Structural Steering—A Path to Productivity

Aimen Amer The number of oil and gas wells drilled horizontally continues to increase as
East Ahmadi, Kuwait
operators strive to maximize contact with target formations, develop more-efficient
Filippo Chinellato completion programs and optimize recovery from complex geologic structures.
Milan, Italy
Structural steering, a method by which operators direct horizontal and high-angle
Steve Collins
Chief Oil & Gas LLC wellbore trajectories, integrates data from deep-reading LWD resistivity tools
Dallas, Texas, USA and high-resolution imaging devices to create structural models of the geologic
Jean-Michel Denichou conditions encountered by the drill bit. This technique allows drillers to correct
Sugar Land, Texas
wellbore trajectories in anticipation of structural changes ahead of the bit and helps
Isabelle Dubourg operators better understand the formations already drilled.
Clamart, France

Roger Griffiths
Petaling Jaya, Selangor, Malaysia The drilling of oil and gas wells today has little in continue to introduce new technologies that help
common with the early days of exploration when operators drill wells that produce longer, drain
Randy Koepsell wildcatters punched holes, sometimes based on the reservoir more completely and improve return
Denver, Colorado, USA seemingly random patterns, in hopes of on investment.
discovering untapped resources. Today, modern The ability to drill high-angle and horizontal
Stig Lyngra
drilling engineers access an array of technologies well trajectories has been one of the most signifi-
Saudi Aramco
to visualize the subsurface and then command cant changes in drilling in recent years. Although
Dhahran, Saudi Arabia
sophisticated downhole hardware to precisely horizontal wells typically average two to three
Philippe Marza target reservoir sections. After analyzing data times the cost of conventional vertical wells and
Aberdeen, Scotland from various sources, well placement engineers involve higher risks, the proportion of horizontal
can adjust wellbore trajectories while drilling to wells continues to increase (below). For exam-
Doug Murray maximize reservoir contact. Service companies ple, in the US at the end of 2012, 63% of the
Abu Dhabi, UAE

70
Iwan (Bob) Roberts
Dhahran, Saudi Arabia
60
Oilfield Review Spring 2013: 25, no. 1.
Horizontal wells in the US, %

Copyright © 2013 Schlumberger. 50

For help in preparation of this article, thanks to Danny
Hamilton, Frisco City, Texas; Remi Hutin, Clamart, France; 40
Emmanuelle Regrain, Houston; and Haifeng Wang,
Stavanger.
CMR, eXpandBG, eXpandGST, FMI, FPWD, MDT, MicroScope, 30
PeriScope, Petrel and PowerDrive are marks of
Schlumberger. 20

10

0
1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012
Year
> Horizontal drilling in the US. The number of wells drilled horizontally, as a
percentage of the total number of wells drilled in the US, has increased
sharply in the past decade. (Data used with permission from Baker Hughes.)

14 Oilfield Review
Spring 2013 15
1,817 wells being drilled were classified as hori-
zontal and another 11% were labeled directional
wells. Only 26% of the wells were classified as
vertical.1 A primary reason for this shift to hori-
zontal and high-angle wells is that they bring
potential short- and long-term rewards that ver-
tical wells cannot typically provide. Through a
single wellbore, high-angle and horizontal wells
can improve drainage, access discrete compart-
ments in complex reservoirs, reduce interven-
tion costs, improve efficiencies and provide
exposure to more of the reservoir. Despite their
higher initial costs, horizontal wellbores often
provide operators with a method for developing
reservoirs that otherwise might not be profit-
able. Horizontal wellbores are especially impor-
tant in unconventional reservoirs where their
use has been a key enabling technology in devel-
oping shale resources.2
These improvements in drilling technologies
and practices have transformed the way operators
approach developing resources.3 In the early days
of horizontal drilling, wells were constructed
based primarily on geometric well plans. When
some wells were not optimally placed in target
zones, results were often disappointing. Today,
geoscientists and drilling engineers understand
that the term horizontal drilling is an
oversimplification of the process, and accurate
well placement involves more than turning a well
90° from vertical. Maximizing reservoir contact
and understanding subsurface geometry and
Horizontal Wells Drilled Worldwide
1,600
Outside North America
1,400
Canada
1,200 United States
1,000
Number of wells
800
600
400
200 145
41 65
0
1986 1987 1988
> Horizontal drilling before 1995. In 1986, only 41 wells worldwide were classified as horizontal. A
dramatic increase in these numbers occurred between 1989 and 1990, driven by technological
improvements and by the resultant increases in production experienced by some operators drilling
horizontal wells. Although at the outset the trend was primarily in the US, operators in other regions,
especially Canada, also drilled more horizontal wells. (Adapted from Kuchuk et al, reference 5.)
16
geology are crucial components of successful
drilling operations.
Technological advances in LWD formation
evaluation tools are also making it possible for
engineers to achieve well placement objectives
more effectively. Tools that probe the formation
some distance from the wellbore allow drilling
engineers to visualize complex subsurface geom-
etry. Engineers use real-time, high-resolution
image data to define structural geometry and pro-
actively adjust drilling programs. LWD measure-
ments also help operators differentiate intervals
that have superior production characteristics
from those that may not be profitable.
To optimize the horizontal drilling process,
well placement engineers have developed work-
flows that help them in the quest to realize drill-
ing objectives. Even before the well is spudded,
these workflows play a role in helping geologists
and engineers identify targets and develop realis-
tic trajectories that avoid unnecessary drilling
complications. While drilling, the well placement
team can update models with real-time informa-
tion utilizing a 3D approach.
These new technologies and workflows may
not always provide the answer. The critical ele-
ment in determining which well type performs
best is the drilling team’s understanding of how
reservoir geology impacts long-term performance
of the well. In many high-permeability reservoirs,
particularly during the early phase of develop-
ment, vertical and horizontal wells may perform
1,570
1,400
1,250
1,190
1,020
257
1989 1990 1991 1992 1993 1994
Year
equally well, which makes vertical wells more
attractive because of lower costs. In a recent
study from a complex, naturally fractured reser-
voir, vertical wells produced with higher oil rates
and lower water cuts than horizontal wells.4 This
phenomenon was a result of waterflood maturity
and unique reservoir geology.
Nevertheless, recent drilling and geosteering
technology developments have allowed access to
resources for which conventional vertical or geo-
metrically drilled horizontal wells would not have
worked as well. This article discusses LWD tools
that provide data directly impacting drilling pro-
grams, software for visualization of subsurface
geometry and a workflow operators use to optimize
well placement through structural steering tech-
niques. Examples from a gas storage project in
Italy, an unconventional reservoir in the US and a
thin carbonate reservoir in the Middle East dem-
onstrate how operators are optimizing well place-
ment and more effectively accessing reservoirs
and resources.
Dimensions of Drilling
Technological advances in drilling were key fac-
tors in the sharp rise in the number of horizontal
and directional wells in the 1990s. In 1986, there
were 41 horizontal wells drilled worldwide
(below left).5 Just four years later, in 1990, there
were 1,190 horizontal wells drilled, the majority
of which were in Texas, USA. More than 20% of
those wells were concentrated in the Austin
Chalk formation. Activity in this formation epito-
mizes the evolution that has occurred in horizon-
tal drilling.
The Austin Chalk is a low-permeability and
low-to-moderate–porosity reservoir. The first
wells were drilled in this formation in the 1920s.
Production of oil and gas in commercial quanti-
ties relied on the well intersecting existing inter-
connected fractures. To enhance production
from vertical wells, operators introduced new
technologies with varying degrees of success.
These technologies included acid treatments to
open pathways from the wellbore to existing frac-
ture networks, seismic interpretation to locate
fracture clusters and hydraulic fracture stimula-
tion to increase wellbore connectivity to natu-
rally occurring networks.
In the 1980s, operators began experimenting
with lateral well extensions, usually by reenter-
ing existing nonproductive wells. Because this
approach allowed the wellbore to contact many
more fractures than was possible with the original
vertical well, spectacular production increases
often ensued.
Oilfield Review
 Offset Log Layer Geologic
Well Log Squaring Column Model Resistivity Log

Markers Markers Surfaces

Modeled
Measured

Horizontal Departure
Formation Formation markers The markers are linked to
resistivity identified surfaces in the geologic model.

