Copyright 2004, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the SPE Annual Technique Conference and
Exhibition held in Houston, Texas, U.S.A., 26-29 September 2004.

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, U.S.A., fax 01-972-952-9435.

Abstract
In this paper, we present a combined geological, geophysical
and rock mechanics approach to natural fractured reservoir
characterization. The local structure entropy analysis on 3D
seismic data is used to detect distributions of fault and sub-
fault systems. The curvature attribute along with modeled
strain and stress field, constrained with the log data measuring
P- and S-wave velocities and rock density and the inverted
elastic modulus from pre-stack seismic data, reveal effects of
the geological structure, bed thickness and lithology on
fracturability of the reservoir layer. These analyses quantify
the relationships between the geologic factors and rock
fracturability and describe physically the weighting factors for
geologic parameters in controlling the rock fracturing. The
comparison of the seismic azimuthal analysis results to these
of geological and rock mechanics modeling provides an
opportunity to verify whether the seismic anisotropy derived
from seismic data is caused by structure related natural
fracture patterns or by other mechanisms. The consistency
among different techniques provides the confidence in the
interpretation of the distribution of fractures induced by
structures. If azimuthal seismic attribute data can be
combined, the application of this procedure results in the
development of the fracture connectivity anisotropy by
considering relationships between the present and palaeo-
stress fields. In addition, the scale depend analysis technique
in this approach can improve the ability to identify the
distribution of fractures with multiple length scales. In this
paper, case studies are used to illustrate applications of these
technologies and their efficiency.
Introduction
Fractures are a crucial factor controlling the well performance
in reservoirs with low permeability. Fractured reservoirs are
highly heterogeneous. A practical approach in modeling of
naturally fractured reservoir is from geology to flow
simulation, which consists of two parts, characterization of
fracture geometry and fracture network dynamic behavior.
The time-independent static data is mainly used for
characterizing fracture distribution. Dependent on scale at
which fracture network is observed, these data include
wireline logs, conventional cores, sub-seismic investigation
(outcrop and structure data) and seismic data. The spatial
distributions of fractures and their geometry at the field scale
derived from static modeling plays important role in the
reservoir modeling.

Studies show that structure, lithology, bed thickness, porosity,
and other geologic factors control the fracture intensity. The
complex interaction between all geologic factors needs to be
considered during the static modeling in order to make these
models to be expected input to the reservoir models. The rock
properties of a reservoir layer depend primarily on the
depositional process. Tectonic events act on the reservoir layer
can lead to the spatial heterogeneous structures which resulted
from the heterogeneously distributed mechanical properties.
The fracturability of a reservoir layer is not only influenced by
mechanical properties, which are controlled by the lithology of
reservoir rocks such as shale content, matrix porosity, and
carbonate contents etc, but also influenced by the tectonic
events that the reservoir rocks have undergone.

To reduce the uncertainties in fractured reservoir modeling,
integrating all available data from geology, geophysics,
petrophysics and engineering is necessary. Since the
geological heterogeneity significantly controls fracture
distributions, the statistic assumptions between two or more
wells are hard to be made. Borehole data and analysis of core
data provide us the direct observation of fractures. The
fracture detecting can benefit from the 3D seismic data and
observed geological data.

Seismic data can be used to get better geologic models and
make geo-mechanical modeling physically efficient and
geologically meaningful. The discontinuity analysis or a
coherence-type attribute is often used to interpret the faults
and fractures at the seismic scale. The local discontinuity
related seismic attribute is computed from the seismic data set
comparing each trace with surrounding traces. An analysis
cube can be selected by the interpreter, according to the type
of geological feature that is of interested.

The curvature analysis at the top of the reservoir can be
performed and a robust curvature map can be extracted from

SPE 90275
A Combined Geological, Geophysical and Rock Mechanics Approach to Naturally
Fractured Reservoir Characterization and Its Applications
Feng Shen and Shuiquan Li, SPE, EP Tech
2 SPE 90275
structural interpretation
1
. The relationship between area of
high curvature and fracture density has been observed in a
large number of reservoirs in the Middle East, in the North
Sea and in North America
2,3
. The positive curvature value may
indicate the possible fracture locations whereas negative
curvature may indicate the non-fracture locations. The
curvature map only reflects the possibility of having fractures.
Resultant variations in curvature, strain and stress are the
results of displacement on faults, folding of the structure and
heterogeneity of lithological properties.

