See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/266670712

Value of Well Test in the Determination and Characterization of Natural

Fractured Reservoir Properties in Large Onshore Abu Dhabi Carbonate Field

Article · November 2012

DOI: 10.2118/162356-MS

CITATION READS

1 130

9 authors, including:

Abdul Samad Manhal Sirat

Occidental Petroleum Corporation Geosolutions Consultancy Sweden
2 PUBLICATIONS 2 CITATIONS 53 PUBLICATIONS 720 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Manhal Sirat on 30 April 2016.

The user has requested enhancement of the downloaded file.

SPE 162356

Value of Well Test in the Determination and Characterization of Natural

Fractured Reservoir Properties in Large Onshore Abu Dhabi Carbonate Field
Abdul Samad Abdul Rehman, Arun Kumar, Riadh Bejaoui, Nurbek Nazarbayev, Mohamed Shuaib, Manhal Sirat,
Maniesh Singh, Taha Al-Dayyani – Abu Dhabi Company for Onshore Oil Operation (ADCO); Cristian Ancuta,
Angelo Farcasanu, Ali Gholipour, Mandefro Belayneh - SENERGY

Copyright 2012, Society of Petroleum Engineers

This paper was prepared for presentation at the Abu Dhabi International Petroleum Exhibition & Conference held in Abu Dhabi, UAE, 11–14 November 2012.

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 have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper 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 SPE copyright.

Abstract
Identifying natural fractures and assessing their impact on flow behavior in hydrocarbon reservoirs have been one of the key
challenges in defining the field development strategy. Failure to correctly represent the natural fractures when working on a
reservoir development plan can lead to poor performance and failure to analyze reservoir response. Pressure transient testing
is widely recognized as a core competency in the oil industry as it provides subsurface specialists with valuable information
regarding the reservoir properties, well connectivity to the reservoir and extent of reach into the reservoir.

When borehole image (BHI) data triggers a signal that fractures are seen in wells but flow rates do not show production
dominated by fractures, the engineers working on the reservoir need to pay attention to pressure transient tests. Depending on
the contrast between fracture permeability and matrix permeability, pressure transient analysis in a fractured reservoir can
show clear dual porosity behavior, or the effect can be masked by the wellbore storage. The latter happens when the contrast
between the matrix permeability and the fracture permeability is small.

The two parameters which characterize a dual porosity model are the storativity ratio, !, and the inter-porosity flow
coefficient, ". The average value of " for the vertical wells studied is 25*10-4, confirming a low contrast between fissure and
matrix permeability. For the storativity ratio, a typical value is between 0.001% and 0.1%. An average of 0.03% has been
calculated from the well tests.

With the results from pressure transient tests, BHI and core description, fractures have been defined as mild with relative
low-intensity. Numerical modeling on a field sector was employed to test the impact of the fractures on production.

Background
Production data at the early phase of development provided limited evidence with respect to the effects of natural fracturing
on reservoir performance. Most of the differences in oil production performance between wells are consistent with diagenetic
changes in the reservoir matrix, with poorer performance from downdip wells in tighter rock. Water & gas movements at the
Water Alternating Gas (WAG) pilots show a thin thief zone which is apparently a horizontally micro-fractured vuggy zone.
Outside the WAG pilot area, there has been limited fluid breakthrough. The drilling mud losses have also had very limited
use in this study. No reliable PLT tests were available for calibrating the fracture intensity and particularly the presence of the
fracture swarm zones in reservoir.

Introduction
Modeling fractured reservoirs still remains a challenge. Depending on type of fractures (Nelson1, 1985), different
development schemes were applied while getting the oil barrels to surface. Integration now is a daily routine in most of the
companies and all the disciplines collect information and provide interpretations. In some cases, all the data align correctly
toward conclusions. In some other cases, the information creates more debates than answers.

According to Allan and Qing Sun2 different recovery factors are to be expected from different types of fractures. Whether
secondary or EOR methods are applicable depends on reservoir description, rock and fluids properties and matrix to fracture
2 SPE 162356

interaction. As no rule of thumb works to development decision, information has to be used at maximum best to mitigate the
uncertainties. Static and dynamic data are used in conjunction to characterize the fracture density and geometry. Well tests
can provide evidence of the fluid flow through fractures. Conventional well test analysis is based on ideal mathematical
formulation rather than geologically realistic fracture networks. Simply matching the models available in the specialized
software library may not reveal the correct nature of pressure response. Engineers have to analyze and use best judgment on
what is behind the matching plots.

As a result of PTA, BHI and core description, the fractures have been characterized as mild with low intensity. The impact of
the fractures has been tested on conceptual sector models.