Geologic Model

> Model, compare and update method of well placement. Log data from offset wells are used to construct geologic and tool
response models. Geologists generate geologic models by first squaring the original data, creating layer columns and labeling
formation markers and surfaces (top left). Formation markers and layering are propagated from offset well data to create a
geologic model (bottom left), which may also be enhanced with seismic data. A forward modeling program predicts how logging
tools, such as a resistivity tool, will respond to formation properties (red, top right). Once the model is generated, a well trajectory
(green, bottom right) is proposed to target specific reservoir layers. As the horizontal well section is drilled, the measured
response (blue, top right) observed from the well is compared with the modeled response. (Adapted from Griffiths, reference 7.)

Some experts have attributed the global
growth in horizontal drilling to the successes and
lessons learned in these and other Texas wells.6
However, not all horizontal wells drilled in the
Austin Chalk formation experienced dramatic
increases in production. Many operators failed to
take into account the complex nature of the for-
mation, which included compartmentalization
and subseismic faults that divided the Austin
Chalk into multiple isolated reservoirs.
Today, operators use extensive reservoir mod-
els to extrapolate surface and downhole data and
predict the formations that horizontal wellbores
will encounter. This approach attempts to iden-
tify rock of better reservoir quality, and where
1. Baker Hughes: “Interactive Rig Counts,” Investor
Relations, http://gis.bakerhughesdirect.com/RigCounts/
(accessed February 13, 2013).
2. Alexander T, Baihly J, Boyer C, Clark B, Waters G,
Jochen V, Le Calvez J, Lewis R, Miller CK, Thaeler J and
Toelle BE: “Shale Gas Revolution,” Oilfield Review 23,
no. 3 (Autumn 2011): 40–55.
3. For more on the evolution and innovations in drilling
technology: Felczak E, Torre A, Godwin ND, Mantle K,
Naganathan S, Hawkins R, Li K, Jones S and Slayden F:
“The Best of Both Worlds—A Hybrid Rotary Steerable
System,” Oilfield Review 23, no. 4 (Winter 2011/2012):
36–44.
applicable, of better completion quality. However,
until a well is drilled, the model, which is a func-
tion of the type and quality of available data,
remains an approximation.
Well Placement Methods
Downhole hardware for drilling horizontal wells
has improved considerably over the past decades,
and well placement methodologies and work-
flows have been developed to capitalize on new
technologies and drilling techniques. With these
improvements, well placement methods continue
to evolve. Today, three complementary methods
are generally used in well placement.7 The first
method is characterized as model, compare and
Bennetzen B, Fuller J, Isevcan E, Krepp T, Meehan R,
Mohammed N, Poupeau J-F and Sonowal K:
“Extended-Reach Wells,” Oilfield Review 22, no. 3
(Autumn 2010): 4–15.
Williams M: “Better Turns for Rotary Steerable Drilling,”
Oilfield Review 16, no. 1 (Spring 2004): 4–9.
Downton G, Hendricks A, Klausen TS and Pafitis D:
“New Directions in Rotary Steerable Drilling,”
Oilfield Review 12, no. 1 (Spring 2000): 18–29.
update and is a reactive drilling process. The sec-
ond relies on estimating and extrapolating the
orientation of bedding planes from formation
dip data, usually with azimuthal measurements
acquired while drilling. The third method relies
on deep-reading directional data for remote
boundary detection to proactively adjust the well-
bore trajectory to maximize reservoir contact and
avoid exiting target zones.
In the model, compare and update method,
the well placement team first generates a model
of logging tool responses based on expected for-
mations observed in offset log data (above).
Seismic data interpretations are included in the
analysis to help geologists estimate the location
4. Widjaja DR, Lyngra S, Al-Ajmi FA, Al-Otaibi UF and
Alhuthali AH: “Vertical Cased Producers Outperform
Horizontal Wells in a Complex Naturally Fractured Low
Permeability Reservoir,” paper SPE 164414, presented at
the SPE Middle East Oil and Gas Show and Conference,
Manama, Bahrain, March 10–13, 2013.
5. Kuchuk F, Nurmi R, Cassell B, Chardac J-L and Maguet P:
“Horizontal Highlights,” Middle East Well Evaluation
Review 16 (1995): 7–25.
6. Kuchuk et al, reference 5.
7. Griffiths R: Well Placement Fundamentals. Sugar Land,
Texas, USA: Schlumberger (2009): 10.

Spring 2013 17
 Real-Time Modeling of formation boundaries. The team may use 3D
200 visualization software that usually includes plan-
Gamma ray, gAPI

Modeled
ahead functionality to develop wellbore trajec-
Measured tory and drilling programs. Real-time data
acquired while drilling either validate the model
0 or are used to update it in response to the new
information (left). The directional driller can
Deep resistivity, ohm.m Shallow resistivity, ohm.m

1,000
then make changes in the wellbore trajectory
100 Modeled
based on the updated model.
Measured
10 The second well placement method requires
an understanding of the orientation and magni-
tude of the formation dip. After interpreting azi-
1,000
muthal data from wellbore images, well
100 Modeled placement engineers are able to estimate and
Measured extrapolate the orientation of the target bed or
10
formation. The bit is steered to remain within the
target. If the bit is no longer in the target reser-
X,450
voir layer, the LWD data can be used to determine
whether the bit has exited the top or the bottom
X,500
of the reservoir, and the directional drilling engi-
True vertical depth, ft

neer can apply corrections to steer the bit back

X,550
toward the target (below). When the wellbore
crosses a fault and leaves the reservoir, this tech-
X,600
nique may not be effective because the engineer
Gamma ray, gAPI
X,650
must know which direction to proceed to recon-
50 75 100 125 150
nect with the target, and azimuthal data alone
X,700 may not provide that information.
1,000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 0 200
In the third method, well placement engi-
True horizontal length, ft Gamma ray, gAPI
neers use remote boundary detection to proac-
> Tracking the model. Well placement engineers and geologists may create software models of tively determine the direction in which to steer
logging tool responses from anticipated subsurface geology. The gamma ray (green, top) and shallow the bit. Deep azimuthal measurements give
resistivity data (blue, middle) are tracking the modeled response (red), which would indicate that the
2D model for well trajectory (bottom, green) is valid. Differences between the modeled and measured
deep resistivity data (log data, bottom) may indicate that the well trajectory will need to be adjusted,
although the deep resistivity data are again tracking at the current bit position.

Well Intersecting Ascending Layer Image log Well Intersecting Descending Layer Image log
Top

Bottom

Top

Measured depth Measured depth

> Well placement using formation dip data. Azimuthal log data in the shapes of smiles and frowns help well placement engineers determine bit corrections.
When a wellbore crosses a bedding plane, an azimuthal logging tool response indicates whether the wellbore is exiting an ascending or descending
geologic layer. When the wellbore cuts an ascending layer (left), the first contact with the formation is at the bottom of the hole (bottom left); when the bit
exits the layer, the last contact will be at the top of the hole. When the bit cuts an ascending layer, the data appear as a frown in the image. Conversely,
measurements from a wellbore that exits a descending bedding plane appear as a smile (right). The bit can be guided up or down based on these
interpretations to ensure that the wellbore remains in or reconnects with a target zone.

