SPE 81526
Seismic Facies Analysis for Fracture Detection : a Powerful Technique
Gérard Bloch, Maged El Deeb, Hussein Badaam, ADCO, UAE, Frédéric Cailly, Gael lecante, Olivier Fonta,
Antoine Meunier, Beicip-Franlab, France
This paper was prepared for presentation at the SPE 13th Middle East Oil Show
& Conference to be held in Bahrain 5-8 April 2003.
This paper was selected for presentation by an SPE Program Committee
following review of information contained in an abstract submitted by the
author(s). Contents of the paper, as presented, have not been reviewed by the
Society of Petroleum Engineers and are subject to correction by the author(s).
The material, as presented, does not necessarily reflect any position of the
Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the
Society of Petroleum Engineers. Electronic reproduction, distribution, or storage
of any part of this paper for commercial purposes without the written consent of
the Society of Petroleum Engineers is prohibited. Permission to reproduce in
print is restricted to an abstract of not more than 300 words; illustrations may not
be copied. The abstract must contain conspicuous acknowledgment of where
and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836,
Richardson, TX 75083-3836, and USA, fax 01-972-952-9435.
Summary
Fractured reservoir analysis requires an accurate
delineation of fracture corridors, as well as an estimate of
the fracture intensity. 3D seismic data is of great help,
fractures being often expressed as very subtle zones of
deformation (discontinuities), with no obvious offset of
seismic reflections, but inducing variations in the seismic
image.
Actually, a large number of seismic attributes are known to
highlight discontinuities. They can be horizon-based, such
as curvature, dip, edge, azimuth, or based on the 3D
seismic trace variability in a given neighborhood such as
“coherence-type” attributes, or computed after Fourier or
Hilbert transform, such as instantaneous phase or
frequency.
This paper proposes to use jointly a set of selected “fracture
relevant” attributes in a multi-variable statistical process
called Seismic Facies Analysis (SFA). This methodology is
demonstrate in a large Upper Cretaceous carbonate
reservoir onshore UAE. We obtain a seismic facies map
that delineates areas spatially oriented towards two
directions consistent with the general structural knowledge
of the field. This map can be interpreted as correlated with
reservoir fracture properties.
Analysis of the seismic attribute distributions within each
seismic facies allows a preliminary interpretation of these
seismic facies in terms of fracture occurrence. The
validation of this interpretation with well data (BHI, core
and dynamic data) delivers a reliable map of fracture
intensity. This map is then used to interpret structural
lineaments, and to constrain stochastic realizations of a
fracture model.
Introduction
An update of the fracture model of an Upper Cretaceous
reservoir was planned in 2002 by Abu Dhabi Onshore Oil
Operations Company (ADCO). The objective of the study
was to compute the equivalent fracture parameters (fracture
porosity, fracture permeability and block sizes) required for
the reservoir simulation. A close integration of geological,
geophysical and dynamic data was carried out using the
workflows implemented in the FRACA software (Cacas et
al., 2001).
The task detailed in this paper corresponds to one of the
beginning steps of this workflow. It aims at providing a
fracture intensity map that will be used to pick a fracture
lineament network, and to seismically constrain a stochastic
fracture model. Our principal objective was to obtain the
maximum benefit from the conventional post-stack 3D
seismic data set.
In this point of view, we use Seismic Facies Analysis
(SFA) as a methodology to integrate information from
different seismic attributes and, to provide spatially
delineated areas (“seismic facies”) related to fracture
properties variations. This seismic facies map is then
interpreted as a fracture intensity map.
The 3D seismic data set (reflectivity, fig.1) has been
acquired and processed in 1994 by ADCO (bin size
25mx25m, 32 fold). A reprocessing took place in 1998. It
was followed by a powerful post-processing noise
reduction step (Shell’s software SOF, “Surface Oriented
Filter” applied in 1999). This latest technique enhances the
signal to noise ratio of the data whilst preserving the subtle
discontinuities (edge preservation algorithm). The quality
of the seismic data has been significantly improved and
different type of fracture-related information can be
reliably extracted from the 3D seismic data set with
different algorithms.
Seismic Facies Analysis Principles
Seismic Facies Analysis (SFA) analyzes automatically the
character of the seismic traces, in a given reservoir
window, to generate “seismic facies” maps (2D analysis) or
2
G. Block, M. El Deeb, H. Badaam (ADCO), F. Cailly, G. Lecante, O. Fonta, A. Meunier (Beicip-Franlab)
cubes (3D analysis). These maps or cubes highlight the
variation of the seismic pattern throughout the 3D seismic
survey. The sub-sequent interpretation step allows to relate
these variations to geological properties variations of the
reservoir, here structural information.
In practical, for 2D analyses, each trace (inline/crossline
location) over the reservoir interval is characterized by a
series of seismic attributes (Déquirez et al., 1995). In this
framework, statistical cluster analyses are carried out for
gathering similar traces (i.e. traces that are sharing similar
global attribute responses). Each group of traces
corresponds to a particular “seismic facies”.
