processes
Article
A Comprehensive Investigation of the Relationship between
Fractures and Oil Production in a Giant Fractured Carbonate Field
Riyaz Kharrat 1, * , Ali Kadkhodaie 2 , Siroos Azizmohammadi 1 , David Misch 3 , Jamshid Moghadasi 4 ,
Hashem Fardin 5 , Ghasem Saedi 5 , Esmaeil Rokni 5 and Holger Ott 1
Citation: Kharrat, R.; Kadkhodaie, A.;
Azizmohammadi, S.; Misch, D.;
Moghadasi, J.; Fardin, H.; Saedi, G.;
Rokni, E.; Ott, H. A Comprehensive
Investigation of the Relationship
between Fractures and Oil Production
in a Giant Fractured Carbonate Field.
Processes 2024, 12, 631. https://
doi.org/10.3390/pr12040631
Academic Editor: Syed Muhammad
Shakil Hussain
Received: 29 January 2024
Revised: 15 March 2024
Accepted: 18 March 2024
Published: 22 March 2024
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Processes 2024, 12, 631. https://doi.org/10.3390/pr12040631
1 Department Geoenergy, Montanuniversität Leoben, 8700 Leoben, Austria;
siroos.azizmohammadi@unileoben.ac.at (S.A.); holger.ott@unileoben.ac.at (H.O.)
2 Earth Sciences Department, Faculty of Natural Sciences, University of Tabriz, Tabriz 51666-16471, Iran;
kadkhodaie_ali@tabrizu.ac.ir
3 Department for Applied Geosciences and Geophysics, Montanuniversität Leoben, 8700 Leoben, Austria;
david.misch@unileoben.ac.at
4 Ahvaz Faculty of Petroleum Engineering, Petroleum University of Technology, Ahvaz 63187-14317, Iran;
j.moghadasi@put.ac.ir
5 Exploitional Geology Section, National Iranian South Oil Company, Ahvaz 61000-619, Iran;
hashemfardin59@gmail.com (H.F.); ghasemsaedi2010@gmail.com (G.S.); rokni.e@nisoc.ir (E.R.)
* Correspondence: riyaz.kharrat@unileoben.ac.at
Abstract: This study examines the connections between various fracture indicators and production
data with an example from one of the giant fields in the Middle East producing complex fractured
carbonate lithologies. The field under study hosts two reservoirs with a long development and
production history, including carbonates from the Asmari and Bangestan Formations. A fracture
intensity map was generated based on the interpretation of image logs from 28 wells drilled within
the field. Mud loss data were collected and mapped based on the geostatistical Gaussian Random
Function Simulation (GRFS) algorithm. Maximum curvature maps were generated based on Asmari
structural surface maps. Comparing the results shows a good agreement between the curvature
map, fault distribution model, mud loss map, fracture intensity map, and productivity index. The
results of image log interpretations led to the identification of four classes of open fractures, including
major open fractures, medium open fractures, minor open fractures, and hairline fractures. Using
the azimuth and dip data of the four fracture sets mentioned above, the fracture intensity log was
generated as a continuous log for each well with available image log data. For this purpose, the
fracture intensity log and a continuous fracture network (CFN) model were generated. The continuous
fracture network model was used to generate a 3D discrete fracture network (DFN) for the Asmari
Formation. Finally, a 3D upscaled model of fracture dip and azimuth, fracture porosity, fracture
permeability, fracture length, fracture aperture, and the sigma parameter (the connectivity index
between matrix and fracture) were obtained. The results of this study can illuminate the modeling
of intricate reservoirs and the associated production challenges, providing insights not only during
the initial production phase but also in the application of advanced oil recovery methods, such as
thermal recovery.
Keywords: continuous fracture network CFN; discrete fracture network DFN; fractured reservoir;
image logs; reservoir modeling; stress analysis
1. Introduction
Unlike siliciclastic reservoirs, where permeability models derived from logging data
typically exhibit reasonable accuracy, fractured reservoirs pose challenges due to their
lithological heterogeneity. Analyzing fractured reservoirs, especially in the context of hy-
drocarbon production mechanisms, is particularly difficult [1,2]. The type and orientation
of fractures notably impact enhanced oil recovery methods, especially in gas and thermal
https://www.mdpi.com/journal/processes
Processes 2024, 12, 631 2 of 20

applications, significantly influencing production mechanisms [3–5]. A significant compli-

cating factor arises from the fact that the permeability of carbonates is influenced either by
secondary dissolution porosity or by fractures that serve as pathways for fluid flow. Such
features are typically hard to predict or describe in generalized models, as their spatial dis-
tribution within the reservoir and their impact on permeability often vary substantially at
the field scale. Hence, a great heterogeneity in well production rates is commonly observed
in naturally fractured carbonate reservoir systems. Different geological processes may
cause further model uncertainty; examples are sedimentary discontinuities, thin-layered
shale intervals acting as semi-permeable membranes, and local facies variations changing
the matrix porosity characteristics of the carbonate strata. Nevertheless, the frequency and
nature of fractures still play paramount roles in the production statistics of a fractured
reservoir. Therefore, algorithm-based methods for detecting layer heterogeneity caused by
fractures can provide a fast method to predict heterogeneity and production behavior and
thereby help to maximize hydrocarbon recovery efficiency [6,7].
Extensive scientific research has been dedicated to advancing methods to achieve
more precise characterization of fractured carbonate reservoirs. Many of these methods
rely on seismic data, wireline logging, and geological borehole data [8–15]. An ideal
characterization integrates all of these data sources. It incorporates additional information
from outcrop analogues, drill core analysis, etc., to create a 3D model that can serve as
a foundation for subsequent reservoir simulation [16–26]. Production data may then be
applied to confirm fracture existence, aperture, and bulk reservoir properties such as
permeability [27].
The fracture characterization and modeling task is a multi-step procedure involv-
ing several static and dynamic reservoir modeling disciplines [28,29]. Discrete Fracture
Network (DFN) modeling can be used to represent hydraulically relevant fractures with
specific characteristics [16,26]. In principle, the DFN is a set of planes representing fractures.
Fractures that share similarities and genetic connections are grouped into a fracture set, and
within each fracture network, there is usually more than one of these distinct sets. Fracture
sets are created from the previously defined fracture based on image logs or seismic data
interpretation. The main contributions of this paper are as follows:
(i) Field stress regime analysis to identify the orientation of fractures into principal
stresses.
(ii) Classification of fractures in terms of their azimuths and apertures.
(iii) Creating continuous and discrete fracture models based on formation image logs,
including Micro-Imaging (FMI), Simultaneous Time and Resistivity (STAR), Ultrasonic
Borehole Imager (UBI), and Circumferential Acoustic Scanning Tool (CAST).
(iv) Comprehensively characterizing the influence of fractures on production by integra-
tion of fracture maps with production data, mud loss maps, and fault data.
(v) Summarizing the benefits of a comprehensive and multi-analytical fracture model-
ing approach to enhance the accuracy of predicting productive zones in fractured
carbonate reservoirs.

2. Geological Setting
The studied oilfield is one of the largest and most complex onshore fields in the Middle
East, with over 400 wells and 90 years of production [30]. The field is located within the
Zagros fold and thrust belt in the Khuzestan province of southern Iran. The field’s NW-SE
trending anticline is approximately 65 km long, with a maximum width of 9 km in its
southeastern part. In 1998, it was proposed that the anticlinal structure is believed to arise
from a detachment level deep beneath the surface, approximately 10 km down, and is
associated with the Hurmuz salt [31]. The tectonic pattern is characterized by overlapping
thrust folds with modest elevations between consecutive folds. The oilfield is subdivided
into four reservoir zones and ten sub-zones of comparable properties, associated with
two main carbonate reservoir horizons within the Oligo-Miocene Asmari and Bangestan
Groups (the Santonian/Campanian Ilam and the Cenomanian/Turonian Sarvak Forma-
 subdivided into four reservoir zones and ten sub-zones of comparable prope
ated with two main carbonate reservoir horizons within the Oligo-Miocene
Bangestan Groups (the Santonian/Campanian Ilam and the Cenomanian/Turo
Processes 2024, 12, 631
Formations). The Asmari reservoir thickness is about 500 m in the3 ofwest 20 and 2
east. It consists mainly of limestone and dolomite with minor siliciclastic in
[32]. The stratigraphic chart of Cretaceous-Tertiary sequences in Dezful Emba
tions). The Asmari reservoir thickness is about 500 m in the west and 250 m in the east. It
tratingofthe
consists mainly Oligo-Miocene
limestone and dolomiteAsmari Formation
with minor is shown
siliciclastic in Figure
intercalations [32].1.The
The uppe
parts are mainly dolomitic limestone. Further elaboration on
stratigraphic chart of Cretaceous-Tertiary sequences in Dezful Embayment illustrating the the stratigraphy
ros region
Oligo-Miocene Asmari is available
Formation in references
is shown [33,34].
in Figure Theupper
1. The Asmariand reservoir
lower partsinare the field
mainly dolomitic
ered in 1937 by completing one of the exploration wells with aboutis 44.5 bil
limestone. Further elaboration on the stratigraphy of the Zagros region
available[35].
in references [33,34]. Theof
The production Asmari reservoir inhas
this reservoir the so
field was
far discovered
exceeded 10inbillion
1937 bybarrels
completing one of the exploration wells with about 44.5 billion Barrels [35]. The production
than 10 trillion cubic feet of gas have been injected into the gas cap during t
of this reservoir has so far exceeded 10 billion barrels of oil. More than 10 trillion cubic feet
project
of gas have since 1977
been injected into [30,36].
the gas cap during the injection project since 1977 [30,36].