18 Oilfield Review
early warning of approaching changes in the
target and surrounding layers (right). This tech- X,606
nique works best when there is sufficient resis-
tivity contrast between the bounding layer and X,608
the target. Drilling programs that optimize

True vertical depth, ft

drainage, access untapped compartments and X,610
steer clear of potential water sources are some Planned trajectory
of the primary beneficiaries of this type of pro- X,612 Drilled trajectory
active drilling. In thick reservoir sections or in
low-contrast environments, this technique may X,614
not be as effective. Complex geologic environ-
ments, such as faulting and folds, are also prob- X,616
lematic for this technique.
Another well placement technique, structural X,618
X,800 X,850 X,900 X,950 Y,000 Y,050 Y,100 Y,150 Y,200 Y,250 Y,300
steering, extends the capabilities of these three
True horizontal length, ft
methods. It replaces geometric assumptions
> Distance to boundary (DTB) technology for well placement. Real-time distance to boundary mapping
about planar surfaces with geologically informed
technology uses directional measurements and large depth of investigation (DOI) to determine the
predictions of structure based on observed well distance to adjacent layers above and below the well path. For DTB technology to be used effectively,
data (below right). Whereas most well placement resistivity contrasts between adjacent beds must be present, and the adjacent beds must be within the
techniques focus on geometry, structural steering measurement window. Resistivity data from deep-reading LWD services, such as the PeriScope tool,
uses some aspects of the traditional methodolo- can be inverted and the values converted to colors. Contrasting colors highlight the differences in
bedding plane properties. Data are processed and presented in such a way as to give the appearance
gies but attempts to resolve geologic complexi- of curtains, giving rise to the name curtain display. When the well position relative to adjacent beds is
ties with LWD data, some of which have only known, the bit can be steered by making adjustments to the drilling assembly to point the bit in the
recently become available in real time. desired direction (blue) so that the wellbore stays within target zones or returns should the trajectory
exit from a target. Had the planned trajectory (green) been followed, this well would have exited the
target zone (light colors).
Structural Steering Workflow
Directional drilling is defined as the science of
steering a wellbore along a planned path to a tar-
get located at a given lateral distance and direc-
tion. Structural steering, which leverages
information from LWD services, is the process of
combining structural analysis and modeling
capabilities with borehole images to create 3D
models that operators use to optimize well place-
ment, often in real time. By incorporating geo-
logic models created with new software tools and
developing greater trust in interpretations that
might not fit original drilling programs, operators
are able to make real-time decisions based on
structural steering methodologies.
One example of software that enables well
placement by means of structural steering com-
bines two plug-ins used in the Petrel E&P soft-
ware platform: eXpandBG near-wellbore to
reservoir scale modeling and the eXpandGST real-
time geosteering module. Real-time data from
the MicroScope resistivity- and imaging-while-
drilling service can be combined with deep mea-
surements from the PeriScope bed boundary
mapper tool to provide structural analysis and
modeling capabilities.
Using data from tools that provide deep-read-
> Structural steering for well placement. Structural steering incorporates reservoir modeling and
ing capabilities along with those that acquire real-
distance to boundary technology in conjunction with high-resolution imaging to manage drilling
time borehole images, geologists at Schlumberger decisions. From these data, geologists create 3D models, such as the one shown, which help
have developed a structural steering workflow directional drillers visualize the formations around and ahead of the bit. This is especially useful for
that provides a framework for well placement predicting subsurface geometry and for guiding the bit in complex reservoirs with faults and folds.

Spring 2013 19
 Prejob Planning decisions (left). The interpreter picks the dis-
tance to boundaries, and the boundaries are dis-
played on an eXpandGST curtain section. Image
data from tools such as the MicroScope service
provide bedding dip, fracture information and
fault detection.
The eXpandBG module imports the LWD log-
ging data, and engineers generate an updated
Structural model Offset well logs
model that includes drilling polarity logs. Polarity
logs indicate whether the well is heading toward
the bottom or toward the top of a structure. The
Correlation and Formation Tops Well Placement Models
software next computes a true stratigraphic
thickness (TST) index; TST is related to the
thickness of the reservoir section. Well place-
ment engineers can compare structural dips
while the drilling progresses with those in the
original model and quickly identify anomalies.
Curtain section The software projects the structural dip away
from the well using stratigraphic horizons, and
the geologist can label formation tops and strati-
Real-Time Dip Analyses
graphic surfaces. Armed with this information,
the well placement team can determine whether
corrections are required and in which direction
to steer.
Two crucial elements for structural drilling
are LWD data that can be used to develop realis-
tic models and software that can provide a
robust solution describing the reservoir. Without
real-time data, engineers and geologists may
TST Profiling Well Placement have difficulty understanding the geometry of
the subsurface and accurately projecting where
the next step should be taken. Unfortunately,
engineers must often make decisions with insuf-
ficient data about complex reservoirs. Until
recently, the tools for resolving these complexi-
ties did not exist for LWD operations, but this is
no longer the case.

Postjob Evaluation and Deliverables

, Structural steering workflows. Prejob planning
for structurally steered wells begins with geologists
creating a structural model from offset well logs,
which may be from a vertical pilot well (top).
Data are correlated to determine the location of
formation tops and geologic markers. These data
are then propagated away from the vertical well,
and a well trajectory is created based on the
expected subsurface geometry. As the well is
drilled, well placement engineers use real-time
analyses of dip data to steer the well. They may
use true stratigraphic thickness (TST) profiles to
determine distances from marker beds and make
adjustments in well placement. When the well
reaches TD, the job is not finished. Models are
updated with the newly acquired data,
completions and designs can be optimized and
new well plans may be developed to maximize
recovery (bottom).

8. Borghi M, Piani E, Barbieri E, Dubourg I, Ortenzi L and presented at the 10th Offshore Mediterranean
Van Os R: “New Logging-While-Drilling Azimuthal Conference and Exhibition, Ravenna, Italy,
Resistivity and High Resolution Imaging in Slim Holes,” March 23–25, 2011.

