SPE 168077

Successful Application of a Novel Mobility Geosteering Technique in a

Stratified Low-Permeability Carbonate Reservoir
Salem H. Al-Suwaidi, SPE, Stig Lyngra, SPE,Saudi Aramco; Iwan Roberts, SPE, Jihad Al-Hussain, SPE, Ihsan
Pasaribu, SPE, Schlumberger; Abdalla S. Laota and Sanggam Hutabarat,Saudi Aramco

Copyright 2013, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Saudi Arabia section Annual Technical Symposium and Exhibition held in Khobar, Saudi Arabia, 19–22 May 2013.

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
A novel hybrid steering methodology employed for a horizontal development well in Saudi Arabia makes use of three
independent and mutually supportive geosteering concepts: mobility steering, a new real-time true stratigraphic thickness
(TST) path-integration method, and distance-to-boundary mapping. This technique, employed for a horizontal development
well in Saudi Arabia, proved the feasibility of mobility steering in a permeable zone of limited thickness, without any
characteristic log markers separating the target zone from the low-permeability rock located above and below, supported only
by the distance-to-boundary mapping and TST path integration techniques.
A mature giant Middle Eastern field contains a previously undeveloped stratigraphically trapped carbonate oil
accumulation discovered in the 1980s. Several delineation wells were drilled at that time, but no recent penetrations exist as
this reservoir is located deeper than the producing horizons. The reservoir of interest contains high-quality oil, but the rock is
predominantly <1 millidarcy (md) permeability.
In early 2012, the first development well was spudded. A 30° slanted pilot hole was first drilled and extensively logged.
The well data confirmed the low permeability, but also demonstrated limited development of higher permeability rock in
vertically restricted zones within a few feet of the top of the reservoir.
A horizontal development sidetrack well was subsequently drilled using advanced real-time logging-while-drilling (LWD)
measurements to steer along the thin higher-quality interval. Potential practical obstacles in achieving optimum well placement
were the lack of traditional log signatures or markers correlating with permeability, uncertainty about lateral and vertical
extent of permeable rock, and no database of existing horizontal well logs and images in this zone.
The LWD measurements required mobility (permeability/fluid viscosity) measurements, a continuous means of
determining the attitude of the strata to steer parallel while maintaining standoff from the top boundary of the reservoir, and a
basic evaluation petrophysical LWD suite.
This LWD string —, the longest successfully used on a kelly driven rig located onshore in Saudi Arabia — consisted of a
total of five LWD tools plus an instrumented rotary steerable tool for precise trajectory control. Careful attention was paid to
operational planning and process workflows, detailed LWD assembly design, and testing, feasibility studies on measurement
responses, data acquisition planning, and directional drilling.
The horizontal lateral, which achieved 100% net-to-gross (N/G) in reservoir comprising >= 5 millidarcy/centipoise (md/cp)
mobility, was geosteered within a narrow, < 2 feet (ft) thick target interval without any distinguishing geosteering log markers.
The target interval was identified and verified using the mobility measurements from the LWD formation tester tool. Some
relatively high real-time mobilities were observed along the lateral, exceeding the expectations set based on the offset pilot
hole data.
The horizontal development sidetrack was put on production in late 2012 producing at an oil rate of approximately (~)
4,000 stock-tank barrels per day (STB/D).

Introduction
The well described in this paper was drilled in a giant mature oil field located in the Eastern Province of the Kingdom of Saudi
Arabia. The field was discovered in the early 1940s (Al-Towailib et al. 2008) and has mainly been produced from two large
fractured carbonate oil reservoirs (Widjaja et al. 2013).
The field contains various other hydrocarbon-bearing reservoirs. Several deep undeveloped significant gas reservoirs cover
most of the extent of area of the field. Many of the hydrocarbon reservoirs are associated with a high-relief dome structure (Al-
2 SPE 168077

Towailib et al. 2008) located toward the southern end of the field. One of these reservoirs is an ample virtually undepleted
heavy oil accumulation located above the two main producing horizons. The last oil discovery in the dome region was made as
late as 2009. Saudi Aramco is currently pursuing further delineation, including pilot production and injection programs, for
several of these secondary reservoirs with the intent that all hydrocarbons in this field should be able to be cost-effectively
produced through the existing infrastructure. As the field is mature and the infrastructure is aging, the optimum secondary
reservoir development expected value can only be achieved by not delaying the investment into secondary reservoirs
development wells too far into the future.
In the 1980s, as part of the drilling program to explore for hydrocarbons located below the main producing reservoirs, two
stratigraphically trapped smaller oil accumulations were discovered more than 15 miles north of the dome area. During 1981-
82, the larger of these two stratigraphic traps, the North reservoir, was delineated, cored and tested with several wells. The
North reservoir contains good quality light oil (41° API) with relatively high gas-oil ratio (~1,500 standard cubic ft / stock-
tank barrel, scf/STB), but the reservoir rock properties are heterogeneous with predominantly low formation permeability.
In early 2012, the first North reservoir development well was drilled as a deepening of an existing dead producer
completed in the main producing horizon. This development well was first drilled as a 30° slanted pilot hole across the
reservoir section. An extensive data acquisition program was carried out on this pilot hole, including coring of the full
reservoir section. The well was subsequently sidetracked as a 2,900 ft single lateral horizontal North reservoir pilot producer.
This producer was put on stream in December 2012 at an oil rate of ~4,000 STB/D.
This paper reviews the results of this first development well with the focus on the novel geosteering technique that was
used for well placement to ensure that the first North reservoir producer was successful.

North Reservoir: Pre 2012 Well Results

In 1981, the North reservoir was discovered by Well A. The Well A log is displayed in Fig. 1. Well A flowed 800 STB/D after
acidizing during a Drill Stem Test (DST). Prior to the stimulation treatment, the well was unable to flow. This low
productivity is consistent with the core permeabilities ranging from 0.1-1.5 md, as shown in Fig. 1. The fluid samples taken in
this well proved the North reservoir hydrocarbon is a light good quality oil (41° API) with relatively high Gas-Oil Ratio
(~1,500 scf/STB).

Fig. 1- Well A formation log analysis

During the 1980s, eight further appraisal wells were drilled to delineate this discovery with core being acquired in four of
these wells. Figure 2 exhibits a core porosity-permeability cross-plot including the data from all cored wells, including Well
A. Table 1 shows a summary of North reservoir DST results.

