Application of 3D seismic visualization techniques for seismic stratigraphy, seismic geomorphology and depositional systems analysis; examples from fluvial to deep-marine depositional environments


Posamentier, H.W.

Geological Society, London, Petroleum Geology Conference Series 6: 1565-1576

2005


In recent years, 3D seismic has become an essential tool for the interpretation of subsurface stratigraphy and depositional systems. Seismic stratigraphy in conjunction with seismic geomorphology, calibrated by borehole data, has elevated the degree to which seismic data can facilitate geological interpretation. 3D seismic data has enabled interpreters to visualize details of complex depositional systems, which can be incorporated into borehole planning for exploration as well as development needs to improve risk management significantly. Common techniques for geological visualization include (1) imaging stratigraphic horizons, (2) time slicing and flattened time slicing, (3) interval attribute analysis, (4) voxbody interpretation and mapping, (5) 3D perspective rendering and (6) opacity rendering. One of the key benefits of modern 3D seismic interpretation is that stratigraphic horizons can be interpreted and horizon attributes (such as reflection amplitude, dip magnitude, dip azimuth, and curvature) can then be imaged directly in 2D or 3D space. Techniques such as variable illumination can enhance geomorphological interpretations, and, when integrated with stratigraphic analyses, can yield insights regarding distribution of source, seal, and reservoir facies. Stratigraphic intervals bracketing sections of geological interest can be evaluated for amplitude and frequency content and can contribute to geological interpretations. Time slices and flattened time slices (also referred to as horizon slices) can bring to light map patterns and geological features that other techniques might overlook. Voxel picking can further bring out features of geological interest. This method involves autopicking of connected voxels of similar seismic character, a technique that can illuminate discrete depositional elements in three dimensions. Similarly, opacity rendering, which makes opaque only those voxels that lie within a certain range of seismic values, can further bring out features of stratigraphic interest. Examples of fluvial, shallow marine, and deep marine depositional environments are shown. A variety of visualization techniques are applied to these examples in an effort to illustrate the variety of interpretation techniques available to the geoscientist. These examples will highlight the integration of seismic stratigraphic and seismic geomorphological analyses essential for maximum benefit to be derived from geological analyses of 3D seismic data.

Application
of
3D
seismic
visualization
techniques
for
seismic
stratigraphy,
seismic geomorphology
and
depositional
systems
analysis:
examples
from
fluvial
to
deep-marine
depositional
environments
H.
W.
POSAMENTIER
Anadarko
Canada
Corporation,
425
1st
Street
SW,
Calgary
T2P
4V4,
Canada
(e-mail:
henry_posamentier@anadarko.corn)
Abstract:
In
recent
years,
3D
seismic
has
become
an
essential
tool
for
the
interpretation
of
subsurface
stratigraphy
and
depositional
systems.
Seismic
stratigraphy
in
conjunction
with
seismic
geomorphology,
calibrated
by
borehole
data,
has
elevated
the
degree
to
which
seismic
data
can
facilitate
geological
interpretation.
3D
seismic
data
has
enabled
interpreters
to
visualize
details
of
complex
depositional
systems,
which
can
be
incorporated
into
borehole
planning
for
exploration
as
well
as
development
needs
to
improve
risk
management
significantly.
Common
techniques
for
geological
visualization
include
(1)
imaging
stratigraphic
horizons,
(2)
time
slicing
and
flattened
time
slicing,
(3)
interval
attribute
analysis,
(4)
voxbody
interpretation
and
mapping,
(5)
3D
perspective
rendering
and
(6)
opacity
rendering.
One
of
the
key
benefits
of
modem
3D
seismic
interpretation
is
that
stratigraphic
horizons
can
be
interpreted
and
horizon
attributes
(such
as
reflection
amplitude,
dip
magnitude,
dip
azimuth,
and
curvature)
can
then
be
imaged
directly
in
2D
or
3D
space.
Techniques
such
as
variable
illumination
can
enhance
geomorphological
inter-
pretations,
and,
when
integrated
with
stratigraphic
analyses,
can
yield
insights
regarding
distribution
of
source,
seal,
and
reservoir
facies.
Stratigraphic
intervals
bracketing
sections
of
geological
interest
can
be
evaluated
for
amplitude
and
frequency
content
and
can
contribute
to
geological
interpretations.
Time
slices
and
flattened
time
slices
(also
referred
to
as
horizon
slices)
can
bring
to
light
map
patterns
and
geological
features
that
other
techniques
might
overlook.
Voxel
picking
can
further
bring
out
features
of
geological
interest.
This
method
involves
auto-
picking
of
connected
voxels
of
similar
seismic
character,
a
technique
that
can
illuminate
discrete
depositional
elements
in
three
dimensions.
Similarly,
opacity
rendering,
which
makes
opaque
only
those
voxels
that
lie
within
a
certain
range
of
seismic
values,
can
further
bring
out
features
of
stratigraphic
interest.
Examples
of
fluvial,
shallow
marine,
and
deep
marine
depositional
environments
are
shown.
A
variety
of
visualization
techniques
are
applied
to
these
examples
in
an
effort
to
illustrate
the
variety
of
interpretation
techniques
available
to
the
geoscientist.
These
examples
will
highlight
the
integration
of
seismic
stratigraphic
and
seismic
geomorphological
analyses
essential
for
maximum
benefit
to
be
derived
from
geological
analyses
of
3D
seismic
data.
Keywords:
seismic
stratigraphy,
seismic
geomorphology,
3D
seismic
visualization
Seismic
data
have
long
been
used
for
lithological
prediction.
Initially,
predictions
of
lithology
were
based
on
analyses
of
2D
seismic
reflection
profiles
and
the
technique
was
referred
to
as
seismic
stratigraphy
(Vail
et
al.