Seismic anisotropy analysis is an efficient tool to characterize
fracture anisotropy. The comparison of the seismic azimuthal
analysis results to these from geological and rock mechanical
studies provides an opportunity to verify whether the seismic
anisotropy derived from seismic data is caused by structure
related natural fracture patterns or by other mechanisms. The
consistency among different techniques provides the
confidence in the interpretation of the distribution of fractures
induced by structures. Characterizing fractured reservoirs
needs integrated seismic analyses with all geological and
mechanical studies.

Characterizing fracture distributions and their geometry is
very important and is a basis for flow simulation of fracture
networks. Most naturally fractured reservoirs include fractures
with multiple length scales. The fracture connectivities and
their distributions will determine fluid transport in a naturally
fractured reservoir
4
. Field characterization studies have shown
that fracture systems are very irregular, often disconnected and
occur in swarms
5
. Incorporating the multiple length scale
fractures into the flow simulation will significantly improve
the modeling technology for fractured reservoirs.

In this paper, we propose a combined geological, geophysical
and rock mechanics approach to naturally fractured reservoir
characterization. With the case study, we demonstrate the
application of our fracture modeling technology to fractured
sandstone reservoir characterization. The scale depend
analysis technique used in this approach greatly improves the
ability to identify the distribution of fractures with multiple
length scales.

3D Seismic Data Multiscale Discontinuity Detection by
Using Local Structure Entropy
One of the most challenging tasks characterizing fractured
reservoirs is locating faults and sub-faults within 3D seismic
volume. The transmissibility features of fault system and their
scales are significant since they are often associated with
reservoir flow simulation. A major step forward in the
interpretation of 3D seismic data was the introduction of the
coherence cube by Bahorich and Farmer
6
. The disadvantage is
its lacking robustness when dealing with noisy data
7
.
Gersztenkorn and Marfurt
8
introduced a coherence estimate
based on an eigenstructure approach. It was shown that this
approach provides a more robust measure of coherence. Its
main drawback is the expansive calculations of dominant
eigenvalues. Cohen and Coifman
9
proposed an analysis
method for the estimation of seismic local structural entropy
which is both robust to noise and computationally efficient.
This method avoids the computation of large covariance
metrics and their dominant eigenvalues. Based on this work,
multiple scale discontinuity analyses are applied to mapping
fault and sub-fault systems.

Large scale structure features such as fault system are
analyzed with larger analysis data cubes. Stratigraphic features
such as channels and sub-faults with shorter vertical duration
need smaller analysis cubes to characterize their spatial
discontinuity. When seismic local structural entropy is
calculated at the small scale data volume, the calculation is
sensitive to smaller-scale discontinuity and provides the
sharper image of seismic discontinuities. When the scale of
data volume is increased, the effect of the large scale faults
becomes dominant. The smaller scales of local structure
entropy calculations should indicate smaller faults and their
features. We use different scale analysis data cubes with
surface windows [2 2], [4 4] and [6 6], where the two numbers
between the square brackets designate, respectively, the
number of seismic inline traces and crossline traces (Figure 1).
The local structural entropy values are mapped; larger values
indicate greater discontinuity. Our results clearly show that the
smallest analysis cube yields the sharpest image of
discontinuity. The detailed structures of fault networks and
their features on the time slice are observable. The
transmissibility of the fault system is related to these features.
Furthermore, when the analysis data cube becomes large, the
large scale fault frames are outlined clearly. The sensitivity to
the size of the data volume depends on the size of the fault
system. The features of each fault segment can be maximized
observed on the discontinuity volume whose volume size is
sensitive to the fault scale. By combining local structure
entropy using various sizes of analysis cubes, high lateral
resolutions can be obtained. The multiscale local structure
entropy volumes emphasize points which likely correspond to
fault surfaces.