Figure 1. A stratigraphic chart of Cretaceous-Tertiary sequences in Dezful Embayment illustrates the

Figure
Oligo-Miocene 1. A Formation
Asmari stratigraphic
as thechart
main of Cretaceous-Tertiary
reservoir sequences in Dezful Embayme
of the studied field.
the Oligo-Miocene Asmari Formation as the main reservoir of the studied field.

Deformation History.
The studied field is subjected to compressional stresses imposed by the n
movement of the Arabian plate pushing against the Zagros trust belt and
Processes 2024, 12, 631 4 of 20

Deformation History
The studied field is subjected to compressional stresses imposed by the northeastern
movement of the Arabian plate pushing against the Zagros trust belt and local stress
distribution in the Dezful embayment. Studying the field stress map can provide insights
into the location of fractured zones and the formation orientation of fractures [37]. Fractures
caused by induction appear in the direction of the highest horizontal stress (Sigma H),
while breakouts form parallel to the lowest horizontal stress (Sigma h). The main stress
paths run parallel and perpendicular to the free surface along the wellbore wall. When the
maximum compressive stress trajectories converge, the stresses are more intense (at the
direction of σH in the case of a vertical well). When the trajectories diverge, the stresses
are less intense (at the azimuth of σH). Geologically, Asmari formation fractures have
been generated in three generic steps, including pre-folding, syn-folding, and post-folding
sets. Pre-folding fractures date back to the Late Oligocene and were formed during the
collision of the Arabian and Central Iran plates. These fractures are distributed parallel
to the maximum horizontal stress (maximum and medium open fractures). Syn-folding
fractures have developed in the Early Miocene (Aquitanian/Burdigalian) due to increased
compressional stress and reactivation of basement faults, causing the opening of N-S, W-E,
N020◦ , and N150◦ trending fracture sub-sets. Major and medium open fractures are hinge-
perpendicular fractures (parallel to the maximum horizontal stress). In contrast, minor
and hairline fractures are parallel to the hinge (parallel to the minimum horizontal stress).
Due to their larger apertures, major and medium open fractures have the most influence
on production.

3. Fracture Characterization of the Asmari Formation

Due to its reservoir lithology being mainly carbonate, the production from the Asmari
Formation heavily relies on the presence of natural fractures [27]. Fractures are flat struc-
tures with no observable movement of blocks along their planes. In compressional stress
regimes, they may dip at either high or low angles. The size of their openings may vary,
being either wide open, narrow (closed), or filled with minerals such as calcite, clay, pyrite,
or anhydrite. This section investigates various geological and reservoir data indicating the
role of natural fractures in field production.

3.1. Curvature Map

By definition, each node of a gridded surface has a certain curvature value for a
given azimuth. In practice, this curvature value corresponds to the second derivative of
the surface topography in the respective orientation. Therefore, a curvature value as a
function of azimuth (orientation) may be computed for any given point on a geological
surface. Curvature analysis is a method to define anomalies in local surface curvature
that may coincide with large-scale structural features such as fractures [38]. Generally,
curvature analysis is performed for geological horizons picked from seismic data sets.
Many authors have also described the relationships between folding and the generation
of fractures on various scales [39,40]. Such features are known as ‘extrados’ fracture sets.
The hypothesis is that fractures perpendicular to the hinge are pre-folding, while fractures
parallel and oblique to the hinge are fold-related or due to post-folding reactions to the
applied extensional stress. In such cases, fracturing is related to spatial variations of
strain. Several researchers have noted a relationship between high fracture intensity and
high curvature areas in the North Sea [39] and North America [40]. The curvature map
(Top Asmari) indicating the location of faults/fractures is represented in Figure 2. High
curvature areas correspond to the faulted and fractured areas, which appear as linear
features along the map.
 Processes 2024, 12, x FOR PEER REVIEW 5 of 20

Processes 2024, 12, x FOR PEER REVIEW 5 of 20

Processes 2024, 12, 631 5 of 20

Figure 2. Curvature map (Top Asmari) indicating the location of faults/fractures.

Figure 2. Curvature map (Top Asmari) indicating the location of faults/fractures.

3.2. Faults
Figure Distribution
2. Curvature MapAsmari) indicating the location of faults/fractures.
map (Top
3.2. To date,
Faults 13 large-scale
Distribution Map faults have been picked in the field structure (Figure 3), dom-
3.2. Faults
inantly Distribution Map
Todistributed in the field’s
date, 13 large-scale western
faults have beenand eastern
picked in thesectors. The preferential
field structure (Figure 3),orientation
domi-
To curvature
ofnantly
the date, 13 large-scale
distributed patterns faultswestern
suggests
in the field’s havethatbeen picked
theyeastern
and in the
represent field
The structure
structural
sectors. (Figure
directions.
preferential For3),
thedom-
orientation main
inantly
of thedistributed
orientation (N130),
curvature initthe
patterns field’s
is clear
suggests western
that thisthey
that and
matches eastern sectors.
verystructural
represent well with The preferential
the structural
directions. For theorientation
axis
mainof the
offield.
the curvature
orientation
The most patterns
(N130),
important suggests
it is clear thatin
fault thisthat
this theyisrepresent
matches
field very well
trust structural
with
fault F2,the directions.
structural
which axis ofFor
separates the main
thesoutheast-
the field.
Thepart
ern most
orientation important
(N130),
from it fault
the main is part.in this
clear Itthatfield
thisismatches
resulted trust
in fault
dipping F2,
very which
thewell separates
with
southern the
flank the southeastern
structural
of Asmari axis
andofBang-
the
part
field. from
estanThe the main part.
mostreservoirs
Group importantup It resulted
faultto in
75thisin dipping
to 80field the southern
is trustItfault
degrees. has F2, flank of
whichthe
affected Asmari
separatesand Bangestan
the southeast-
communication of the
Group reservoirs up topart.
ern part from
reservoir’s the main
eastern and75western
to 80 degrees. It
It resulted
sectors,inhas affected
dipping
separating thethe communication
southern
the Asmari flank ofofAsmari
reservoir the reservoir’s
into and
twoBang-
south-
eastern and western sectors, separating the Asmari reservoir into two southeastern and
estan
easternGroup
and reservoirs
main sectors. up toThe 75 eastern
to 80 degrees.
part is Itmore
has affected
complicated the communication
than the other of the
sectors
main sectors. The eastern part is more complicated than the other sectors (same as Asmari)
reservoir’s
(same eastern dueand western sectors, separating the Asmari reservoir into two south-
due toastrust
Asmari)
faults and tothetrust faults
Kharg-Mish and the Kharg-Mish
structure. structure.
eastern and main sectors. The eastern part is more complicated than the other sectors
(same as Asmari) due to trust faults and the Kharg-Mish structure.

Figure 3. Location map of the faults of the studied field structure, Asmari top.
Figure 3. Location map of the faults of the studied field structure, Asmari top.

Figure 3. Location map of the faults of the studied field structure, Asmari top.
 Processes 2024, 12, x FOR PEER REVIEW 6 of 20
Processes 2024, 12, 631 6 of 20

3.3. Image Log-Based Fracture Intensity Map

3.3. Image Log-Based
Image Fracture
logs serve Intensity
as advanced Map
logs for imaging a borehole’s physical properties (elec-
tricalImage logs serve
resistivity as advanced
or acoustic logs forThey
impedance). imaging
can bea borehole’s physicaldifferent
used to interpret properties (elec-
types of
trical resistivity
fractures in the or acousticbased
formation impedance). They can
on resistivity be usedcontrasts.
or acoustic to interpret different
Fracture types of
information
fractures
such as openin theand
formation based on
filled fractures, resistivity
fracture dip or
and acoustic
azimuth, contrasts.
fractureFracture
spacing,information
and fracture
such as open
intensity and filled
is derived fromfractures, fracture dipofand
the interpretation azimuth,
image fracture
logs. There spacing,
are 28 wells and
in fracture
the field
intensity is derived from the interpretation of image logs. There are 28
with good-quality image interpretation results. An example of continuous open fractures wells in the field with
good-quality image interpretation results. An example of continuous
from the middle part of the Asmari Formation on FMI logs of the field is shown in Figure open fractures from
the middle
4. Each part ofisthe
fracture Asmari Formation
represented onaFMI
spatially as planelogs of the
with field is shown
its azimuth and dipin Figure 4. Each
attributes. The
fracture
intensity is of
represented
open fracturesspatially
wasascomputed
a plane withas its
theazimuth and dip attributes.
area of fractures The intensity
per unit volume (1/ft),
of open fractures
represented was(3computed
by P32 refers to as the area
a unit of fractures
volume of rock, per
andunit volume
2 refers to a(1/ft), represented
2D fracture area).
by P32 (3 refers to a unit volume of rock, and 2 refers to a 2D fracture
The fracture intensity map (P32) of the Asmari reservoir in the oilfield is shown in Figurearea). The fracture
intensity maphigh
5. As is seen, (P32)fracture
of the Asmari reservoir
intensity in the oilfield
areas correspond is shown
to fault in Figure
locations and high5. As is seen,
curvature
high fracture
intensity. intensity areas correspond to fault locations and high curvature intensity.