Two complementary approaches are possible. Firstly, the
supervised analysis consists in the use of geological
information, through training traces around well locations,
to guide the facies determination. This approach allows a
straightforward interpretation of the resulting seismic facies
maps, but requires a sufficient number of wells to be
carried out. Conversely, the non-supervised analysis does
not use a geological a-priori and is based on cluster
analysis, carried out in the attribute space. In this case, well
information is used for a-posteriori interpretation of the
obtained seismic facies maps.
In this study, we focus on the non-supervised scheme.
Seismic Facies Map Building
Extracting and selecting a set of seismic attributes is the
first task of the SFA. A rather limited number of welladapted attributes is preferred to a large amount of non
relevant, redundant information. Different sets of attributes
have hence to be tested.
SPE 81526
Dip, edge and azimuth have the advantage of being easily
and quickly available from any seismic interpretation
workstation. They are usually computed from the top of the
reservoir interpretation.
Different
coherency-type
attributes
exist.
The
“dissimilarity” we used is computed from the seismic data
set by comparing each trace with all eight surrounding
traces through 3D semblance analysis, in the reservoir
window. It is also extracted from usual workstations.
All these attributes can be related to the occurrence of
fractures. However, their variations do not depend on
fractures only, and may be partially corrupted by seismic
noise or other (geological?) effects. For this reason, the
characterization of “sub-seismic faults” is not based on one
of these attributes but on their joint analysis in order to
extract their relevant part only. By characterizing each
seismic trace according to all its attribute responses, SFA is
a powerful tool for this integrated interpretation.
Several tests allowed to identify an optimum set of seismic
attributes, composed of the edge, the 3D trace dissimilarity
and the curvature (fig.2). Dip information was not used
because it was too redundant with edge (the correlation
coefficient between these two variables was up to 0.95).
A non supervised SFA was run and the resulting 8 seismic
facies map is shown on figure 3.
The seismic facies map highlights structures (essentially
seismic facies 2, 3, 7 and 8) spatially oriented towards two
directions (N40 and N70). These directions are consistent
with the general structural knowledge of the field, and the
map can be assumed to be correlated with reservoir fracture
properties.
Seismic facies map interpretation
Classical attributes based on trace amplitude or spectrum
characteristics were tested and quickly interpreted as non
relevant for fracture detection in this reservoir. They
appeared to be much more well-suited for predicting the
lateral lithofacies distribution.
An extraction of more “fracture-related” attributes was
therefore necessary.
Since the relationship between areas of high curvature and
fracture density have been observed in a large number of
reservoirs (Rijks & Jauffred, 1991), a precise curvature
analysis at the top reservoir was performed. It consist in the
extraction of a robust curvature map after an adapted
horizon smoothing. Curvature artifacts due to seismic fault
throws are analyzed and withdrawn. The final curvature is
relevant of folding and must highlight fractures as
developed in the “extrado” fracture set concept.
Several other seismic attributes as dip (magnitude of the
time gradient), edge (differences in dip across a horizon),
azimuth (direction of maximum dip) and coherency are
well known by the geophysicist for structural interpretation.
The interpretation of each seismic facies in terms of
fracture meaning must be done using well data (BHI, core
and dynamic data). However, the number of wells is
limited and some facies can correspond to areas that have
not been drilled. It is therefore quite difficult to deliver an
exhaustive interpretation of all the identified facies.
The first analysis we propose takes into consideration the
fracture meaning of each seismic attributes that have been
used in the SFA procedure:
This interpretation is based on 3 main assessments :
• High positive values of curvature can be associated
to fracture;
• Fracture can be expressed on seismic data by a high
dissimilarity of neighborhood seismic traces;
• High values of edge can be correlated to fracture
zones.
SPE 81526
_
Seismic Facies Analysis for Fracture Detection: A Powerful Technique
Obviously, for reasons already detailed above, the
interpretation of the seismic facies based on the separated
use of each of these assessments can be misleading and
may lead to erroneous conclusions. However, their
combination is an interesting index, which is much more
promising.
Thus, the distribution of seismic attributes within each
seismic facies is analyzed (fig.3) and interpreted in terms of
probable high or low fracture content :
• seismic facies 4 (cyan) and seismic facies 5 (pink)
always display low attribute values (low or
negative curvature, low dissimilarity, low dip and
edge values). They are assumed to correspond to
non fractured areas.
• seismic facies 2 (gray) and 7 (green) display high
positive curvature values and high dissimilarity
values. They correspond to fractured zones. This
is confirmed by high values of dissimilarity.
• seismic facies 6 (violet) is characterized by high
negative values of curvature. It is interpreted as
non- or weakly fractured areas. However medium
values of other indexes highlight that the fracture
potential is nevertheless not null. It has to be
more investigated through a comparison with
well data.
• seismic facies 8 (dark blue) displays globally null
curvature values. Based on a sole curvature
analysis (that is sometimes run), this seismic
facies would have been considered as non
fractured. However it shows high seismic
attribute values (mainly dissimilarity). Therefore,
it might be interpreted as potentially fractured
areas, to be confirmed by well data.
• seismic facies 3 (brown) displays relatively high
positive curvature values. However low values of
other attributes imply that this seismic facies
might not be heavily fractured.