Multiscale Curvature Analysis and Related Strain
Properties
Almost all of techniques related to fracture detection consider
the present structures interpreted from the 3D seismic data. It
is difficult for us to have the knowledge regarding the exact
the tectonic history of the reservoir layer, while searching the
detailed information related to fracture distribution from the
present structural of the formation is possible. Curvature
analysis has a long history of use in prediction of fractures
1,10
.
It is usually accepted that where curvature exists, the fracture
can exist. The principal curvatures are directly proportional to
the principal strains, if it is assumed that the rock is elastic and
that all of the deformation associated with the curvature is
dilatational.

When the curvatures are calculated at the smallest mapping
scale, they are greatly affected by the displacement of fault
systems. When the curvature calculation scale is increased, the
fault effect is reduced and large scale structures play important
roles. The large scales of curvature calculations should
indicate the occurrence of small fractures and related to
regional structural features. On the other hand, the small scales
of curvature calculations are closely related to local structures.
SPE 90275 3
To identify multiple sizes of fractures, various scales of
curvatures should be considered. Our calculations show that
the combination of different scales can reveal the possibility of
occurred fractures.

To investigate the effects of a specific fault on fracture
distributions and fluid flow transmissibility, scale-dependent
seismic discontinuity measures need to be combined with
mapped curvatures along fault segments. The curvature maps
in specified directions and sizes can improve the ability to
analyze and interpret the fault properties. The calculated 3D
curvatures at different scales and directions improve our
ability to detect the fractures induced by the fault system.

Strain is linearly related to surface curvature and distance from
the neutral surface within the reservoir layer
11
. The positive
strain value may indicate the possible fracture locations
whereas extension may occur. With the increasing strain,
rocks undergoing fracturing exhibit increased fracture
intensity. The areas where faults could influence reservoir
properties should be mapped in the fractured reservoir
characterization, when faults are suspected to induce
significant strain accommodated, or when heterogeneous
strain is accommodated by a combination of faulting. Fault
orientation and its network structures contain useful
information. In addition, the effects of fault network on the
deformation are characterized by a fault density map
computed by summing the fault lengths weighted by their
throws in a moving window
12
. This measure has the same
dimension as strain. The detailed interpretation of the fault
related strain field is given by Gauthier at al
12
.

In our study, faults are mapped from 3D seismic data with
multiple sizes of data volumes. The scale dependent
discontinuity analysis is able to reveal subtle lineaments than
any of the other conventional coherence methods in the
reservoir layer and is able efficiently to highlight fault network
geometry and spatial variations. The calculated strain from
curvatures at different scales and directions can be combined
with these constructed by the fault throw along fault traces to
predict variations in fracture distributions. Using this
approach, the observed and quantified fractures from wells
and image log interpretations can be used to constraint the
interpretation of fracture presence. Figure 2 shows the strain
map.

Effects of Lithology Heterogeneity on Stress Field
The fracturability of the reservoir rocks is a very important
indicator for the fractured reservoir characterization. Nelson
13

documented the geological influences on the fracture intensity
in sedimentary layered rock such as composition, grain size,
porosity, bed thickness and structural position. In addition to
the effects of structures, lithology/facies, shale contents and
porosity play an important role in determining the fracture
intensity and controlling the rock mechanical properties. If the
fractures are induced by extensional deformation, fractures
could better develop in the brittle rocks than in ductile
reservoir rocks. The rock fracturability increases with the
increasing of rock brittleness characterized by large elastic
modulus of rock matrix. The increase of dolomitic contents
and decrease in shale contents and porosity can enhance the
rock fracturability. The heterogeneity of rock fracturability
leads to the spatial variations of fracture intensity.

The fracture intensity is related to strain energy originally
stored in the rock, which is defined as the half of stress
component multiplied with resulting strain component. In
general, the rock with relatively high strain energy will have
large fracture intensity than a rock of equal thickness with
relatively low calculated strain energy. The historical stress
magnitudes and directions are related to heterogeneous
lithology/facies distribution and stain. Mapping the
lithology/facies of the reservoir layer, i.e. distributions of 3D
mechanical properties, can make the geomechanical modeling
geologically meaningful. The 3D heterogeneity of
lithology/facies can be described with elastic modulus of
rocks, i.e. with P- and S-wave velocities and rock density. To
obtain spatial variations of P- and S-wave velocities and rock
density of the reservoir, elastic impedance inversion proposed
by Connolley
14
provides a consistent and absolute framework
to calibrate and invert prestack seismic data to obtain the rock
elastic modulus.