Oil Rate Water Rate

Comments
STB/D B/D
Well-A 800 No Acidized, flow not stable
Well-B No flow to surface Water, oil traces in string
Well-C 1,600 No Acidized, stable flow rate
Well-D No flow to surface No fluids recovered
Well-F 160 200 Acidized, flow not stable

Table 1- North reservoir well test (DST) results

Fig. 2- North reservoir core porosity-permeability cross plot

SPE 168077 3

It is evident from this data set that the matrix permeability is predominantly less than 1 md with the occasional interval of
higher permeability. In fact, ~40% of the core plugs are reported with 0.1 md permeability. This is consistent with DST results
that indicate that only one of the five wells tested was capable of achieving stable post stimulation flow.

North Reservoir Geology

The North reservoir is a ~50 ft thick Middle Jurassic carbonate interval. The reservoir consists of several coarsening upwards
cycles, these cycles start with wispy laminated wackestone to mud-dominated packstone with the occasional rip-up clasts at
the base, followed by thin peloid-dominated packstones and fragmented fossil-dominated, burrowed wackestone, including
foraminifera, echinoids, bivalves, and brachiopods. Each cycle is capped by a grainstone with coated grains, sometimes ooid
dominated with local anhydrite nodules. The interval is characterized by moldic and interparticle porosity.
The average high-frequency sequence thickness is about 5-10 ft, each sequence consists of a porous interval on top of a
tight interval. Stylolites are common in the interval and open and filled fractures are observed. Reservoir quality deteriorates
significantly from north to south resulting in the up dip stratigraphic southern seal. The reservoir interval has been divided into
five zones, of which the upper most units are slightly thicker and more porous. Lateral seal is provided by deteriorating
porosity to the south and the top seal is the overlying transgressive-associated mud-dominated sediments.

The Pilot Hole

In March 2012, an existing dead producer in the main producing formation, the Upper reservoir, was plugged back and
sidetracked through the deeper North reservoir oil accumulation as a 30° inclination pilot hole. The full North reservoir section
was cored and an extensive wireline formation evaluation program was carried out. The primary data acquisition objective was
to assess the productivity of the hydrocarbon-bearing interval. This assessment was required to allow optimum well placement
for a planned subsequent sidetrack of a 2,900 ft single-lateral horizontal pilot producer. In addition to the standard
petrophysical suite, the pilot hole formation evaluation logging program also comprised nuclear magnetic resonance (NMR)
logging for mobile fluid detection and a formation tester program to obtain formation pressures, fluid gradients, formation
mobility and fluid samples.
Figure 3 presents the pilot hole petrophysical interpretation based on resistivity, density and neutron porosity, combined
with free fluid (FF) and bound fluid (BF) from the NMR T2 relaxation data. The interpretation indicates target interest zones
for the pilot producer sidetrack with 10-15% porosity and NMR permeability ranging from 1 to 10 md.

Fig. 3- Pilot hole petrophysical analysis

The different zones with elevated NMR permeability indicators were further characterized through formation tester
measurements with the aim of determining whether the interpreted permeable intervals from the conventional petrophysical
interpretation and NMR information were valid. A total of 25 pressure-mobility points were attempted with the formation
tester. The results were as often seen with low-permeability reservoirs, 18 tight tests or lost seal, five supercharged tests and
only two tests with stabilized pressure. As presented in Fig. 4, an interval of elevated permeability found to be around ~6 ft
below the top reservoir reading 2.2 md/cp drawdown mobility, improving to 34 md/cp after pumping for an extended time
during sampling with a probe formation tester. Using the oil viscosity measured from the laboratory PVT analyses of the fluid
samples taken in the pilot hole, these mobilities correspond to a permeability of 15 md and 225 md, respectively.
4 SPE 168077

Fig. 4- Pilot hole pressure and mobility data

The formation tester results demonstrate that the intervals with rock of producible quality are more vertically restricted
than was apparent from the distribution of the free fluid NMR interpretation. As the density-neutron and NMR logs did not
exhibit straightforward correlations to the formation tester mobility results for identification of potential targets with
producible characteristics, the target selection for geosteering of the horizontal production hole was restricted to this elevated
formation mobility interval.

Technical Challenges Driving Measurement Selection for Well Placement of Horizontal Producer
The logs from the pilot hole had revealed the presence of a thin interval of elevated permeability. Since the surrounding
pretests showed drawdown mobilities of less than 1 md/cp, the well placement team conceived the idea of drilling a producing
lateral targeting the higher permeability interval. A number of practical difficulties in achieving a horizontal well effectively
exposing this stratum of permeable rock were apparent from examination of the pilot hole logs. This mandated an informed
technical discussion around the design of the optimal LWD logging program to be able to successfully place this lateral. The
exact thickness of the permeable interval was not known, but it was expected not to exceed 2 to 3 ft true stratigraphic thickness
(TST). The lateral continuity of the permeability was therefore not certain; however, a strong lateral continuity in basic log
properties was observed from the zonal correlation of the offset wells, as previously described in the North Reservoir Geology
section. The dip was expected to be gentle and fairly consistent at this structural level in this part of the field, but the
possibility of local variations could not be ruled out.
Even if the lateral continuity of the permeability was assumed, no obvious correlating signature or log markers that could
act as mobility proxy were evident from the pilot hole triple combo logs. Consequently, no obvious “landmarks” to guide the
horizontal drilling could be expected inside the targeted reservoir interval. This meant that triple combo LWD log
measurements would lack the capability to identify, correlate and follow the permeable but petrophysically cryptic target
interval. Consequently, it was necessary to specify LWD measurements to profile, identify and quantify the permeability as a
function of its position in true stratigraphic thickness terms and thereby remain within the target interval using appropriate
steering concepts.
The LWD measurements considered necessary to facilitate successful geosteering in this thin layer were as follows:
• Distance to upper boundary detection
• Resistivity images for dip determination
• Permeability proxy (mobility indicator)
Each one of these LWD measurements is described further below. All three measurements were considered a prerequisite for
successful geosteering within the thin target interval due to lack of any other petrophysical distinguishing features.

Distance to Upper Boundary

A limited, but consistent, conductivity contrast at a boundary some 6 ft TST above existed and the well placement team
conceived the idea of using this as a steering reference by employing deep directional electromagnetic (EM) LWD distance to
boundary mapping techniques. The feasibility of this approach needed to be tested. Since the triple combo LWD log
measurements would lack the capability to identify, correlate and follow such a permeable interval, it was necessary to specify
LWD measurements to profile, identify and quantify the permeability as a function of its position in true stratigraphic
thickness terms and thereby remain in the target interval using appropriate steering concepts.

