You are on page 1of 16

SPE

SPE 22389
Reservoir Evaluation of Horizontal Bakken Well Performance on
the Southwestern Flank of the Williston Basin
M.R. Reisz, Union Texas Petroleum
SPE Member
Copyright 1992, Society of Petroleum Engineers, Inc.
This paper was prepared for presentation at the SPE International Meeting on Petroleum Engineering held in Beijing, China, 24-27 March 1992.
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 mate,rial, as presented, does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Papers pressnted at SPE meetings are subject to publication review by Editorial Committees of the Society
of Petroleum Engineers. Permission to copy Is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstmct shOUld contain conspicuous acknowledgment
of where and by whom the paper is presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836 U.S.A. Telex, 730989 SPEDAL.
ABSTRACT
This paper presents the results of a
reservoi r performance study for hori zonta1
and vertical Bakken wells in the "Fairway" of
the Williston Basin. A combination of
forecasting methods were utilized, including
decline curve analysis, material balance,
analytical solutions, and empirical
correlations.
The results indicate that recoverable
reserves from hori zonta1 well sin the
"Fairway" are 2.5 to 3.0 times a vertical
well for 1.5 to 2.0 times the cost .. Simple
theoretical calculations combined with field
data are useful in making initial estimates
of original oil in place, recoverable
reserves, and drainage area.
INTRODUCTION
The Bakken formation, located in northwestern
North Dakota and northeastern Montana
(Figure 1), is the source of a large portion
of the oil generated and produced in the
Williston Basin. The Mississippian Devonian
age Bakken shale is a naturally fractured
formation that is both source and reservoir
rock. The Bakken consists of three members.
References and illustrations at end of paper.
9
The upper and lower members are black shales,
and the middle member is a dolomitic shaley
si 1tstone. Located above the Bakken is the
Lodgepole (dense lime), and below is the
Three Forks sand. The overpressured Bakken,
found at approximately 10,000' [3048 m] with
a virgin reservoir pressure corresponding to
a 0.6 to 0.7 psi/ft gradient [13.6 - 15.8
kPa/m], has g n r t ~ over 100 billion
barrels [15.9 E+09 m] of oil based on
industry estimates.
The key to Bakk,en product i on, in the mature
portion of the Williston Basin, is having a
permeable interval in which to put the oil
after it has been generated. Both the
Lodgepole and Three Forks have potential oil
storage capacity. The Antelope field, located
on the Nesson Anticline in northeastern
McKenzie County, is a good example of
increasing the available storage capacity
with the presence of a permeable sand below
the Bakken.
Figure 2 shows the location of the study area
in Billings and McKenzie counties along the
Billings Nose Trend. This area is referred
to as the "Fainilay". In the "Fairway" only
the upper Bakken is considered net pay, and
future discuss i on will refer to the upper
Bakken as Bakken.
2 RESERVOIR EVALUATION OF HORIZONTAL BAKKEN WELL PERFORMANCE SPE 22389
ORIGINAL OIL IN PLACE
For si ngl e phase flow another form of the
material balance equation is
Calculation of the original Bakken oil in
place is a non-unique answer, due to the
number of unknowns in a fractured reservoir.
In a conventional homogenous reservoir,
original oil in place can be determined
volumetrically from the relationship 1,2
... (1)
.... (2)
.... (3)
N= NpBo / Bo - Boi
N 7758A he (l-Sw)
Bo
In the Bakken, there is uncertainty in both
area (A) and thickness (h). Therefore,
another approach is needed to arrive at
satisfactory answers. The obvious method is
a form of the material balance equation,
which assumes that reservoir voidage caused
by production of reservoir fluids is equal to
expansion of reservoir fluids due to a drop
in pressure. By definition original oil in
place (N) can be represented in simplest form
by the equation 1,3
CUM OIL
Qt=Qi[l- (OOIP) (RF) ]
This equation 4 is helpful in reservoirs with
single phase flow where little is known about
the reservoir properties. By utilizing the
relationship in equation 3 and solving for
OOIP, one gets
OBJECTIVES
Several key points are supported by
performance data: an average initial decline
of 40-45%, a final decl ine' of 25-35%, a
breakeven point of approximately 150 MBO
[23.8 E+03 m
3
] at NPV(15) = 0, recoverable
reserves 2.5-3.0 times a vertical well, and
20-25% of recoverable reserves produced in
first year.
Since 1ate 1987 the Bakken pl ay has been
domi nated by' hori zonta1 dri ll-j ng. There are
approximately 140 horizontal wells that have
produced oil from the Bakken formation, and
an equal number of vertical wells.
The real challenge, in an area of horizontal
development, is to make an early and accurate
eva1uat i on of performance characteri sti cs,
recoverable reserves, and drainage area.
Management needs thi s data to make a
determi nat i on of the economi c: vi abil ity for
horizontal development in a play such as the
Bakken. The factors which control Bakken
production were studied, and key reservoir
parameters were identified for early analysis
of recoverable reserves. The range of
ultimate recoveries for Bakken drainholes
indicate that no single factor can accurately
predi ct future performance. Results from
decl ine curve analysis were compared with
analytical solutions and material balance to
gain a higher confidence level in the range
of estimated values.