1977).
The
approach
that
was
used
involved
the
identification
of
reflection
terminations
indicative
of
stratigraphic
discontinuities,
the
description
of
reflection
geome-
tries
between
discontinuity
surfaces,
and
mapping
of
the
amplitude,
continuity,
and
frequency
of
reflections,
all
on
seismic
reflection
profiles
(Fig.
la).
Integration
of
these
observations
into
seismic
facies
maps
(Fig.
ib)
provided
the
basis
for
interpretation
of
depositional
environment
and
lithology.
With
the
development
of
3D
seismic
acquisition
techniques,
the
opportunity
to
image
geological
features
in
map
view
has
opened
up
new
approaches
to
geological
prediction
(e.g.
Weimer
&
Davis
1996).
Various
seismic
reflection
attributes
such
as
amplitude,
dip
magnitude
and
azimuth,
time/depth
structure
and
curvature,
to
name
a
few,
can
yield
direct
images
of
depositionally
and
structurally
significant
features.
In
addition,
analysis
of
seismic
attributes
over
multi-cycle
seismic
intervals
can
lend
further
insights
to
such
features.
The
study
of
depositional
systems
using
3D
seismic
derived
plan-view
images
has
been
referred
to
as
seismic
geomorphology
(Posamentier
2000).
This
represents
a
significant
step
change
in
how
seismic
interpreters
use
3D
seismic
data
for
the
analysis
of
depositional
systems.
Previously,
using
2D
seismic,
depositional
environments
and
lithologies
were
inferred
on
the
basis
of
cross-section
derived
seismic
reflection
geometries
and
associated
map
patterns
derived
from
time-consuming
seismic
facies
mapping
(Fig.
lb).
With
the
advent
of
seismic
geomorphology,
discrete,
detailed
depositional
subenvironments
and
depositional
elements
could
be
interpreted
directly
from
map
view
images
such
as
time
slices
or
horizon
slices
(e.g.
Fig.
2)
leading
to
much
more
accurate
understanding
of
lithological
distribution
patterns
and
enhanced
prediction
of
the
distribution
of
reservoir,
source
and
seal
fades.
Techniques
Numerous
volume-based
data
manipulation
techniques
exist
for
the
extraction
of
stratigraphic
information
from
3D
seismic
data.
These
techniques
range
from
analyses
of
discrete
horizons
to
analyses
of
discrete
sub-volumes.
In
each
instance,
3D
visualiza-
tion
techniques
can
play
a
significant
role
in
assisting
with
interpretation.
With
the
advent
of
high-speed,
affordable
hardware
and
software,
many
techniques
and
avenues
of
inquiry
that
in
the
past
might
have
been
unrealistic
to
pursue
have
now
become
mainstream
tools
in
the
interpreter's
toolkit.
This
evolution
of
interpretation
techniques,
which
has
afforded
interpreters
the
opportunity
to
ask
more
of
their
data,
has
necessitated
an
additional
required
skill
set
on
the
part
of
the
interpreter:
the
ability
to
recognize
and
interpret
landforms
seen
in
map
view,
and
POSAMENTIER,
H.
W.
2005.
3D
seismic
visualization
techniques.
In:
Dost,
A.
G.
&
VINING,
B.
A.
(eds)
Petroleum
Geology:
North-West
Europe
and
Global
Perspectives—Proceedings
of
the
6th
Petroleum
Geology
Conference,
1565-1576.
©
Petroleum
Geology
Conferences
Ltd.
Published
by
the
Geological
Society,
London.
0-
T
IME
-
SEC
OND
S
1-
~-
tb
2
64i
5
L
ill
Iii
2
_
150
1(12
_
4
it,3_26
0
5
KM
0
5
MILES
SEISMIC
_MT.
-
TOO,
0:
SAND
,cIES
C
T
F
Slump
Scars
at
Shelf
Edge
.1,
0
km
5km
(b)
Carbonate
Patch
Reef
:
0
km
5km
(b)
(a)
Incised
Valley
near
Shelf
Edge
(a)
Fig.
1.
(a)
Seismic
reflection
profile
from
offshore
Morocco
illustrating
the
application
of
classical
seismic
stratigraphic
interpretation
(from
Vail
et
al.
1977);
(b)
Seismic
facies
map
of
the
Eocene
Frigg
submarine
fan
offshore
Norway,
based
on
interpretation
of
2D
seismic
reflection
profiles
(from
McGovney
&
Radovitch
1985).
Fig.
2.
(a)
Seismic
horizon
slice
illustrating
outer
shelf
and
upper
slope
geomorphology
of
offshore
Indonesia.
Features
of
interest
imaged
here
include:
(1)
a
small
incised
valley
that
terminates
just
inboard
of
the
shelf
edge,
and
(2)
a
series
of
slump
scars
at
the
shelf
edge,
with
associated
slump
debris
downslope.
This
image
was
produced
by
flattening
on
an
overlying
seismic
horizon
and
then
extracting
the
reflection
amplitude
information
36
ms
below
that
horizon;
(b)
Miocene
carbonate
patch
reefs
offshore
northwest
Java,
Indonesia.
This
image
was
produced
by
flattening
on
a
horizon
below
the
reefs
and
slicing
at
a
level
120
ms
above,
and
subsequently
extracting
the
reflection
amplitude
at
that
level.
41tiO
.
Horizontal
Planar
Slice
(i.e.,
Time
Slice)
tr
4
--r
ti
Inclined
Planar
Slice
(i.e.,
Dipping
Time
Slice)
Horizon
Parallel
Slice
(i.e.,
Horizon
Slice)
0
km
(b)
(c)
1567
to
infer
their
stratigraphic
significance.