Resistivity Images for Dip Determination

The expected LWD resistivity image signatures in this layer were poorly understood, as no such images had previously been
acquired in the reservoir. It was not known to what extent the image character of this interval would reveal compositional or
SPE 168077 5

textural layering or bedding at the scale of the measurement, or whether the image response would be massive. The Saudi
Arabian carbonate experience is mixed in this regard with some intervals displaying good layering suitable as geosteering
input while other reservoirs demonstrate massive textured intervals lacking internal laminae, which makes this approach not
suitable. Natural fractures were not expected to be significant in this reservoir. In the event that bedding features were
detectable, there was a prospect that these features would provide the geosteering team with periodic indications of the
orientation of the bedding, and provided the layering was parallel to the local structural dip, this data could be useful in
geosteering the well. During the landing of the well, the dip data would provide a precise calibration of approach angle, and
the resistivity contrast at the top of the reservoir was likely to provide at least one dip data point at the crucial target entry stage
and this dip data could be used to gauge the precise landing position in TST terms.

Permeability Proxy
To profile and verify the exposure of higher permeability rock, an LWD measurement sensitive to permeability was required.
There are no fully satisfactory quantitative permeability measurements available in LWD, and those that exist are indirectly
derived, resting on assumptions that tend to limit applicability. Traditional proxies for permeability employed in transforms
such as porosity measurements are limited in scope as these empirical relationships are tenuous at best. In this case, it was
known from the pilot hole data that the there is no good permeability indicator from the triple combo or NMR data, which will
necessitate to include an LWD formation tester in the tool string to verify mobility real time in the horizontal hole section.

Pre-Job Geosteering Feasibility Assessment

The geosteering challenge for placement of the 2,900 ft single lateral producer, as laid out in the previous section, was to:
1. Land a horizontal producer precisely in a 2 to 3 ft TST layer that is lacking distinguishing petrophysical
characteristics from triple combo and NMR logs
2. In real time, verify its elevated permeability in relevant time to guide the drilling assembly
3. Stay in the target layer even if changes of dip or layer thickness occur
The LWD measurements deemed necessary to be able to meet this challenge were as follows: (1) Distance-to-boundary
mapping, (2) LWD resistivity image real-time dip analysis, and (3) advanced LWD measurements sensitive to mobility or
permeability to detect and steer for the permeable streak. Prior to including these costly petrophysical tools in the bottom-hole
assembly (BHA), a feasibility study was carried out based on the input data acquired in the pilot hole and from the earlier well
penetrations to determine if the expected output data from the selected tools would be sufficient to meet this sophisticated
geosteering challenge. The following section addresses the predrill planning feasibility analyses for each of the three main
geosteering components.

Distance to Boundary Feasibility

The technique of distance-to-boundary mapping relies on detectable and predictable (and ideally monotonic) variation of the
deep EM tools’ symmetrized directional signals as a function of distance from the boundary or boundaries in question. Since
the magnitude of such signals is governed by the size of the conductivity contrast at the boundary, it was recognized that the
limited contrast at the boundary might pose a challenge for this technique. Moreover, as the well would also be situated on the
less favorable conductive side of the boundary “looking” toward more resistive rock, the boundary was expected to be difficult
to map. Detailed modeling was performed to determine whether the data would be sufficiently sensitive from the boundary to
enable meaningful inversion for distance-to-boundary at the ~6+ ft true vertical thickness (TVT) range implied by the distance
of a tool located within the target zone and parallel to the resistive boundary. The results of this modeling are presented in Fig.
5.
Since the shoulder bed was significantly more resistive than the ambient resistivity of the target reservoir, it was
anticipated that boundary detection and mapping would be more challenging than the more usual case of resistive target and
conductive shoulder. In summary, for this specific resistivity profile, the tool was found to respond in a predictable fashion to
a band of resistive lithology at or near the top of the reservoir up to 6 ft away. Monotonically varying simulated responses of
several different transmitter-receiver spacings and frequencies were observed in the forward modeling simulations and suitable
measurements were selected for real-time transmission on the basis of these forward modeling responses.
The forward modeling inversion simulations were performed prior to the drilling of the well. These simulations used as
their basis an earth model made from a resistivity property column gleaned from the pilot-hole log, propagated through a
structural section along the planned well trajectory. The forward modeling was performed with custom-made software that
utilized a sequential 1D dip code that simulated the responses of the LWD tools taking into account the shoulder beds with
their resistivities, relative angles and distances from the trajectory. These simulations indicated that that selected outputs of the
deep resistivity LWD tool at several different transmitter-receiver spacings and frequencies would respond in a predictable and
monotonic manner to the boundary while in the target zone. The tool located at 6 ft position at the target standoff would be on
or close to the very edge of detectability of the upper boundary. Hence, it was known before drilling the well that the distance-
to-boundary inversion would likely be intermittent, depending on tool standoff from the upper boundary, but that valuable
information would still be found in the symmetrized directional signals from the tool. It was also known that manual distance-
to-boundary computations could be deployed where a valid response was present, but geological noise defeated the real time
inversion.
6 SPE 168077

The methodology employed in real time for solving for distance to the boundary (Omeragic et al. 2006, 2009) uses as
inputs multiple symmetrized directional EM measurements (both phase and attenuation measurements) of multiple transmitter-
receiver spacings and three frequencies together with resistivity measurements. In this case, both propagation and laterolog
resistivity cases were run simultaneously. The unknowns in the model are the TVT distances to the boundaries, the resistivity
of each shoulder bed and the vertical and horizontal resistivity components of the bed. Iterative modeling simultaneously
solves for a several variants of the model using an automated model selection process. This approach warrants — using
iterative techniques involving forward modeling calls and comparison techniques — that the likely model is identified that
best explains all of the output responses acquired by the deep resistivity tool.

Forward Modelling
Simulated Logs

Earth Model based on

offset well properties

Inversion result using forward-

modeled curve simulations as input

Top reservoir
boundary @ 6 ft

Fig. 5- Deep directional EM forward modeling realization

To determine the feasibility of this inversion technique for this case, the inversion was run on subsets of the forward-model
simulated responses to (a) determine the best subset of measurements to specify in the real time data frame to be transmitted
up hole by the LWD tool and (b) establish which boundaries would be resolved by the technique, which would allow an
assessment of their usefulness for the specific well placement objectives.
The detectability of the top of the reservoir was in this fashion determined as up to ~6 ft standoff, a similar standoff to the
position of the permeable target zone.