20 Oilfield Review
Tools of the Trade
Along with modeling software and drilling hard-
ware, LWD tools are experiencing an evolution in
design and function. Originally, LWD tools repli-
cated information available from conventional
wireline logging tools, which are designed pri-
marily to obtain high-quality petrophysical data
essential for reservoir characterization. Modern
LWD tools still provide petrophysical information
and have the advantage that data are acquired
before the formations have been exposed to drill-
ing fluids, which over time can alter rock and
fluid properties. However, the real-time aspect of
LWD operations is creating a divergent path in
tool development.
Service companies are introducing LWD tools
that probe regions deeper in the formation than
is customary with wireline tools. New tool designs
also make it possible to acquire data at the bit.
With a wide swath of the formation around the
wellbore illuminated by these measurements,
drillers can precisely position the wellbore to
optimize production or injection performance.
Application of data from these new tools has
the potential to fundamentally change the way
directional wells are drilled. For example, well
placement engineers can use information pro-
vided by deep-reading tools to help steer the well
within a narrowly defined target zone. Using deep-
reading measurements and TST processing, engi-
neers can now manage the direction of wellbore
trajectories based on surrounding structures
rather than conditions very close to the wellbore.
For fractured reservoirs, such as the Austin
Chalk and many shale plays, the reservoir section
may be tens or even hundreds of meters thick,
and well placement may be more focused on
intersecting fracture networks than staying in a
narrow zone. Imaging tools that provide high-
resolution measurements can confirm the pres-
ence of fractures and perhaps lead to redirecting
or redrilling wellbore sections that are not opti-
mally placed. Conversely, to prevent early-onset
water production, fracture and fault avoidance
may be the objective in some reservoirs.
Engineers use these same imaging tools to iden-
tify fractures and faults and to accurately charac-
terize their orientation.
Because these measurements, especially
wellbore imaging, involve large amounts of data,
and LWD data transmission rates are orders of
magnitude below those of wireline logging sys-
tems, the primary source for imaging data has
been wireline tools. Recently, LWD data trans-
mission systems and imaging tools have been
introduced that can replicate the capabilities of
Spring 2013
Mud Pulse Telemetry Data Rate
10 3
Data transmission rate, bit/s
10 2
10 1
10 0
10 –1
1970 1980 1990 2000
Year
> Transmission rates with mud pulse telemetry (MPT). LWD tools transmit
and receive data using mud pulses encoded with data. Early MPT systems
had transmission rates in the single-digit bit/s range. Although these rates
have increased in the past few decades, LWD data transmission rates are
two orders of magnitude below that of wireline logging tools. For some
operations, LWD data must be stored in memory and retrieved when the
tools are back at the surface. Downhole storage is typical for LWD
measurements that require large amounts of data such as high-resolution
images. New data compression techniques, combined with higher data
transmission rates, now make it possible to acquire some of these data in
real time for operations such as structural steering.
wireline tools for detecting fractures and faults wireline counterparts, service companies have
and determining their orientation. No longer are found ways to overcome inherently low MPT
separate logging runs required to obtain this rates. However, data-intensive measurements,
information, and drillers can make decisions such as those associated with borehole imaging,
while the drilling assembly is still in the hole. were almost always performed with wireline log-
ging tools because the logging cable offered the
Resolution Evolution ability to transmit data at sufficiently high rates.
Most LWD tools transmit data to the surface Modern LWD MPT systems transmit at higher
using mud pulse telemetry (MPT). Although rates—some systems can approach 128 bps.
today, data rates are often given in megabits/s These enhanced transmission speeds, together
and terabits/s, mud pulse telemetry systems orig- with new methods for data compression, have
inally offered data rates in the single-digit bit/s opened up a new world of possibilities for real-
(bps) range (above). Because LWD tools have the time data acquisition. One tool that has benefited
ability to continuously transmit data uphole from higher data transmission capabilities is the
while drilling, giving them the benefit of having MicroScope service (below).8 Acquiring data
more time to acquire and send data than their from focused azimuthal sensors while rotating,
Azimuthal gamma ray
Two high-resolution
button sensors
Inclination
Bit resistivity
+ Laterolog resistivity at four DOI
+ Borehole images at four DOI
+ Toroidal resistivity at two DOI
+ Mud resistivity
> MicroScope tool. The MicroScope LWD logging tool provides four laterolog resistivities, four
borehole image measurements and two toroidal resistivities. The tool also includes inclination and
azimuthal gamma ray measurements. High-resolution borehole images from the tool can be used to
define structural conditions. From these data, faults as well as fractures can be identified. Because
the tool provides high-resolution information at multiple depths of investigation (DOIs), natural
fractures can often be differentiated from shallow drilling-induced fractures. Toroidal resistivity
measurements are useful for determining drilling conditions and identifying formations at the bit.
21

FMI Images Depth, MicroScope Images
ft
X,333
X,334
X,335
X,336
X,337
X,338
X,339
X,340
> Wireline versus LWD imaging. Data from wireline imaging tools, such as
the FMI tool (left), have been the standard, even though running these tools
in horizontal wells can be time consuming and may result in added risk of
sticking. Recently introduced LWD imaging tools, such as the MicroScope
tool, provide images (right) with a quality comparable to that of wireline
tools, often in real time or stored for surface retrieval. (Adapted from
Allouche et al, reference 10.)
the tool provides images of the borehole compa- measurement of current is a function of the for-
rable to those of wireline tools such as the FMI mation conductivity (and its reciprocal, resistiv-
fullbore formation microimager. An added bene- ity). The buttons measure azimuthal resistivity in
fit of the MicroScope tool is that it can provide 56 separate bins distributed around the borehole
high-resolution resistivity images at different circumference, and the orientation of the button
radial depths of investigation, which allows engi- measurements is determined with respect to the
neers to distinguish natural fractures from drilling- Earth’s magnetic field, which is measured with
induced fractures. an azimuthal orientation system mounted per-
The tool uses toroidal antennas as transmit- pendicular to the tool’s axis.
ters to send axial currents along the collar and The full array of measurements has depths of
into the formation for resistivity measurements. investigation approximately 2.5, 7.6, 13 and
Two electrode buttons mounted at opposite sides 15 cm [1, 3, 5 and 6 in.], measured radially out-
of the collar provide borehole coverage as the ward from the tool surface. These data can
tool rotates. The current leaves the tool surface, resolve bedding planes and features as small as
is directed through the conductive drilling fluid an inch. Although FMI images can resolve
into the formation and returns to the button elec- smaller features, which are useful for texture
trodes.9 Once corrected for borehole effects, the analysis and characterization of fractures,
MicroScope image data compare favorably with
FMI images (above).
22
A bit resistivity measurement, derived from
two antennas at the bottom of the tool, is also
available. One antenna acts as a transmitter and
the other as a monitor. Current flows out from the
bit and returns farther up the toolstring. The
drillstring below the antennas acts as an elec-
trode, and the measured current depends on the
formation resistivity and mud properties.
Other than the large amount of data needed
to provide images, one of the biggest challenges
of producing high-resolution images using LWD
tools is the conversion of time-based to depth-
based data. Traditional LWD measurements
are indexed to pipe movement observed at the
drilling floor. This technique is not adequate
for detection of small formation features
because drillpipe movement at the surface may
not reflect small tool movements downhole.
Scientists at Schlumberger have introduced a
new algorithm to derive local depth information
based on tool revolutions rather than observed
pipe movement.10
For this technique, high-resolution data,
along with magnetometer-based tool orientation,
are recorded versus time. These data can be
viewed as strips with a constant and known thick-
ness. Converting the time-based measurements
to a depth-indexed image requires precise esti-
mations of the azimuthal and axial position of the
sensors. As the tool advances, overlapping strips
are merged and then correlated to axial tool
movement. The technique provides a high-resolu-
tion depth match (next page, top). The images
are then transmitted to the surface with minimal
resolution degradation.
Well placement engineers also use measure-
ments with greater depths of investigation than
those of wireline logging tools to identify dis-
tances to top and bottom boundaries of reservoir
sections. These measurements help engineers
plan wellbore trajectories so they remain within
target intervals. The PeriScope bed boundary
mapper makes a 360° measurement and can
detect beds as far as 6.4 m [21 ft] from the bore-
hole. Tilted receiver coils that have directional
sensitivity can determine bed orientation. As
long as there is sufficient resistivity contrast
between target beds and those adjacent to the
zones of interest, the PeriScope tool can provide
crucial information about the position of the
wellbore in the formation.11
Modern well placement requires more than
determining the location and orientation of the
bit within a target zone. If faults are encoun-
tered, well placement engineers may not have
Oilfield Review
 Noncorrelated Correlated
Images Images

2,944
Correlation depth
2,945 Surface-measured
depth
2,946

Depth

Depth, ft
2,947
Image
strip
2,948

2,949

2,950

2,951
1,498 1,499 1,500 1,501 1,502 1,503 1,504 1,505
Time, 100 s
> Correlating high-resolution measurements to depth. LWD logging depths are referenced to pipe measurements taken at the surface. For most data, this is
an acceptable acquisition method. However, the accuracy of this method is not sufficient for high-resolution measurements. To compensate for
shortcomings of traditional depth measurements, engineers at Schlumberger developed a technique that uses overlapping strips from images (left) to
create an internal depth reference based on known fixed distances between sensor buttons on the tool. Correlation takes into account the mismatch
between tool movement downhole (middle, blue) and surface movement (black). The resulting correlated images (right) are much improved compared with
the noncorrelated images. (Adapted from Borghi et al, reference 8.)

