiment, the invention comprises directly enhancing the
amplitudes of the peaks and troughs on zero-phase data traces so that the
peaks and
troughs are easier to identify. As noted above, this enhancement may be
applied to
either prestack or poststack data traces.
This embodiment of the invention is illustrated in FIGS. lA and 1B. FIG. lA
illustrates a hypothetical zero-phase seismic data trace 10 (either prestack
or
poststack) having four peaks 12a - 12d and four troughs 14a - 14d. The
amplitudes of
peaks 12a and 12b are quite small compared to those of peaks 12c and 12d.
Similarly,
the amplitudes of troughs 14b and 14d are small compared to those of troughs
14a and
14c. FIG. 1B illustrates the same data trace 10' after application of the
present
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
_g_
invention. The amplitude of each of the peaks 12a' - 12d' has been adjusted to
equal a
constant value x, and the amplitude of each of the troughs 14a' - 14d' has
been
adjusted to equal a constant value y, which may or may not be equal to x.
Thus, peaks
and troughs in the hypothetical trace 10 (FIG. lA) having small amplitudes
have been
enhanced to permit easy identif cation.
In an alternate embodiment (not illustrated), the value of x is equal to the
maximum peak amplitude of the unenhanced peaks 12a - 12d (FIG. lA) and the
value
of y is equal to the maximum trough amplitude of the unenhanced troughs 14a -
14d
(FIG. lA). In other words, the amplitude of peaks 12a, 12b, and 12c would be
enhanced to be equal to the amplitude of peak 12d, and the amplitude of
troughs 14b,
14c, and 14d would be enhanced to be equal to the amplitude of trough 14a. In
another alternate embodiment (not illustrated), minimum amplitude values for
the
peaks and troughs are specified, and all peaks and troughs having amplitudes
less than
the specified minimums are identified and enhanced to the specified minimum
values.
Other methods for enhancing the amplitude values of the peaks and troughs
will be apparent to persons skilled in the art. For example, the enhancement
could be
based on the local phase of the zero-phase data trace. The concept of local
phase is
reached by comparing the trace to the cosine function, i.e., local phase is
zero at
peaks, ~ at troughs, and ~/2 or 3~/2 at inflection points. The cosine of local
phase is 1
at peaks, -1 at troughs, and 0 at inflection points. Thus, each peak on the
zero-phase
data trace would be assigned an amplitude value of 1 in the enhanced trace,
each
trough would he assigned any amplitude value of -1, and each inflection point
would
be assigned an amplitude value of 0.
FIGS. 2 and 3 illustrate application of the invention to an actual data set.
FIG.2 is a zero-phase seismic section for a particular survey line prior to
enhancement. FIG. 3 shows the same zero-phase seismic section after
enhancement
of the peaks and troughs in the manner described above with respect to FIGS.
lA and
1B. FIG. 3 shows the remnants of an ancient stream channel (reference numeral
17)
that is not visible in FIG. 2 (reference number 1 S).
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
-9-
In another embodiment of the invention, the seismic data traces (either
prestack or poststack) are enhanced by converting them to "curvature" traces.
These
curvature traces are then used to construct "curvature" sections for use in
the data
interpretation process.
As used herein, "curvature" is a measure of the concavity or convexity of an
arc; i.e., it is the inverse of the radius of curvature of the arc (i.e., the
radius of an
inscribed circle). Sharp turns have larger curvatures than blunt turns because
a circle
inscribed in a sharp turn will have a smaller radius (and, therefore, a larger
curvature)
than a circle inscribed in a blunt turn.
With respect to a seismic data trace, curvature is a measure of the rate of
bending in the trace as a function of two-way seismic signal traveltime.
Curvature
may be used to enhance peaks and troughs in seismic data traces, and sometimes
kinks or bends as well. Curvature also has a sign. Where a trace is a concave
left,
curvature is positive, regardless of whether the concavity is located on the
positive or
negative side of zero amplitude. Hence, even relative peaks located on the
negative
side of zero amplitude show up as positive peaks on a curvature trace.
Mathematically, for a continuous function in the time domain, curvature is
defined by the second derivative with respect to time of the function.
However, as
will be well known to persons skilled in the art, digital seismic data
actually
comprises a series of discrete samples (typically at 2 or 4 millisecond
intervals) of the
amplitude of the seismic reflection. Computing a true second derivative for
such
discretized data requires spline-fitting or some other approximation to obtain
a
continuous interpolated trace defined at every time point. Methods for
creating such
an interpolated trace are well known to persons skilled in the art and,
accordingly, will
not be described herein. Care should be exercised in creating the interpolated
trace to
avoid potential aliasing problems. The second derivative with respect to time
of the
interpolated trace is then computed to obtain a curvature trace.