Feasibility of Use Resistivity Images for Dip Determination

A fundamental step in the well placement process is to deduce or determine stratigraphic position of the well (or standoff from
a marker) and its relative angle with respect to the local structural dip. It has long been possible to determine dip rapidly from
real time LWD images, using data streaming techniques to make the images easily available in an interpretation software
package, and using sinusoid-matching techniques to determine relative angle when beds or laminae of contrasting properties
are crossed by the borehole. Their projection as apparent dip sticks in the geosteering models is a feature in most geosteering
software applications used within the industry today. This allows an interpreter to compare apparent dips from images with
structural dip of the model, and where appropriate, update the model to accord with the dips.
In this instance, the requirement was not only to steer the well using the dips to inform the real time model update, but also
use the dip and trajectory inclinometer data to compute a relative true stratigraphic TST index so that the LWD responses,
SPE 168077 7

including drawdown mobility, could be accurately profiled for real time geosteering purposes. Using a new application built
into a 3D reservoir modeling platform, a new software application was used for the structural interpretation workflow. Drilling
polarity plots were made along the trajectory using the dip data. These drilling polarity plots helped the well placement team to
understand whether the well was being drilled stratigraphically up- or down-section. The drilling polarity plots are useful for
well placement and geosteering decisions in stratigraphically-controlled pay zones within structurally complex areas, but in
this case, the application provided improved precision for well placement in a structurally simple reservoir, but with a thin
target interval with cryptic petrophysical responses. In real time, dip sequence analyses was performed and used in generating
a TST index log (Fig. 6). The TST index log was used to complement the estimate of bit distance-to-boundary from deep
resistivity modeling and then integrated with a curtain section model to allow constantly updated knowledge of the precise
stratigraphic position and dip to be taken into account in the geosteering decisions. Also from the dip sequence analyses, 3D
near-wellbore structural modeling was done creating surfaces (Fig. 7) that respect true structural dip as recorded around the
wellbore.

Fig. 6- Horizontal lateral LWD logs and mobility data indexed in True Stratigraphic Thickness (TST) relative to the top of the reservoir

Fig. 7- Three dimensional visualization of real-time dip data, with drilling polarity indicated on trajectory
8 SPE 168077

Permeability Proxy
A case could be made for each of three possible advanced LWD methods to provide a suitable proxy for permeability: (1)
Stoneley permeability, (2) NMR relaxation data modeling, and (3) formation-tester drawdown mobility estimation. Each of
these methods and their particular suitability for this real-time geosteering application is described below.
The Stoneley permeability is a continuous curve measured by the reduction in amplitude or increase in slowness of the
acoustic Stoneley wave generated in the borehole due to the presence of mobile fluids in the rock. In practice, the mobility
needs to be reasonably high for the method to be accurate. In this case, the mobility was expected to be less than 10 md/cp,
which is below the minimum limit of accuracy for this method which renders this method unsuitable for this application.
NMR relaxation models, e.g., Timur-Coates or SDR (Schlumberger-Doll Research) formulations (Sigal 2002) and their
many variants, use the log mean of the T2 distribution or free/bound fluid partitioning of the distribution as ingredients in the
divination of a continuous computed permeability. The relationship of the pilot hole probe based drawdown formation tester
permeability (mobility) data and NMR log derived permeability data was not particularly good. As the 1980s core data and
production test results were more in line with the mobilities observed from the formation tester, the NMR approach was
discarded as not applicable at this time. This method may be suitable in the future, but it will require careful core-log
integration study to provide the required accuracy, which was not available for the North reservoir at the time of drilling the
horizontal production hole.
LWD formation tester drawdown mobility estimation is a technology similar to the pilot hole wireline formation tester.
The LWD version does not permit taking fluid samples, but it is a well-established technology that is known to be able to
differentiate between permeable and impermeable rock. This is the most direct method as it relates to actual flow of fluid
through the formation during a pretest. As this is a not continuous station measurement, the acquisition program would need
to carefully account for operational and acquisition limitations (Pop et al. 2005) to ensure adequate data for profiling,
identifying and validating the expected permeable interval in support of the well placement process. As the available core and
production test data from prior North reservoir wells were in line with the formation tester data acquired in the pilot hole, it
was decided to utilize an LWD formation tester as the real-time permeability proxy during the geosteering of the horizontal
producer.

Other Pre Job Planning Considerations and Contingencies

Appropriate geological and petrophysical fabric data were scarce for this reservoir prior to drilling of the horizontal section.
Hence, the LWD logging program needed to provide for contingencies relating to the applicability of well placement process
input data. One such contingency was whether the resistivity imaging tool could detect laminations within the reservoir well
enough for effective and frequent real time dip determination or whether it would be necessary to rely partly or solely on
distance-to-boundary techniques for well placement. In the case, it would be required to determine the feasibility and
limitations on boundary detection imposed by the actual resistivity profile acquired during drilling.
Another requirement of the LWD logging program was to acquire an enhanced set of formation evaluation data along the
trajectory, providing interpretive self-sufficiency while drilling, for evaluation and characterization purposes and also to gain
full understanding of the actual placement of the well.
The trajectory planning and well placement process utilized for this well is similar to previous cases documented from
Saudi Arabia (Al-Fawwaz et al. 2004, Al-Maskeen et al. 2013); however, the execution and BHA requirements were more
complex due to the nondescript reservoir target zone.