Figure4.4.Representation
Figure Representationof
ofmajor
majorand
andmedium
mediumopen
openfractures
fractureson
oncore
coresamples
samplesand
andFMI
FMIlog
log[27].
[27].
Processes 2024, 12, x FOR PEER REVIEW 7 of 20
Processes 2024, 12, x FOR PEER REVIEW 7 of 20
Processes 2024, 12, 631 7 of 20

Figure 5. Average fracture intensity map (P32) in Asmari reservoir.

Figure 5.
Figure Average fracture
5. Average fracture intensity
intensity map
map (P32)
(P32) in
in Asmari
Asmari reservoir.
reservoir.
3.4.
3.4.Mud
MudLoss LossMap
Map
3.4. Mud Loss Map
Mud
Mud loss data
loss provide very
data provide veryuseful
usefulproxies
proxies forfor
thethe fracture
fracture characterization
characterization of car-
of carbonate
bonate Mud loss data
reservoirs. Wellsprovide
with verymud
high useful
loss proxies
are for the
usually fracturetocharacterization
correlated fractured zones. ofThis
car-
reservoirs. Wells with high mud loss are usually correlated to fractured zones. This study
bonate
study reservoirs. Wells with high mud loss are usually correlated to fractured zones. This
used used
mud loss muddatalossfrom
data261 from 261to
wells wells to generate
generate the mud thelossmudmap. lossInmap.
someIn some
cases, mudcases,
loss
study
mud used mud loss data from 261 wells to generate the mud loss map. In some cases,
ratesloss
were rates were
given in given in inconsistent
inconsistent units. Inunits.
other In otheronly
cases, cases,
theonly
bulkthe bulk volume
volume of the mud of
mud
the mudloss rates waswere given infirst inconsistent units. In other tocases, only
unitthe bulk volume of
loss waslossgiven. given.
The first The
step in our step in our
efforts was efforts was unit
to obtain obtain
consistency. consistency.
The secondThe step
the mud loss
second was given. Thethe first steploss
in our efforts was drilling
to obtain unit consistency. The
was tostep was
estimate to
theestimate
mud loss duration mud fromduration
drillingfrom
reports to convert reports theto volumes
convert the into
second
volumes step
into was to
rates. estimate
The mud the mud
loss rate forloss duration from drilling reports to convert the
rates. The mud loss rate for complete losscomplete loss was
was reported reported
as 100 STB/hour,as 100 STB/hour,
equivalent to
volumes
equivalent into rates.
to 2400 The
STB/D. mud
However, loss rate
whilefor complete loss was reported as 100 STB/hour,
2400 STB/D. However, while searching thesearching the drilling
drilling records, it was records,
seen that it was seen that
the circulation
equivalent
the circulation to 2400
rate, STB/D.
in someHowever, while gosearching theSTB/D,
drilling records, it wasthanseen that
rate, in some cases, could gocases,
up tocould
5000 STB/D, up towhich
5000 is morewhich than 200 is more
STB/hour. 200 The
the circulation
STB/hour. rate, in some cases, could go up to 5000 STB/D, which is more than 200
Gaussian The Gaussian
random functionrandom function
simulation simulation
(GRFS) algorithm(GRFS) algorithm
generated a mud generated
loss mapa for mud the
STB/hour.
loss The
map Formation Gaussian
for the Asmari random
Formation function
(Figure simulation
6). It should (GRFS) algorithm generated a mud
Asmari (Figure 6). It should be noted that thebesource
notedofthat the source
“complete of “com-
losses” could
loss
plete map fractures
losses”
be either forcould
the Asmari
be Formation
either
or localized fractures (Figure
cavities,orand 6). Ittend
localized
they should
cavities, beand
to occur noted
inthey that
tend
geologicalthetosource
occurof
formationsin“com-
geo-
that
plete
logical losses”
formationscouldthatbe either
have
have pre-existing fractures [32,34]. fractures
pre-existing or localized
fractures cavities,
[32,34]. and they tend to occur in geo-
logical formations that have pre-existing fractures [32,34].

Mudloss
Figure6.6.Mud
Figure lossmap
mapininthe
theAsmari
Asmarireservoir.
reservoir.
Figure 6. Mud loss map in the Asmari reservoir.
Processes 2024, 12, x FOR PEER REVIEW 8 of 20

Processes 2024, 12, 631 8 of 20

3.5. Productivity Index (PI) Map

3.5. Productivity Index (PI)
The productivity indexMap is a significant measure that indicates the ability of an oil or
gas well to produce. It is calculated
The productivity index is a significant by dividing
measure thethat
total liquid flow
indicates rate atof
the ability theansurface by
oil or gas
the pressure drawdown at the midpoint of the productive depth range.
well to produce. It is calculated by dividing the total liquid flow rate at the surface by the In essence, it rep-
resents the
pressure thickness
drawdown of the
at the formation
midpoint of theresponsible for production.
productive depth Therefore,
range. In essence, it is im-
it represents
portant
the to determine
thickness this thickness
of the formation to characterize
responsible a given well’s
for production. deliverability
Therefore, appropri-
it is important to
ately. The PI
determine is thickness
this usually expressed in STB/D/psi.
to characterize a givenReservoirs may be classified
well’s deliverability as cross-flow
appropriately. The
and
PI iscommingled systemsinfrom
usually expressed the standpoint
STB/D/psi. of reservoir
Reservoirs may be dynamics.
classifiedInasthecross-flow
latter case,and the
commingled
production issystems
commingled fromonlythe standpoint
in the common of reservoir
wellbore,dynamics. In the latter
and the reservoir layerscase,
havetheno
production is commingled
vertical communication. only in the common
In commingled systems,wellbore, and layer
the PI of each the reservoir
dependslayerson its have
char-
no vertical The
acteristics. communication.
overall PI of In thecommingled
reservoir depends systems, on the
the PIsumof ofeach
thelayer
PI’s of depends
the layerson
its characteristics.
exposed The overall
to the wellbore. In the PI of the reservoir
cross-flow system, depends on the sum
the PI is unique due toofinterlayer
the PI’s ofcom- the
layers exposed
munication. Theto system
the wellbore.
behaves In the
likecross-flow
one, even system, the PI is penetrated
when partially unique due[32,34].
to interlayer
Since
communication. The system behaves like one, even when partially penetrated
the field is a fractured reservoir, the PI estimates represent the whole Asmari interval and [32,34]. Since
the field is a fractured
a combination of layersreservoir,
from Asmari the PIto estimates represent
the Bangestan Group,the whole
whereAsmari interval
appropriate. The andPI
adatabase
combination of layers from Asmari to the Bangestan Group, where
included 177 digitized build-up data from 55 wells and 196 PI tests. A bubble appropriate. The PI
database
map showingincluded
the 177 digitized build-up
productivity index (PI)data from
in the 55 wells and
field/Asmari is 196 PI tests.inAFigure
presented bubble7.map As-
showing
suming the the poor
productivity
and good index (PI) in the
reservoir field/Asmari
quality zones areisdirectly
presented in Figure
linked to PI 7. Assuming
distribution,
the
the poor
mostand good reservoir
productive areas are quality
located zones
in thearecentral
directly linked
and to PI distribution,
northwestern parts of the the struc-
most
productive
ture. areas are located in the central and northwestern parts of the structure.

PI (stb/d/psi)
3200000

3180000
8000
3160000

3140000
3000
3120000
200
Y (m)

3100000

3080000

3060000

3040000

3020000

3000000
6600000 6650000 6700000 6750000 6800000 6850000
X (m)

Figure 7.
Figure 7. The
The bubble
bubble map
map shows
shows the
the productivity
productivity index
index (PI)
(PI)in
inthe
theAsmari
Asmarireservoir.
reservoir.