Figure 3 shows the distributions of maximum principal stress
magnitudes and directions. Compared to Figure 2, the
different properties shown on Figure 3 indicate the effects of
heterogeneity of 3D mechanical properties of reservoir rocks.
In general, fractures will align with the maximum stress
directions. The present-day maximum horizontal stress can
modify the anisotropy of fracture connectivity. If the present-
day maximum horizontal stresses are in the same directions as
calculated stresses derived from structure and lithology data,
the connectivity of fractures is enhanced. On the other hand, if
the present-day maximum horizontal stresses are
perpendicular, the fracture connectivity will be reduced.

Fracture Detection Based on 3D Azimuthal Seismic Data
Analysis
Numerical simulations and physical experiments show that the
existence of vertically aligned fractures leads to azimuthal
amplitude versus offset (AVO) variations, which can be used
to determine the principal direction of fractures. Lynn et
al.
15
present field data examples showing a correlation between
azimuthal differences in P-wave AVO and S-wave anisotropy
and demonstrating the potential of using P-waves to
characterize fractured reservoirs. Based on a 3-D AVO
technique, Ramos and Davis
16
show spatial variations of
fracture intensity in coal-bed methane reservoirs. Shen et
al.
17
use azimuthal offset dependence seismic amplitude and
frequency attributes to detect fractures in carbonate reservoirs.

The azimuthal variation of P-wave reflection amplitudes
(coefficients) is an important seismic characteristic for
detecting fractures. Seismic anisotropy analysis is an efficient
tool to characterize fracture distributions. Because of the
effects of elastic properties of background rocks, saturated
fluids and overburden anisotropy, anisotropy caused by high
angle fractures does not necessarily correlate with the
magnitude of fracture density. However, the seismic
anisotropy can be used to detect the presence of fractures and
4 SPE 90275
map fracture distributions spatially. Due to the limited
frequency band, seismic data can provide limited lateral and
vertical resolution and help us identify fractures with the
dominant effects. However, geologic data indicates that
fractures usually occur in multiple sets, and engineering
models require that more than one fracture set occurs before
fractures form networks capable of transmitting fluids.
Considering the uncertainties in seismic data analysis and
interpretation, fractured reservoir characterization by using
azimuthal amplitude variations in structural reservoirs needs to
combine other data sources to quantify weather spatial
variations of seismic anisotropy were caused by the natural
fracture patterns or were an artifact of the processing. To
better interpret the seismic anisotropy for fracture distribution,
it needs to understand natural fracture spatial geometry and
geological mechanism.

The comparison of the seismic azimuthal analysis results to
these of geological and rock mechanics modeling provides an
opportunity to verify whether the seismic anisotropy derived
from seismic data is caused by structure related natural
fracture patterns or by other mechanisms. The consistency
among different techniques provides the confidence in the
interpretation of the distribution of fractures induced by
structures. Characterizing fractured reservoirs needs integrated
seismic analysis tools to characterize all possible geological
and mechanical influences.

Application of the Combined Geological, Geophysical and
Rock Mechanics Approach to Naturally Fractured
Reservoir Characterization

Geological Modeling of Fractured Sandstone Reservoir
The fractured sandstone reservoirs to be studied in our
working area are distributed in the Quantou Formation,
deposited in the lower Cretaceous in the Songliao Basin of
China. 3-D seismic surveys are acquired over the area. The
reservoir is located at 1200-1500 m in depth and its total
thickness is about 130 m. The reservoir intervals consist of
sandstone and shale and belong to channel and delta deposits.
The structural high is cut by normal faults trending north
south. Faulting system within the Quantou Formations is
complex. Drilling results and production data reveal the
fractured reservoirs developed in sandstone reservoirs.
Geological modeling of sandstone reservoirs in the Quantou
Formation is very important for detecting and predicting the
occurrence of fractures, understanding the effects of fractures
on reservoir properties and helping reservoir evaluation and
management.