The
disciplines
of
seismic
stratigraphy
and
seismic
geomorphology
go
hand
in
hand
in
the
geological
interpretation
of
3D
seismic
data.
Features
observed
in
plan
view
should
be
corroborated
by
cross-section
view
images.
Hence
the
integration
of
stratigraphy
(i.e.
section
view)
with
geomorphology
(i.e.
plan
view).
Time
slices
and
horizon
slices
Time
slices
and
horizon
slices
are
excellent
for
initial
reconnais-
sance
through
a
3D
volume.
Time
slices
represent
horizontal
slices
through
the
seismic
volume.
From
a
stratigraphic
perspective,
these
yield
the
most
meaningful
images
in
data
sets
where
the
strata
are
close
to
horizontally
bedded
(Fig.
3a
and
b).
In
data
sets
where
time
slices
are
not
parallel
to
seismic
reflections,
hints
of
depositional
features
of
interest
nonetheless
may
be
observed
on
time
slices,
but
may
be
more
fully
revealed
using
other
techniques.
With
variably
dipping
seismic
reflections,
slicing
parallel
to
reflections
commonly
yields
the
best
insights
(Fig.
3c).
These
images
are
referred
to
as
horizon
slices
or
flattened/datumed
time
slices.
Using
either
of
these
techniques,
reflection
amplitude
map
patterns
can
indicate
the
presence
of
discrete
depositional
elements
such
as
channels
or
reefs.
When
slices
reveal
the
presence
of
features
of
interest
in
map
view,
it
is
essential
to
examine
the
feature
in
cross
section
to
confirm
that
what
has
been
observed
is
(a)
stratigraphic
rather
than
structural
in
origin,
and
that
the
feature
is
not
a
seismic
data
acquisition
or
processing
artefact
(see
discussion
below).
Horizon
attributes
Once
features
of
interest
are
identified
using
time
slices
or
other
reconnaissance
techniques,
specific
reflections
associated
with
these
features
are
interpreted.
The
interpreted
horizons
can
be
examined
using
a
variety
of
attributes.
These
attributes
can
include
structure,
amplitude,
dip
azimuth,
dip
magnitude,
curvature
and
roughness
(Fig.
4).
Each
attribute
can
reveal
different
character-
istics
of
the
feature
of
interest,
with
each
characteristic
aiding
in
the
interpretation
of
the
geological
feature
and
ultimately
the
prediction
of
lithologies
in
the
associated
section.
Simply
lighting
a
surface
from
various
directions
can
also
aid
in
interpretation
and
afford
significantly
different
insights
(Fig.
5).
Interval
attributes
In
some
instances,
seismic
interval
attribute
extractions
rather
than
horizon
attributes
are
the
tool
of
choice.
This
is
especially
true
where
features
are
subtle
and
discrete
seismic
reflections
are
difficult
to
interpret
and
map.
This
approach
involves
the
seismic
analysis
of
stratigraphic
intervals
or
seismic
'slabs'
that
bracket
features
of
interest;
attribute
analyses
include
characterization
of
amplitude,
frequency,
and
seismic
waveforms
(Fig.
6).
Voxbody
detection
Features
of
interest
also
can
be
imaged
by
detecting,
mapping,
and
visualizing
voxels
in
3D
views.
In
this
approach,
voxels
(i.e.
three-
dimensional
pixels)
are
highlighted
according
to
amplitude
(or
some
other
seismic attribute)
range
and
then
tracked
in
three
dimensions
(Fig.
7).
The
seismic
attribute
range
can
be
selected
to
correspond
to
a
specific
lithology,
the
presence
of
hydrocarbons,
or
simply
a
specific
seismic
reflection.
Identification
of
depositional
elements,
fluid
contacts,
and
interpretation
of
seismic
reflections
can
be
facilitated
through
this
approach.
Opacity
rendering
3D
seismic
volumes
can
be
rendered
transparent
or
partially
transparent
according
to
the
interpreter's
display
criteria
In
this
way,
the
interpreter
can
'see
through'
the
seismic
volume
and
concentrate
on
features
of
interest
that
can
be
imaged
within
a
certain
amplitude
range,
which
the
interpreter
has
rendered
opaque
(Fig.
8).
The
interpreter
can
quickly
target
features
of
interest
for
follow-up
analyses.
This
technique
can
benefit
from
volume
`sculpting',
which
can
isolate
parts
of
the
3D
seismic
volume
for
closer
inspection.
This
is
especially
useful
where
high
amplitude
reflections
just
above
or
below
target
zones
can
be
'sculpted'
out
leaving
behind
only
those
sections
where
opacity
rendering
is
meant
to
highlight
subtle
stratigraphic
detail.
3D
perspective
rendering
Interpretation
of
depositional
elements
can
be
facilitated
by
viewing
horizons
in
perspective
view.
Such
views
are
especially
useful
in
illustrating
spatial
relationships
between
different
elements,
which
can
assist
in
interpretations
and
communication
of
ideas
(Fig.
9a).
Perspective
views
are
also
useful
in
those
instances
where
depositional
settings
are
characterized
by
erosional
or
depositional
relief
(Fig.
9b).
5
km
Fig.
3.
Three
views
of
a
Pleistocene
leveed
channel
and
associated
Discussion
frontal
splay/lobe
on
the
ultra-deep
basin
floor
of
the
Gulf
of
Mexico:
(a)
horizontal
time
slice;
(b)
a
planar
surface,
dipping
slightly
Features
imaged
in
map
view
can
be
analyzed
using
basic
prin-
(approximately
at
a
one
degree
dip);
and
(c)
horizon
parallel
slice.
ciples
of
geomorphology
and
stratigraphy.
Map
representations
1568
(a)
(b)
(C)
".