Fracture System Description

Usually the availability of direct data on fracturing is minimal in all fractured reservoirs. This scarcity of data creates lots of
challenges in characterizing the fractures. Therefore, availability and suitability of the seismic, dynamic, and geological data,
as the regional drivers that carry the fracture information, are essential for fracture characterization in these reservoirs.

The overview of structural geology identified the major deformational events that have either direct or indirect influence on
the formation of the structure and associated fractures.

Structuration analysis, in conjunction with a study of 886 ft of core from 10 wells, BHI log interpretation and thin sections,
led to three-fold classifications of fractures: (1) Stylolite-related vertical fractures, and horizontal fractures related to uplift,
erosion and releasing excess pore pressure, (2) fold-related fractures, and (3) fault-related fractures. Fault-related fractures are
the most abundant of all fractures in the studied field (for details see Sirat3 et al., 2012).

A geomechanical review showed that the in situ stress interpretation is still an issue. On a field scale, the maximum
horizontal stress interpreted from well data may deviate from the regional orientation to lie parallel to some of the major
faults. Existing faults, particularly those oriented obliquely to the regional maximum horizontal stress can cause local
perturbations of the stress and induce local stress re-orientation as a result of irregularities and discontinuities over the
structure, have been confirmed by studies. The data from observed stresses in a particular area can be explained as the global
stress field overprinted by regional and local effects and variations can occur from well to well.

Additional rock mechanical test data were analyzed and used, to verify the relationship between the fracture intensity and
rock mechanical properties.

Five types of fractures (F1 to F5) which are different in size, connectivity, and intensity, have been identified throughout the
reservoir. For modeling purposes they have been categorized into two groups ( Fig. 1):

- Group one consists of large, more connected and more intense fractures called fault-related fractures and fold-related
fractures (F1 & F2 types).
- Group two consists of small, less connected, less intense and more diffused fractures (F3, F4 and F5).

Based on the groups defined, a conceptual model was created which looks like (Fig. 2). The fracture intensity model
confirmed that the open fractures are mainly associated with the major faults with the NW-SE trend (Fig. 3).

Pressure Transient Analysis in Naturally Fractured Reservoirs

Pressure transient analysis (PTA) has been a core competency of the oil and gas industry for a long time and it provides
engineers with valuable information necessary to develop a better understanding of reservoir characterization and
performance predictions. PTA is an approach which allows testing of the reservoir in dynamic conditions.

Naturally fractured reservoirs (NFR) are recognized as challenging reservoirs to manage and contain fractures created by
nature which could have an effect, either positive or negative, on fluid flow. NFR has two different porous media, the matrix,
which in general has high storage but low flow capacity, and the fractures, which provide high flow capacity but low storage
capacity.

The main objectives of PTA in a natural fractured system are to determine the system permeability, skin factor and to
estimate the dual porosity parameters. Pressure transient approach in well tests, appear to be a powerful way of achieving this
objective.

In a fractured reservoir, the evaluation of flow network characteristics is very important as it can have an impact on the early
appraisal stage as well as in mature field development stage. Flow characteristics of the fractures themselves (permeability,
porosity, compressibility, etc.) are better determined from sensitivity studies on well test data than from geological
SPE 162356 3

observations of well fracture aperture, mineralization reactivation, roughness and other fracture characteristics liable to
influence fluid flow [1]. That is why a good understanding of the fracture system is essential for a successful field
development plan since the fissures have a significant impact on the fluid flow and well productivity. Accurate fracture
description in reservoir simulation models is critical for reliable forecasts.

Natural fractured reservoirs are controlled by two parameters, namely the storativity ratio (!) and the interporosity flow
coefficient ("). Storativity ratio is the fraction of fluid stored in the fissures system and is given by:
!"!! !
!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!"!! !!!"

Interporosity flow coefficient characterizes the communication between matrix blocks and fractures and it is a function of
shape factor (matrix block size) and fracture matrix permeability ratio and it is given by the equation:
!!"
! ! !!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!"

This coefficient controls the total time of V-shaped transition ( Fig. 4). A low ! imply a tight matrix for which a larger
drawdown is needed to be established before the matrix blocks start feeding the fractures. In this case the V-shaped transition
starts later. The higher the values for ", the smaller occurring time for the V-shaped transition and most likely the dual
porosity transition is hidden by the wellbore storage.

Depending on the " values, PTA in NFR can show a clear dual porosity behavior or the effect can be completely masked by
the wellbore storage. The latter happens when the contrast between the matrix permeability and the fracture permeability is
small and/or there is high shape factor (low values for fracture spacing) which imply large value for ".