LWD BHA Design

An extensive combination of real-time measurements for geosteering and formation evaluation was deployed for drilling of the
pilot producer. The initial BHA was the longest and most technically advanced BHA ever used successfully on a kelly-driven
land rig onshore Saudi Arabia.
An optimal combination of LWD tools which would satisfy these data acquisition requirements is not a straight forward
design problem. The initial design task was to establish priorities for positioning of the tools in the BHA, as bit-to-sensor
distance is an important factor in a functioning geosteering BHA design. The downhole power supply also needed careful
attention as the LWD tools were powered from a single measurement-while-drilling (MWD) tool with two power supplies, one
up hole and the other downhole, each with a maximum power rating that could not be exceeded.
The most critical measurements for steering reactively were determined to be the resistivity image and the formation tester
mobility. Consequently, these tools were placed closer to the bit, the resistivity image ahead of the formation tester to allow
probe-placement to be optimized using the real-time image. As the density-neutron tool contained a fishable chemical source
assembly, this tool needed to be placed at the back of the BHA. As a result, the deep resistivity EM mapper was placed behind
the MWD, distant from the bit.
It was understood from the pre job modeling that when drilling parallel to the strata, the distance to boundary measurement
would still have value owing to its deep depth of investigation. To demonstrate this point, the following illustration is
provided. It assumes a boundary detection distance of 6 ft TVT and a well initially drilling parallel to dip at 5 ft standoff from
the upper boundary. An abrupt change of dip by +0.5° would imply a relative closure angle of -0.5° between the trajectory and
the structure. If left uncorrected, the trajectory would take 572 ft to intersect with the boundary, whereas the distance to
boundary measurement would start responding to the change in the deep resistivity LWD measurement after 116 ft measured
SPE 168077 9

drilled depth, representing an ~ 1 ft deviation in true stratigraphic position before remedial geosteering action could be
initiated based on the measured distance-to-boundary data. Thus, a feedback loop between measurement detection and steering
response relevant to the objective was preserved during the operation.
The nominal power demand of each tool was known, but to guard against any possible deviations, the proposed BHA
power budget balance was verified with a full system basket test in the vendor’s workshop to ensure that the was no flaw or
oversight in the design.
Deciding on the telemetry scheme and real-time data frame list were also important design tasks, as the volume of
available data was large and the bandwidth of this generation of MWD tool was finite at 6 bits per second physical telemetry
rate. The well placement team needed to select the most relevant measurements for real-time transmission. Three alternate
repeating data frame configurations were programmed, allowing the field engineer to downlink a command to the MWD tool,
if required, to switch to an alternate configuration, e.g., one option provided for substitution of the resistivity image with a
density image if required in case the resistivity data were compromised.
The primary frame consisted of a high-resolution resistivity image, laterolog and propagation resistivities, six symmetrized
deep directional EM measurements, gamma ray, bulk density, bulk density correction, photo electric factor, neutron, rotary
steerable system (RSS) parameters, continuous near bit inclination and azimuth, as well as a number of ancillary drilling
mechanics and engineering measurements. A utility frame included status information for each tool, and an on demand frame
transmitted a suite of data points from the formation tester data. Downhole data compression was applied to a number of the
measurements, enhancing the effective data rate.
The power steering device was a push-the-bit system designed for full directional control while rotating the drill string.
Automatic inclination hold and efficient downlink functions allowed maintenance of directional control while drilling ahead.
The downlink systems and automatic inclination hold closed-loop functionality assisted in achieving fine control of the
inclination and azimuth of the well. Near-bit inclination and azimuth were measured and transmitted to surface continuously,
enabling the directional driller and the well placement team to scrutinize the effects of steering commands, and adjust
whenever necessary. These elements were critical for accurately executing the trajectory decisions, avoiding overreaction and
preventing undesirable roller-coaster trajectory profiles.
Figure 8 depicts the lower BHA, including all LWD tools.

Fig. 8- Horizontal Lateral BHA deploying measurements from a total of six tools

Real-Time Operation, Data Handling, Interpretation and Decision Making

Saudi Aramco’s Geosteering Operations Center (GOC) handled the geosteering operation. The real-time data transmission, via
a data streaming application, from the well site to the GOC was fundamental to the real time process, which fully integrated
the skills and knowledge of the operating company and the technology service provider experts. GOC is staffed with
geosteering geoscientists 24-7, which is critical for this type of operations where real-time decisions based on models are made
around the clock with very little response time. As expected from the pre-job assessment and analyses, the placement of the
horizontal producing lateral was a very challenging operation that required team work and total focus from all personnel
involved in this operation to deliver a successful outcome.
Real-time inversion on streaming directional data was performed using a geosteering software platform and dip analyses
were performed with service company software. The mobility results from the LWD formation tester were integrated fully
utilizing the petrophysical expertise of Saudi Aramco’s Reservoir Description Division and Schlumberger’s formation tester
experts to arrive at informed recommendations. The geosteering decisions were communicated promptly by the GOC via
secure chat to the rig site. The responsible GOC geosteering geoscientist was able to observe the service provider’s real time
deep resistivity boundary detection interpretation results and integrate these results with the rest of the data available to make
10 SPE 168077

final decisions for proactive well placement. The continuous support throughout the well placement operation from Saudi
Aramco’s assigned field reservoir engineers and geoscientist was critical for a successful outcome.
The real-time downhole LWD data were streamed remotely from the acquisition computers at the well site to the GOC.
The real-time data were displayed, interpreted, visualized and employed by a multidisciplinary well placement team to make
evidence-supported decisions on choices such as pretest point frequency and selection, target stratum, and geosteering
commands.
The geosteering process followed by the real-time well placement team was as follows:
• The well was landed softly with minimal closure angle at reservoir entry to manage descent slowly enough to profile the
upper few feet of the reservoir using the drawdown mobility stations at intervals of ~45 ft (twice per stand of drill pipe, as
per the survey program).
• Incremental and integrated TST using path-integration techniques were computed in real time.
• The drawdown mobility data in the TST was indexed for accurate profiling of permeability.
• Once a suitable mobility interval was identified, steps were taken to land the well parallel to and within the identified
permeable stratum.
• The deep-directional EM measurements were providing distance-to-boundary computations providing accurate and
consistent TST position independent of the image-based methods.
• The laterolog resistivity measurements were providing boundary-independent resistivity measurements for enhanced
distance-to-boundary capabilities.
• LWD resistivity images were employed in computing and controlling precise standoff from the top marker boundary to
stay in the most mobile available stratum.
• The density and neutron measurements in the pilot hole exhibited little character over the target interval, but these were
also monitored in real time during the geosteering for possible identification of any correlation that may aid the well
placement.
This process can be boiled down to implementation of a hybrid of three different geosteering concepts (Fig. 9):
1. The mobility steering concept.
2. The TST integration steering method
3. Distance-to-boundary steering

Fig. 9- The geosteering concepts employed for placement of the North reservoir producer
SPE 168077 11

The Desert Ant – What Can We Learn?