•••
Dip
Azimuth
Roughness
-
4
%.-
Structure
(e)
(f)
,
r
f
A'
I
-
Dip
Magnitude
7
\
-
-
v•-•
Curvature
5
km
Fig.
4.
Multiple
visualizations
of
a
Pleistocene
channel-levee
complex
on
the
ultra-deep
basin
floor
of
the
Gulf
of
Mexico.
The
visualizations
include
(a)
dip
azimuth;
(b)
roughness;
(c)
structure;
(d)
dip
magnitude;
(e)
curvature;
and
(f)
perspective.
Dip
azimuth
maps
are
maps
that
assign
value
to
a
surface
area
as
a
function
of
the
direction
that
surface
faces.
Roughness
maps
describe
the
degree
of
irregularity
or
'bumpiness'
of
surface
areas.
Structure
maps
illustrate
the
time
(or
depth)
map
for
a
given
horizon.
Dip
magnitude
maps
display
the
angle
of
inclination
of
surface
areas.
Curvature
maps
illustrate
the
deviation
from
a
planar
surface
characterizing
small
surface
areas.
Perspective
views
portray
seismic
horizons
in
three-dimensional
space.
of
depositional
elements
comprise
a
view
of
the
landscape
or
seabed
as
it
existed
in
the
distant
past,
with
one
minor
caveat.
That
is,
what
is
seismically
imaged
actually
represents
that
part
of
the
landscape
or
seabed
that
has
been
preserved
subsequent
to
post-
depositional
erosion.
For
example,
fluvial
channels
are
dynamic
in
nature
and,
because
of
repeated
reactivation
and
potential
erosion,
present-day
landscapes
are
subject
to
later
modification.
Land-
scapes
observed
on
seismic
data
represent
the
end
product
of
these
dynamic
processes.
The
integration
of
seismic
geomorphology
and
seismic
stratigraphy
is
a
critical
part
of
the
workflow
designed
to
extract
geological
information
from
seismic
data.
Once
features
of
interest
are
identified
in
map
view,
it
is
essential
to
examine
these
features
in
cross
section
to
obtain
stratigraphic
confirmation
of
depositional
origin
if
possible.
3D
seismic
data
make
possible
any
orientation
of
cross
section
desired.
These
cross
sections
can
be
used
to
evaluate
stratigraphic
architecture
and
confirm
geological
interpretations
based
initially
on
recognition
of
depositional
elements
in
plan
e
do
7
oh
2
A!"
K
5
km
II)
Light
Source
Fig.
5.
A
seascape
in
the
deep
water
of
the
Gulf
of
Mexico
with
lighting
from
four
different
directions.
Note
that
different
lighting
angles
highlight
different
geomorphic
details,
in
particular
with
respect
to
the
slope
and
basin
floor
channels.
t..
Itagamil
1
kin
1569
Distributary
Channel
.
_
Pinnacle
Reefs
Delta
Plain
Channels
n
Fig.
6.
Two
examples
of
interval
attribute
images
from
the
western
Canada
sedimentary
basin.
(a)
Interval
attribute
of
a
Cretaceous
distributary
channel;
the
attribute
illustrates
the
amplitude
strength
of
a
40
ms
interval;
(b)
Interval
attribute
of
two
pinnacle
Devonian
reefs;
the
attribute
illustrates
the
peak
amplitude
within
a
60
ms
interval
that
brackets
the
reefs.
Channel
1
km
(b)
Crevasse
Splay
Voxbody
Fig.
7.
Voxbody
interpretation
of
a
deep-water
Pleistocene
channel
crevasse
splay
on
the
basin
floor
of
the
Gulf
of
Mexico.
The
splay
is
approximately
4
km
wide.
This
image
was
produced
by
'seeding'
a
voxel
of
a
specific
amplitude
and
then
allowing
the
autopicker
to
find
all
adjacent
voxels
of
similar
or
higher
amplitude.
(a)
.r
Channel
Bell
Fig.
8.
Opacity
rendering
of
a
shelf
channel,
western
Canada
sedimentary
basin.
The
channel
is
approximately
one
kilometre
wide.
Two
prominent
horizons
bracketing
the
channel
were
selected
for
horizon
sculpting,
a
process
by
which
all
voxels
above
the
upper
horizon
and
all
voxels
below
the
lower
horizon
are
rendered
transparent.
Subsequently,
the
interval
between
the
two
horizons
is
rendered
partially
transparent
whereby
only
the
extreme
amplitudes
are
left
opaque.
(b)
Fig.
9.
(a)
Time
structure
perspective
view
of
a
Pleistocene
channel-levee
complex
on
the
basin
floor
of
the
Gulf
of
Mexico
illustrating
several
depositional
elements
including
channel,
channel
belt,
and
overbank/levee.
The
channel
belt
is
approximately
3.5
km
wide;
(b)
Time
structure
perspective
view
of
the
base
Cretaceous
unconformity
separating
Cretaceous
fluvial
to
estuarine
deposits
from
underlying
Mississippian
carbonates
in
the
western
Canada
sedimentary
basin.
The
channel
in
the
foreground
is
approximately
400
m
wide.
f.
1570
(a)
,
V
(b)
X
,
-
..111P
Lateral
ACcfetr6n
Surfaces
Apr
'
1
km
1
km
Fig.
10.
(a)
Time
slice
through
an
Upper
Cretaceous
fluvial
channel
in
the
western
Canada
sedimentary
basin.
Flow
direction
is
inferred
from
lower
right
to
upper
left.
The
arcuate
forms
are
interpreted
as
point
bar
deposits;
(b)
Seismic
reflection
profile
across
fluvial
channel
shown
in
A.
