SPE 99668
Application of Horizontal Waterflooding To Improve Oil Recovery From Old Oil Fields
R.V. Westermark, SPE, and J. Schmeling, Grand Directions; D.L. Dauben, SPE, Dauben Intl. Energy Consultants Inc.;
S. Robinowitz, Grand Resources Inc.; and H.V. Weyland, U.S. Dept. of Energy
Copyright 2006, Society of Petroleum Engineers
This paper was prepared for presentation at the 2006 SPE/DOE Symposium on Improved Oil
Recovery held in Tulsa, Oklahoma, U.S.A., 22–26 April 2006.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
This paper discusses the results of the application of horizontal
waterflooding technology to improve the recovery of oil from
older fields. A multi-disciplinary approach combining
geological interpretations, rock mechanics evaluations,
reservoir simulation studies, and drilling technologies was
employed. A range of reservoir conditions in which the
technology is viable is discussed and the results of the field
tests are presented.
The major focus has been on the shallow sandstone
reservoirs in northeast Oklahoma and southeast Kansas. It has
been demonstrated that additional oil can be recovered in
several different reservoir environments in these older fields,
by utilizing a combination of horizontal and vertical wells.
Two previous SPE papers described the results of a
Department of Energy (DOE) supported horizontal
waterflooding pilot located in the Wolco Field, Osage County,
Oklahoma.
The technology has continued to evolve over time. The
key to assessing the potential for a horizontal well application
is a thorough understanding of the reservoir description,
including the oil saturation, depositional environment,
reservoir pressure, permeability, and an evaluation of any
prior production and/or injection. Other parameters include the
length, spacing, positioning and orientation of the injectors
and horizontal producers in the target zone. Use of rotary-
steerable drilling equipment has made it possible to drill
underbalanced, short-radius horizontal laterals at a low cost.
Special procedures have been adopted to permit the transport
of openhole logging equipment through the short-radius curve
to the end of the lateral, providing a good description of
reservoir properties and detecting the presence and orientation
of natural fractures.
Results from several successful horizontal well projects are
discussed. These include (1) two horizontal waterfloods in
shallow Bartlesville sandstone reservoirs, (2) recovery from a
rim of moderately heavy oil underlain by water, and (3) an
application of horizontal waterflooding in a moderately heavy
oil reservoir.
Introduction
The major objectives of this paper are to discuss the industry
trends in the application of horizontal wells in flooding
operations and to describe the specific experiences of Grand
Directions, LLC (Grand), a division of Grand Resources, Inc.
Grand has been developing and applying horizontal well
technology during the past four years, with a focus on the
Pennsylvanian sands in the mid-continent area. Significant
advancements have been made in a variety of reservoir
environments1-3.
Waterflooding has been successfully used for many years
in the recovery of oil from petroleum reservoirs. Successful
waterfloods often recover as much or more oil than in the
primary recovery phase. In spite of the many successes,
waterfloods have not always been effective from a technical
and/or economic basis. Common problems have included
reservoir heterogeneities that cause the channeling of water;
low permeability which limits the rate at which water can be
injected below the fracture-parting pressure into the reservoir,
and high infrastructure costs for applications in reservoirs that
are deeper or located offshore.
The concept of horizontal waterflooding was introduced by
Taber4 in 1992 as a method for improving the performance of
conventional waterfloods. It was pointed out that water can be
injected at much higher rates and lower pressures in horizontal
wells than in vertical wells, allowing oil to be recovered
quicker. It was further concluded that areal sweep efficiency
would be improved by the line-drive geometry as compared to
the sweep patterns that develop between vertical wells.
Because of the accelerated rate of recovery and the potential
for increased total recovery, it was speculated that horizontal
waterflooding could offer significant benefits over
conventional waterflooding. Joshi was also a pioneer in the
2 SPE 99668
development and application of horizontal well technology5
and recently provided an analysis of the costs and benefits of
horizontal wells used in a variety of applications, including
waterflooding and enhanced oil recovery6.
Grand continues to monitor and take advantage of the
ongoing industry advancements that have been made. This
discussion will show that Grand has advanced the technology
during the process of implementing a number of projects. Fig.
1 shows the location of the projects discussed in this paper.
Discussion
Industry Trends.
Horizontal waterflooding has been under development for
the past 15 years. Texaco has been a leader in the testing of
the technology with applications in the New Hope Field
(Texas)7,8, Aneth Field (Utah)9,10, and the Captain Field (North
Sea)11. Table 1 provides the descriptions and applications of
the major projects that have been conducted.
In a horizontal well, the curve is referred to as the heel
while the end of the lateral is referred to as the toe. A
horizontal waterflood, as originally conceived, consists of at
least one horizontal injection well and one or more adjacent,
parallel horizontal producing wells, heels and toes in the same
orientation, as shown in Fig. 2a. With time, the concept of
horizontal waterflooding has been expanded to include various
combinations of horizontal and vertical wells, depending upon
the particular application. In flooding applications, horizontal
wells are aligned parallel to the principal orientation of natural
fractures so that oil can be mobilized and produced from the
matrix. This is in contrast to many primary recovery
operations in which the horizontal laterals intersect the natural
fractures to maximize productivity.
Field Applications.
The major objective in the earlier applications was to use