Real-Time Images Used for Dip Picking

sufficient information from deep-reading tools
alone to understand the geometry required to
guide the bit back to the target. Integrating
high-resolution image data with data from deep-
reading tools helps geologists construct a 3D
Bed dip
picture of the structure surrounding the well- X,300
bore and can often help directional drillers Target formation Actual trajectory
True vertical depth, ft

decide where to go next and how to reconnect X,310

with the reservoir if the wellbore trajectory
exits the target interval (right). X,320

Resolving Complexities X,330

During the past decade, gas production from
organic-rich shales has become a global pursuit;
this phenomenon has been driven in large part by
hydraulic stimulation and horizontal drilling. The
conventional approach to developing these
resources is to drill a vertical pilot well followed
by a horizontal sidetrack targeting the shale
interval. Because of the complex geologic struc-
tural settings in many of these plays, some well-
bores may exit the pay zone or encounter rock
with poor reservoir quality. Although seismic
data are frequently used to resolve reservoir com-
plexities, in many cases these data lack the reso-
lution to adequately define subsurface features.
A new 3D structural technique, which includes
the application of eXpandBG modeling, was
recently utilized in a Marcellus Shale well oper-
ated by Chief Oil & Gas LLC.12
Planned trajectory
Microseismic fault
X,340
X,800 Y,000 Y,200 Y,400 Y,600 Y,800 Z,000 Z,200 Z,400 Z,600 Z,800 XX,000
True vertical length, ft
> Data integration. Geologists use real-time images (top) to identify faults and determine dip direction;
these data are then used to explain geologic conditions. Geologists may also use DTB measurements
to help generate models of subsurface layers (bottom). The integration of these data allows directional
drillers to modify planned well trajectories (green) to maximize reservoir contact and determine the
optimal path (blue) to return the wellbore to the target (yellow) should the wellbore encounter unexpected
conditions such as faults and folding.
9. The MicroScope tool is designed for use in conductive muds. Bourgeois D, Tribe I, Christensen R, Durbin P, Kumar S,
Skinner G and Wharton D: “Improving Well Placement
10. Allouche M, Chow S, Dubourg I, Ortenzi L and van Os R:
“High-Resolution Images and Formation Evaluation in with Modeling While Drilling,” Oilfield Review 18, no. 4
Slim Holes from a New Logging-While-Drilling Azimuthal (Winter 2006/2007): 20–29.
Laterolog Device,” paper SPE 131513, presented at the 12. Amer A, Collins S, Hamilton D, Gamero H, Contreras C
SPE EUROPEC/EAGE Annual Conference and Exhibition, and Singh M: “A New 3D Structural Modeling Technique
Barcelona, Spain, June 14–17, 2010. Unravels Complex Structures Within the Marcellus
11. For more on the PeriScope tool and bed boundary Shale: Utilizing Borehole Image Logs,” presented at the
mapping: Chou L, Li Q, Darquin A, Denichou J-M, AAPG Eastern Section Meeting, Washington, DC,
Griffiths R, Hart N, McInally A, Templeton G, Omeragic D, September 25–27, 2011.
Tribe I, Watson K and Wiig M: “Steering Toward
Enhanced Production,” Oilfield Review 17, no. 3
(Autumn 2005): 54–63.

Spring 2013 23
 Based on the dips identified in the vertical
well, the formation was assumed to be gently dip-
ping to the NNW at 5°. The well was to be landed
Anticline in the target interval and follow this trend.
Structural Model Produced Using eXpandBG Processing
Syncline However, the lateral was actually landed at a fold
NW SE
axis where the structure immediately turned
down to the south at 24°. The well soon exited the
Vertical well 8°
reservoir section and crossed a fault, eventually
5° Fold axis reconnecting with the reservoir but in a section
25° that dipped in the opposite direction, north at
24° Marcellus Shale
25°. The well encountered a second fault and was
Horizontal well
then back in the target formation; drilling contin-
ued along the path indicated by the original
structural model. Unfortunately, because the for-
Marcellus Shale mation was dipping more steeply than had been
Gamma ray log
modeled, the well exited the bottom of the
Marcellus Shale earlier than expected.
Major fault line A review of the well path validates the need
Gamma ray
log X° Average dip and for real-time structural data while drilling. With
direction
only azimuthal gamma ray data available for
Highly interpreting the formation structure, the drilling
tilted
Asymmetrical anticline block Gently tilted block program did not produce an optimal well path.
Had imaging and deep resistivity data been
> Complex Marcellus Shale model. The original reservoir model (inset) was created by propagating acquired with LWD tools in real time, the resul-
formation tops picked from pilot hole data, and well placement engineers designed a well trajectory tant drilling and completion programs may have
based on the model. The actual structure was quite different. The lateral well was landed in what was been quite different.
thought to be a gently upward sloping (5° NNW) layer cake formation. However, before reaching One final step in the modeling process
horizontal, the well crossed a fold axis with steeply dipping beds, and then exited the target interval. involves validation with Petrel geologic recon-
Although the directional driller hoped that continuing to drill would help determine which way to steer
the well, the well soon crossed a fault; the fault block was lifted and tilted relative to the previous struction software. This suite enables restora-
section and was also dipping in the opposite direction. The well reconnected with the target interval tion and forward modeling of complex folded
only to cross a second fault. Fortunately, the well was still within the Marcellus Shale, and drilling and faulted geologic models. By simulating
continued close to the originally projected angle based on the pilot hole, from which it was assumed
mechanical rock behavior with a comprehensive
that the formation dipped 5° NNW. The well exited the bottom of the target earlier than expected
because the formation was dipping more steeply (8° NNW) than projected. The directional driller had set of boundary conditions, the software allows
little help determining how best to steer this well because only azimuthal gamma ray information was the user to analyze complex structures. The
available for guidance. Geologists later loaded image data, acquired with an FMI tool, into the software confirmed the viability of the complex
eXpandBG module of Petrel modeling software, which generated an interpretation that explained why
the well failed to encounter the reservoir as expected.
present-day interpretation (next page, top).
Newly acquired 3D seismic data also validated
the structural model.
Production from this well was classified as dis-
Constructed with data from the pilot well, the structural models using multiwell data for input. appointing compared with that in nearby wells.
original model indicated a syncline plunging to In this example, formation dip was computed Had the structural model been updated using
the northwest and an anticline plunging to the using the local curvature axes technique to pro- LWD images, the well path may have been modi-
southeast toward the toe of the proposed lateral duce the eXpandBG structural model.13 Solutions fied or perhaps redrilled based on the new model.
section. Engineers at Chief Oil & Gas proposed a using this technique may not be unique; however, Similarly, the four-stage stimulation design may
well trajectory based on this interpretation from the interpreting geologists may intervene and have been more effective (next page, bottom).
the vertical pilot well. adjust the solution to fit with other data such as Only Stages 2 and 3 were completely within the
The horizontal section was drilled using only 3D seismic interpretations. target zone. Stages 1 and 4 covered zones that
LWD azimuthal gamma ray data. After the lateral Geologists with Chief and Schlumberger were in the target for only half the interval.
was drilled, borehole images were acquired on wire- analyzed dip sequences in both the pilot well Additionally, a section of the Marcellus Shale at
line with an FMI tool. The interpretation provided and the subsequent horizontal section and clas- the heel of the well was not stimulated, although
by eXpandBG processing highlighted significant dif- sified structures using the local curvature axes it coincides with a highly stressed interval around
ferences between the original reservoir model and technique on a Schmidt plot to resolve the the fold where FMI data indicated the presence of
the geologic structure observed in the well. structural complexity. Contrary to the original natural fractures, which often enhance produc-
The eXpandBG approach solves for geometric model, the new model revealed three distinct tion in shale reservoirs. In this case, real-time
complexities without the need for extensive input sections: an asymmetrical anticline, a highly structural data may have resulted in a well trajec-
by the interpreter. The software can also create tilted block and a third section characterized as tory that contacted more of the target formation
gently tilted (above). and led to better well production.