Note
the
lateral
accretion
surfaces
leading
to
an
interpretation
of
point
bar
deposits
consistent
with
the
interpretation
of
the
plan
view
image
in
A.
view.
For
example,
as
illustrated
in
Figure
10a,
a
meandering
fluvial
channel
observed
in
plan
view
is
characterized
by
a
succession
of
meander
loops
and
possible
scroll
bars.
This
same
feature
viewed
in
cross
section
(Fig.
10b)
is
characterized
by
lateral
accretion
sets
suggesting
the
presence
of
point
bar
deposits,
an
interpretation
that
is
consistent
with
the
interpretation
of
a
meandering
fluvial
channel
derived
from
plan
view
images.
In
this
way,
the
stratigraphy
and
geomorphology
constitute
'converging
lines
of
evidence'
that
yield
a
more
robust
geological
interpret-
ation
than
is
possible
with
either
approach
alone.
Further
confirmation
of
this
interpretation
comes
from
borehole
infor-
mation,
where
well
logs
exhibit
a
fining-upward
trend,
typical
of
point
bar
deposits.
Another
reason
for
integrating
plan
view
and
section
images
is
to
confirm
that
the
origin
of
a
given
map
pattern
is
in
fact
stratigraphic
rather
than
a
data
processing
or
acquisition
artefact,
a
structural
feature,
a
map
of
fluid
distribution,
or
an
artefact
of
an
unusual
slice
through
the
data.
For
example,
horizon
slices
that
cut
across
seismic
reflections
at
a
low
angle
can
result
in
a
map
pattern
similar
to
that
of
a
high-sinuosity
meandering
channel.
Viewing
such
features
in
cross
section
immediately
confirms
that
this
pattern
is
not
indicative
of
a
specific
depositional
element
but rather
is
an
artefact
of
the
analysis
technique
Similarly,
cross
sections
can
quickly
reveal
a
structural
explanation
for
features
that
might
initially
be
inter-
preted
as
stratigraphic
in
nature.
Depositional
elements:
examples
Seismic
geomorphological
analyses
require
specific
skill
sets.
It
is
critical
for
interpreters
to
be
able
to
recognize
geologically
significant
map
patterns.
They
must
be
able
to
distinguish
seismic
noise
from
seismically
imaged
depositional
elements.
In
order
to
do
that,
the
interpreter
must
be
familiar
with
the
geomorphology
of
as
broad
a
range
of
depositional
elements
as
possible.
Of
course,
the
interpreter
must
also
be
familiar
with
various
expressions of
seismic
noise
(whether
due
to
data
acquisition
or
data
processing
issues).
A
Cretaceous-aged,
kilometre-wide
channel
crosscut
by
two
lesser
channels
is
imaged
in
a
variety
of
plan
and
section
views
(Figs
11
and
12).
Figure
1
la
is
a
horizon
slice
or
flattened
time
slice,
whereby
a
reflection
32
ms
above
was
interpreted
and
used
as
a
reference
horizon
for
the
purpose
of
slicing
through
the
3D
seismic
volume.
Figure
11b
is
a
reflection
amplitude
extraction
of
(a)
(b)
SCALE
N.!
Distributary
Channel
Point
Bar
Deposits
Delta
Plain
Channels
1
km
.11
Fig.
11.
Seismic
images
of
a
distributary
channel
crosscut
by
later,
smaller
delta
plain
channels,
western
Canada
basin
(compare
with
Fig.
6A).
(a)
Horizon
slice
or
flattened
time
slice
and
subsequent
amplitude
extraction
across
distributary
channel.
Note
the
lineaments
within
the
channel,
interpreted
as
point
bar
or
side
bar
deposits.
A
smaller
crosscutting
channel
can
be
observed
in
the
central
part
of
the
image;
(b)
Reflection
amplitude
map
of
reflections
immediately
below
the
level
associated
with
the
channel
(the
NRG
attribute
of
Paradigm
Geophyscial's
Stratimagic
application).
Note
that
this
image
brings
to
light
another
crosscutting
channel
in
the
lower
part
of
the
image.
(a)
(b)
x
50
ms
(c)
1
km
1571
Lateral
Accretion
surface
"
00
004
.1.401mor
Lateral
Accretion
Surfaces
........
,
.....
,
t,'.4:
...
........
.
.
...
!I,
.....
:11,1".'.
'VI
11
zone
of
least
compaction
(d)
0/aired
lateral
accrahon
De
compacted
channel
configuration
Most
sand-prone
clinolonns
Fig.
12.
(a)
Time
thickness
map
of
the
distributary
channel
shown
in
Figure
11;
blue
colours
represent
thicks
and
red
colours
represent
thins;
(b)
Transverse
seismic
profile
across
distributary
channel
shown
in
A.
Note
the
lateral
accretion
surfaces
suggesting
point
bar
accretion
from
left
to
right;
(c)
Line
drawing
of
distributary
channel
stratigraphy
based
on
seismic
reflection
geometry.
Note
the
bump
along
the
upper
bounding
surface,
likely
corresponding
to
a
differential
compaction
effect;
(d)
Channel
stratigraphic
architecture
reconstructed
by
decompacting
the
channel
fill.
Note
that
the
shingles
(i.e.
point
bar
deposits)
that
contribute
to
the
differential
compactional
bump
are
the
most
sand
prone.
reflections
immediately
below
the
reflection
associated
with
the
channel.
Each
images
the
channels
in
a
different
way,
with
different
details
brought
out
by
the
two
display
styles.
Both
show
linear
features
within
the
large
channel,
which
can
be
interpreted
as
possible
point
bar
deposits.
Both
show
a
crosscutting
and
therefore
younger
channel
in
the
middle
of
the
illustration.