horizontal wells to improve waterflood performance in deeper,
low permeability reservoirs. The low matrix permeability
mandated the use of smaller well patterns to recover the oil
within a reasonable time frame. At the same time, the cost of
drilling the deeper wells was too high to justify their use.
Several field tests established that one horizontal well in such
a reservoir could replace several vertical wells in
effectiveness, with a considerable savings in drilling costs.
Good examples are the Texaco projects conducted in the New
Hope Shallow Unit7,8 and the Aneth Field9,10.
Horizontal waterflooding has also been applied in
reservoirs containing moderately viscous oil. This process
accelerates and increases oil recovery by the improvement of
sweep efficiency and was successfully applied by Texaco in
the Captain Field, located in the North Sea11. Fig. 3 depicts
the theoretical improvement in sweep efficiency achieved by
horizontal wells12. As shown, the predicted breakthrough
volumetric sweep for a conventional 5-spot waterflood is 72%
pore volume for a mobility ratio of 1, compared to 83% pore
volume for a horizontal waterflood. This higher sweep
efficiency predicts that a larger amount of oil can be recovered
within the normal life of a project.
Grand focused its early attention on the application of
horizontal waterflooding in the 25-50 millidarcy (md)
permeability portion of the shallow Bartlesville formation1-3.
Conventional waterfloods had often failed because of the
inability to inject water at sufficient rates while staying below
the fracture-parting pressure. By contrast, sufficient volumes
of water can be injected below fracture-parting pressure into
horizontal injection wells. Horizontal producing wells are then
used to capture the mobilized oil. The University of Tulsa,
together with the DOE, also designed a similar project in the
Glenn Pool Field in Oklahoma13.
Recent laboratory studies and field projects evaluated the
immiscible water-alternating-gas (WAG) process using
horizontal wells in reservoirs containing moderately viscous
oil. Key targets were the heavier oil reservoirs in the North
Slope of Alaska. A recent DOE study estimates that 10-20
billion barrels of moderately heavy oil are contained in
reservoirs such as Ugnu, West Sak, and Shrader Bluff14.
British Petroleum (BP) is evaluating an immiscible WAG
process using horizontal wells as a means of recovering a
portion of the moderately viscous oils in the North Slope15.
Horizontal injection wells will permit the high-rate injection
of fluids into the reservoir and horizontal producing wells will
capture the oil. A soluble gas such as carbon dioxide (CO2)
will dissolve in the heavier oil and reduce its viscosity. The
combined effects will permit the recovery of oil that may not
otherwise be possible. Thermal projects are not applicable due
to the permafrost conditions.
Horizontal injection and producing wells are also being
used in low permeability reservoirs where miscible gas
flooding is underway. These miscible gas projects are
conducted in light oil reservoirs where miscibility with the
crude oil can be established with the injection of gasses such
as CO2 or an enriched hydrocarbon gas. Mobil used horizontal
injection wells as an alternative to infill drilling in a CO2
project in the Levelland Field16. Mobil also reported plans to
use a CO2 horizontal flooding process in the Aneth Field to
improve the economics of the project17. EnCana (formerly
PanCanadian) used a combination of vertical and horizontal
wells to maximize the performance of a CO2 flood in the
Weyburn Field, Canada18. Phillips evaluated horizontal
injection wells to accelerate the recovery of oil from the South
Cowden Unit and to permit the centralization of field handling
facilities19. ARCO used a combination of vertical and
horizontal wells in the implementation of an enriched gas
drive project in Prudhoe Bay20.
Well Patterns.
Experience has shown that the pressure drop through a
horizontal injector can be significant relative to the pressure
drop between wells within the reservoir. Vicente21 performed
SPE 99668
a numerical model study to couple the well flow dynamics of
the horizontal wellbore and the reservoir, determining that
reservoir models need to incorporate the effects of pressure
drop through the horizontal lateral and the transient saturation
effects that occur in the reservoir around the wellbore. Steady
state models are considered appropriate in low-permeability
reservoirs where the pressure drop through the lateral is
negligible compared to the pressure drop between wells. In
other situations, the reservoir model needs to include a
wellbore model along with local grid refinement around the
lateral to accurately predict fluid flow through the reservoir. In
such cases, there is a tendency for fluids to flow from the heel
of a horizontal injector toward the heel of a horizontal
producing well. The net result is a lower sweep efficiency and
recovery than might be anticipated. Various investigators have
studied the problem of non-uniform flow out of and into
horizontal laterals and have made recommendations on
completions and well configurations that would minimize the
problem. Key findings include:
• Lach11 discusses the use of a “tuned” liner to
encourage the distribution of injected water toward the
toe of the injection well. A combination of
interspersed open wire wrapped screen and blank
casing was used to promote a more even inflow
profile.
• Popa22 proposed the drilling of horizontal injectors
and producers in a toe-to-heel (TTH) orientation, as
shown in Fig. 2b, to promote better sweep between
wells. He also proposed the use of horizontal
producing wells along with vertical injection wells,
pointing out that much better control on flow
distribution can be achieved in vertical injection wells
rather than in horizontal injection wells. A vertical
injection well aligned with the toe(s) of horizontal
producing well(s) is also considered TTH and is
shown in Fig. 2c-1.
• In a later publication, Popa23 continues to advocate the
drilling of horizontal injectors and producers in a TTH
configuration. He further discusses a variety of other
field patterns designed to enhance sweep efficiency,
noting that horizontal producers are always
recommended and that the injectors should be either a