24 Oilfield Review
Steering for Storage Step 1: After Deposition
A structural steering workflow that did use real-
time LWD data and eXpandBG processing was
recently employed in an underground gas storage
project carried out by Stoccaggi Gas Italia
(Stogit) SpA, the gas storage division of Società
Nazionale Metanodotti (Snam). The multifield,
Step 2: Early Stage of Deformation
multiwell project was developed with the techni-
cal contribution of Eni SpA specialists. The objec-
tive of the drilling program was to expose as Maximum
much reservoir section with optimal properties compresssion

as possible in the shortest well length.14 To that

end, wells were drilled and steered using real-
time LWD data.
As is the case in many areas of Italy, horizon-
Step 3: Present Day
tal drilling is challenging because of steeply dip-
ping beds, faulting and abrupt stratigraphic Fold axis

changes. The reservoir section of the Furci field

is characterized as a limited extension Pliocene
turbidite system. It includes several sand bodies
with smaller interbedded laminations. The pro-
cedure for drilling wells in the field followed a
predetermined workflow. The operator chose
horizontal targets, and well placement engineers Uplift results in
localized extension
loaded the well plan into the eXpandGST module
in the Petrel E&P software platform, which was
> Structural balancing and restoration. To confirm the validity of an interpretation created using the
populated with log properties from a vertical
eXpandBG model, structural balancing and restoration modeling must be performed. Assuming the
pilot well. The program created a forward mod- original layer cake geometry (top), the model is exposed to postdepositional loading using Petrel
eled log to predict log responses for several sce- geologic reconstruction software. Early-stage compression (middle) produces the complex geometry
narios such as a formation dip that was higher or observed, and later uplift explains the present-day condition (bottom). This last modeling step
validates the interpretation generated by the eXpandBG software.
lower than expected. These scenarios would indi-
cate that the well was in a different part of the
reservoir than planned. Hydraulic Fracture Stages
For the second of two wells drilled in the
field, the target consisted of two large sand lobes
separated by two shale beds. The objective was to
drill through the shallow sand lobe, cross the thin
shale beds and navigate into the deeper sand
lobe. The operator drilled the vertical pilot hole
Stage 4
as planned and then began the horizontal section Stage 3
Stage 2
following the predetermined trajectory. Stage 1
Geologists determined formation dips by
using two independent tool systems: a deep-read-
Marcellus Shale
ing bed boundary PeriScope tool and a borehole
imaging MicroScope tool. These measurements
provided information about faults and bedding
that cut across the borehole. As horizontal drill-
ing commenced, the PeriScope tool indicated flat
dip and then a slightly rising inclination. A sud-
den decrease in resistivity appeared to indicate
13. Amer et al, reference 12.
14. Borghi M, Loi D, Cagneschi S, Mazzoni S, Donà E, > Completion results. The operator designed the stimulation program for the Marcellus Shale well
Zanchi A, Boiocchi D, Gremillion J, Chinellato F, based on interpretation of the azimuthal gamma ray data; the program was developed before the
Lebnane N, Lepp R, Chow S and Squaranti S: “Well
Placement Using Borehole Images and Bed Boundary revised structural model, shown here, was created. Of the four stages shown (magenta), only Stages 2
Mapping in an Underground Gas Storage Project in and 3 were wholly within the target zone. Stages 1 and 4 were only partially in the Marcellus Shale. No
Italy,” presented at the 10th Offshore Mediterranean treatment was applied to the heel of the well (dashed white oval) where fractures were identified,
Conference and Exhibition, Ravenna, Italy, which engineers viewed as a missed stimulation opportunity. The operator considered the performance
March 23–25, 2011. of this well disappointing compared with that in offset wells.

Spring 2013 25
 LWD Data
MicroScope
Resistivity Images

PeriScope Data

PeriScope
Resistivity

ROP
Gamma Ray
700 800 900 1,000

X,004
Shale layers
Planned trajectory
X,006 Upper lobe Drilled trajectory

X,008
True vertical depth, m

X,010

X,012

X,014

X,016
Lower lobe Possible subseismic fault,
apparent dip
X,018
700 800 900 1,000
True horizontal length, m

> Complex geology in an Italian gas storage well. The targets for this horizontal well were two sand lobes separated by shale layers. The original model
assumed a layer cake geology, and the drilling engineers developed a trajectory (green) to pass through the upper sand lobe, cross the shale layers and
terminate in the lower lobe. Instead of encountering continuous layers, the actual trajectory (blue) encountered a fault, entered an uplifted section below
the target sand lobes and crossed another fault before reconnecting with the lower sand lobe in a downthrown section. After drilling through a section of
the lower sand lobe, the well placement engineers turned the well upward and then it again crossed the shale layer and reconnected with the downward-
dipping upper sand lobe. Traditional petrophysical measurements—resistivity (Tracks 2 and 3), gamma ray (Track 4) and formation density (not shown)—did
not provide much directional guidance. In addition, had engineers used only DTB data from the PeriScope tool (red and blue dots, bottom), they would have
had difficulty determining directional adjustments. Image data from the MicroScope tool (top) allowed geologists to identify the faults and determine dip
direction and correctly steer the well. Without the two complementary measurement systems, engineers would have had difficulty determining in which
direction to steer the well after it crossed either fault. (Adapted from Borghi et al, reference 14.)

that the trajectory had crossed the upper lobe
and the middle shale sections and was approach-
ing the lower lobe; however, geologists inter-
preted the image data as indicating the well had
crossed an unexpected fault and was now in an
upthrown section (above).
Drilling through the upthrown section con-
firmed the fault interpretation, and eventually
the wellbore crossed a second fault and again
encountered the reservoir section. Log data
acquired after the trajectory crossed the second
fault indicated that the well was penetrating the
lower sand lobe, and the drilling team made the
decision to incline the bit to more than 90° and
reconnect with the upper lobe.
The well crossed back through the shale lay-
ers and eventually reached a downdip section of
the upper sand lobe. After crossing the first fault,
the original well trajectory would have missed
the lower lobe entirely, and a large portion of the
well would have been drilled through the shale
layers that separated the lobes. The updated
trajectory, modified using real-time LWD data,
intersected both lobes and maximized wellbore
contact with the reservoir section.
Shale Drilling
In a Niobrara Shale exploration well in northeast
Colorado and southeastern Wyoming, USA, engi-
neers used the structural drilling workflow to
resolve complex geologic conditions in a shale
well.15 The Niobrara Shale play is an upper
Cretaceous calcareous shale that produces oil
and gas. The shales are composed of argillaceous
limestones with interblended chalk, marl and
bentonite. Because of the rock’s low permeability
and porosity, production is generally higher in
zones with natural fractures enhanced by hydrau-
lic stimulation.
The typical development scenario for
Niobrara Shale wells is to drill a vertical pilot
hole and acquire petrophysical data using logging
tools on wireline. For evaluation wells, the log-
ging program usually consists of resistivity, neu-
tron and density porosity, elemental capture
spectroscopy and nuclear magnetic resonance
(NMR) tools. Borehole image logs are run for
fracture identification and geologic characteriza-
tion. Acoustic logs may be run for mechanical
properties, which are used in fracture stimula-
tion design and wellbore stability estimations.
Operators often include conventional coring for
the pilot wells to determine lithology and
describe fractures. Data from the pilot holes are
used to characterize the reservoir, define the ori-
entation of target zones and identify the optimal
depth for landing the lateral section.
15. Koepsell R, Han SY, Kok J, Munari M and Tollefsen E:
“Advanced LWD Imaging Technology in the Niobrara—
Case Study,” paper SPE 143828, presented at the SPE
North American Unconventional Gas Conference and
Exhibition, The Woodlands, Texas, June 14–16, 2011.