However
Figure
11b
shows
another
smaller
channel
crosscutting
the
larger
channel
towards
the
bottom
of
the
illustration,
not
apparent
in
Figure
1
la.
This
underlines
the
fact
that
multiple
ways
of
imaging
the
same
feature
can
commonly
bear
fruit
insofar
as
subtle
features
might
not
be
imaged
on
every
form
of
display.
The
integration
of
seismic
geomorphology
and
seismic
stratigraphy
is
illustrated
in
Figure
12.
Inclined
reflections
within
the
interpreted
channel
fill
can
be
observed
on
the
reflection
profile
oriented
normal
to
the
long
axis
of
the
large
channel
(Fig.
12b).
(b)
(a)
Scroll
Bars
1
km
Scroll
Bars
(c)
X
x'
< -
1
Horizon
Slice
1
0110
.0
6
1
11041049."4111
°
411
/
4
1.1.1.14104Zt
50
msl
1
km
Fig.
13.
(a)
Meander
loops
and
scroll
bars
on
the
modern
Mississippi
River
flood
plain;
(b)
Cretaceous-age
meander
loops
and
scroll
bars
on
a
seismic
horizon
slice
(western
Canada
sedimentary
basin);
(c)
Seismic
reflection
profile
transverse
to
meander
loop
shown
in
(b);
Note
position
of
the
horizon
slice
shown
in
(b).
This
profile
shows
no
obvious
stratigraphic
expression
suggesting
the
presence
of
point
bar
deposits.
r
,
,4
.
if
.
1572
1
.
..
FluviJ1
..
1
km
Fig.
14.
Dip
azimuth
map
of
the
base
Cretaceous
unconformity
in
the
western
Canada
sedimentary
basin.
Note
the
high-sinuosity
fluvial
channels
entrenched
in
Mississippian-aged
carbonates.
(
a)
These
reflections
can
be
interpreted
to
represent
lateral
accretion
surfaces
associated
with
point
bar
deposition
within
the
channel
(Fig.
12c
and
d).
The
time
thickness
map
indicates
the
presence
of
a
thicker
channel
fill
on
the
southwestern
side
of
the
channel
(Fig.
12a
and
b).
The
seismic
profile
reveals
that
the
thicker
part
of
the
channel
does
not
correspond
to
a
deeper
channel
thalweg,
but
rather
is
associated
with
a
'bump'
across
part
of
the
channel.
This
`bump'
is
interpreted
to
be
associated
with
a
fill
that
is
less
compactable
than
the
other
part
of
the
channel
fill
(Fig.
12c).
This
less
compactable
section
would
suggest
the
presence
of
lateral
accretion
sets
that
are
sand-rich,
sand
being
less
compactable
than
silt
or
shale
(Fig.
12d).
Fluvial
systems
characterized
by
high-sinuosity
channel
belts
are
illustrated
in
Figure
13.
The
concentric
arcs
imaged
in
map
view
(Fig.
13b)
represent
sections
through
point
bar
deposits
and
may
represent
scroll
bars.
Figure
13a
illustrates
an
analogous
modern
feature
from
the
Mississippi
floodplain
for
comparison.
Examination
of
the
reflection
profile
shown
in
Figure
13
illustrates
a
stratigraphic
representation
of
such
deposits;
interpretation
of
the
correct
depositional
element
would
probably
not
have
been
possible
if
only
the
reflection
profile
were
available.
Fluvial
systems
overlying
a
major
unconformity
surface
are
illustrated
in
Figs
9b
and
14.
In
this
instance,
Cretaceous
fluvial
channel
fill
deposits
directly
overlie
Mississippian-aged
(b)
MK,
IC)
(d)
L
4
I
';
f
t
.
;di
g
(e)
r'
1
km
I
20ms
Base
Cretaceous
Inconformity
1
km
Fig.
15.
Four
map
views
and
one
section
view
of
the
base
Cretaceous
unconformity,
western
Canada
sedimentary
basin.
(a)
Dip
magnitude
map
illustrating
incised
Cretaceous-aged
drainage
pattern;
(b)
Time
structure
map
of
the
same
surface
highlighting
the
dendritic
drainage
of
an
ancient
highland
area
on
the
right
side
of
the
image
(compare
with
perspective
view
of
Fig. 9b);
(c)
Co-rendered
image
of
dip
magnitude
and
time
structure
bringing
together
benefits
of
both
(a)
and
(b);
(d)
Dip
azimuth
map
of
the
same
surface
bringing
to
light
more
obscure
lineaments
that
correspond
to
smaller
channels;
(e)
Seismic
reflection
profile
across
the
study
area
illustrating
the
stratigraphic
view
of
the
same
surface
shown
in
(a—d).
Note
the
reflection
terminations
that
define
the
unconformity.
Note
also
the
onlapping
fill
section
above
the
unconformity.
'
rte.
Iir
.... d
o
ii
.
!btl'•
p
..,
4.
%
.107ro.
f
Iki
.,7
.
.
,
.
ert
-
.1
-
CtIU
a
.
-
"
.
P
i.
.
,
.•
1
n"
4
lit
r
.;
,
.
fr.
'•
Pr
A
..
op
:,
••
-..
:
..
1
;4'
:
t
.
4"
i:ii
.
'
•••••
. ,
•%.1
.
.
.
*4
-
:4
4
`;
'
47;
r.:
-
-
,
..
.1
.
7
••
I
11
1.
,
,
mot
1573
(a)
Slope
Channel
(b)
Shelf
Edge
Slope
Gullies
Fig.
16.
Perspective
views
of
a
Late
Pleistocene
upper
slope
environment.
(a)
Turbidity
flow
slope
channel
on
the
upper
slope.