vertical well or a relatively short horizontal well, as
shown in Fig. 2d.
• Zhao24 proposes a toe-to-heel waterflooding process to
recover heavier oils, in a pattern that utilizes a
horizontal producer and a vertical injector. The
horizontal producers are located at the top of the
formation and the vertical injection well is placed near
the toes in a staggered line drive configuration, as
shown in Fig. 2c-2. The injected water tends to flow
near the bottom of the formation by gravity drainage
and the pressure support pushes the oil upward toward
the horizontal producing well.
3
Technical Approach.
Grand has made continuous improvement over the past
four years in the development of the technology needed for the
successful application of horizontal wells in waterflooding.
This success is attributed to the technical team and to
innovations in the technology.
Screening of Prospective Reservoirs.
Grand uses a team of professionals to make an initial
evaluation of a prospective reservoir as a candidate for
horizontal waterflooding. This team consists of a geologist,
rock mechanics expert, reservoir engineer, and drilling
engineer.
Geologic studies focus on the vertical and areal
distribution of porosity and permeability. Isopach, structure,
and stratigraphic maps are generated to determine the
suitability of a prospect for waterflooding and the use of
horizontal wells. The reservoir needs to have lateral
continuity, sufficient vertical permeability to permit the use of
horizontal wells, and sufficient oil saturation to permit an
economic application.
A rock mechanics analysis is performed to evaluate the
properties of the rock where the curve and the horizontal
lateral will be drilled. This evaluation will determine if the
rock has adequate strength to maintain hole stability during
drilling and to permit the openhole completion of the lateral. A
cased hole completion or liner may be required if borehole
stability is an issue.
A reservoir engineering evaluation is made to determine if
the prospective reservoir appears suitable based upon prior
experiences.
The drilling engineer makes an assessment of all of the
issues related to the drilling, logging, and completion of the
well. He also evaluates drilling the well underbalanced, to help
minimize formation damage.
Approximately 70% of projects do not pass the initial
screening and those that do move on to a more rigorous
screening process.
Detailed Planning.
The same team of professionals performs the detailed
analysis of those potential projects that passed the initial
screening.
A major activity during this stage is the use of a 3-
dimensional, 3-phase black oil simulator to confirm the
suitability and to assist in the design of the project. A reservoir
description is initially built into the model by using the
previously generated geologic description and by history
matching past performance. Once a history match is achieved,
a number of simulations are performed to evaluate the impact
of different horizontal well design parameters. These include
the placement of horizontal and/or vertical wells, the lengths
4 SPE 99668
of horizontal laterals, spacing between wells, placement within
the vertical interval, and the orientation of wells. Economic
analyses are performed on the proposed pattern to be used in
the project to confirm viability.
A large number of simulations performed under a variety
of conditions have resulted in some general observations.
These include: (1) horizontal injection and producing wells
should be drilled in a toe-to-heel orientation, (2) horizontal
well patterns should be drilled parallel to the prevailing natural
fracture orientation in waterflooding operations, (3) horizontal
producers should be placed near the top of the reservoir where
the oil saturation is the highest, (4) pressure support is needed
to sustain production from a horizontal producing well, (5)
horizontal injectors should be placed lower within the section,
and (6) vertical injection wells are sufficient where good
injectivity is possible.
A detailed well plan is developed to specify the kick off
point, end of curve, trajectory, orientation, and length of the
horizontal wells to be used in the project.
Drilling.
Grand uses a low-cost, rotary-steerable drilling system to
drill the short-radius curves discussed in this paper. This
system was developed and licensed by Amoco Production
Company (now BP). Grand has advanced the technology as a
result of drilling these projects.
All existing wells in a prospective project area are
reviewed for possible use. If the risk-weighted cost of exiting
a candidate well exceeds the cost of a new well, which occurs
frequently in this shallow production region, a new well is
drilled.
When a new well is required, outside contractors are used
to drill and cement casing in the vertical hole to a specified
depth. In some cases, the vertical portion of the well is initially
drilled through the target formation for the purpose of
obtaining additional information about the reservoir. Openhole
well logs and a borehole televiewer can be run to evaluate
formation characteristics, porosity, fluid saturations, and
fracture orientation. In these instances, the well is then
plugged back with cement and the horizontal lateral kicked off
in an orientation parallel to the observed natural fractures.
There are two drilling assemblies used to drill a horizontal
well: the curve drilling assembly (CDA) and the lateral
drilling assembly. A gyroscopic surveying tool is utilized to
orient the CDA, which drills a reliable curve based upon the
tool configuration, typically a 70 ft turning radius. This means
that the wells go from vertical to horizontal, 90 degrees, in 70
ft of true vertical depth (TVD), a total drilled length of
approximately 110 ft. The curve is typically drilled with water
or mud. The CDA is pulled from the hole after the curve has
been completed.
A modified air hammer drilling assembly, which produces
a good penetration rate, is run into the hole to drill the
horizontal lateral to the desired length. An air/foam mixture is
used for circulation to permit underbalanced drilling, thereby
minimizing formation damage. Precautions are taken to avoid
the use of surfactants that tend to emulsify with the oil, which