26 Oilfield Review
 95°
Resistive Fracture

Fault
True Dip Angle

Bed Boundary

Fault Orientation

Microfault Orientation
–5°
Top
Bed Boundary
Dynamic Image

Fault
Bottom
Resistive Fracture

Microfault
Top
1,000
Deep Resistivity
ohm.m
Shallow Resistivity
0.1
Top
Static Image

Bottom

Top
150
Gamma Ray gAPI

0
Measured
X,940

X,950

X,960

X,970

X,980

X,990

Y,000

Y,010

Y,020

Y,030

Y,040

Y,050
Depth, ft

> Resolving complex geology. Image data from the MicroScope tool can be presented in dynamic (Track 2) or static (Track 4) mode. Formation dip (top,
Track 1) can be handpicked from images or computed from these data. The green tadpoles indicate the down direction of the dip, 0° to 360° clockwise
around each tadpole, which represents north-east-south-west-north. The magnitude of dip is also computed and can be read from the log. The magenta
tadpole indicates a fracture and provides its orientation. The image data can be presented in a wrap mode that simulates the horizontal well (inset).
Bedding planes (green), faults (magenta), open fractures (blue) and healed fractures (cyan) can be visualized as they appear in the horizontal wellbore.
(Adapted from Koepsell et al, reference 15.)

Logging objectives in the horizontal section
differ from those of the vertical pilot wells.
Fracture population density, type and orientation
are needed for stimulation design. Measuring the
wellbore path and orientation are crucial, espe-
cially when the well crosses in and out of reser-
voir layers. Identifying faults and determining
their location and orientation aid completion
design. Engineers identify zones with structural
complexity to help them keep the wellbore in the
reservoir unit or determine the best path to
reconnect if the well exits the target. Depositional
variations can be determined with LWD tools,
and geologists use these data to adjust models
that extrapolate properties from the pilot well.
Traditional methods of acquiring image data
in the lateral section require that drillpipe-
conveyed wireline tools be deployed. Geologists
used these data to identify the presence of natu-
ral fractures and quantify their orientation and
density. However, high-resolution image data
from the MicroScope tool eliminate the need for
separate wireline logging runs (above).

Spring 2013 27
 An operator in the Denver-Julesburg basin target layer known as the C bench. The image tures at an azimuth of 104° and with a length of
began a campaign to develop the Niobrara data indicated open fractures with strike popula- approximately 2,400 ft [730 m].
formation with multistage hydraulic fracturing in tions oriented NW-SE and mineralized fractures The MicroScope tool provided real-time high-
horizontal wells. From the vertical pilot well logs, striking NNE-SSW. To maximize intersection with resolution images for structural and fracture anal-
geologists were able to confirm the presence of a the natural fractures, the horizontal well section ysis. Engineers created 3D models using eXpandBG
was planned perpendicular to the natural frac- software, which helped them optimize well place-
Raw dips ment and design hydraulic stimulation operations.
N
W E
Vertical pilot hole The MicroScope image data were particularly use-
S
0 90 ful in revealing the complex structural setting. In
X,200
Open fracture
addition to numerous open and healed fractures,
Healed fracture geologists identified numerous faults, a missing
X,300 Fault section and structurally deformed beds (below).
Structural dip The planned well trajectory, developed from
X,400 pilot hole and surface seismic data, resulted in
the well crossing a fault and exiting the target
Niobrara formation
C bench Fort Hays zone into unproductive marl sections below the
X,500 Limestone target. The last half of the well was below the C
bench and was drilled mostly in nonreservoir
Structural dip Codell
X,600 siltstone quality ductile shale. After these data were ana-
Depth, ft

lyzed, the well was sidetracked and redrilled

through most of the interval and steered higher
X,700
Marl in the structure based on the new model.
Chalk The logging results affected a number of the
X,800 Ductile shale engineers’ decisions for the completion program
Fault for the sidetrack well. For instance, the program
called for openhole packers for isolation.
X,900 Carlile Engineers identified washed out and elliptical
Shale
borehole sections and avoided setting packers in
Y,000 these zones. Packers were not set near faults,
Greenhorn
Limestone which can affect the quality of the seal as well as
impact stimulation results. For similar reasons,
Y,100
packers were not set in open natural fractures,
Dip Data
X,100 Y,100 Z,100
90
Raw dips
E
N

S
W

Expanded Section, C Bench Target Structural dip Fault

Well in lowest part of C bench Well below C bench

C bench 0
target
270 90
0

180
270 90
0 0 0

180
270 90 270 90 270 90

180 180 180

Ductile shale Chalk Marl Fault

> Stepping out from vertical. Geologists can identify the location and orientation of bedding planes and faults in vertical wells and project them away from
the wellbore, but horizontal wells often encounter unexpected geologic geometry. In this vertical well section (top), geologists identified several geologic
sections, including the target reservoir C bench section, which is a mixture of chalk and marl bounded by ductile shales and unproductive chalks and is part
of the Niobrara formation. Well placement engineers developed a trajectory to follow the target, and directional engineers landed the lateral well in the C
bench (bottom, expanded section). Horizontal drilling proceeded for approximately 2,350 ft [716 m] and the well encountered structural geometry that
differed from geologists’ expectations. The well (black) crossed at least seven major faults (magenta lines). After the first set of faults, the C bench was
found to be upthrown, which positioned the well in the lowest part of the reservoir. As drilling progressed, the well crossed a fifth major fault and was
below the target formation and completely out of the reservoir. After geologists developed the new model of the horizontal well, the operator pulled back to
the first fault section and redrilled the horizontal section with an orientation (not shown) that carried the well above the original trajectory; this repositioning
allowed the well to remain in the target interval. Geologists can also use dip data to identify other features. The stereonet plots shown across five of the
intervals can be used to identify fracture and fault type and orientation. The far left plot shows the NNW by SSE orientation of healed fractures, probably
resulting from folding. (Adapted from Koepsell et al, reference 15.)

28 Oilfield Review
 Resistivity Water
90-in. Array
Oil
0.2 ohm.m 2,000

60-in. Array Bound Fluid

0.2 ohm.m 2,000 0 % 25

30-in. Array Bulk Density CMR Porosity T2 Distributions

Volumes
0.2 ohm.m 2,000 1.95 g/cm3 2.95 0 % 25 0.3 ms 5,000
T2 Cutoff Calcite
20-in. Array PEF Total Porosity Hydrocarbon
0.2 ohm.m 2,000 0 10 0 % 25 0.3 ms 5,000 Water
Hydrocarbon
Gamma Ray 10-in. Array Neutron Porosity Water Volume T2 Log Mean Saturation Oil
Depth,
ft 0 gAPI 200 0.2 ohm.m 2,000 45 % –15 0 % 25 0.3 ms 5,000 0 % 100 0 % 100

X,725

X,750

X,775

> Pilot hole logs. An oil-bearing carbonate reservoir section was identified from openhole logs acquired in a pilot hole drilled at a 30°
angle. The pay zone (gray shading) was less than 20 ft [6 m] thick based on measured depth. Corrected for well deviation, the true
vertical thickness would be even less than shown. Fluid mobility, determined from NMR (Track 5) and MDT data (not shown), indicated
a small permeable streak within the interval. The well placement team proposed drilling a horizontal well that would be steered using
PeriScope and MicroScope tools. Directional drillers use petrophysical measurements to guide them in steering wells, but there was
insufficient variation in the gamma ray (Track 1), resistivity (Track 2) or porosity (Track 3) within the target zone for these to be used. A
high-resistivity caprock overlying the zone, however, provided a boundary layer for reference in steering the well.