Spectral decomposition provides a tool for imaging and
mapping temporal bed thickness and geological discontinuities
over large 3-D seismic surveys
18
. This technique has been
used to delineate facies and depositional environments such as
channel sands and incised valley-fill sands
19
. We apply our
instantaneous spectral analysis technique based on wavelet
transform algorithm to image the reservoir distribution. The
methodology of the technology is discussed by Shen et al
20
.
Mono-frequency energy data generated with this technology
from the seismic data are used to delineate temporal reservoir
thickness variability and reveal lateral geological
discontinuities. We consider the optimum frequency band of
each set of reservoirs based on our well log modeling and
quantify them in the time and frequency domains. Figure 4
shows imaged reservoirs with energy spectrum in the
Quandou Formation. The sandstone reservoirs are composed
of two channel systems located in the north and south,
respectively. Channels are oriented in the east-west direction,
which is consistent with the data of paleo-current studies in
this field. The impedance inversion technique is applied to
obtaining the rock mechanical properties.

The seismic discontinuity analysis reveals the fault structures
on the top of the reservoir (Figure 5). The curvature attribute
along with modeled strain and stress field, constrained with
the geological structure, lithology and layer thickness, are
used for fracture detection. Strain distribution is localized
along faults in the N-S direction. There is localized strain
oriented along the E-W direction, which is observable in the
southern part (Figure 6). E-W orientated strain distribution
indicates that local structures could be responsible for not only
north-south oriented fractures but also E-W oriented ones.

Seismic Derived Geophysical Modeling of Fractured
Sandstone Reservoir
Our seismic data processing focuses on preserving the seismic
amplitudes. Based on the offset and azimuth coverage of
seismic data on the target zone, the field seismic data is sorted
as six-azimuth data. Understanding how the fractures cause
azimuthal amplitude variations is essential to the analysis of
seismic anisotropy and to the determination of the fracture
orientation. To identify the effects of fracture density and fluid
content on the azimuthal AVO responses, the forward seismic
modeling is very helpful to relate the seismic responses to
fracture properties. The modeled results will provide us how
the fluid contents and fracture density influence the azimuthal
seismic responses and what relationships between the seismic
anisotropy and fracture orientation could be.

The seismic and log tie shows that the strong reflectivity
occurs at the top of fractured reservoirs. The modeled
azimuthal amplitude versus offset (AVO) responses show that
the presence of fractures causes the largest amplitude decay
with offset (Figure 7). The maximum AVO gradient occurs in
the fracture strike direction. The modeled results show that the
orientation of amplitude anisotropy indicates the fracture
normal direction and oil saturated fractured reservoirs have
smaller seismic anisotropy than that of water saturated ones
when the fracture density is the same. With the modeled
studies, the fracture orientations are consistent with the
directions of minimum axis of a amplitude ellipse.
The orientation of fractures can be derived from the azimuthal
analysis of calibrated seismic amplitudes in six azimuths. The
orientation of the minimum axis of the seismic amplitude
ellipse indicates the fracture orientation. Meanwhile, the ratio
of elliptic maximum axis to the minimum axis is one of the
main seismic attributes to the determination of fracture
density.


SPE 90275 5
I ntegrated Fracture Modeling in the Sandstone Reservoir
To characterize fracture density is a challenge process and
needs to relate quantitatively seismic attributes to fractured
reservoir properties. Seismic attributes used to quantify
fracture density should be statistically significant and robust,
geologically meaningful and physically effective. Theoretical
studies on scattering characteristics in the heterogeneously
fractured reservoirs provide a basis to relate the seismic
attenuation to the spatial variations in fracture density field
21
.
In addition to amplitude variations with offset, the
heterogeneity of vertically aligned fractures will lead to the
attenuation of the high frequency of seismic energy at the far
offset at the top of fractured reservoirs. The offset-depend
amplitude and frequency attributes can also be used to
characterize fractures
22
. In addition to the attribute of
anisotropic ratios extracted from seismic amplitude anisotropy
and modeled strain field, attributes related to wave attenuation
caused by fractures and their heterogeneity can also be used to
characterize fracture density.