This real life case from the animal world is included to explain how independent navigation concepts are sometimes required
to reduce the navigation error to meet the overall navigation objective. This example is a simple illustration of the concept that
the different geosteering methods support each other which leads to less error in the overall well placement.
Ants of the genus Cataglyphis, living in the sandy deserts and sabkhas of Arabia survive in a very challenging environment
due to high temperatures, shifting sands and natural predators. The ants occupy an underground home burrow, and forage for
food in the surrounding sand. These ants are reputedly called “the thief of the cooking pot” in Arabic for their habit of rapidly
removing small pieces of food from campsites. Since there are effectively no permanent landmarks in the natural environment
of these ants, and the burrow is invisible to the miniscule ant from more than a few inches away, the animal requires effective
means of navigating home after a foraging trip. The ant forages in complex random paths but retain the almost uncanny ability
to return in a straight line to their starting position or fix with only a little error.
Detailed experiments have revealed that the ants have developed a several independent means of navigation (Darwin 1873,
Műller and Wehner, 1988):
1. The small brain of the ant has means of counting the steps taken, a measure of distance.
2. The ant uses a “sky compass” to determine its orientation by detecting the polarization direction of the light, which is
constant throughout the day.
3. Using some simple but elegant algorithm, the ant tots up the distance and direction and integrates the total distance and
direction travelled and computes a running update of a “homing vector” consisting of a distance in steps and a straight line
direction, stored in the ant’s tiny brain. The homing vector is instantly called upon when the ant needs to return to its
burrow, whether because it has located food or to escape from a predator.
The further the ant has travelled, however, the less accurate the homing vector is, and the greater the error. When the ant
reaches the end of the homing vector path, it then switches to a different search method, which consists of a tight, expanding
spiral path, which soon brings the ant back to its burrow (Fig. 10).

Fig. 10- Navigation method of the ant Cataglyphis

Well Placement Steering Concepts

The story of the desert ant’s path-integration navigation techniques was included to illustrate the challenge of geosteering
without obvious landmarks. Just like the desert ant’s path-integration navigation method (Fig. 10), in this case, it was
necessary to combine more than one geosteering approach to develop independent real-time workflows, combined into an
integrated model update for well placement decision making.
For this particular geosteering challenge, three separate steering concepts were used. These methods are illustrated in
cartoon form in Fig. 9.
1. Mobility steering used frequent formation tester drawdown mobility measurements to profile the target and locate the
target interval, then land in the target, reducing the sampling interval of drawdown mobility measurements to verify the
mobility of the exposed rock (Seifert et al. 2008) when steering parallel to dip.
2. TST integration, used dips picked from real time resistivity images and trajectory survey data to precisely integrate the
distance crossed in TST relative to a reference stratum coinciding with a resistive band above the target. The logs and
drawdown mobility measurements could thus be displayed relative to a TST index, rather than true vertical depth (TVD)
or measured depth (MD), both of which would have distorted the true thickness relationships, but it was understood that
as with the desert ant’s navigation, the path integration technique would be subject to a compounding error. The TST
integration steering method resembles the nautical concept of dead-reckoning in which the position of the ship is fixed at a
known reference point, and the current position is calculated relative to the fix by integrating incremental shifts in position
based on known or estimated data without reference to external landmarks. In this case our “fixed point” was at the entry
to the target reservoir and our projected TST increments relative to the target entry were computed by relating the
trajectory angle to bedding dip angles computed from the LWD image analysis. As the trajectory advances away from the
fixed point, the cumulative errors increase in proportion to the errors in the inputs.
12 SPE 168077

3. Distance to boundary mapping technique (Li et al. 2005, Seydoux et al. 2004), although close to its limit in this
environment, provided a more accurate but less precise estimate of the distance to the same resistive reference stratum.
Moreover, as the error would not compound, a useful cross reference to the TST technique would be provided. This
distance-to-boundary technique would also work if there were intervals where the reservoir was massive and insufficient
laminations existed for dip picking, leading to a breakdown of the TST technique. This concept relies on algorithms which
solve for distance to an adjacent boundary using deep-directional EM measurements. This method requires that the
trajectory to be within the effective reach of the depth of investigation of the deep resistivity tool response, but provided
the lateral resistivity profile does not change radically, its accuracy at any given standoff is unlikely to change. Thus, the
accuracy inherent in this method complements the greater precision of the TST path-integration method while the absence
of compounding errors acts as a recurrent reality check on the TST method.
The workflow architecture for integrating such a wealth of data and applying it to effective geosteering of the horizontal
well incorporated:
• Pre-job workflows such as: feasibility studies, sensitivity analyses, measurement selection and BHA design, key
performance indices development and process definition.
• Real-time data quality control such as: validation of tool operation, signal demodulation quality, data integrity, drilling
dynamics analysis/feedback to drilling parameters, data-streaming verification, survey validation, directional drilling
computation validation, measurement depth control and internal and external consistency to offset reference well(s).
• Real-time interpretation workflows: including correlation, dip-picking and analysis, formation tester and mobility
processing, TST profiling and distance-to-boundary inversion.
• Real-relevant time integration and evaluation in the appropriate 3D models used for each of the steering concepts.
• Decision making: recommendations transmitted to the well site by secure chat in a data streaming system.
The end result, a very successful well drilled for about 2,900 ft in a 1-2 ft TST interval, would have been impossible to
track using any single conventional method due to its cryptic conventional log signatures, but was successfully exploited using
a multidisciplinary integrated workflow drawing on a rich dataset.