can also cause formation damage. Surveys are run frequently
to confirm that the wellbore direction and inclination are
within plan. While lengths of up to 1000 ft have been drilled,
experience has shown it is not the length of the lateral but how
many compartments of high oil saturation are encountered that
determine the potential production from the well.
A geologist is on site during the drilling of the well. He
analyzes the cuttings from the well and provides an
interpretation of the lithology, sand quality, and the presence
of oil. Observation of fluids circulated to the pit also provides
valuable information on the contents of the reservoir being
drilled. Attention is given to the first fluids to the surface after
surveying, which acts as a mini drill stem test (DST), as the
reservoir has had approximately one hour to fill the wellbore
with natural fluids.
Formation Evaluation.
Because conventional well logging tools will not pass
through the short radius curve, Grand has developed a
procedure to log short radius horizontal wells. Logging tools
were modified and are conveyed by conventional sucker rods.
Because friction limits the distance to which the logging tools
can be pushed with sucker rods only, roller rod guides are
used, when necessary, to move the logging tools to a longer
distance. This equipment facilitates the acquisition of a suite
of logs, including gamma ray, density, induction, and acoustic
borehole televiewer for fracture identification.
Completion.
Because of the short-radius of the curve, it is possible to
use a conventional downhole pump assembly, set in the
vertical section above the curve, with relatively low pressure
head on the formation. The initial production provides an early
indicator of the amounts of oil, water, and gas that will be
produced from the well. It has been noted that there is a good
correlation between the amount of oil circulated to the pit
during drilling and the stabilized oil production rate that is
ultimately achieved from the well. The curve and horizontal
portions of the well are openhole completed, with no
stimulation required.
Field Applications.
Grand has applied horizontal well technology in the
following reservoir conditions:
• Shallow, low permeability (< 50md) reservoir
• Shallow, low permeability zone (< 50md)
with an underlying high permeability (> 100md) zone
• Shallow, moderately viscous (52 cp) oil reservoir
underlain by water
• Shallow, moderately viscous (70 cp) oil reservoir
SPE 99668
Bartlesville Sandstone Reservoirs.
The Bartlesville sandstone in northeastern Oklahoma has
been one of the most prolific oil producing formations in the
United States. Ye25 reports that 1.5 billion barrels of oil have
been produced from the Bartlesville sandstone through the
1960s. The Bartlesville remains an important producing
horizon, even though it is in a mature stage of depletion. In
spite of the large cumulative production from the Bartlesville,
the recovery efficiency is often less than 15% of the original-
oil-in-place (OOIP). Recovery during the primary recovery
phase is low as a consequence of the low initial pressure and
the inefficient solution gas drive mechanism.
Secondary recovery operations are often not effective or
economic due to the shallow depth, low permeability, and the
natural fractures. The Bartlesville sandstone across northeast
Oklahoma ranges west to east from approximately 3000 ft to a
surface outcrop. The formation typically has a permeability of
less than 50 md, contains natural fractures and/or is found with
an underlying 200-600 md aquifer. The fracture-parting
pressure is low, which makes it difficult to flood the reservoir
at sufficiently high injection rates. Operators usually inject
above the fracture-parting pressure in an attempt to get
sufficient injectivity and the result is often unfavorable, as the
water tends to channel through the fractures and bypass a
significant amount of oil. Fracture stimulation treatments in
injection wells often result in an interconnection with the
lower aquifer while stimulation treatments in producing wells
often lead to excessive water production. Development of
small patterns with closer well spacing can lead to improved
recovery, but the economics are negatively impacted due to
the cost of drilling the additional wells.
Wolco Pilot.
Grand’s first horizontal waterflood project, the Wolco
Pilot, was supported by the DOE and is summarized here. For
additional details refer to earlier technical papers1-3.
The objective of the Wolco project was to demonstrate that
a horizontal waterflood could recover a significant amount of
oil from a portion of the reservoir that had not been previously
flooded. Vertical well logs from the early 1980s indicated a
thick, continuous sand with 50%+ oil saturation, making the
pilot appear suitable. Although the pilot area had not been
previously flooded it was located adjacent to an area that had
been. The pilot was also located close to an area of the
reservoir that contains a high-energy, lower member of the
Bartlesville formation. That zone, referred to as the “D” zone
is characterized by high porosity and permeability in the
several hundred millidarcy range, while the target zone above
it, the “C” zone, has lower porosity and permeability.
5
Thickness Porosity Permeability
Sand
(ft) (% pv) (md)
Upper
40-100 17 25-50
“C” zone
Lower
“D” Zone
0-50 28 800
The porosity log for well 1-A, Fig. 4, clearly shows these
two zones. The “D” zone was not considered to have a major
impact since it was located at a sufficient distance from the
pilot. Also, it was anticipated that the pressure drawdown from
the two horizontal producing wells would adequately control
reservoir fluid movement.
Fig. 5a shows the layout of the original pilot This figure
shows the location of a central horizontal injection well, two
adjacent horizontal producing wells and the location of the
nearby “D” zone. Wells were drilled in a toe-to-heel
orientation, consistent with recommendations made in the
literature22-24. Produced water was disposed of into the high
energy “D” zone through the nearby 1-A vertical well.
Tests on the horizontal injection well provided useful
information. An injection rate of 2000 barrels of water per day
(BWPD) was achieved on vacuum or at a zero surface
