About This Item
- Full text of this item is not available.
- Abstract PDFAbstract PDF(no subscription required)
Share This Item
The AAPG/Datapages Combined Publications Database
Houston Geological Society Bulletin
Abstract
Abstract: Two Alternative Seismic Fault Interpretation
Techniques—Part One
Consulting Geophysicist, T/X Resources
This is Part One of an expanded article describing two alternative seismic fault interpretation techniques. These were summarized in my “Rocks-2-Digits” web log; however, I also thought that they might be of interest to those who may not have seen them already. Part Two, which will be published next month, will show how shaded relief maps can be used to delineate very small faults.
Technique #1—Horizontal Fault Interpretation
As a standard practice, interpreting faults on a seismic workstation
is usually done on vertical seismic displays (VSDs) of 3D
seismic volumes—that is, on inline, crossline or arbitrary line
presentations. However, this can sometimes be problematic
and/or very time
consuming in a complexly-faulted area.
One factor is that faults are best seen on lines oriented in a dip
direction, and in a complex area there may be several different
fault orientations within a seismic volume, causing the interpreter
to change line directions frequently. For a correct interpretation,
the initial fault interpretation picks must be connected properly.
Figure 1. An
oblique view of a 3D seismic volume, illustrating the regularly spaced,
horizontal fault interpretations (brown lines). These were picked on
time
slices
such as the one shown at 1.8 seconds, against the backdrop of a Coblended (migration
and Ridge Enhancement) vertical seismic display.
End_Page 45---------------
An easier, quicker and more accurate method than using VSDs for this is to interpret faults in the horizontal domain, as illustrated in Figure 1.
Here’s a brief description of the method that I use for this:
1) Generate and/or load a coherency-type seismic attribute
volume into the workstation project. I usually create a Ridge
Enhancement attribute (in dGB’s OpendTect software), an
alternative to a coherency volume, as shown on the time
slice in
Figure 1.
2) I then use Seismic Micro-Technology’s (SMT) 2d/3dPak to
interpret the fault discontinuities (black areas of minimum
similarity in the example) on time
slices
at 100-millisecond (ms)
increments. The
time
slices
should be spaced according to the
nature of the fault—some faults require closer
time
spacing,
whereas on other faults, a larger
time
spacing
may be sufficient.
3) The next step is to switch to the seismic
visualization module (I use SMT’s VuPak
module) and scroll through the time
slices
to assign the faults that are connected and
continuous to an already named fault—for
example, the RED FAULT. Note that the
fault interpretation done in step 2 can also be done in the visualization
mode. Any necessary fault editing can also be done in this
step.
4) While still in the visualization module, display the fault interpretation in a continuous surface mode, and check for any fault “kinks” (unrealizable geometries). A correct fault surface should be relatively smooth, whereas improperly interpreted or assigned fault segments are immediately apparent, with dramatic kinks or gyrations in the fault surface.
5) The final step, while not absolutely essential, is to scroll through the vertical seismic lines as another quality control process.
When this workflow is completed, you have a very robust fault plane map. This can then be used to accurately determine the intersection of the horizon and fault surfaces, which by definition is the fault polygon outline, indicating the missing section on a structure map of that horizon.
An additional animated illustration for the Horizontal Fault Interpretation Technique, which couldn’t be properly displayed here, was posted on my “Rocks-2-Digits” blog on 12/11/07. Go to http://txresources.com/rocks2digits/, scroll the calendar back to December, and then click on the 11th. The direct address is http://txresources.com/rocks2digits/2007/12/11/horizontal-faultinterpretation/.
End_of_Record - Last_Page 47---------------