4.
4. Classification
Classificationof ofFracture
FractureTypes
Types
On
On the
the FMI
FMI images,
images, fractures
fractures are
are seen
seen asas linear
linear features
features with
with aa steeper
steeper dip
dip than
than the
the
structural
structural dip.
dip. Open fractures
fractures have
have aa conductive
conductive appearance
appearance on on the
the images
images of of aa clay-free
clay-free
formation
formationasastheir
theiraperture
apertureis invaded
is invadedwithwith
conductive
conductivedrilling fluid. fluid.
drilling Mineralized or sealed
Mineralized or
fractures, such as anhydrite or calcite, seem resistive if their filling material
sealed fractures, such as anhydrite or calcite, seem resistive if their filling material is dense. The is
fractures
dense. The containing
fracturesacontaining
pyrite or clay fillingorshow
a pyrite clay afilling
conductive
show aresponse.
conductive To response.
differentiate To
between pyrite/clay-filled
differentiate conductive fractures
between pyrite/clay-filled conductive andfractures
mud-filled and fractures, knowing
mud-filled the
fractures,
depositional
knowing the setting of thesetting
depositional study ofarea
theisstudy
necessary.
area isOpen-hole logs can also
necessary. Open-hole logsbecanhighly
also
beneficial
be highly for this kind
beneficial forofthis
differentiation.
kind of differentiation.
In
In specific zones of thefield
specific zones of the fieldformations,
formations,the theFMI
FMIlog logrevealed
revealedthe thepresence
presence ofofeither
eithera
limited or extensive number of fractures. Generally, fractures can be
a limited or extensive number of fractures. Generally, fractures can be classified into classified into closed
or continuous/discontinuous
closed or continuous/discontinuous conductive traces,traces,
conductive calledcalled
open open
fractures. MudMud
fractures. invasion
inva-
into the open parts of the fractures’ apertures makes them conductive.
sion into the open parts of the fractures’ apertures makes them conductive. Water-based Water-based mud
invasion into fractures causes a decrease in the reading of resistivity buttons that appeared
as black-colored sinusoids in the FMI logs.
ER REVIEW Processes 2024, 12, x FOR PEER REVIEW 9 of 20 9 o

Processes 2024, 12, 631 9 of 20

mud invasion into fractures causes a decrease mud in invasion
the readinginto fractures
of resistivitycauses a decrease
buttons that in the reading of resistivity buttons t
appeared
appeared as black-colored sinusoids in the FMI logs. as black-colored sinusoids in the FMI logs.
Open fractures can further be grouped Open
into more fractures
classesbe can further
regarding be grouped into more classes regarding their continu
Open fractures can further grouped their continuity
into more classes regarding their continuity and
and appearance within the wellbore. and
This study
appearance withinappearance
classified within
open
the wellbore. thestudy
fractures
This wellbore. This
intoclassified
four study
open classified
catego- fractures open fractures
into four into four cate
categories
based on image log interpretation results: major open, medium open, minor open, andminor op
ries based on image log interpretation ries
results: based
major on
open,imagemediumlog interpretation
open, minor results:
open, major open, medium open,
and hairline. Other open fractures, such
hairline. and
as
Other hairline.
vuggy,
open Othersuch
partially
fractures, open
closed,as fractures,
and partially
vuggy, such open
partially as vuggy,
closed, andpartially
partiallyclosed, and partially op
open fractures,
fractures, were also consideredwere
whenalso fractures,
interpreting
considered were
image
whenlogs. also considered
interpreting image logs. when interpreting image logs.
Based on FMI log interpretations, induced
Based on FMI Based
fractures onoccur
FMI log interpretations,
in two
log interpretations, opposite induced
directions.
induced fractures fractures
occur in twooccur in two
opposite opposite directio
directions.
Most induced fractures indicateMosta NE-SW
induced Most
strike induced tofractures
(N015 indicate
fractures N030),a while indicate
NE-SW aminority
thestrikeNE-SW(N015 strike
rep- (N015while
to N030), to N030), while the
the minority minority r
rep-
resent an NWW-SEE strike (N285 resent an NWW-SEE strike (N285 to N300). Such stress behavior indicates the rotation rotation
to N300). resent
Such an NWW-SEE
stress behavior strike
indicates(N285 the to N300).
rotation Such
of stress behavior indicates the
maximum horizontal stress from of the original
maximum maximum horizontal
NE-SW (parallel
horizontal tostress
stress from thethe from
direction theoforiginal
original the Ara-NE-SW
NE-SW (parallel(parallel to the direction
to the direction of the of the A
bian plate push against the Zagros Trustplate
Arabian bian
Belt)push plate
to the push
recentthe
against against
NW-SEZagros the Zagros
trends
Trust[41]. Trust Belt) to the recent
Belt) to the recent NW-SE trends [41]. NW-SE trends [41].
Figure 8 represents the rose diagram
Figureof8all Figure 8
open fractures
represents represents
the rose indiagram the
the studiedrose diagram
oilfield.
of all of all open fractures in
Open in the studied oilfield. Open
open fractures the studied oilfield. Op
fractures propagate in all directions, parallel
fractures fractures inpropagate
or perpendicular
propagate tointhe
all directions, allparallel
directions,
maximum parallel or perpendicular
horizontal
or perpendicular to the maximum to thehorizontal
maximum horizon
stress. Rose diagrams of major stress. Rose stress.
open fractures, Rose
medium
diagrams diagrams
openopen
of major offractures,
fractures, major open
minor fractures,
open
medium open medium
frac- fractures,open
minorfractures, minor open fr
open fractures,
tures, and hairline fractures areand
shown in tures,
Figure and
9a–d.hairline
The fractures
important are shown
understanding in
hairline fractures are shown in Figure 9a–d. The important understanding of the Figure
of 9a–d. The important understanding
the extracted rose diagrams is extracted
the bipolarity thediagrams
rose extracted
of differentis rose diagrams
fracture
the bipolarity sets.ofisMajor
the bipolarity
different open and ofsets.
fracture different
Majorfracture
open andsets. Major open a
medium
open fractures
medium open fractures are developed medium
parallelare open fractures
developedhorizontal
to maximum are developed
parallel tostress. maximum parallel
horizontal stress. In contrast, minor In contr
In contrast, to maximum horizontal stress.
minor open and hairline fracturesopenare minor fractures
andperpendicular
hairline opentoand hairline
are
maximum fractures to
perpendicular
horizontal are perpendicular
maximum
stress (or horizontal to maximum
stress (orhorizontal
parallel stress
parallel to minimum horizontaltostress).
minimum It isparallel
horizontal
important to minimum
tostress).
note that horizontal
It is important
caution stress).
mustto beItnote
is important
exer- that caution to note
must that
be caution
exercised must be ex
when making
cised when making blanket statements aboutcised thewhen
blanket makingbetween
statements
correlation blanket statements
aboutfracture
the correlation
types about the correlation
between
and between
fracture types andfracture
stress types a
directions
stress directions for all wells in the field.for stress
Dueall to directions
wellsthein the for
local field.
stressallDuewells toin
variation,thethe field.
local
some Due
stress
stress to the local
variation, some stress
stressvariation,
rotationssome str
rotations may occur. may occur. rotations may occur.

8. Rose
Figure of
Figure 8. Rose diagram diagram
all open of all
fractures Figure
open
(Asmari 8. Rose(Asmari
fractures diagramreservoir).
reservoir). of all open fractures (Asmari reservoir).

(a) (b) (a)

(c) (b)
(d) (c) (d)
Figure
Figure 9. Rose diagram of (a) major, Figure
9. Rose
(b) medium, (c) 9. Rose
diagram
minor, of
anddiagram
(a)(d) of (b)
major, (a) fractures
hairline major, (b)(Asmari
medium, medium,
(c) minor,
res-(c)
andminor, and (d)fractures
(d) hairline hairline fractures
(Asmari (Asmari
ervoir). reservoir). ervoir).