injection pressure. However, a step-rate test using a
bottomhole pressure sensor indicated that a slope change in
the pressure/rate chart occurred at 0.35 psi per ft. It appeared
that this represented the fracture-parting pressure being
exceeded at a pressure less than the hydrostatic head of water,
making it necessary to limit the injection rate to below the
critical injection rate at which the apparent parting occurred.
An injection profile survey determined that all of the
injected water was leaving the wellbore within the area of the
curve. The tendency to preferentially flow out of the heel of
the injector is consistent with industry experience23,24.
However, during the drilling of the curve circulation was lost,
suggesting that fractures may have also played a role.
The horizontal waterflood project was successfully drilled
and performed as expected from the standpoint of total fluid
injection and withdrawal. Both the productivity index and the
injection index exceeded original estimates, however, the oil
production response was disappointing. New horizontal well
logs indicated that the principal cause for the poor response
was an oil saturation that was greatly below the original
projections determined by the 1980s logs.
The pilot configuration was revised, as shown in Fig. 5b.
The horizontal injection well was shut-in and each of the
horizontal producing wells was plugged back and new laterals
were drilled in the opposite direction. The horizontal
producing wells were drilled in the top of the formation (Fig.
2c-2) which contained a thin rim of oil. A new pilot was
therefore created, consisting of the two horizontal producing
laterals and the 1-A vertical disposal well completed in the
high permeability “D” zone. Most of the produced water is
6 SPE 99668
entering the “D” zone and providing the pressure support for
the upward movement of oil toward the horizontal producing
wells.
While drilling the new horizontal producers, the drilling
returns together with openhole well logs indicated that the oil
saturation is not uniform throughout the horizontal lateral. The
fluid returns to the pit further indicated that oil and water
saturations can change rapidly as the horizontal lateral is
drilled. These rapid changes suggest that the reservoir is
compartmentalized and much more complex than considered
in the geological and simulation studies. Such observations
reinforce the merits of horizontal wells because of their ability
to penetrate a large portion of the reservoir with a single
wellbore.
The new Wolco pilot has been successful, with producing
rates that have averaged 15 barrels of oil per day (BOPD). An
economic project has been achieved, even with oil saturations
and rates below expectations. The current high oil prices have
been a key factor in this success.
The following lessons were learned from this project.
• Oil can migrate over a period of years due to regional
pressure gradients.
• High injection rates can be achieved in horizontal
injection wells; however, diagnostic testing is needed
to determine the fracture-parting pressure and to
assess the flow distribution around the lateral.
• Because the horizontals are drilled from below the
bottom of the casing, it is relatively inexpensive to
plug back an ineffective lateral and drill a new one
from the vertical wellbore, reinforcing the merits of
using a short-radius drilling system.
• The Bartlesville sandstone reservoir proved to be quite
heterogeneous, with strong evidence of compartment-
alization that tends to isolate parts of the reservoir
resulting in low recovery rates.
• The favorable results being achieved by the current
pilot configuration indicates the viability of a
horizontal well application in a reservoir containing an
oil rim over a high permeability zone.
Avant Field.
This project, located to the east of the Wolco pilot project,
demonstrates how the drilling of horizontal laterals in the
upper portion of a pressured up reservoir containing a high
permeability lower zone can lead to significant oil production.
An isopach map of the “D” sand in the southwest quarter
of section 19 is shown in Fig. 6. For approximately 20 years,
water has been injected into the seven vertical injection wells
at a rate of over 4000 BWPD. At the initiation of this project,
approximately 25 BOPD was being produced at a watercut of
approximately 98.5% from the eight producing wells. In this
area, the high permeability “D” zone has a thickness ranging
from a few feet up to 30 ft while the “C” zone it above ranges
from 40 to 100 ft.
Water has been over-injected into this quarter-section for
years, with an injection-withdrawal ratio of approximately 2.
This practice has led to the pressurization of the reservoir in
the area surrounding the quarter section which was confirmed
by fluid level tests run in some of the peripheral wells.
Because of the extreme contrasts of permeability within the
reservoir, most of the injected water is presumed to have
flowed through the “D” sand, causing much of the oil in the
upper portion of the reservoir to be bypassed. The horizontal
wells were drilled to recover this bypassed oil.
Simulations studies indicated that horizontal producing
wells drilled near the top of the Bartlesville sand would be
highly productive as a result of unswept oil saturation and
good reservoir pressure. Sensitivity cases indicated:
• The horizontal lateral should be placed in the top of
the zone with the toe a reasonable distance from the
vertical injectors.
• High oil recovery can be achieved with multiple
horizontal wells in different locations. However,
interference effects ultimately reduce recovery if too
many horizontals are used or if located too close to
each other.
• Oil recovery is higher with longer laterals due to
greater exposure to the formation and the tendency to
encounter more compartments of higher oil saturation.
The simulations indicate that oil is recovered by the action
of the water pushing the oil upward toward the pressure sink
created by the horizontal lateral. The simulations assumed that
the flow of fluids within the reservoir is controlled by the
matrix, although in reality, fractures may also be influencing
the flow to an unknown degree.
Wells 325 and 328 were originally drilled and completed