Case Study
The main objective of this paper is to characterize the natural fractures in an oil-bearing reservoir, onshore Abu Dhabi. The
study was conducted by analyzing the pressure transient data from 29 vertical wells (oil producers) and a water injector in a
WAG pilot area. In many cases a homogeneous response can be read. The pressure response will be matched with four
interpretation models such as: homogeneous reservoir, fractured well (finite/infinite conductivity), dual porosity reservoir and
double permeability. All the analyzed wells show a high negative skin from -3 to -5.5 (Fig. 5), indicating there is an
enhancement which can be due to strong stimulation or fissures presence. To sustain a high negative skin due to stimulation
all the wells have to have large acid jobs. On checking the wells history, no record of stimulation was found.

Naturally occurring fractures were identified on BHI and cores, and PTA data was used to take a closer look at their
characteristics.

The PTA of the water injector well in the WAG Pilot was carefully analyzed over the water cycles. The analysis revealed that
the wellbore storage coefficient (WBS) is nearly two orders of magnitude higher (Table 1) than the WBS calculated from
wellbore geometry. As the wellbore storage represents the connected volume, the difference between the WBS calculated
based on the well geometry and WBS from interpretation suggests the presence of fractures network intercepting the
wellbore. A negative skin associated with a high wellbore storage usually indicates a fissured reservoir even if the pressure
behavior appears to be homogeneous (Gringarten A., 1984). The average value of " for the vertical wells studied is 2.5E-3
confirming a low contrast between fissures and matrix permeability.

From the BHI and core study the equivalent average matrix block size is approximately 20 ft for this reservoir. The plot of
matrix blocks size versus !/rw2 for different fracture-matrix permeability ratios, Fig. 6, indicates a permeability ratio in
between 1 and 10.

As a direct consequence of high ! coupled with a moderate to high wellbore storage (wells are shut-in on the surface), the
dual porosity V-shaped transition is most likely masked by the WBS effects on the log-log plot (Fig. 7). For the storativity
ratio, a typical value is between 0.001% and 0.1%. An average of 0.03% has been calculated from the well tests
interpretation of the 29 vertical wells.

The workflow followed in this case study is summarized in the following steps (Table 2 for exemplification):
Step 1 from PTA => ", !
Step 2 from " => k_PTA/km
Step 3 from k_PTA/kma => kf/km
Step 4 from k_PTA/kma => Effective Matrix Perm, kma
4 SPE 162356

Step 5 from k_PTA => Effective Fracture Perm, kf

Step 6 calculate aperture wf
Step 7 calculate fracture porosity

Numerical Modeling
Numerical modeling approach has been largely employed to validate the PTA well test interpretation model by using a radial
simulation model. A layer cake model has been built for a water injector well by keeping the number of layers and properties
identical with the 3D static model. In accordance with the PTA results, a dual porosity – dual permeability concept has been
considered.

The fracture properties as permeability, aperture and porosity have been estimated as follows:

- Effective fracture permeability:

!!!" ! !"! ! ! !
!!"!! ! ! !!! ! ! ! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! !! !! !!

- Fracture aperture:

!
! ! ! !"# ! ! ! ! !"
!! ! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!" ! !"! ! ! !
! ! ! ! !!!
! !! !! !!
where:
!
!
!!! !" ! !!"!!"#"$%!!"#$% ! !!" !!" ! !! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!
!
! !
!!! !"# ! !!"!!"#"$%!!"#$% ! !!" !!" ! !!"!! ! !! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!

The average fracture aperture calculated is less than 0.1 mm.

A radial model was set up for the injector and pressure transient response matching was intended. The model was built using
information from BHI (Fig. 8) and the result of history matching is presented in Fig. 9.

After matching the pressure response, the radial model has been used to assess the impact of the fractures on fluid movement
(Fig. 10). As expected the fluids were seen following the fractures even though the fractures permeability values are only
twice as high as the matrix permeability. No attempt was made to account for wettability effect as no conclusive SCAL data
was available.

Once the model properties have been confirmed through history matching, a small sector model (quarter of a 5 spot pattern)
has been built using the same matrix and fracture properties as used for the injector to assess the impact of fractures on
waterflooding. An identical sector model stripped by fractures was used for comparison purpose (Fig. 11 and Fig. 12). Clear
faster movement of the water through the fracture network resulted in an early water breakthrough for the fractured model
versus the homogeneous model (Fig. 13).

Data Integration
Seismic data, core data, log data and dynamic data were all used to build and confirm the understanding of fracture network
system. Fig. 14 summarizes the inputs into fracture modeling and is briefly described in the following paragraphs.

Information from core and BHI, followed by Rock Mechanical Test (RMT) and Petrophysics, helped in defining the rock
mechanical properties and their control on the rock strength. The seismic data and structurisation history helped in defining
fracture type, origin, distribution of faults and associated fractures. The PTA and diagnostic test defined the effective
permeability and impact on fluid flow.