5. Results and Discussion 5. Results and Discussion

5. Results and Discussion
5.1. The Relationship between Fractures
5.1. The Relationship betweenandFractures
Production
andData
Production Data
5.1. The Relationship between Fractures and Production Data
This section This
investigates
section the relationship
investigates between fractures and actual production data.
the relationship
This section investigates the relationship between fractures and actual production between fractures and actual product
Figure 10 shows a generally
data. Figure good
10 shows agreement
a generally between
good fracture intensity,
betweenmud loss,intensity,
and the mud lo
data. Figure 10 shows a generally good agreement between fracture intensity, mudagreement
loss, fracture
PI in the Asmari reservoir parts of the studied field. Notably, the highest values of thehighest
PI
and the PI in the Asmari reservoir parts ofand the the PI infield.
studied the Asmari
Notably,reservoir partsvalues
the highest of the studied field. Notably, the val
in the northwestern part of the study area do not correlate well with the other available
of the PI in the northwestern part of
of the PI in the northwestern part of the study area do not correlate well with the other the study area do not correlate well with the ot
data. This means other geological complexities probably control high mud losses in this
zone. Irrespective of the northwestern part of the field, the areas with high production rates
almost correspond to those with high mud losses and high curvature in the structural map,
 Processes 2024, 12, x FOR PEER REVIEW 10 of 20

Processes 2024, 12, x FOR PEER REVIEW 10 of 20

Processes 2024, 12, 631 available data. This means other geological complexities probably control high mud10losses of 20
in this zone. Irrespective of the northwestern part of the field, the areas with high produc-
available data. This means other geological complexities probably control high mud losses
tion rates almost correspond to those with high mud losses and high curvature in the
in this zone. Irrespective of the northwestern part of the field, the areas with high produc-
structural of
indicative map,
faultedindicative
zones of withfaulted
highzones
fracture with high fracture intensity in their Two surround-
tion rates almost correspond to those with highintensity
mud losses in their
and surroundings.
high curvature in the mud
ings. Two
loss clusters mud loss
canindicative clusters
be detected can be
in the field detected
structure, in the
one field structure,
in the central andone in the central and
structural map, of faulted zones with high fracture intensity in northwestern
their surround- parts
northwestern
of theTwo
ings. fieldmud (highparts
lossmud of the field
flowcan
clusters rates) (high mud
and the in
be detected flow
other rates) and
in thestructure,
the field the
southeastern other
one inarea in the southeastern
(minor and
the central losses).
areamost
The (minor
northwestern losses).
parts of The
productive areas
the most productive
are(high
field distributed
mud flow areas
in the are
rates) distributed
central
and and
the ininthe thecentral
northwestern
other and of
parts
southeastern north-
the
western
field, whereparts highof the
mud field,
loss where
and high
fracture mud loss
intensity and fracture
coincide.
area (minor losses). The most productive areas are distributed in the central and north- intensity
The results coincide.
show thatThe
the results
mud
show
loss
western that
zones
parts theofmud
match the loss zones
reasonably
field, where match
well
high withreasonably
mud the major
loss andwell withintensity
fractured
fracture the major
zones, fractured
likely
coincide. Thezones,
to show likely
the most
results
to
show show
intense the most
deformation
that the mud lossand, intense
zones deformation
consequently,
match reasonably and,
high fractureconsequently,
well withintensity. high
the majorAlthough fracture
fractured the intensity.
zones, likely Alt-
relationship
hough
between
to the
show the relationship
the most
mud intenseloss map between the mud
and productivity
deformation loss map and productivity
index is scattered,
and, consequently, index
it can generally
high fracture is scattered,
intensity. be Alt-seenit
can generally
as a direct
hough be seen
correlationbetween
the relationship as a direct
betweenthe correlation
themudPI andloss mud between
map and losses the PI and
(Figure 11a).
productivity mud losses
indexFractures (Figure
is scattered, it11a).
enhance
Fractures
can generally
reservoir enhance
be seenreservoir
permeability as aindirect permeability
correlation
horizontal in horizontal
between
and vertical the PIand
directions. andvertical
mud directions.
losses
Studying (Figure
the Studying
11a).
relationship
the relationship
Fractures
between enhancebetween
horizontal reservoir horizontal
permeability (Kh)permeability
permeability and inproductivity
horizontal (Kh) andand
index productivity
vertical directions.
(PI) for index
the Asmari (PI)
Studying for the
reservoir
Asmari
the
in the reservoir
relationship
studied field in
between thehorizontal
confirms studied field
the strong confirms
permeability
dependence the of
(Kh) strong
and dependence
theproductivity
productivity of the
index
index (PI)
on productivity
for the
horizontal
Asmari
index on reservoir
permeability horizontal in the
(Figure studied
permeability
11b). field(Figure
Permeability confirms 11b).
is thePermeability
highly strong dependence
correlated iswith
highly of the productivity
correlated
fracture with and
aperture frac-
index
ture on horizontal
aperture and permeability
spacing. The (Figure
fractures’ 11b). Permeability
presence is
spacing. The fractures’ presence is directly related to interval permeability data (mD- is
directlyhighly correlated
related to with
interval frac-
permea-
ture
bility
ft) aperture
derived and formation
data (mD-ft)
from spacing.
derived Thefromfractures’
tests formation presence
(e.g., Drill tests isTest,
Stem(e.g., directly
Drill
DST). related
Stem Care tomust
Test, interval
DST). permea-
be Care
taken must
not to be
bility
use
taken data
matrix (mD-ft)
not topermeability derived from formation
for calibration
use matrix permeability tests
for with (e.g.,
fracture
calibration Drill
with Stem
data, Test, DST).
as it does
fracture data, not Care
as ithave
doesmust behave
intrinsic
not
taken
fracturenotinformation.
intrinsic to use matrix
fracture permeability
Wells
information. with Wells forwith
a high calibration
PIaare high with
expected
PI arefracture data,
to produce
expected as
from
to produceit does
both not
from have
the matrix
both the
intrinsic
and
matrix fracture
fractures. information.
Accordingly,
and fractures. Wells with a
higher productivity
Accordingly, high PI
higher productivity are expected
index values to
index areproduce
values directlyfrom both the
correlated
are directly corre-to
matrix
horizontal
lated to and fractures.permeability
permeability
horizontal Accordingly,
and the direct higher
and role
the productivity
role of index
of fractures
direct in field
fractures values are directly
production.
in field production. corre-
lated to horizontal permeability and the direct role of fractures in field production.

Figure
Figure 10.AA
10.
Figure10. good
Agood
good agreement
agreement
agreement between
between
between fracture
fracture
fracture intensity,
intensity,
intensity, mudmud
lossloss
mud lossmap,
map, andand
map, andproductivity
productivity
productivity index
index (PI) (PI)
index (PI)
inthe
in
in theAsmari
the Asmarireservoir.
Asmari reservoir.
reservoir.

Figure 11. (a) Cross plot showing the relationship between mud loss and PI (b) Cross plot showing
Figure 11. (a) Cross plot showing the relationship between mud loss and PI (b) Cross plot showing
Figure
the 11. (a) Cross
relationship plot Kh
between showing the
andPIPIin relationship
inAsmari
Asmari between mud loss and PI (b) Cross plot showing
reservoir.
the relationship between Kh and reservoir.
the relationship between Kh and PI in Asmari reservoir.
5.2. Field Stress Analysis & Geomechanical Sectors Classification
5.2. Field Stress Analysis & Geomechanical Sectors Classification
5.2. Field StressisAnalysis
The field exposed&toGeomechanical
compressional Sectors Classification
forces imposed by the NW movement of the
The field is exposed to compressional forces imposed by the NW movement of the
ArabianTheplate
fieldpushing against
is exposed the Zagros thrust
to compressional forcesbelt (ZTB) by
imposed andthe
theNW
distribution
movement of of
local
the
Arabian plate pushing against the Zagros thrust belt (ZTB) and the distribution of local
stress
Arabianin the Dezful
plate embayment.
pushing According
against the to the field’s
Zagros thrust reverse
belt (ZTB) andand
the thrust faults system
distribution of local
reported [42], the magnitude of maximum horizontal stress (σH ) is greater than minimum
horizontal stress (σh ) and vertical stress, respectively (σH > σh > σv ). As is seen in Figure 12a,
the northwestern and southeastern main fault zones of the field structure have created
 stress in the Dezful embayment. According to the field’s reverse and thrust faults system
reported [42], the magnitude of maximum horizontal stress ( σ H ) is greater than mini-
mum horizontal stress ( σ h ) and vertical stress, respectively ( σ H > σh > σ v ). As is seen
Processes 2024, 12, 631 11 of 20
in Figure 12a, the northwestern and southeastern main fault zones of the field structure
have created three geomechanical sectors, including the northwestern (geomechanical
sector 1), central (geomechanical sector 2), and southeastern sectors (geomechanical sector
three
3). Togeomechanical
perform a more sectors, including
in-depth analysistheofnorthwestern (geomechanical
the stress conditions, wells 337, sector 1),and
316, central
325
(geomechanical sector 2), and southeastern sectors (geomechanical sector
were selected as the representative wells for geomechanical sectors 1, 2, and 3, respec- 3). To perform a
more in-depth analysis of the stress conditions, wells 337, 316, and 325
tively, as shown in Figure 12b. Major open fractures in geomechanical sectors 1 and 3 have were selected as
the representative
a similar azimuth,wells
whileforthegeomechanical
stress regime in sectors 1, 2, and
the middle 3, respectively,
sector (geomechanical as shown
sectorin 2)
Figure 12b. Major open fractures in geomechanical sectors 1 and 3 have a
differs. The strike of maximum horizontal stress in sector 2 is NW-SE (Figure 13). The field similar azimuth,
while
stressthe stressused
analysis regime in the
more middle
wells from sector (geomechanical
the mentioned sectors sector 2) differs.
to validate The strike of
the geomechanical
maximum horizontal stress in sector 2 is NW-SE (Figure 13). The field stress analysis used
sectors identified in this study.
more wells from the mentioned sectors to validate the geomechanical sectors identified in
The classification scheme, including four classes of open fracture sets used in this
this study.
study, is consistent with Price’s classification of natural fractures [43,44]. Major and me-
The classification scheme, including four classes of open fracture sets used in this study,
isdium open fractures
consistent are aligned
with Price’s with the
classification of maximum horizontal
natural fractures stressMajor
[43,44]. ( σ H ),and
while minor
medium
open fractures
and hairline are aligned
fractures formwith the to
parallel maximum horizontal
the minimum stressstress
horizontal σ h ). minor
(σH ), (while This showsand
hairline fractures form parallel to the minimum horizontal
that two generations of fractures, syn-folding and post-folding, occur stress (σ h ). This shows that
in the studied struc-two
generations
ture. of fractures, syn-folding and post-folding, occur in the studied structure.