in 1914 and were re-entered and drilled horizontally. As
shown in Fig. 6, the two laterals are located close to one
another. They have responded favorably, with a combined
production of approximately 40 BOPD and 600 BWPD,
although the close proximity of the wells will likely result in
interference effects over the long term.
Well 55 has not responded favorably to date. Good oil
saturation in the “C” zone was indicated on the logs; however,
the well produces a large volume of water. Although the
source has not yet been identified, it is speculated that fracture
communication with a water source, possibly an offending
injection well, is the cause.
The following lessons were learned from this project:
SPE 99668
• The favorable results being achieved indicate the
viability of a horizontal well application in a pressured
up reservoir containing a high permeability zone at the
base of the pay. Using vertical injection wells and
horizontal producing wells appears to be effective.
• The Bartlesville sand appears quite heterogeneous, as
indicated by the extreme contrasts in permeability,
variability in thickness, compartmentalization, and the
apparent influence of natural fractures.
Bird Creek Field.
This project demonstrates the application of horizontal
wells to improve recovery in a reservoir where moderately
heavy oil is underlain by water. Water coning frequently
occurs in vertical wells due to the unfavorable mobility ratio
between the oil and water. The use of horizontal wells to
reduce the effects of water coning is one of the applications
discussed by Joshi6.
The Bird Creek Field, discovered in 1904, produces from
both the Bartlesville sandstone and the underlying Tucker
sand. The Tucker sand, located about 70 ft below the
Bartlesville, has been the principal target for the application of
horizontal well technology. The Tucker sand is characterized
by an average porosity of 18%, a permeability of
approximately 300 md, and an oil viscosity of 52 cp at
reservoir temperature. The thickness of the oil column is
controlled by the structure and the oil is underlain by water.
Current reservoir pressure is 115 psi.
Conventional vertical wells produce 2-3 BOPD at a 98
percent water cut. Water coning occurs rapidly, even though
the wells are completed in the oil column, as shown in Fig. 7a.
Conventional waterflooding of such reservoirs is difficult due
to the strong tendency of injected water to flow principally
through the water-bearing portions of the reservoir.
The basic procedure in the Tucker Sand is to drill
horizontal producing wells in the crest of the formation where
the oil column is the thickest, as depicted in Fig. 7b.
Reservoir pressure is maintained by the injection of produced
water within the water column away from the horizontal wells.
A total of seven horizontal wells have been drilled to date.
Some of these were new wells, while the rest were old
Bartlesville wells that had been deepened and used for the
horizontal laterals in the deeper Tucker formation. The laterals
have been in the 300-400 ft range.
Reservoir simulation studies were conducted to determine
the critical design parameters needed for success. These
studies indicated the need to place the lateral as high as
possible within the oil column and to provide pressure
support. The longer horizontals produce oil at higher rates.
In this project, the successful horizontal wells are
producing 15 BOPD at an oil cut of approximately 25%. This
increased oil rate and oil cut are attributed to the horizontal
7
geometry that exposes more of the oil column to the wellbore
and also reduces the tendency for coning.
It has been difficult to define the geology in portions of the
reservoir where there have been few penetrations of the sand.
Because of this, some of the horizontal wells were drilled too
close to the water/oil contact and have not been successful.
The following lessons were learned from the project.
• Recovery can be increased by the use of horizontal
wells in reservoirs containing moderately heavy oil
underlain by water.
• A good understanding of reservoir geology is critical
to insure that horizontal wells are properly placed
within the reservoir.
West Savonburg Field.
Another horizontal well project has been initiated in the
Manson Lease, West Savonburg Field, Allen County, Kansas
as a joint venture between Grand and Verde Oil Company.
The project is being conducted in a Pennsylvanian sand
known locally as the Tucker reservoir, located at depth of
approximately 800 ft. The reservoir has an average thickness
of 25 ft, an average porosity of 21%, and an average
permeability of 300 md. The oil viscosity is 70 cp at reservoir
temperature. Primary recovery from the field with wells on 5-
acre spacing was approximately 5.5% of the OOIP. The low
oil recovery is attributed to the shallow depth and
corresponding low pressure, the high oil viscosity, and an
inefficient solution gas drive mechanism.
A polymer flood was conducted in 2000-03 on an 80 acre
tract in the Manson Lease. The rationale for using polymers
was to increase oil recovery by improving the mobility ratio.
The polymer flood project produced an oil recovery of
approximately 3.3% of the OOIP, which was below
expectations.
A horizontal waterflood project was considered on the
basis of the potential for improved sweep efficiency (Fig. 3)
and the successful field demonstrations with moderately
viscous oil reservoirs in Grand’s Bird Creek Field, the North
Sea11 and the North Slope15. In the West Savonburg Field, the
moderate oil viscosity, high oil saturation, and pressured-up
condition were positive contributing factors in deciding to
evaluate the application of a horizontal waterflood.
Simulation studies indicated that a 5-well project would be
effective in increasing the producing rates and recovery from
the field. There were concerns about the large amount of water
and polymer already injected in the lease, but the relatively
low oil recovery indicated that considerable oil saturation still
existed within the reservoir. Fig. 8 provides a conceptual
layout of the wells, consisting of two relatively short
horizontal injection wells and three horizontal producing
wells. The plans to drill the wells in a TTH orientation are
consistent with industry recommendations22-24.
8 SPE 99668