This
channel
is
approximately
1.8
km
wide
and
is
flanked
by
levees
on
both
sides;
(b)
Small
slope
gullies
approximately
200-300
m
wide
located
at
the
uppermost
slope.
These
features
are
associated
with
the
presence
of
a
shelf
edge
delta
that
was
active
during
Pleistocene
sea-level
lowstand.
Fig.
17.
Co-rendered
seismic
coherence
and
amplitude
for
a
deep-water
turbidite
channel-levee complex.
The
coherence
attributes
emphasizes
reflection
discontinuities
and
therefore
enhances
channel
edges,
whereas
the
amplitude
attribute
is
an
indication
of
lithology
distribution
and
helps
define
potential
sand
and
mud.
Note
the
multiple
channel
threads
that
characterize
this
system.
These
channel
threads
are
indicative
of
meander
loop
evolution
suggesting
flow
from
left
to
right.
1
km
Fig.
18.
Seismic
amplitude
extraction
of
basement
reflection,
western
Canada
sedimentary
basin.
Basement
lithology
here
likely
is
metamorphic.
carbonates.
Figure
15
illustrates
several
horizon
attributes
as
well
as
a
seismic
profile
showing
the
stratigraphic
discontinuity
bet-
ween
the
Cretaceous
and
Mississippian
deposits.
Note
the
apparent
dendritic
drainage
pattern
off the
highland
area
at
the
right
side
of
Figures
9b
and
15.
Each
image
portrays
the
deposi
ional
elements
somewhat
differently.
In
comparing
Figures
9b
and
15b,
note
that
the
perspective
view
(Fig.
9b)
allows
for
enhanced
understanding
of
the
landscape
elements.
Figures
9a
and
16
illustrate
additional
examples
of
how
perspective
views
enhance
the
understanding
of
the
distribution
of
depositional
elements.
Co-rendering
of
different
attributes,
whereby
two
or
more
attribute
maps
are
overlain
and
imaged
simultaneously,
can
also
serve
to
enhance
geological
features
(Figs
15c
and
17).
Images
that
provide
complementary
information
can
synergistically
combine
to
bring
together
the
best
of
each
individual
image.
The
image
of
a
deep-marine
leveed
channel
shown
in
Figure
17
combines
the
edge
A
Note
positive
relief
Avulsion
Channel
Note
positive
relief
1
km
(c)
I
Channel
Fill
Eroded
Note
negative
relief
Note
negative
relief
B'
Knickpoints
(a)
1574
(b)
Fig.
19.
(a)
Perspective
view
of
Late
Pleistocene
sea
floor
in
the
eastern
Gulf
of
Mexico,
characterized
by
a
high-sinuosity
leveed
channel
(see
Fig.
20
for
location)
and
an
associated
avulsion
channel.
In
the
main
channel,
updip
of
the
avulsion
channel,
associated
knickpoints
representing
the
upstream
end
of
channel
incision
associated
with
the
avulsion
event
can
be
observed.
Note
that
the
channel
upstream
of
the
knickpoint;
(b)
is
characterized
by
a
convex-up
profile
suggesting
that
the
channel
is
sand
filled
there,
and
downstream
of
the
knickpoint
(c)
is
characterized
by
a
concave-up
profile
suggesting
that
the
channel
is
mud
filled
there.
detection
capabilities
of
a
'coherence'
map
with
the
lithology
indications
inherent
in
a
seismic
amplitude
map
to
portray
clearly
both
the
margins
of
the
channel
as
well
as
the
inferred
type
of
fill
in
the
channel.
In
certain
instances,
where
basement
reflections
are
well
defined,
subcrop
seismic
expression
can
provide
significant
insight
with
regard
to
basement
lithologies.
Figure
18
illustrates
subcrop
seismic
amplitude
expression
and
styles
of
deformation
(i.e.
the
folded
banding
in
the
subcrop)
indicative
of
a
metamorphic
basement
terrain.
The
application
of
the
3D-based
seismic
geomorphology
approach
has
been
particularly
important
for
improved
understanding
of
sedimentary
processes
in
deep-water
deposi-
tional
environments.
These
settings
are
difficult
to
study
first
hand
because
of
their
remoteness
and
inaccessibility.
3D
seismic
data
affords
section
and
plan
view
images
that
have
significantly
enhanced
our
understanding
of
spatial
and
temporal
distribution
as
well
as
lithology
of
deep-water
depositional
elements.
Figure
19
illustrates
a
high
sinuosity
Pleistocene
channel
deposited
in
the
eastern
Gulf
of
Mexico
in
water
depths
greater
than
2500
m.
This
channel,
described
by
Posamentier
&
Kolla
(2003),
appears
to
be
sand
filled
in
part,
as
indicated
by
the
positive
relief
of
the
channel
fill,
and
mud
filled
in
part,
where
the
channel
fill
is
characterized
by
a
concave-up
upper
surface
(Fig.
19). This
(a)
Figure
23
LI
9
FlowD
ir
e
c
t
i
o
n
(b)
g.g
7
1100
ms
5
km
Fig.
20.
(a)
Dip
magnitude
map
of
a
Late
Pleistocene
high
sinuosity
leveed
channel,
eastern
Gulf
of
Mexico;
(b)
Axial
transect
through
leveed
channel
shown
in
(a).
Light
grey
arrows
indicate
locations
where
transect
crosses
the
channel;
dark
grey
arrows
indicate
the
direction
of
migration
of
the
channel
through
time
as
the
system
aggraded.
In
most
instances
they
indicate
a
down-system
meander
loop
migration.
I
Flow
Direction
Air
aak-.•••••••-,
b)
1575
rn
Mr••••
"
''
(a)
I
plir/OFff
47
Chann
Le
ee
res't
Flow
Direction
0
in
(c)
2
3
Sweep
Swing
Fig.