Figure12.
Figure 12.(a)
(a) Three
Three geomechanical
geomechanicalsectors
sectorswere
wereidentified
identifiedin
inthis
thisstudy
studybased
basedononthe
thefield
field structure’s
structure’s
NW and SE main faults. (b) Wells 337, 316, and 325 were chosen as the representative
NW and SE main faults. (b) Wells 337, 316, and 325 were chosen as the representative wells corre- wells
sponding to geomechanical sectors 1, 2, and 3, respectively. The red lines are the two major faults in
the field.
cesses 2024, 12, x FOR PEER REVIEW 12 o

Processes 2024, 12, 631 corresponding to geomechanical sectors 1, 2, and 3, respectively. The red lines are
12 ofthe
20 two m
faults in the field.

Figure
Figure 13. Strike
13. Strike of of openfractures
open fractures inin
representative wells wells
representative from geomechanical sector 1 (337),
from geomechanical sector1 2(337), se
sector
(316), and sector 3 (325). The red lines are the two major faults in the field.
2 (316), and sector 3 (325). The red lines are the two major faults in the field
5.3. Continuous Fracture Network Modeling
5.3. Continuous Fracture
The fracture Network
dip and Modeling
azimuth values derived from image log interpretation were
used to build a plane in 3D space for each fracture. Fracture intensity was calculated
The fracture dip and azimuth values derived from image log interpretation w
using the fracture plane data (fracture dip and azimuth) for each well as a P32 log, the
usedfracture
to build a plane
area in 3D
per rock unitspace for Accordingly,
volume. each fracture. P32Fracture
logs wereintensity
generatedwas calculated us
as fracture
the fracture
intensity plane
logs fordata
all 28(fracture dip
wells of the and
field azimuth)
with image logfor each well results.
interpretation as a P32 log,the
Using the fract
Gaussian Random Function Simulation (GRFS) algorithm, a 3D continuous
area per rock unit volume. Accordingly, P32 logs were generated as fracture intensity lfracture model
for28
for all allwells
types of
offractures
the field was prepared
with image (Figure 14a). A longitudinal
log interpretation and transverse
results. Using thesection
Gaussian R
through the fracture intensity model is illustrated in Figure 14b,c. Fracture intensity maps
domofFunction Simulation (GRFS) algorithm, a 3D continuous fracture model for all ty
Asmari A1 (a), C2 (b), and D2 (c) zones are shown in Figure 15a–c. The generated
of fractures was prepared
fracture intensity volume (Figure
indicates14a). A longitudinal
a higher and transverse
stress concentration section through
in areas corresponding
fracture
to theintensity
main faultmodel
zones. is illustrated
The generated inCFNFigure 14b,c. Fracture
was employed intensity
as an input maps of Asm
for the discrete
fracture modeling process.
A1 (a), C2 (b), and D2 (c) zones are shown in Figure 15a–c. The generated fracture intens
volume indicates
5.4. Discrete a higher
Fracture stress
Network concentration
Modeling for a Prototype in areas corresponding to the main fa
Sector
zones. The generated CFN was employed as an input
A model of the fractures in a prototype area was created for the discrete
using fracture
available data to model
process.
examine the impact of different fracture types on production. The southeastern sector of the
field has good coverage of both image logs and production data. Therefore, a small sector
within the southeastern flank of the field structure was chosen for the prototype modeling
(Figure 16). The modeling area includes wells 245, 347, and 363 with image logs and wells
033 and 041 with production data (Figure 16). When constructing the model, a static model
was trimmed to fit the designated fracture modeling region, which encompasses the oil-
water contact (OWC) and gas-oil contact (GOC) of the reservoir. Afterward, a continuous
and discrete fracture model of the prototype area was built. The DFN model was created
for the prototype model based on the CFN model. The maximum and minimum fracture
aperture was set to 0.005 mm and 0.000075 mm, respectively. The fracture porosity and
 Processes 2024, 12, 631 13 of 20

Processes 2024, 12, x FOR PEER REVIEW permeability parameters are listed in Tables 1 and 2, respectively. Considering moderate 13
to of 20
relatively high fracture porosity and permeability, the Asmari Formation falls in Class I
and II of the Nelson classification of fractured reservoirs [45].

(a)

(b) (c)

Figure 14. Fracture Intensity Model Analysis; (a) Asmari Reservoir: (b) Sections of Fracture Intensity
Figure 14.in
Model Fracture Intensity
Longitudinal Model
(c) and Analysis;
Transverse (a) Asmari
Directions. Reservoir:
Higher (b) Sections
fracture intensities of Fracture
highlighted Intensity
in blue
Model in Longitudinal (c) and Transverse Directions.
and green represent A1, C2, and D2 reservoir zones. Higher fracture intensities highlighted in
blue and green represent A1, C2, and D2 reservoir zones.
Processes 2024, 12, x FOR PEER REVIEW 14 of
Processes 2024, 12, 631 14 of 20

(a)

(b)

(c)

Figure Fracture
15. 15.
Figure intensity
Fracture map, Asmari
intensity A1 (a), A1
map, Asmari C2 (b),
(a), and D2 reservoir
C2 (b), zones (c). zones (c).
and D2 reservoir

5.4. Discrete Fracture Network Modeling for a Prototype Sector

A model of the fractures in a prototype area was created using available data to e
amine the impact of different fracture types on production. The southeastern sector of t
field has good coverage of both image logs and production data. Therefore, a small sec
within the southeastern flank of the field structure was chosen for the prototype modeli
(Figure 16). The modeling area includes wells 245, 347, and 363 with image logs and we
033 and 041 with production data (Figure 16). When constructing the model, a static mod
was trimmed to fit the designated fracture modeling region, which encompasses the o
 and modeled permeability proves the reliability of the image log interpretation and re-
sulting fracture network modeling. Hence, the prototype model results shown for the
studied part of the Asmari reservoir in the field may be used to promote a more realistic
and less uncertain dynamic reservoir model. Furthermore, the integrated fracture study
Processes 2024, 12, 631
workflow outlined in this research could greatly increase the predictive power of15future
of 20

fractured reservoir models.

Figure16.
Figure 16.3D
3Dvisualization
visualizationofofthe
theprototype
prototypearea
areaofofthe
thefield
fieldstructure.
structure.The
Theyellow
yellowand
andblue
bluesurfaces
surfaces
show oil-water and gas-oil contact, respectively.
show oil-water and gas-oil contact, respectively.

Table 1. Modeled fracture porosity derived from image logs for different sectors before the history
Table 1. Modeled fracture porosity derived from image logs for different sectors before the his-
match.
tory match.
Sector 1 2 3 4 5 6
Sector
Maximum porosity (fr) 1
0.0091 2
0.00678 0.004993 4
0.007317 5
0.01321 6
0.00461
Mean porosity
Maximum porosity (fr) 0.00117
0.0091 0.000971
0.00678 0.000879
0.00499 0.00115
0.007317 0.001494
0.01321 0.001157
0.00461
Mean porosity (fr) (fr)
Minimum porosity 0.00117
0.000027 0.000971
0.0000418 0.000879
0.000077 0.00115
0.000028 0.001494
0.000011 0.001157
0.000087
Minimum porosity (fr) 0.000027 0.0000418 0.000077 0.000028 0.000011 0.000087

Table 2. Modeled fracture permeability was derived from DST data for different sectors before the
history match.