For discretized seismic data, trace curvature may be approximated by the
negative second difference (-OZ) of the data, which is actually a measure of
numeric
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
-10-
acceleration. If a" a2, and a3 are successive sample amplitudes on a seismic
data trace,
then the negative second difference at sample a2 is defined by the following
equation:
-02 =-(a3 +a~ -2a2)
The "negative" second difference is used in order to compensate for a
180° phase
rotation (polarity reversal) resulting from the second difference calculation.
Negative second difference may be used to enhance subtle features (e.g., bends
or kinks) in the data, as well as peaks and troughs, so that their true
continuity can be
identified. FIG. 4 shows five data samples of a seismic data trace 16, which
may be
any type of seismic data trace. TABLE 1 below gives the time and amplitude
values
for each of the five data samples, as well as the negative second difference
(calculated
according to the above formula) for the middle three data samples.
TABLE 1
3146 -200 __
3148 -50 80
3150 20 -60
3152 150 gp
3154 200 __
The negative second difference 18 is also plotted on FIG. 4.
It can be seen from FIG. 4 that negative second difference can be used to
highlight subtle features of seismic character, as well as peaks and troughs.
The bend
in trace 16 at 3150 milliseconds may be a muted expression of an impedance
boundary. It may also be noise. The negative second difference calculation can
be
used to enhance this feature so that it can be identified and traced laterally
through the
curvature section, but its meaning requires careful interpretation.
FIGS. 5 and 6 illustrate the results of the negative second difference
calculation for two longer trace segments. FIG. 5 shows seismic data trace 20,
which
may be any type of seismic data trace, and its related curvature trace 22.
Trace 20
CA 02325874 2000-07-28
WO 99/41623 PCT/US99/02711
-11-
contains a number of relative peaks (e.g., relative peaks 24 and 26) located
on the
negative side of zero amplitude. On curvature trace 22, the corresponding
peaks 28
and 30 are located on the positive side of zero amplitude. FIG. 6 shows a
seismic data
trace 32 and its corresponding curvature trace 34. It can be seen that the
peaks and
troughs of the two traces generally track each other, with the curvature trace
being
more sensitive.
FIGS. 7A, 7B, 8A, and 8B further illustrate the utility of curvature traces.
FIG. 7A shows 23 zero-phase stacked data traces for a particular survey line,
and FIG.
7B shows the corresponding curvature traces. In FIG. 7A, box 36 shows a peak
event
that seems to split. However, in box 38 of FIG. 7B, it can be clearly seen
that there
are actually two separate events. Also, in FIG. 7A, the peaks within area 40
are
difficult to follow, while the same peaks on the coiTesponding curvature
traces (area
42 of FIG. 7B) are quite easy to identify and follow laterally. FIG. 8A is the
conventional stacked seismic section, and FIG. 8B is the corresponding
curvature
section. Sequence boundaries and other geologically significant features are
much
easier to identify and interpret in the curvature section. The ability to
identify such
features is critical to sequence stratigraphy. Changing the gain and making
other
changes in the conventional stacked seismic section did not bring these
features
forward. Only the curvature section permitted them to be identified and
interpreted.
FIG. 9 illustrates another application of the present invention. In FIG. 9,
the
left panel shows an impedance curve 44 which was derived from well log data
obtained from a well. The right panel shows a zero-phase quadrature data trace
46 for
a location near the well in question. The center panel shows a curvature trace
48
obtained from zero-phase quadrature data trace 46 using the negative second
difference calculation described above. Note that the curvature trace 48
mimics the
well log impedance curve 44 much better than does the zero-phase quadrature
data
trace 46.
The foregoing description is directed to particular embodiments of the present
invention for the purpose of illustrating the invention. It will be apparent,
however, to
one skilled in the art that many modifications and variations to the
embodiments
CA 02325874 2000-07-28
WO 99/41623 PCTNS99/02711
-12-
described herein are possible. For example, a curvature trace may be generated
by
transforming a seismic data trace to the frequency domain to obtain an
amplitude
spectrum (i.e. a plot of amplitude versus frequency), multiplying each point
on the
amplitude spectrum by the associated frequency squared (w2), and then inverse
S transforming the result back to the time domain. Implementation of the
invention in
other data domains may also be possible. All such modifications and variations
are
intended to be within the scope of the present invention, as defined by the
appended
claims.