21.
(a)
Time
slice
of
a
Late
Pleistocene
high
sinuosity
leveed
channel,
eastern
Gulf
of
Mexico;
(b)
Detail
of
this
channel
indicates
a
progressive
shift
of
the
channel
through
time
as
shown
in
the
line
drawing
interpretation
showing
meander
loop
evolution
(c).
Both
meander
expansion
(i.e.
swing)
and
down-system
migration
(i.e.
sweep)
through
time
can
be
observed.
(a)
Abandoned
Meander
Loop
Slice
B
100
ms
(b)
Slice
C
1
km
(c)
41*
Flow
Direction
Slice
B
Slice
C
Fig.
22.
(a)
Seismic
reflection
transect
through
same
channel
shown
in
Figure
20.
Note
abandoned
meander
loop;
(b)
Seismic
horizon
slice
taken
near
top
of
meandering
channel
section
illustrating
position
of
channel
axis
at
level
B
shown
in
(a).
The
dashed
line
in
(b)
represents
the
channel
position
at
the
lower
level
C
as
shown
in
the
seismic
horizon
slice
C.
Fig.
23.
Perspective
view
of
the
late
Pleistocene
channel
shown
in
Figure
20.
The
channel
is
approximately
600
m
wide.
Note
the
raised
profile
of
the
channel
indicating
the
presence
of
sand
there.
Note
also
the
enhanced
height
of
the
levees
along
each
outer
channel
bend.
channel
was
probably
characterized
by
channel
avulsion
in
its
late
stage
of
development,
resulting
in
an
upstream-migrating
knickpoint.
Cross-section
views
show
how
this
channel
evolved
through
time
and
space
(Fig.
20).
Figure
21
illustrates
progressive
down-system
meander
loop
migration
in
map
view,
whereas
Figure
22
illustrates
map
views
of
this
channel
at
different
times
in
its
evolution.
Note
also
the
greater
height
of
the
levees
on
each
outer
meander
loop
(Fig.
23),
a
characteristic
consistent
with
partially
channelized
sediment
gravity
flow
deposits.
Conclusions
The
study
of
depositional
systems
in
time
and
space
has
benefited
greatly
from
analyses
of
3D
seismic
data.
Through
a
combination
of
plan
views
and
section
views,
discrete
depositional
elements
can
be
identified
and
visualized,
and
ultimately
interpreted
with
regard
to
paleogeography,
temporal
evolution
and
lithology.
This
approach
constitutes
an
integration
of
seismic
geomorphology
1576
(i.e.
plan
views)
and
seismic
stratigraphy
(i.e.
section
views).
Interpretation
of
3D
seismic
volumes
can
involve
a
variety
of
analytical
techniques
ranging
from
mapping,
visualizing
and
characterizing
attributes
of
seismic
horizons,
to
performing
amplitude
extractions
from
seismic
slices,
to
characterizing
seismic
attributes
of
seismic
intervals.
Knowledge
of
map
view
expression
of
a
range
of
depositional
elements
as
well
as
a
sound
understanding
of
stratigraphic
architecture
is
essential
for
this
approach.
From
an
exploration
perspective,
the
identification
of
depositional
elements
facilitates
lithology
prediction
because
most
depositional
elements
have
a
somewhat
predictable
distribution
of
rock
types.
I
wish
to
thank
reviewers
C.
Rossen
and
R.
Davies
for
their
comments,
which
helped
to
improve
the
manuscript.
I
would
also
like
to
thank
Paradigm
Geophysical
and
in
particular
P.
Lepper
for
their
patient
support
and
guidance
in
the
use
of
their
interpretation
applications,
Stratimagic
and
VoxelGeo.
Thanks
are
also
due
to
Anadarko
Petroleum
Corporation
for
permission
to
publish
this
manuscript.
Finally,
to
T.
Lemon,
for
helping
me
through
many
a
software
and
hardware
maze,
always
with
good
cheer
and
humour,
thanks
for
all
the
help.
References
McGovney,
J.
E.
&
Radovitch,
B.
J.
1985.
Seismic
stratigraphy
and
facies
of
the
Frigg
Fan
complex.
In:
Berg,
0.
R.
&
Woolverton,
D.
G.
(eds)
Seismic
Stratigraphy
II:
an
Integrated
Approach
to
Hydrocarbon
Exploration.
American
Association
of
Petroleum
Geologists
Memoir,
39,
139-154.
Posamentier,
H.W.
2000.
Seismic
stratigraphy
into
the
next
millenium;
a
focus
on
3D
seismic
data.
American
Association
of
Petroleum
Geologists
Annual
Conference,
New
Orleans,
IA,
April
16-19,
2000,
A118.
Posamentier,
H.
W.
&
Kolla,
V.
2003.
Seismic
Geomorphology
and
stratigraphy
of
depositional
elements
in
deep-water
settings.
Journal
of
Sedimentary
Research,
73,
367-388.
Vail,
P.
R.,
Mitchum,
R.
M.
Jr.
&
Thompson,
S.
BI
1977.
Seismic
stratigraphy
and
global
changes
of
sea
level,
part
3:
relative
changes
of
sea
level
from
coastal
onlap.
In:
Payton,
C.
E.
(ed.)
Seismic
Stratigraphy—Applications
to
Hydrocarbon
Exploration.
American
Association
of
Petroleum
Geologists
Memoir,
26,
63-81.
Weimer,
P.
&
Davis,
T.
L.
1996.
Applications
of
3-D
seismic
data
to
exploration
and
production.
American
Association
of
Petroleum
Geologists
Studies
in
Geology,
No.
42,
Society
of
Exploration
Geophysicists
Geophysical
Development
Series,
5.