Figure 8 shows the map of fracture density and orientation in
the working area. Our results show that fracture distribution is
mainly controlled by local structure properties and reservoir
lihtological properties. E-W orientated fractures play
important roles in connecting flow units within the channel
reservoirs. Although N-S orientated fractures, perpendicular to
the channel direction, are well developed, effects of these
fractures on the fluid flow within reservoirs are very limited.
Production data also show that wells located in the area with
highly developed E-W orientated fractures have high fluid
production rate.

Discussion and Conclusions
To reduce the uncertainties in fractured reservoir modeling,
integrating all available data from geology, geophysics,
petrophysics and engineering is necessary. Since the fracture
distributions are characterized by multiple length scales and
orientations, the scale depend analysis technique should be
considered to improve the fracture detection. The fracture
detecting can benefit from the geomechanical modeling,
discontinuity detection and azimuthal seismic anisotropy of
3D seismic data. The consistent results from different data
resources and techniques provide the confidence in the
interpretation of the distribution of fractures induced by
structures.
The case study has demonstrated our workflow and
technologies used in fractured reservoir characterization.
Fracturing is a complicated geological phenomenon controlled
by geological and geomechanical factors. Modeling reservoir
sedimentary facies, distribution and spatial lithology
properties is the first step before fractured reservoir
characterization. In our working area, fractures are a major
contribution to azimuthal AVO variations. Our modeled
results show that seismic anisotropy increases with fracture
density and that the fracture striking direction corresponds to
the minimum gradient of amplitude decay with offset. The
heterogeneity of lithology is one of the major factors
influencing fracture distributions. Integrating rock physics
modeling and borehole data, attributes obtained from seismic
anisotropy analysis, geomechanical modeling and seismic
attenuation attributes could be used to calibrate and quantify
fractures under the well constraints. Our results show that
fracture orientation and density distributions are controlled by
structures and suggest that fractures can strongly influence
well performance. The case study shows that the detailed
fracture distributions can be extracted from 3-D prestack
seismic and log data using a rigorous processing and that the
integrating engineering data and geological can help us
achieve more useful information beyond the fracture
distributions.

Acknowledgements
The authors wish to thank EP Tech for their permission to
publish this work. The authors also wish to acknowledge the
support of colleagues in PetroChina.