The cores and BHI logs data are the most credible hard data that have been used in this study. The following main parameters
that are essential for fracture modeling have been measured from these two data sets:
• Fracture types Core & BHI
SPE 162356 5

• Fracture sets BHI

• Fracture orientation BHI
• Fracture aperture Core & BHI
• Fracture length Core & BHI
• Fracture density Core & BHI

Structural history in the area and particularly the evolution of the structure in integration with the study of cores and BHI
logs enabled us to propose a three-fold classification of fractures. This classification is: (1) Pre- and syn-folding tectonic
fractures including fold-related fractures formed during the collision of the Oman micro-plate with Arabia, (2) Fault-related
fractures (Fig. 15), and (3) Fractures associated with burial, uplift and un-roofing. In this study three origins have been
identified for fractures:
• Tectonic
• Unloading/Stress and pore pressure release
• Diagenesis

PTA results were compared with the discrete fracture distribution resulting from the above work and a reasonable match was
found (Fig. 16).

The purpose of modeling was to build an improved geomodel including fractures and test the effect of fractures on fluids
flow. Sector models have been built, as described earlier, using the fracture effective permeability from PTA and fracture
density from geomodeling ground work.

Conclusions
Core and BHI data analyzed as part of this case study demonstrated the existence of natural fractures within the subject
reservoir although the production data shows limited evidence with respect to their effects on reservoir performance. A
reinterpretation of the PTA data was done in the light of information from cores and BHI and a fractured system was
discovered and characterized. The reinterpretation of PTA data provided valuable insights with regards to the characterization
of these fractures and was found to be generally consistent with the core & BHI observations of a low-intensity naturally
fractured reservoir. It also indicated some localized parts of the reservoir where fracture networks were interconnected over
larger distances. Flow patterns and sweep efficiency, expected to be influenced by fracture flow in those localized higher-
permeability fractured areas, and as observed at the WAG pilots was verified by the numerical sector modeling.

Acknowledgment
The authors would like to thank the management from ADCO and ADNOC for giving permission to use their data and to
publish this paper. ADCO Management is thanked for their continuous support.
6 SPE 162356

Nomenclature
c = compressibility, 1/psi
C = roughness coefficient (96 " C " 400)
h = thickness, ft
k = permeability, mD
L = matrix block size, ft
n = number of grid cells in z direction
r = radius, ft
Swi = connate water saturation, fraction
V = volume, ft
w = fracture aperture, mm
! = geometric parameter (is related to the fissure network geometry), 1/ft2
! = porosity, fraction
" = interporosity flow coefficient
# = ratio of fracture storativity to total system storativity

Subscripts
ef = effective
ma = matrix
f = fracture
o = oil
PTA = pressure transient analysis
r = relative
T = total
w = well
x = in the x direction
y = in the y direction
z = in the z direction

References
1) Nelson, R. A.: Geological Analysis of Naturally Fractured Reservoirs, Gulf Publishing Co., Houston (1985).
2) Jack Allan and S. Qing Sun “Controls on Recovery Factor in Fractured Reservoirs: Lessons Learned from 100
Fractured Fields” Petroleum Exploration and Development, vol-30, No-6, 129
3) Manhal Sirat, Taha Al-Dayyani, Maniesh Singh, Abdul Samad Abdul Rehman, Sabry Lotfy Mahmoud, Naema
Obaid Al-Zaabi, Mohamed Shuaib and Michel Moge (2012). Characterization of Fractures in Large Onshore Abu
Dhabi Carbonate Field. SPE-162342, P. 1–14
SPE 162356 7

Table 1 - Water Injector – Wellbore Storage Coefficient

Table 2 – Workflow Table summarizing steps and results

Fig. 1 – Dominant fracture types and sets Fig. 2 - Conceptual fracturation model
8 SPE 162356

Fig. 3 - Fractures trend

Fig. 4 - Dual Porosity Reservoir – V shaped transition Fig. 5 - Wells location and skin category

Fig. 6 - Fracture – Matrix Permeability Ratio Fig. 7 - Log-Log Plot - Interporosity Flow Sensitivity
SPE 162356 9

Fig 8 - Fracture distribution from BHI

Fig. 9 - Numerical model well test matching Fig. 10 - Effect of fractures on water injection

Fig 11 - Water saturation profile – homogeneous model Fig. 12 – Water saturation profile – DPDP model
 10 SPE 162356

Fig 13 – WCT behavior – Homogeneous vs DPDP model

Fig 14 - Fracture modeling Fig. 15 – Fracture types and possible origins

Fig. 16 - Data Integration: seismic, BHI and PTA

View publication stats