First introduced to our industry in the late
1980s, by the early 1990s, 3-D seismic
data had become routine with most
oil companies. The acceptance of this new
technology data was due to the significant
effort invested in demonstrating the value of
volumetric interpretation of 3-D seismic data.
The interpretation of vertical, time and horizon
slices through corendered multiattribute
volumes using HSL and RGB color models
(see our Geophysical Corner articles from
the Dec. 2019 and Jan. 2020 issues) was
augmented by another new technology: the
isolation, detection, and display of geobodies
in 3-D.
The geobody tool provides a means for an
interpreter to rapidly visualize the extent and
orientation of anomalous geologic features of
interest. However, the last decade has seen
an exponential growth in both the number
and size of 3-D seismic surveys. Augmented
by multiple attribute volumes for each survey,
these large data volumes provide both an aid
and a burden on the interpreter, whose goal is
to wade through all these data with the goal of
extracting patterns that correlate to a geologic
model, which can then be used for oil and gas
exploration and development.
As many of the world’s oil and gas
resources lie beneath the oceans, the
advances in exploration, drilling and
production technologies have also focused
in those areas. Oil and gas production started
on land, and then moved to first shallow,
then moderate, and for the past 20 years,
deepwater environments. Unlike land data,
deepwater marine data do not suffer from
statics and heterogeneity in the near surface.
In general, the data quality of deepwater
data is superior to land surveys attempting
to image similar geologic targets, with better
preservation of amplitudes and less sourcecorrelated
noise. Because of the preservation
of relative amplitudes, seismic amplitude
anomalies associated with hydrocarbon
accumulation or impedance changes as well
as the vertical and lateral changes in lithology
are often amenable to interpretation as 3-D
seismic geobodies.
Defining a ‘Geobody’
A seismic “geobody” refers to an
interpreted 3-D object comprising voxels
exhibiting a similar range of amplitudes or
attributes. In this short article, we illustrate
two key aspects of deepwater seismic data
interpretation – 3-D visualization and seismic
geobody picking of the observed high seismic
amplitude anomalies. As the interpreters
navigate their way through gigabytes of
seismic data by way of visualization, they
are usually looking for unique features in
their broad zone of interest, which might
include turbidite channels and fans, mass
transport deposits, current-generated bars
and other geologic features that help define
the environment of deposition. Once seen on
a 2-D section, the interpreter can use opacity
(the opposite of transparency) to better
delineate such features in 3-D.
Figure 1a shows a segment of a seismic
section from the deepwater East Breaks
Alaminos Canyon area of the western U.S. Gulf
of Mexico that exhibits a number of strong
amplitude anomalies. The seafloor reflection
is characterized by a red-blue-red amplitude
sequence (figure 1c), while the bright spots
show the opposite (blue-red-blue) amplitude
sequence, suggesting a phase reversal. The
amplitude anomaly to the left (figure 1a)
seems to be following the sloping bed and
abuts against a salt body farther to the left.
The high amplitude anomaly to the right is
also following the bedding, but is strongest
toward the graben faults, which could be a
possible scenario for entrapment of fluids.
Depending on the thickness of the geologic
feature of interest, a selection can be made for
the input data to be used for geobody tracking.
In general, the seismic amplitude itself is not
the best candidate for defining a geobody. For
example, consider gas-saturated sandstone.
The top of the anomaly in this data volume is a
trough, and the base a peak. There is no single
range of amplitude values that can separate
the top and base response from the geologic
background at the same time. By using the
instantaneous envelope (also called the
reflection strength) we can avoid this problem
– both strongly negative and strongly seismic
amplitudes give rise to a strong envelope
anomaly. Acoustic impedance, which
transforms the response of the upper and
lower reflectors to that of the low impedance
interval, is better still. Figure 1b shows the
instantaneous envelope corresponding to
the seismic amplitude data shown in figure
1a. Notice, how it would be easier to pick the
geobody on the envelope than the seismic.
Alternatively, it is also possible to use both
envelope as well as impedance, and other
similar attributes for geobody tracking.
Figure 2 shows a sub-volume of the
seismic data restricted to the areal extent of
the two seismic anomalies but using opacity
and exhibiting their volumetric disposition.
Both these anomalies could be picked up
for more detailed analysis, but first their
volumetric definition could be refined. This
could be done by rejecting the smaller high
envelope events and picking the larger ones
with seismic geobody picking.
There are a few ways in which seismic
geobody picking can be carried out and
different software packages could offer
different options. Seismic geobody picking can
be carried out by using (1) the original seismic
amplitude or a derived attribute volume, (2) the
seismic volume in addition to derived attribute
volumes, and (3) a derived facies classification
volume. Once the input data volume(s)
has been selected, next the parameters for
geobody picking need to be selected.
Geobody Picking
For geobody picking the temporal window
can be restricted either within a simple time
window, or with a little more work, between
two bounding horizons. The lateral range
selection can be carried out by specifying the
inline and crossline numbers or by making
polygon selection which can be drawn on a time or a horizon slice. For seismic amplitude
and attributes, extreme (either anomalously
high or anomalously low) values provide the
easiest way to define a geobody. A seed point
is then inserted in the anomaly and the range
of upper and lower amplitude threshold values
are specified. Starting at the seed voxels, the
seed tracker will search for connected voxels
that satisfy the user-defined search criteria.
While searching for the adjacent voxels that
have the valid amplitude values falling in the
specified range, a couple of options could be
utilized, which will have a bearing on the time
taken for the geobody picking.
The first option would search for the
adjacent voxels to the left and right, front
and back, as well as top and bottom, or six
voxel searches, while the second option
would search for the adjacent voxels in 26
voxel searches. By doing so the defined
volume of the input data is scanned for
detection of the geobody amplitude values
of interest. Depending on the input attribute
used, the interpreter can generate geobodies
corresponding to a porous sand accumulation
(using Poisson’s ratio), a salt body (using a
texture attribute) or the channel-fill within
a drainage system (using impedances or
spectral components) can be delineated. The
geobodies shown in figure 3 have been picked
using the second option.
The Usefulness of Geobody Tracking
In the next article we will discuss the use of
more than one attribute for geobody picking
and how to bring out its advantages, as well
as picking geobodies on a seismic facies
classification volume.
Picking geobodies is an important step
toward volumetric interpretation of seismic
data. In comparison with picking of faults
and horizons, which define boundaries of a
reservoir structure, picking geobodies is the
3-D mapping of a reservoir itself, such as
sandy channels and depositional lobes.
Thus, geobody picking can delineate
geometric, structural and lithological patterns
in a reservoir, and can help with visualization
of complex geological settings. Should
such picked geobodies be prospective and
the data imaged in or converted to depth,
the calculation of original oil in place or the
original gas in place can be used for reserve
estimation. Such computations could be
carried out with the knowledge of some of the
reservoir parameters for example porosity,
water-saturation and formation volume
factors for oil and gas. Thus, geobody-tracking
can help in assessing the hydrocarbon
reserves in each prospect.