Steering the Well

After sidetracking the original well, short landing the 8-½ inches (in) curve section a short distance stratigraphically above the
target and setting and cementing a 7 in liner, the LWD assembly was picked up to drill the lateral, guided by the well
placement plans previously discussed. The well placement team used the boundary mapper response to map entry into the
reservoir and used this response, together with the resistivity image, to determine the apparent dip at the reservoir top. On
entering the reservoir, the well was gently brought down section while taking frequent formation tester measurements, to
profile the mobility and identify the most mobile zone for landing.
The eight major geosteering decisions are numbered chronologically and labeled on the well path shown in Fig. 11. A
discussion of each of the featured decisions and the interpretive context thereof is included below:
1. The top of the reservoir formation was picked and formation apparent dip of 0.3° up-dip was estimated from the high-
resolution resistivity image log. A decision was taken to maintain an inclination at 89° while using formation mobility
from station formation tester measurements to define a possible target. At XX76 ft MD, a fluid mobility of 2.6 md/cp was
obtained and at XX05 ft MD another real-time pressure test gave a fluid mobility of 17.9 md/cp. A decision was taken to
continue drilling with inclination of 89° so as to locate an interval with the highest mobility.
2. At this point in the well a fluid mobility of 1.4 md/cp was obtained and in the next test a fluid mobility of 6.0 md/cp was
also obtained from the analysis of the real-time formation pressure test. Hence, a decision was taken to build to 91° to
place the well within the TST interval that had registered mobility of 17.9 md/cp. The thickness of that zone from the
resistivity image and model update was estimated to be ~ 0.5ft TST.
3. At this point a mobility 432.9 md/cp was obtained from a real-time pressure test. This measured mobility is significantly
higher than any other mobility ever measured in the North reservoir, but the other measured formation tester data suggest
that this data point is valid. At present time, no futher data is available that can explain this mobility anomaly. The
formation apparent dip was estimated from the real time resistivity image log to be ~ -0.4° dipping toward the wellhead.
Consequently, a decision was made to keep the inclination at 90.5° to place the well in this high mobility layer. From the
distance-to-boundary inversion and TST index log, the layer was evaluated to be at ~ 4.7 ft TST from the reservoir top.
4. For ~ 600 ft MD the well had been drilling with ~90° at the bit with intention to place the well within a high-mobility
layer. The dip of the formation estimated from resistivity image log indicated the formation apparent dip was likely in the
range of flat to about 0.3° up. At this point, the mobility of 3.9 md/cp was recorded from a real-time pressure test.
Correlation using real-time distance-to-boundary inversion and the resistivity image and other logs indicated that the high
mobility was below the bit. As a result, a decision was taken to hold 90° to gradually drill stratigraphically downward into
the target.
5. At this juncture, because of problems with recurrent sticking pipe, interpreted as differential sticking, in the context of
limited scope for reaming with a kelly rig, the BHA was pulled out of the hole and the re-run BHA was without the
formation tester tool due to the hole condition. A decision was taken to drop inclination to 89.5° based on correlation from
the TST index log and deep resistivity inversion.
SPE 168077 13

6. As the resistivity image had indicated the well had been drilled back into the target layer, and a decision was taken to
build and hold inclination at 90°. The formation dip was also estimated to be almost flat.
7. At this point, the formation apparent dip was estimated to be dipping away from the wellhead at about +0.5°, therefore
decision was taken to drop inclination to 89.5°.
8. Since the formation dip was estimated to be flat, a decision was taken to keep inclination at 90°. The well was drilled with
90° until well total depth (TD).

Fig. 11- Geosteering summary

The distance-to-boundary determination was likely to be difficult, as illustrated by the prejob modeling, but it was
understood that even an intermittent result would be useful since this technique does not have the compound error of the
integration techniques. During the real-time job, in addition to the standard commercial distance to boundary inversion, a new
multilayer statistical inversion prototype was also field tested, as shown in Fig. 12 as the 2nd track from the bottom. In these
inversion results, using a rainbow color map with blue and green denoting low resistivities, and yellows and reds indicating
high resistivity, the definition of the upper resistive boundary was somewhat more accurate than with the standard inversion;
displayed the 3rd track from the bottom with dark color representing conductive rock and light color representing resistive
formation.

Evaluation of the Lateral

In the horizontal lateral, a comprehensive LWD data set was acquired for the purpose of geosteering and formation evaluation.
The measurements used for formation evaluation were natural gamma ray, laterolog resistivities at four depths of
investigation, phase and attenuation resistivities at five transmitter-receiver spacings at 2MHz and 400 KHz, azimuthal
density, thermal neutron porosity and formation pressure while drilling for mobility and formation pressure. Figure 13
presents the results from the horizontal section petrophysical evaluation.
After entry to the reservoir, the basic log responses were very constant and lacking any petrophysical character that could
be used to identify the target zone. The drawdown mobility measurements showed considerable variation, and this had been
used as the primary criterion for identification and verification of the higher mobility target zone. When looking at the
responses in more detail, it is evident that there are subtle differences in the separations of the shallow-reading laterolog
resistivities that may be significant as they may reveal differences in the progress of filtrate invasion at the time of drilling,
which could be caused in part by differences in mobility.
To systematize these differences, an invasion profile computation was performed, derived from multiple depth of
investigation laterolog-type resistivity measurements, the technique being based on inversion of a step-profile invasion model.
The model is analogous to an intuitively equivalent circuit model that can accurately reproduce the response of a
focused array laterolog tool (Li et al. 1999). This inversion solves for formation true resistivity (Rt), and flushed zone
resistivity (Rxo) displayed on log plot track-2 and invasion diameter (Di) displayed in log tracks 8 and 9.
As a post-drilling experiment, using the invasion profile data as an input, a qualitative indicator of mobility (K* index) was
computed at each level from the LWD recorded memory data. For the purposes of this calculation, fluid viscosity was
considered a constant, fixed at the oil viscosity measured from pilot hole samples. The local Archie parameters were also
14 SPE 168077

assumed: cementation exponent (m) = saturation exponent (n) = 2. K* index is computed in accordance with an isotropic
application of Darcy’s law, using Rt /Rxo ratio in conjunction with mud filtrate resistivity (Rmf) and water resisitivity (Rw) as an
indicator of moveable fluid, corrected for geometrical effects using the volume of invasion. A further normalization for
exposure time was applied. A similar approach was employed by Semmelbeck et al. (1995) using resistivity measurements
from an array induction tool (AIT).
In early time, prior to filter cake formation, the qualitative mobility indicator (K*) derived from the laterolog resistivity
would respond to a function of connected porosity, Rt and Rxo ratio and diameter of invasion. The K* index curve, in track 6,
was calibrated with drawdown mobility from formation pressure while drilling pretests by applying empirical coefficients to
bring the K* curve into agreement with the drawdown mobility station measurements. The purpose of this exercise was to
provide a continuous curve to connect the dots between the station measurements of mobility. As there are obvious limitations
to this technique, it is suspected that the physicality response is meaningful, especially where good agreement with the
drawdown data was seen, although there were also some obvious discrepancies. This approach neglects the influence of
mudcake permeability and mechanical effects of stabilisers on mudcake, which may benefit from further investigation to
expand and increase the validity of this technique.
In track 6, the K* index curve shows some correlation between elevated fluid mobility with the computed depth of invasion
at the time of logging, though there are also clear deviations.