which were also identified in image data.
Fracture stimulations were performed in stages,
and the stages were designed to target similar
rock types identified from petrophysical data.
Stimulation designs also included consideration
of local stresses that resulted from formation
structural complexity.
Drilling Between the Lines
Unconventional resources may require approaches
for drilling and completions that differ from those
for conventional reservoirs, but conventional res-
ervoirs can benefit from application of unconven-
tional solutions. Saudi Aramco used the real-time
structural steering workflow with eXpandBG and
eXpandGST processing to access resources that oth-
erwise would have been difficult to produce eco-
nomically. The carbonate reservoir consisted of a main producing horizon. The pilot hole was
thin, permeable layer sandwiched between low- drilled as a 30° slanted well across the reservoir
permeability intervals overlain by a thick, nonpo- section, and Saudi Aramco carried out an exten-
rous carbonate caprock. sive data acquisition program that included cor-
The well was drilled in a mature giant field ing the full reservoir interval.
located in Saudi Arabia.16 Historically, this field Porosity and resistivity in the zone of interest
has produced mainly
Oilfield from two major carbonate
Review were fairly uniform; the operator used a CMR
reservoirs.SPRING 13 1980s, two smaller strati-
In the early combinable magnetic resonance logging tool to
graphic oilStructural Steering
accumulations Fig.discovered.
were 18 The identify the presence of movable oil (above). An
ORSPG
example well was13-STCTSTR
drilled in the18larger of these MDT modular formation dynamics tester con-
two reservoirs. The discovery was further delin- firmed that only a thin layer within the zone had
eated and tested by several vertical wells. The 16. Al-Suwaidi SH, Lyngra S, Roberts I, Al-Hussain J,
low-permeability reservoir contains good quality Pasaribu I, Laota AS and Hutabarat S: “Successful
Application of a Novel Mobility Geosteering Technique
light oil with a relatively high gas/oil ratio. in a Stratified Low-Permeability Carbonate Reservoir,”
In early 2012, Saudi Aramco drilled the first presented at the SPE Saudi Arabia Section Annual
Technical Symposium and Exhibition, Al-Khobar, Saudi
reservoir development well, deepening an exist- Arabia, May 19–22, 2013.
ing dead producer originally completed in the

Spring 2013 29
 Mobility

Formation
Dip

MicroScope
Images

Resistivity

Gamma Ray
Inclination
X,000 Y,000 Z,000

Drilling
Polarity

Caprock layer Well trajectory

PeriScope Inversion

Top of reservoir

PeriScope Real-Time
Boundary Inversion
Curtain

Well trajectory
Top of reservoir
Integrated
Curtain
Section

Inferred high-mobility layer

> Well placement and job execution. Within the oil-bearing carbonate zone of interest, engineers confined the target to a narrow permeable streak (bottom,
yellow) bounded by lower-permeability oil-bearing layers (tan). The drilling objective was to guide the well maintaining a constant distance from the
high-resistivity, low-porosity caprock (green) overlying the reservoir. Resistivity (Track 4) and porosity (not shown) data exhibited little variation across the
interval. For guidance, engineers used PeriScope curtain data (Tracks 7 and 8) to maintain the DTB. Geologists also used MicroScope image data (Track 3)
to detect subtle changes in orientation and formation dip (Track 2). Well placement engineers proactively corrected the well trajectory based on polarity
data (Track 6, red indicates drilling up structure, green indicates drilling down structure). Because fluid mobility and permeability were the properties that
differentiated the target interval from the rest of the zone of interest, an FPWD tool was included in the LWD logging suite. Mobility measurements were
acquired at irregular intervals along the well (blue circles, Track 1), but after validating the presence of fluid mobility for approximately 1,700 ft [520 m],
engineers removed the FPWD tool from the string because of concerns about tool sticking. The well placement team steered the well for approximately
2,900 ft [884 m] (bottom, blue) and stayed within the narrow window throughout the interval.

30 Oilfield Review
good mobility and would produce oil. From the
log data, petrophysicists determined that the
permeable layer was less than 10 ft [3 m] thick
and was positioned about 6 ft [1.8 m] below the
high-resistivity caprock layer. Log analysts were
uncertain whether the zone with high mobility
extended farther out in the reservoir or was sim-
ply a stratigraphic anomaly.
Even if the zone extended into the reservoir,
engineers knew that effectively producing from
such a small interval in the pilot well would be
difficult. Thus, they designed a horizontal pilot
producer to more effectively drain the reservoir.
Challenges included using real-time data from
LWD tools to verify the presence of the high-
mobility zone and stay within this narrow, high-
permeability window. Porosity and resistivity logs
provided little help in identifying the zone with
the best mobility.
The technical team determined that the best
course of action was to drill the well with a trajec-
tory that maintained a constant standoff or dis-
tance from the overlying caprock. The standoff
was based on distance to boundary (DTB) mea-
surements computed from a PeriScope tool. The
team relied on true stratigraphic thickness (TST)
data to maintain a constant position relative to
the caprock location. Well placement engineers
with Schlumberger were able to compute TST in
real time using eXpandBG processing of formation
dips picked from MicroScope images. Saudi
Aramco personnel used these interpretations to
instruct the directional driller in the proper
direction to guide the PowerDrive rotary steer-
able system.
Based on results from the pilot hole, NMR
data were judged to be insufficient to identify
the zone with mobility. Consequently, an FPWD
formation pressure while drilling tool was used
to confirm that the wellbore trajectory remained
in the high-mobility streak. To ensure that the
wellbore followed subtle changes in dip and
direction, the geosteering staff used interpreta-
tions from borehole images acquired with a
MicroScope tool.
Geologists created a 2D structural model
from pilot well data and forward modeled logging
responses for the LWD tools. The well placement
team landed the well near the interval, steering
the well stratigraphically upward to attain the
required distance to the upper boundary. Once
the data from eXpandBG processing confirmed
the required trajectory, the well was drilled main-
taining the proper orientation (previous page).
Spring 2013
FPWD data were acquired for the first 1,700 ft Structural steering involves more tools and
[520 m] and confirmed that the chosen path was requires more data for analysis than conven-
following the high-permeability streak. Each tional drilling; in addition, the costs of structural
FPWD mobility test required leaving the drilling steering are higher. But the answers provided by
assembly stationary for 20 min. Significant over- the tools and data to engineers and geologists
pulls began to occur after each mobility test, and have the potential to reveal better access to
the FPWD tool was removed because of opera- more of the reservoir, enhance recovery and pro-
tional concerns related to hole conditions and duce more hydrocarbons. Structural steering
sticking. The remainder of the well was then may not be the answer for every well, but the
drilled using only DTB and TST data from opportunity to resolve the complexities of down-
eXpandBG and eXpandGST processing to deter- hole geology offers operators a tremendous tool
mine corrections to the wellbore trajectory. for enhancing resource recovery. —TS
Images from the MicroScope tool helped estab-
lish the formation dip and were a key input in the
interpretation. The horizontal interval covered
approximately 2,900 ft [884 m] and remained
within a 4-ft [1.2-m] sweet spot window for the
entire interval.
The well confirmed that the high-permeabil-
ity streak was not a stratigraphic anomaly and
extended far out into the reservoir. The well was
tested after completion and produced at a rate of
several thousand bbl/d. Further evaluation is
ongoing, but early analysis confirms that because
the well followed the high-permeability path,
resources were accessed that otherwise might
have been difficult to produce economically.
Knowledge Is Power
At one time, horizontal drilling was an exercise in
geometry and drilling technology. However, as
well placement techniques and practices have
evolved, LWD tools have been introduced that
provide well placement teams with a better grasp
of geologic and subsurface structural conditions.
Integrating downhole data into modeling soft-
ware provides operators with the ability to visual-
ize subsurface complexities. This knowledge
gives operators powerful tools to modify drilling
plans, alter wellbore trajectories and optimize
completion programs.
Service companies continue to add to the
assortment of LWD tools that may have been
considered impractical for the drilling environ-
ment in the past. Pressure sampling, downhole
seismic acquisition and acoustic logging devices
were once considered to be beyond the capabili-
ties of tools used while drilling. Just as these
services have been accepted by the industry,
high-resolution measurements that image the
borehole and result in large amounts of data are
now becoming available. Proper interpretation
of these data has the potential to alter the way
wells are drilled; such drilling is no longer based
primarily on geometry but optimized for down-
hole structural conditions.
31