• seismic facies 1 (yellow) is more difficult to
interpret. It is mainly located in the west part of
the area and associated to an important dip
(relatively to other part of the field). Curvature
values are small, it does not display any
preferential trend and consequently it is
interpreted as a non-fractured facies.
This analysis, based on geophysical data only, provides a
quick and exhaustive interpretation of each seismic facies.
However, without a well calibration it should remain
doubtful.
This essential validation step is facilitated on this data set
by the presence of several wells with cores, BHI and
dynamic results.
The well trajectories are projected on the seismic facies
map. Fracture orientation and intensity along well
3
trajectories are analyzed for each encountered seismic
facies (fig.4).
It confirms the reliability of the preliminary interpretation:
1. trajectories of highly fractured wells are across
seismic facies 2, 7 or 8, and trajectories of poorly
fractured wells are across seismic facies 1, 4, 5 or
6;
2. The fracture orientations detected at wells are
consistent with the main orientation of the
seismic facies distribution (N40 and N70);
3. Seismic facies 2 is confirmed to be the most
fractured;
4. Seismic facies 8 is a fractured facies although it
exhibits low curvature values.
According to this interpretation, a fracture intensity index
can be given to each seismic facies to generate a fracture
intensity map, that is used in FRACA software, to pick
precisely fracture lineaments and to constrain a stochastic
fracture modeling.
Conclusion
Seismic Facies Analysis is successful in identifying and
characterizing the fractured zones in one single process
using a multi-attribute analysis. The obtained seismic facies
map is expected to be more reliable than any single
attribute map, including curvature. Such a map also brings
more information than just the location of the fractured
zones, as it can also deliver a qualitative estimate of the
fracture intensity.
The seismic facies map highlights the variation of the
fracture properties within the reservoir. It is then used in
FRACA software, (1) for a precise picking of fracture
lineaments and (2) as a constraint for a stochastic fracture
modeling step.
The fracture-related seismic attributes we used are
extracted from conventional post-stack seismic data only. It
allows the methodology to be used with any 3D seismic
data set. However, the SFA can be run with pre-stack
seismic attributes and benefits from seismic anisotropy and
AVAz effects when such data are available.
References
Cacas, M.C., Daniel, J.M. and Letouzey, J., 2001, Nested
geological modeling of naturally fractured reservoirs,
Petroleum Geosciences, vol. 7 2001, pp. S43-S52.
Déquirez, P.-Y., Fournier, F., Blanchet, C., Feuchtwanger,
T. and Torriero, D., 1995, Integrated stratigraphic and
lithologic interpretation of the East-Senlac heavy oil pool,
65th Annual SEG Meeting Expanded Abstracts, 104-107.
4
G. Block, M. El Deeb, H. Badaam (ADCO), F. Cailly, G. Lecante, O. Fonta, A. Meunier (Beicip-Franlab)
Rijks E.J.H. and Jauffred J.C.E.M, 1991, Attribute
extraction : An important application in any detailed 3D
NW
interpretation study, The Leading Edge, 10, 11-19.
Probable Fault (sub-seismic)
Probable Fracture Zone
Fault
SPE 81526
SE
2000 ft
TWT
ms
Well A
Horiz.
Well A
Vert.
Well C
Vert.
Well B
Vert.
Top
Reservoir
TD
TD
Base
Reservoir
TD
Reproc. 99 Reflectivity Data
Figure 1 Seismic cross-line (Reflectivity) with well data
Curvature
N
Dissimilarity
Edge
1250
1250
1250
1250
1250
1250
1000
1000
1000
1000
1000
1000
750
750
750
750
750
750
500
500
500
500
500
500
250
250
250
250
250
250
High
Low
cdp
150
250
350
line
Figure 2 Seismic attribute maps
450
150
250
350
450
150
250
350
450
SPE 81526
_
N
1250
1250
Seismic Facies Analysis for Fracture Detection: A Powerful Technique
Seismic
Facies
1000
Globally null
1 yellow
2 gray
750
750
500
Fracture
potential
Characteristics
Curvature
1000
5
Dissimilarity
Edge
Dip
Medium
High
High
Very high
positive
Very high
Medium
positive
Low to
medium
3 brown
Medium to
high
Low
Medium to
high
Very high
Low
Low
Low
Low
Low
Very low
Very low
Low
4 cyan
Low to medium
negative
5 pink
Globally null
Medium
Medium
Medium
6 violet
High negative
Medium
Medium
Medium
Low
7 green
High positive
Medium
Medium
High
8 dark
blue
Globally null
Medium
Medium
High
500
250
250
cdp
line 150 250
350
Medium to
high
High
450
Figure 3 Seismic facies map and seismic attribute distribution analysis per seismic facies
Seismic facies with high potential
of fracture occurrence
1250
1250
1000
1000
N
Seismic cross-line (fig.1)
N
750
non fractured
PI = 7
750
moderatly fractured
PI = 22
High water cut
500
500
highly fractured
PI = 33
250
250
poorly fractured
Water at the last 500ft
cdp
line
150
250
350
450
Figure 4 Seismic facies validation with well data
highly fractured
PI = 78