References
1. Cetin Ozgen, Tuba Firincioglu, and Alejandro Albertoni,
Fracture Characterization for Integrated Studies: A new
approach and its applications, paper SPE 84413,
presentation at the 2003 SPE Annual Technical
Conference and Exhibition held in Denver, Colorado,
October 2003.
2. Zellou, A.M., Ouenes, A., and Banik, A.K., Improved
Fractured Reservoir Characterization Using Neural
Networks, Geomechanics and 3-D Seismic, paper SPE
30722, presentation at the 2000 SPE Annual Technical
Conference and Exhibition held in Dallas, Texas, October
1995.
3. Ouenes, A., and Hartley, L.J., Integrated Fractured
Reservoir Modeling Using Both Discrete and Continuum
Approaches, paper SPE 62939, Presented at the SPE
Annual Technical Conference and Exhibition held in
Dallas, Texas, October 2000.
4. Lee, S. H., Jensen, C. L., and Lough, M. F, An Efficient
Finite Difference Model For Flow In a Reservoir With
Multiple Length-Scale Fractures, paper SPE 56752,
presentation at the 1999 SPE Annual Technical
Conference and Exhibition held in Houston, Texas,
October 1999.
5. Lorenz, J. and Hill, R.: "Measurement and Analysis of
Fractures in Core," in Geological Studies Relevant to
Horizontal Drilling: Examples from Western North
America, Schmoker J., Coalson, E., Brown, C. (Eds.)
Rocky Mountain Association of Geologists, 1992.
6. Bahorich, M. S., and Farmer, S. L., 1995, 3-D seismic
discontinuity for faults and stratigraphic features: The
Leading Edge, 14, 10531058.
7. Marfurt, K. J., Kirlin, R. L., Farmer, S. L., and Bahorich,
M. S., 1998, 3-D seismic attributes using a semblance-
based coherency algorithm: Geophysics, 63, 11501165.
8. Gersztenkorn, A., and Marfurt, K. J., 1999,
Eigenstructure-based coherence computations as an aid to
3-D structural and stratigraphic mapping: Geophysics, 64,
14681479.
9. Israel Cohen and Ronald R. Coifman, 2002, Local
discontinuity measures for 3-D seismic data, Geophysics,
67, 19331945.
10. Heffer, K., King, P., and Jones, A.:"Fracture Modeling as
Part of Integrated Reservoir Characterization," paper SPE
6 SPE 90275
53347 presented at the 1999 SPE Middle East Oil Show,
Feb. 20-23.
11. Couples, G., H. Lewis and P.W. Tanner, 1998, Strain
partitioning during flexural-slip folding. In: Coward,
M.P., T.S. Daltaban and H. Johnson, Eds: Structual
Geology in Reservoir Characterization, Geological
Society Special Publications, Vol. 127, pp.149-165.
12. Gauthier, B., Zellou, A., Toublanc, A., Garcia, and J.M
Daniel:Integrated Fractured Reservoir Characterization:
a Case Study in a North Africa Field, paper SPE 65118
presented at the 2000 European Petroleum Conference,
Paris, Oct.24-25.
13. Nelson, Ronald A., 2001, Geological Analysis of
Naturally Fractured Reservoirs, second edition, Gulf
Professional Publishing, p332.
14. Connolley, P., 1999, Elastic Impedance, The Leading
Edge, 18, 438-452.
15. Lynn, H. B., Bates, C. R., Simon, K. M., and van Dok, R.,
1995, The effects of azimuthal anisotropy in P-wave 3-D
seismic: 65th Ann. Internat. Mtg., Soc. Expl. Geophys.,
Expanded Abstracts, 723730.
16. Ramos, A. C. B., and Davis, T. L., 1997, 3-D AVO
analysis and modeling applied to fracture detection in
coalbed methane reservoirs: Geophysics, 62, 16831695.
17. Feng Shen, Sierra, J., D.R. Burns and Nafi Toksoz, 2002,
Azimuthal offset-dependent attributes (AVO and FVO)
applied to Fracture detection, Geophysics, Vol. 67, 355-
364.
18. Partyka, G., Gridley, J., and Lopez, J., 1999,
Interpretational applications of spectral decomposition in
reservoir characterization, The Leading Edge, Vol.18,
353-360.
19. Peyton, L., Bottjer, R. and Partyka, G., 1998,
Intepretation of incised valleys using new 3-D seismic
techniques: A case history using spectral decomposition
and coherency, The leading Edge, Vol.17, 1294-1298.
20. Feng Shen, Gary C. Robinson and Tao Jiang, 2003,
Instantaneous Spectral Analysis Applied to Reservoir
Imaging and Producibility Characterization, 73th Ann.
Internat. Mtg., Soc. Expl. Geophys, 1406-1409.
21. Feng Shen and Nafi Toksoz, 2000Scattering
Characteristics in Heterogeneous Fractured Reservoirs
From Waveform Estimation, Geophysical Journal
International, Vol. 140, 251-265.
22. Feng Shen, Xiang Zhu and Nafi Toksoz, 2002, Effects of
Fractures on P-wave NMO Velocity and P-wave
azimuthal Amplitude Versus Offset (AVO) response, Vol.
67, 711-726, GEOPHYSICS.













SPE 90275 7
















Figure 1a Interpreted Structure Map

















Figure 1c Local structure entropy with horizontal size [4 4]


















Figure 2 Calculated principal strain distribution.























Figure 1b Local structure entropy with horizontal size [2 2]

















Figure 1d Local structure entropy with horizontal size [6 6]

















Figure 3 Calculated principal stress distribution.






8 SPE 90275
















Figure 4 Mapped channel sandstone distribution.
















Figure 6 Calculated principal strain on the top of the
reservoir.
















Figure 8a Fracture distributions on the top of reservoir.


















Figure 5 The discontinuity from local structure entropy for on
the top of reservoir
















Figure 7 Modeling the azimuthal AVA responses and
building the relationships between fracture parameters and
azimuthal amplitude variations.















Figure 8b Amplified Fracture distributions on the top of
reservoir.