DOE = Department of Energy

The initial well was drilled to a length of approximately WAG = water-alternating-gas
350 ft. A pressured up condition was encountered during the
CO2 = carbon dioxide
drilling, with the well flowing at the rate of 100 barrels of
fluid per day. A fracture was intersected during the course of TTH = Toe-to-Heel
drilling, which initially produced oil at high rates but later CDA = curve drilling assembly
turned to water. It appears that the fracture is connected to OOIP = original oil in place
nearby injection wells.

The initial production data indicates an oil rate of 10

BOPD and a water rate of 100 BWPD. The project is in the
Metric Conversion Factors
early phases of evaluation. Additional wells will be drilled
contingent upon the longer-term response of the initial well.
bbl X 1.589 873 E – 01 = m3
ft X 3.048 E – 01 =m
inch X 2.54 E + 00 = cm
Conclusions
md X 9.869 233 E – 04 = m2
Successful field demonstrations of horizontal well technology psi X 6.894 757 E + 00 = kPa
have been conducted, including two projects in a light oil
reservoir containing an underlying high permeability zone and
one project in a reservoir containing moderately heavy oil
underlain by water. A fourth project in a moderately heavy oil References
reservoir is currently under evaluation.
1. Westermark, R. V.; Dauben, D.; Robinowitz, S.: “Enhanced
The two major factors contributing to success include (1) Oil Recovery with Horizontal Waterflooding, Osage County,
reservoir evaluation studies to determine the suitability of a Oklahoma”, SPE 89373, presented at the 2004 SPE/DOE
Fourteenth Symposium on Improved Oil Recovery, Tulsa,
project and (2) a reliable, low cost method for the drilling of
OK, April 17-21, 2004.
short-radius horizontal wells.
2. Westermark, R. V.; Schmeling, J.; Dauben, D.; Robinowitz,
Numerous reservoir simulations and field experience S.; Weyland H. V.: “Increased Production Results From Pilot
indicate that: Horizontal Waterflood in Osage County, Oklahoma”, SPE
94094. presented at the 2005 SPE Production and Operations
• In a waterflood operation, horizontal wells should be Symposium, Oklahoma City, Oklahoma, April 17-19, 2005.
aligned parallel to the principal fracture orientation to
maximize the amount of oil that can be displaced 3. Westermark, R. V.; Robinowitz, S.; Weyland, H. V.:
through the matrix. “Horizontal Waterflooding Increases Injectivity and
Accelerates Recovery”, World Oil V 225, No 3, pp 81-82,
• Horizontal producing wells should be placed in the March 2004.
portion of the sand where the oil saturation is the
highest. 4. Taber, J. J.; Seright, R. H.: “Horizontal Injection and
Production Wells for EOR or Waterflooding”, SPE 23952,
• Pressure support is needed to sustain the production presented at the 1992 SPE Permian Basin Oil and Gas
Recovery Conference, Midland, Texas, March 18-20, 1992.
from a horizontal producing well.
5. Joshi, S. D.: “Horizontal Well Technology”, published by
•Horizontal injection wells should be drilled in a toe- PennWell, Tulsa, OK, 1991.
to-heel orientation with horizontal producing wells to
counter the tendency of injected fluids to flow from 6. Joshi, S. D.: “Cost/Benefits of Horizontal Wells”, SPE
the heel of the injector toward the heel of the 83621, presented at the SPE Western Regional/AAPG
producer. Vertical or shorter length horizontal Pacific Section Joint Meeting held in Long Beach,
injectors should be located toward the toe of the California, May 19-24, 2003.
horizontal producer.
Nomenclature 7. Huang, W. S.; Kaetzer, T. R.; Bowlin, K. R.: “Design and
Performance of a Horizontal Well Waterflood Project in New
md = millidarcy Hope Shallow Unit, Franklin County, Texas”, SPE No
24949, presented at the 67th Annual SPE Technical
BOPD = barrels of oil per day [m3/d oil] Conference, 1992, Washington, D. C.
BWPD = barrels of water per day [m3/d water]
8. Huang, W. S.; Kaetzer, T. R.; Bowlin, K. R.: “Case History
psi = pounds/square inch
of a Horizontal Well Waterflood Project – New Hope
cp = viscosity Shallow Unit, Texas – An Update”, presented at the
SPE = Society of Petroleum Engineers Canadian SPE/CIM/CANMET International Conference on
SPE 99668 9

Recent Advances in Horizontal Well Applications, Calgary, Bay”, SPE 39628, presented at the SPE/DOE Improved Oil
Alberta, Canada, March 20-23, 1994. Recovery Symposium. Tulsa Oklahoma, April 19-22, 1998.

9. Hall, S. D. “Multi-Lateral Horizontal Wells Optimizing a 5-
Spot Waterflood”, SPE 35210, SPE Permian Basin Oil and
Gas Recovery Conference, Midland, Texas, 1996, pp 535-
543.
10. Hall, Scott D.: “Multilaterals Convert 5-Spot Line Drive
Waterflood in SE Utah”, SPE 48869, presented at the 1998
SPE International Conference and Exhibition in China,
Beijing, China, November 2-6, 1998.
11. Lach, J. R.: “Captain Field Reservoir Development Planning
and Horizontal Well Performance”, OTC 8508, 1997
Offshore Technology Conference, Houston, Texas, May 5-6,
1997.
12. Ferreira, Horacio; Mamora, Daulat D.; Startzman, R.
A.:”Simulation Compares 5-Spot to Horizontal Wells for
Waterflood”, Petroleum Engineer, November 1998, pp 43-
50.
13. Kelkar, M. C.; Liner, C.; Kerr, D.: “Integrated Approach
Towards the Application of Horizontal Wells to Improve
Waterflooding Performance”, Final Report, DOE/BC/14951-
16, October 1999.
21. Vicente, R.; Sarica, C.; Ertekin, T.: “A Numerical Model
Coupling Reservoir and Horizontal Well Flow Dynamics:
Transient Behavior of Single-Phase Liquid and Gas Flow”,
SPE/CIM 65508, presented at the 2000 SPE/CIM
International Conference on Horizontal Well Technology,
Calgary, Alberta, Canada November 6-8, November 2000.
22. Popa, C. G.; Chipea, M.: “Improved Waterflooding
Efficiency by Horizontal Wells”, SPE 50400, presented at
the 1998 SPE International Conference on Horizontal Well
Technology, Calgary, Alberta, Canada, November 1-4, 1998.
23. Popa, Constantin; Turta, Alex T: “Waterflooding by
Horizontal Injectors and Producers”, SPE 78989, presented
at the 2002 SPE International Thermal Operations and Heavy
Oil Symposium and International and Horizontal Well
Technology Conference, Calgary, Alberta, November 4-7,
2002.
24. Zhao, Litong; Turta, Alex: “Toe-To-Heel Waterflooding:
Field Scale Numerical Simulation Study”, SPE 89380,
presented at the 2004 SPE/DOE Fourteen Symposium on
Improved Oil Recovery, Tulsa, Oklahoma, April 17-21,
2004.