Sector Kx (md) Ky (md) Kz (md)

Max 2307 5000 2380.9
1 Mean 391 486.4 659.8
Min 17.44 8.95 7.9
Max 1618.6 5000 3653.4
2 Mean 345.6 415.8 445.8
Min 3.97 14 10
Max 1338.6 4030 1767.6
3 Mean 335.97 434 411.3
Min 32 32.8 34.3
Max 3058 5000 3043.2
4 Mean 404.8 538.8 543.4
Min 6 5.96 12.16
Max 3675.8 5000 4499.5
5 Mean 500 672.7 664.8
Min 4.8 4 3.75
Max 1347 1521 3318.4
6 Mean 423.7 398.9 500.8
Min 22.7 35.5 30

By upscaling the DFN to the simulation grid, fracture porosity, permeability, sigma,
and block height models were generated. The fracture permeability model for major
open fractures before (a) and after (b) upscaling is shown in Figure 17 for illustration. The
permeability decreases from major open fractures to hairline fractures (Figure 18). However,
as permeability is a vector parameter showing variations in the simulation cell’s i, j, and k
directions, the minor and hairline fractures contribute to fluid flow perpendicular to the
direction of major open fracture sets.
 5 Mean 500 672.7 664.8
Min 4.8 4 3.75
Max 1347 1521 3318.4
6 Mean 423.7 398.9 500.8
Processes 2024, 12, 631 Min 22.7 35.5 30 16 of 20

Processes 2024, 12, x FOR PEER REVIEW 17 of 20

Fracturepermeability
Figure17.17.Fracture
Figure permeability model for
for major
majoropen
openfractures before
fractures (top)
before andand
(top) afterafter
(bottom) upscaling.
(bottom)
upscaling.

Figure18.
Figure 18. Fracture
Fracture permeability
permeability histograms
histograms for major
for major open, medium
open, medium open,and
open, minor, minor, andfrac-
hairline hairline
tures. Permeability
fractures. showsshows
Permeability a clearadecreasing trend from
clear decreasing trendmajor
fromopen
majortoopen
hairline fractures.
to hairline fractures.

5.5. Limitations of the Methodology

Although the integrated methodology adopted in this study shows an advantageous
workflow for fracture characterization and modeling, it has some limitations, as follows:
(i) The CFN and DFN models heavily rely on the interpretation results of image logs
Processes 2024, 12, 631 17 of 20

On the other hand, Price’s classification scheme was used for field fracture classifi-
cation. Accordingly, major and medium open fractures form parallel to the maximum
horizontal stress direction, while minor and hairline fractures are aligned with the minimum
horizontal stress. Such good agreement between the proposed open fracture classification
and modeled permeability proves the reliability of the image log interpretation and result-
ing fracture network modeling. Hence, the prototype model results shown for the studied
part of the Asmari reservoir in the field may be used to promote a more realistic and less
uncertain dynamic reservoir model. Furthermore, the integrated fracture study workflow
outlined in this research could greatly increase the predictive power of future fractured
reservoir models.

5.5. Limitations of the Methodology

Although the integrated methodology adopted in this study shows an advantageous
workflow for fracture characterization and modeling, it has some limitations, as follows:
(i) The CFN and DFN models heavily rely on the interpretation results of image logs such
as FMI, CAST, and UBI. An adequate number of wells should have image logs that
are geologically well distributed on the structure. Without adequate, good-quality
image logs, fracture models based on geological data will be highly uncertain.
(ii) The image log interpreter should have sufficient experience interpreting complex and
heterogeneous carbonate reservoirs. Otherwise, some important fracture information
may be misinterpreted, causing uncertainty in the fracture model.
(iii) Field observation data must be incorporated into the fracture modeling software. For
example, if the maximum fracture length is 100 m based on field observations, then
the outputs of fracture models can be validated. Otherwise, the wrong parameter
setting of a fracture model will not honor production data.
(iv) The fracture modeler must be familiar with geological, petrophysical, and reservoir
simulation disciplines. For example, a fracture modeler can be a geologist with a good
background in reservoir dynamic model parameters.

6. Conclusions
This study investigated the correlation between fractures and oil production in a giant,
fractured carbonate field using image logs, curvature maps, stress maps, mud loss data, and
production data. Four fracture sets were identified based on the interpretation of the image
logs of the studied field, including major open fractures, medium open fractures, minor
open fractures, and hairline fractures. Using the UGC maps, a curvature map was generated
for the reservoir units in the Asmari Formation. Subsequently, an investigation was
conducted to analyze the correlation between fracture characteristics and production data.
A relatively good correlation was found between the fracture model, curvature map, mud
loss map, fracture intensity, and productivity index. Studying the stress regime showed that
three individual geomechanical sectors occur in the field structure, showing different stress
directions. Major open and medium open fractures align well with the maximum horizontal
stress direction. In contrast, minor open fractures and hairline fractures are parallel to the
direction of minimum horizontal stress. A small sector in the southeastern part of the field
was selected to build a prototype fracture model. Both continuous fracture network (CFN)
and discrete fracture network (DFN) models were developed for the prototype model.
Fracture length was modeled by a normal distribution function. Fracture orientation
was modeled by using the Fisher probability distribution model. Finally, 3D models of
fracture dip and azimuth, fracture porosity, fracture permeability, fracture length, fracture
aperture, and the sigma parameter were computed for the prototype model. The DFN
model established for the prototype area could serve as a baseline to investigate the impact
of various fracture types on the dynamic simulation model and to address the frequent
production challenges encountered in the field under examination.
The results show a good agreement between high fracture intensity zones and produc-
tivity index maps. Mud loss data also confirms the fracture model results. The methodology
Processes 2024, 12, 631 18 of 20

adopted in this study can be applied to other complex carbonate reservoirs to address the
heterogeneities resulting from variations in fracture parameters such as aperture, length,
density, and continuity. The prototype model performs well in only one sector of the field.
Following the same methodology in all sectors and feeding the outputs to the dynamic
model will provide a more realistic dynamic model. This is compared to the case in which
only a constant value for fracture porosity, permeability, and sigma is assumed within
each sector.

Author Contributions: Conceptualization, R.K., H.O. and A.K.; methodology, R.K., A.K., S.A. and
D.M.; formal analysis, H.F., G.S., E.R., A.K. and S.A.; investigation, R.K., A.K., S.A. and D.M.;
software, A.K., S.A., H.F., G.S. and E.R.; supervision, H.O., R.K. and J.M.; writing—review and
editing, R.K., A.K., H.O., D.M. and S.A. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by the Petroleum University of Technology—grant number
GSPUT20C01.
Data Availability Statement: Data are contained within the article.
Acknowledgments: We express our gratitude for the support and permission to publish the results,
which were made possible through the generous support of the National Iranian South Oil Company
and the Petroleum University of Technology.
Conflicts of Interest: Authors Hashem Fardin, Ghasem Saedi, and Esmaeil Rokni were employed
by NISOC. The remaining authors declare that the research was conducted in the absence of any
commercial or financial relationships that could be construed as a potential conflict of interest.

References
1. Bratton, T.; Hunt, D.W.; Gillespie, P.A.; Li, B. The naturally fractured reservoirs. Oilfield Rev. 2006, 18, 4–23.
2. Mason, H.E.; Smith, M.M.; Carroll, S.A. Calibration of NMR porosity to estimate permeability in carbonate reservoirs. Int. J.
Greenh. Gas Control 2019, 87, 19–26. [CrossRef]
3. Kharrat, R.; Zallaghi, M.; Ott, H. Performance Quantification of Enhanced Oil Recovery Methods in Fractured Reservoirs. Energies
2021, 14, 4739. [CrossRef]
4. Kharrat, R.; Ott, H. A Comprehensive Review of Fracture Characterization and Its Impact on Oil Production in Naturally
Fractured Reservoirs. Energies 2023, 16, 3437. [CrossRef]
5. Kharrat, R.; Alalim, N.; Ott, H. Assessing the Influence of Fracture Networks on Gas-Based Enhanced Oil Recovery Methods.
Energies 2023, 16, 6364. [CrossRef]
6. Wynn, T.J.; Stewart, S.A. Fracture and In-Situ Characterization of Hydrocarbon Reservoirs. Geol. Soc. Lond. Spec. Publ. 2003, 209,
127–143. [CrossRef]
7. Verney, P. Interpretation geologique de donnees sismiques par une methode supervisee basee sur la vision cognitive. Ph.D. Thesis,
Ecole Nationale Superieure des Mines de Paris, Paris, France, 2009; 190p.
8. Berberian, M. Master “blind” thrust faults hidden under the Zagros folds: Active basement tectonics and surface morphotectonics.
Tectonophysics 1995, 241, 193–224. [CrossRef]
9. Aydin, A. Fractures, faults, and hydrocarbon entrapment, migration, and flow. Mar. Pet. Geol. 2000, 17, 797–814. [CrossRef]
10. Barton, C.A.; Zoback, M.D. Discrimination of natural fractures from drilling-induced wellbore failures in wellbore image
data-implications for reservoir permeability. SPE Reserv. Eval. Eng. 2002, 5, 249–254. [CrossRef]
11. Rajabi, M.; Sherkati, S.; Bohloli, B.; Tingay, M. Subsurface fracture analysis and determination of in-situ stress direction using FMI
logs: An example from the Santonian carbonates (Ilam Formation) in the Abadan Plain, Iran. Tectonophysics 2010, 492, 192–200.
[CrossRef]
12. Spence, G.; Couples, G.; Bevan, T.G.; Redfern, J. Advances in the study of naturally fractured hydrocarbon reservoirs: A broad
integrated interdisciplinary applied topic. Geol. Soc. Lond. Spec. Publ. 2014, 374, 1–22. [CrossRef]
13. Abdel Azim, R. Prediction of Naturally Fractured Reservoir Performance using Novel Integrated Workflow. Int. J. Adv. Comput.
Sci. Appl. 2017, 8, 115–122. [CrossRef]
14. Pejic, M.; Kharrat, R.; Kadkhodaaie, A.; Azizmohammadi Ott, H. Impact of fracture types on the oil recovery in naturally fractured
reservoirs. Energies 2022, 15, 7321. [CrossRef]
15. Kolapo, P.; Ogunsola, N.O.; Munemo, P.; Alewi, D.; Komolafe, K.; Giwa-Bioku, A. DFN: An Emerging Tool for Stochastic
Modelling and Geomechanical Design. Eng 2023, 4, 174–205. [CrossRef]
16. Jing, L.; Stephansson, O. 10-Discrete Fracture Network (DFN) Method. Dev. Geotech. Eng. 2007, 85, 365–398.
17. Tokhmchi, B.; Memarian, H.; Rezaee, M.R. Estimation of the fracture density in fractured zones using petrophysical logs. J. Pet.
Sci. Eng. 2010, 72, 206–213. [CrossRef]
Processes 2024, 12, 631 19 of 20