Fig. 12- Summary section montage with distance to boundary inversions and selected LWD data
SPE 168077 15

Fig. 13- LWD log analysis of lateral section

Well Placement Results and Well Rate

The horizontal interval covered ~ 2,900 ft with a 100 % net-to-gross in terms of reservoir exposure. The pilot producer was put
on production producing an oil rate of ~4,000 STB/D. Production logs and pressure transient analysis confirm that the well
remained within a 4 ft sweet-spot window for the entire horizontal section.

Conclusions
The first North reservoir horizontal single lateral producer was placed with a thin interval of higher hydrocarbon mobility
using a novel methodology, including real time mobility measurements as an indication of the permeable interval. The LWD
measurements, including electromagnetic boundary mapping inversion, real-time dip evaluation from high resolution
resistivity image and real time fluid mobility evaluation from formation tester pretests, were integrated proactively to achieve
successful well placement. Dip sequence analyses were periodically performed in real time and the TST index log was used to
ensure that the well was placed within the high mobility interval relative to a resistive top of formation. Distance-to-boundary
measurements were used to validate the TST integration. A 100% net-to-gross reservoir exposure was achieved. This was also
a record breaking longest BHA used successfully on a kelly driven rig onshore Saudi Arabia. The well delivered an oil rate of
~4,000 STB/D when it was pit on stream.

Acknowledgements
The authors thank Saudi Aramco and Schlumberger for permission to publish this paper.
A special thanks to Yousef Al-Mousa for his meticulous core descriptions and zonation work that formed the basis for the
North reservoir correlation work.
16 SPE 168077

References
Al-Fawwaz, A, Al-Yosef, O., Al-Qudaihy, D., Al-Shobaili, Y., Al-Faraj, H., Maeso, C. and Roberts, I. 2004. Increased Net to Gross Ratio as
the Result of an Advanced Well Placement Process Utilizing Real-Time Density Images. Paper IADC/SPE 87979 presented at the Asia
Pacific Drilling Technology Conference and Exhibition, K.L., Malaysia, 13-15 September.
Al-Maskeen, A.A., Sung, R.R. and Ali, S.S. 2013. Horizontal Well Correlation Using Real Time Data and Log Prediction in Geosteering
Complex Reservoirs of Saudi Arabia. Paper SPE 164151 presented at the 18th SPE Middle East Oil & Gas Show and Conference,
Manama, Bahrain, 10–13 March.
Al-Towailib, A.I., Lyngra, S. and Al-Otaibi, U.F. 2008. Advanced Completion Technologies Result In Successful Extraction Of Attic Oil
Reserves In A Mature Giant Carbonate Field. Paper WPC19-2451 presented at the 19th World Petroleum Congress, Madrid, Spain, 29
June – 3 July.
Darwin, C. 1873. Origin of Certain Instincts,. Nature 7 (179): 417–418
Li, Q., Rasmus, J. and Cannon, D. 1999. A Novel Inversion Method for Interpretation of a Focused Multisensor LWD Laterolog Resistivity
Tool. Paper presented at the SPWLA 40th Annual Logging Symposium, 30 May – 3 June.
Li, Q., Omeragic, D., Chou, L., Yang, L., Duong, K., Smits, J., Yang, J., Lau, T., Liu, C.B., Dworak, R., Dreuillault, V., and Ye, H. 2005.
New Directional Electromagnetic Tool for Proactive Geosteering and Accurate Formation Evaluation While Drilling. Paper presented
at the SPWLA 46th Annual Logging Symposium, New Orleans, LA, USA, 26-29 June.
Műller, M and Wehner, R 1988. Path Integration in Desert Ants, Cataglyphis Fortis. Proceedings of the National Academy of Sciences of the
United States of America, v. 85, p. 5287-5290, July 1988 Issue Neurobiology.
Omeragic, D., Dumont, A., Esmersoy, C., Habashy, T., Li, Q., Minerbo, G., Rosthal, R., Smits, J. and Tabanou, J. 2006. Sensitivities of
Directional Electromagnetic Measurements For Well Placement And Formation Evaluation While Drilling. Paper presented at the 2006
SEG Annual Meeting, New Orleans, LA, USA, 1-6 October.
Omeragic, D., Habashy, T., Chen, Y.-H., Polyakov, V., Kuo, C.-h., Altman, R., Hupp D. and Maeso, C. 2009. 3D Reservoir Characterization
and Well Placement in Complex Scenarios Using Azimuthal Measurements While Drilling. Paper presented at the SPWLA 50th Annual
Logging Symposium, Woodlands, TX, USA, 21-24 June.
Pop J., Laastad, H., Eriksen, K.O., O’Keefe, M., Follini, J.-M. and Dahle,T. 2005. Operational Aspects of Formation Pressure Measurements
While Drilling. Paper SPE/IADC 92494 presented at the SPE/IADC Drilling Conference, Amsterdam, The Netherlands, 23-25
February.
Seifert, D.J., Neumann, P.M., Al-Nasser, M.N., Al-Dossari, S.M, and Hibler, A.P. 2008. Optimizing Horizontal Well Placement Using
Formation Pressure While Drilling Measurements. Paper presented at the SPWLA 49th Annual Logging Symposium, Edinburgh,
Scotland, 25-28 May.
Semmelbeck, M. E., Dewan, J. T. and Holditch, S. A. 1995. Invasion-Based Method for Estimating Permeability from Logs. Paper 30581
presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, 22-25 October.
Seydoux, J., Tabanou, J., Ortenzi, L., Denichou, J., De Laet,Y., Omeragic, D., Iversen, M and Fejerskov, M. 2004. A Deep-resistivity
Logging-while-drilling Device for Proactive Geosteering. The Leading Edge, v. 23, n. 6. June 2004 Issue, p. 581-586.
Sigal, R. 2002. Coates And Sdr Permeability: Two Variations On The Same Theme. Petrophysics Journal, Society of Petrophysicists and
Well-Log Analysts, v. 43, n.1, January- February 2002 Issue, p. 38-46.
Widjaja, D.R., Lyngra, S., Al-Ajmi, F.A., Al-Otaibi, U.F. and Alhuthali, A.H. 2013. Vertical Cased Producers Outperform Horizontal Wells
in a Complex Naturally Fractured Low Permeability Reservoir. Paper SPE 164414 presented at the 18th SPE Middle East Oil & Gas
Show and Conference, Manama, Bahrain, 10–13 March.