14. Mohanty, Kishore K.: “Development of Shallow Viscous Oil 25. Ye, Liangmiao; Kerr, Dennis: “Sequence Stratigraphy of the
Reserves in North Slope, Final Report, DE-FC26- Middle Pennsylvanian Bartlesville Sandstone, Northeastern
01BC15186, December 2004. Oklahoma: A Case of an Underfilled Incised Valley”, AAPG
Bulletin, Vol. 84, No. 8, August 2000, pp 1185-1204.
15. McGuire, P. L.; Redman, R. S.; Jhaveri, B. S.; Yancey, E.;
Ning, S. X.: “Viscosity Reduction WAG: An Effective EOR
Process for North Slope Viscous Oils”, SPE 93914,
presented at the 2005 SPE Western Regional Meeting, Irvine,
CA, March 30-April 1, 2005.

16. Collier, F. B.; Sharma, A. K.; Stewart T. L.; Pickett, S. L.:

“Application of Horizontal Wells in Waterflood
Optimization at the North Central Levelland Unit in Cochran
and Hockley Counties, Texas”, 42nd Annual Southwestern
Petroleum Short Course Association, April 19-20, 1995, pp
293-306.

17. Dollens, Kimberly B.; Larson, Dana C.; Anthony, Terry L.;
Howell, David A.; Jones, Tim A.: “The Redevelopment of a
Mature Waterflood Through the Application of Horizontal
Multilateral Drilling Technology in Preparation for Tertiary
CO2 WAG Operations”, May 15-18, 1999.

18. Hancock, Susan K.: “Project Brings Commercial-Scale CO2

Miscible Flooding to Canada”, Oil and Gas Journal, October
18, 1999, pp 70-75

19. Wier, D. R.; Chimahusky, J. S.; Czirr, K. L.; Hallenbeck, L.;

Gerard, M. G.; Dollens, K. B.; Owen, R.; Gaddis, M.;
Moshell, M. K. : “Design and Implementation of a CO2
Flood Utilizing Advanced Reservoir Characterization and
Horizontal Injection Wells in a Shallow Shelf Carbonate
Approaching Waterflood Depletion”, DOE/BC/14991-23,
Final Report, July 2, 2001.

20. McGuire, P. L.; Redman, R. S.; Mathews, W. L.; Carhart, S.

R.: “Unconventional Miscible EOR Experience at Prudhoe
10 SPE 99668

Table 1

Applications of Horizontal Wells in Waterflooding and

Enhanced Oil Recovery Projects

Depth Perm. API

Field Description Application
(ft) (md) Gravity

New Hope Improved economics and recovery; one

Waterflood in deep, low
Shallow Unit, 7965 2 46 horizontal well can replace several vertical
7,8 permeability reservoirs
Texas wells.

Improved economics and recovery; one

Aneth Field, Waterflood in deep, low
9,10 5400 0.2-5 38 horizontal well can replace several vertical
Utah permeability reservoirs
wells

Acceleration and increase of oil recovery;

Captain Field, Waterflood in moderately
11 2800 8500 19.3 higher injection rates and improved areal
North Sea viscous oil reservoirs
sweep efficiency through mobility control.

Increase of oil recovery; improvement of areal

Immiscible WAG in
West Sak, sweep efficiency by a viscosity reduction WAG
15 3900 50-400 16-23 moderately viscous oil
Alaska process using horizontal injectors and
reservoirs
producers.

Levelland Unit,
Increase injectivity and accelerate oil recovery. Miscible WAG in tight, light
Slaughter Field, 4750 1.8 31
16 Reduce need for infill drilling. oil reservoirs
Texas

Aneth Field, Increase of oil recovery; increased injection Miscible WAG in tight, light
17 5400 0.2-5 38
Utah rates in miscible CO2 WAG process. oil reservoirs

Weyburn Field, Maximize recovery in CO2 project using various Miscible WAG in tight, light
18 4655 10 28
Canada combinations of vertical and horizontal wells. oil reservoirs

South Cowden
Accelerate recovery in CO2 project; horizontal Miscible WAG in tight, light
Sand Andres 4550 3.7 36
19 injection wells to increase injection rates oil reservoirs
Unit, Texas

Maximize recovery in an enriched gas drive

Prudhoe Bay, Miscible WAG in light oil
20 8800 400 27 project by using a combination of vertical and
Alaska reservoirs
horizontal wells.
SPE 99668 11

Figure 1: Location of Projects

Wolco Avant Field Bird Creek Field West Savonburg Field

Figure 2: Various Well Configurations Used in Horizontal Waterflooding

12 SPE 99668

Figure 3: Comparison of Breakthrough Volumetric Sweep Efficiency for 5-Spot and Horizontal Well Patterns

Figure 4: Density Log Showing Extreme Porosity Contrasts Between “C” and “D” Zones
SPE 99668 13

Figure 5a: Well Configuration in Original Wolco Pilot Figure 5b: Well Configuration in Modified Wolco Pilot

“D” Isopach

Figure 6: Isopach Map of the “D” Sand in the Avant Field Showing the Location of the Injectors, Producers and Horizontal Wells
14 SPE 99668

Figure 7a: Conventional Vertical Well Completion Depicting Water Coning

Figure 7b: Horizontal Well Completion Depicting Reduced Tendencies for Water Coning

Figure 8: Conceptual Layout of Horizontal Wells for the West Savonburg Field