18. Jafari, A.; Babadagli, T. Estimation of equivalent fracture network permeability using fractal and statistical network properties. J.
Pet. Sci. Eng. 2012, 92–93, 110–123. [CrossRef]
19. Zazoun, R.S. Fracture density estimation from core and conventional well logs data using artificial neural networks: The
Cambro-Ordovician reservoir of Mesdar oil field, Algeria. J. Afr. Earth Sci. 2013, 83, 55–73. [CrossRef]
20. Nie, X.; Zou, C.; Pan, L.; Huang, Z.; Liu, D. Fracture analysis and determination of in-situ stress direction from resistivity and
acoustic image logs and core data in the Wenchuan Earthquake Fault Scientific Drilling Borehole-2 (50–1370 m). Tectonophysics
2013, 593, 161–171. [CrossRef]
21. Wennberg, O.P.; Casini, G.; Jonoud, S.; Peacock, D.C.P. The characteristics of open fractures in carbonate reservoirs and their
impact on fluid flow: A discussion. Pet. Geosci. 2016, 22, 91–104. [CrossRef]
22. Kosari, E.; Ghareh-Cheloo, S.; Kadkhodaie, A.; Bahroudi, A. Fracture characterization by fusion of geophysical and geomechanical
data: A case study from the Asmari reservoir, the Central Zagros fold-thrust belt. J. Geophys. Eng. 2015, 12, 130–143. [CrossRef]
23. Kosari, E.; Kadkhodaie, A.; Bahroudi, A.; Chehrazi, A.; Talebian, M. An integrated approach to study the impact of fractures
distribution on the Ilam-Sarvak carbonate reservoirs: A case study from the Strait of Hormuz, the Persian Gulf. J. Pet. Sci. Eng.
2017, 152, 104–115. [CrossRef]
24. Fang, J.; Zhou, F.; Tang, Z. Discrete fracture network modelling in a naturally fractured carbonate reservoir in the Jingbei Oilfield,
China. Energies 2017, 10, 183. [CrossRef]
25. Correa, R.S.M.; Pereira, C.E.L.; Cruz, F.A.S.; Lisboa, S.N.D.; Junior, M.P.A.; Carvalho, B.R.B.M.; Souza, V.H.P.; Rocha, C.H.A.;
Araujo, F.G. Integrated Seismic-Log-Core-Test Fracture Characterization, Barra Velha Formation, Pre-salt of Santos Basin; AAPG
Search and Discovery Article #42425. In Proceedings of the AAPG Annual Convention and Exhibition, San Antonio, TX, USA,
19–22 May 2019.
26. Hosseini, E.; Sarmadivaleh, M.; Chen, Z. Developing a new algorithm for numerical modeling of discrete fracture network (DFN)
for anisotropic rock and percolation properties. J. Pet. Explor. Prod. 2021, 11, 839–856. [CrossRef]
27. Aghli, G.; Moussavi-Harami, R.; Mohammadian, R. Reservoir heterogeneity and fracture parameter determination using electrical
image logs and petrophysical data (A case study, carbonate Asmari Formation, Zagros Basin, SW Iran). Pet. Sci. 2020, 17, 51–69.
[CrossRef]
28. Ja’fari, A.; Kadkhodaie, A.; Sharghi, Y.; Ghanavati, K. Fracture density estimation from petrophysical log data using Adaptive
Neuro-Fuzzy Inference System. J. Geophys. Eng. 2012, 9, 105–114. [CrossRef]
29. Ja’fari, A.; Kadkhodaie, A.; Sharghi, Y.; Ghaedi, M. Integration of Adaptive Neuro-Fuzzy Inference System, Neural Networks
and Geostatistical Methods for Fracture Density Modeling. Oil Gas Sci. Technol. Rev. IFP Energ. Nouvelles 2014, 69, 1143–1154.
[CrossRef]
30. NISOC. National Iranian South Oil Company’s Official Website. 2023. Available online: www.nisoc.ir (accessed on 22 March 2023).
31. Talbot, C.J. Extrusions of Hormuz salt in Iran. Geol. Soc. Lond. Spec. Publ. 1998, 143, 315–334. [CrossRef]
32. Sarvandani, M.M.; Kalateh, A.N.; Ghaedrahmati, R.; Majidi, A. Investigating subsurface structures of Gachsaran oil field in Iran
using 2D inversion of magnetotelluric data. Explor. Geophys. 2018, 48, 148–162. [CrossRef]
33. Alavi, M. Regional stratigraphy of the Zagros fold-thrust belt of Iran and its proforeland evolution. Am. J. Sci. 2004, 304, 1–20.
[CrossRef]
34. Gholami Vijouyeh, A.; Kadkhodaie, A.; Hassanpour Sedghi, M.; Hamed Gholami Vijouyeh, H. A committee machine with
intelligent experts (CMIE) for estimation of fast and slow shear wave velocities utilizing petrophysical logs. Comput. Geosci. 2022,
165, 105149. [CrossRef]
35. Slinger, F.G.P.; Crichton, J.G. The geology and development of the Gachsaran field, southwest Iran. In Proceedings of the Fifth
World Petroleum Congress, New York, NY, USA, 30 May–5 June 1959; Volume 18, pp. 349–375.
36. Shammas, P. Iran: Review of Petroleum Developments and Assessments of the Oil and Gas Fields. Energy Explor. Exploit. 2001, 19,
207–260. [CrossRef]
37. Zoback, M.D. Reservoir Geomechanics; Cambridge University Press: Cambridge, UK, 2007.
38. Stewart, S.A.; Podolski, R. Curvature analysis of gridded geological surfaces. in Coward, M.P., Daltaban, T.S., Johnson, H., eds.,
structural geology in reservoir characterization. Geol. Soc. (Lond.) Spec. Publ. 1998, 209, 133–147. [CrossRef]
39. Antonellini, M.; Aydin, A. Effect of faulting on fluid flow in porous sandstones: Geometry and spatial distribution. Am. Assoc.
Pet. Geol. Bull. 1995, 79, 642–671.
40. Rijks, E.J.H.; Jauffred, J.C.E.M. Attribute extraction: An important application in any detailed 3D interpretation study. Geophysics
1991, 10, 11–19.
41. Karimnejad Lalami, H.R.; Hajialibeigi, H.; Sherkati, S.S.; Adabi, M. Tectonic evolution of the Zagros foreland basin since Early
Cretaceous, SW Iran: Regional tectonic implications from subsidence analysis. J. Asian Earth Sci. 2020, 204, 104550. [CrossRef]
42. Farshi, M.; Moussavi-Harami, R.; Mahboubi, A.; Khanehbad, M. Reservoir rock typing using integrating geological and
petrophysical properties for the Asmari Formation in the Gachsaran oil field, Zagros basin. J. Pet. Sci. Eng. 2018, 176, 161–171.
[CrossRef]
43. Price, N. Fault and Joint Development in Brittle and Semi-Brittle Rock; Pergamon Press: Oxford, UK, 1966; p. 176.
Processes 2024, 12, 631 20 of 20

44. Price, N.; Cosgrove, J. Analysis of Geological Structures; Cambridge University Press: Cambridge, UK, 1990; p. 502.
45. Nelson, R. Geologic Analysis of Naturally Fractured Reservoirs; Elsevier: Amsterdam, The Netherlands, 2001; p. 352.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.