Enhancing Discontinuities in Seismic Data and Automated Fault Mapping
Enhancing Discontinuities in Seismic Data and Automated Fault Mapping
Enhancing Discontinuities in Seismic Data and Automated Fault Mapping
According to the Schlumberger’s Petrel help manual, the process “can be divided into
four activities:
1. Seismic conditioning (structural smoothing)
2. Edge detection (chaos and variance)
3. Edge enhancement (ant tracking)
4. Interactive interpretation of surfaces, which can include fault extraction (from fault
patches).” Fault patches are described in lecture 5.
The tracking is based on measures of variance or chaos in the attribute. The ant-tracking
process is usually applied to secondary 3D volumes that provide measure of variance or
chaos in the 3D attribute being evaluate. In the Petrel help files these are simply
described as follows.
1
Ant Tracking Workflow
Ant tracking is performed within Petrel’s Geophysics > Volume Attributes process. The
ant tracking process is preceded by smoothing and variance or chaos processes. Each
time one of these algorithms is run, a new “output” file is inserted in the input tree. Each
of these new 3D volumes serves as input to subsequent processing in the flow.
The flow chart below illustrates typical ant-track workflows. The smoothed volume is
then either subjected to a Variance or Chaos analysis. The resulting variance or chaos
volume is then used in the ant-track computations. The ant track process is usually run
twice.
The different output volumes can be superimposed on each other for comparison using
the seismic calculator. Use of the seismic calculator is illustrated in the following notes,
but will not be demonstrated in class.
2
3
4
5
6
PART 2. Ant track workflow.
7
The workflow involves a series of
volume attribute calculations
performed on the 3D seismic cube
(either in time or depth) from the
Process Tab>
Geophysics>Volume attributes
Original Seismic
Smoothing
8
Make Variance (after structural Smoothing)
Original Seismic
Smoothing
Variance
Original Seismic
Smoothing
Variance
Ants
9
Make Chaos (after structural Smoothing)
Original Seismic
Smoothing
Chaos
Original Seismic
Smoothing
Chaos
Ants
10
Ants Ants (after Chaos)
Original Seismic
Smoothing
Chaos
Ants
Ants
As with any other seismic volume, the output data are viewable in 3D Windows and
Interpretation Windows.
11
Variance-Ants-Ants on Seismic
Using the seismic calculator, you can add or superimpose the “variance-ants-ants” onto
the original seismic data. This can reveal faults right upon the 3D seismic rendering.
12
Chaos-Ants-Ants on Seismic (SKIP - will not be covered in class)
Using the seismic calculator, you can add or superimpose the “chaos-ants-ants” onto the
original seismic data. This allows you to inspect interrelationships between input and
output data sets.
13
Chaos-Ants-Ants on Seismic
Shown below is a time slice at ~ 1100 ms with “Chaos-Ants-Ants on Seismic”. Fault
sticks from original fault interpretations are superimposed upon this realization. Note
how the algorithms have indeed resolved the major faults. This is especially true of Fault
S0 and S1 in this image, followed by S3 which is a little more shallow. The S2 fault
doesn’t run as deep as 1100 ms so there isn’t a strong connection in this time slice.
14
Sum of All Ants -Ants-Ants
Variance-Ants-Ants + Chaos-Ants-Ants
15
Conclusion
The procedures discussed above are summarized in the image below taken from the
Petrel software help manual.
The edge detection volumes generated in the preceding section also serve as input files
for another process called Automatic Fault Extraction. This process is introduced in the
last half of today’s workshop.
16
Fault Patches
In this exercise you will learn how to:
- Generate fault patches using the ant-track volume as input.
- Use interactive fault patch filtering tools.
One could also try this using the chaos and variance seismic volumes. The ant-track
volume is ideally suited for the automatic fault detection process since its purpose is to
enhance discontinuities in the seismic data that might be associated with faults or fracture
zones. The fault extraction process will provide an integrated 3D view of the distributions
of these features.
Extracted fault patches can also be interactively filtered as described in the following
pages.
17
Numerous fault patches may be present and can overwhelm the 3D view and the
interpreter. There are a couple of interactive filtering
tools that are used to selectively reduce the number
of faults in the display
18
In this example from Teapot Dome, the “Stereonet” tool
was used to focus on the steeply dipping patches. These
surfaces will have poles that are nearly horizontal in
orientation.
19