The main goals of this work included:
(4)
1. Estimation of original oil in place,
recoverab1e reserves, and drai nage areas.
2. Early evaluation of plays through the
identification of key parameters
affecting well performance.
A total of 21 wells from 7 fields in the
"Fairway" were studied in deta.il. This group
of wells are referred to as Group "A". The
selection criteria for the study group
included: a minimum of 1 year production
history, and data availablle on azimuth,
length of drainhole, path of drainhole, and
net pay.
The following sections discuss initial
estimates and results obtained in this study.
OOIP= CUM OIL
(l-Qt/Qi) (RF)
This equation is valid for L/2Xe (penetration
ratio) > 0.5 and Qt/Qi ratio> 0.5 (early
time data). One can now calculate OOIP with
an equation that utilizes rate-time
relationships, and compare the results with
other methods. One can also set equation 4
equal to equation 1 and solve for drainage
area (A). This equation is represented by
Area = CUM OIL (BO)
(I-Qt/Qi) (RF) (0)(h) (I-Sw) (7758)
.... (5)
10
SPE 22389 M. R. REISZ 3
The results obtained from these equations are
in the range 2.0 to 2.7 MMBO [318-429 E+03
m
3
] per square mile [2.6 km
2
]. Table 1
provides OOIP and other pertinent information
on Bakken drainho1es selected from different
fields throughout the "Fairway". Production
data has been updated through February 1991.
The magnitude of fracture volume as a percent
of total oil in place is an important number.
An examination of storage capacity in
fractures indicates that less than 10% of the
total oil in place may be stored in the
fractures of the upper Bakken member. This
number is consistent with empirical
correlations from Nelson 5.
In summary, a number of different approaches
were used to obtain estimates of OOIP,
including volumetric, material balance 3,6 and
Joshi's OOIP equation 4. The range of 2.0 to
2.7 MMBO [318-429 E+03 m
3
] per square mile
[2.6 km
2
] is a reasonable approximation.
RECOVERABLE RESERVES
Vertical Wells
Good agreement has been obtained in the
estimation of recoverable oil from vertical
Bakken wells using both decline curve
ana1ys is, and log of pressure versus
cumulative production plots. Vertical Bakken
reserves average 108 MBO per well [17.2 E+03
m
3
] from a data set of 119 wells in the
"Fairway".
The decline characteristics of the vertical
wells were very predictable. The vast
majority of vertical Bakken wells had decline
rates between 15-17% with a range of 10-20%.
Wells with a 10-12% decline and good
productivity were connected to an effective
fracture system and a large drainage area.
Horizontal Wells
Horizontal drainho1es exhibit a number of
decline characteristics resulting from a
variety of factors. The initial decline is
associated with the effectiveness of the
fracture system and size of the area being
drained. Figures 3, 4, and 5 show individual
performance data from Table 1. These three
wells have estimated recoverable reserves in
11
excess of 400 per well [63.6 E+03 m
3
] ,
and should give insight concerning horizontal
Bakken performance characteristics.
Figure 3 shows Meridian #33-11H in the
Elkhorn Ranch area. This well has a
production of 280 MBO [44.5 E+03
m] through February of 1991 and estimated
recoverable reserves of 442 MBO [70.3 E+03
m
3
]. The initial stabilized rate was 300
BOPO [47.7 m
3
/d] followed by a 55% decline.
The current dec"line rate is 25%. The well in
Figure 4 has excellent production for the
first six months with ini}ia1 production in
excess of 600 BOPO [95.4 m/d]. The #1-7 has
an overall decll ine rate of 56% with a 25%
decline for the! last six months. The well
has a cumulative of 131 MBO [20.8 E+03 m
3
]
and estimated recoverable reserves of 405 MBO
[64.4 E+03 m
3
] Figure 5 shows a consistent
31% decline, a cumulative of 160 MBO [25.4
E+03 m
3
J. ' and an estimated 425 MBO [67.6
E+03 m] of reserves from the #14-27H.
Dec1i ne performance is a funct i on of the
effective areal extent of the fracture system
and the permeability of the fractures
intersected by the we11bore.
Table 1 shows calculated drainage areas
rangi ng from 469 acres [1898 E+03 m
2
] for
#14-27H to 664 acres [2687 E+03m
2
] for
#33-11H. Three wells listed in Table 1 from
the Bicentennial area are estimated to have
recoveries in the 150-200 MBO [23.8-31.8 E+03
m
3
] range with calculated drainage areas of
210-230 acres [fl50-931 E+03 m
2
]
Figure 6 is a plot of rate versus time for
the entire Rough Rider Field. Normal izing
and grouping of production data can yield
predictive characteristics early in the life
of wells. The Rider Field has averaged
286 BOPO [45.5 m/d] from 10 wells in the
first month, and has a decline rate of 25%.
Figure 7 is a normalized plot of the entire
Group "A" wells, and contains performance
trends representative of a Bakken well
located in the "Fairway". Figure 7
illustrates an initial decline of 40-45% from
21 wells in the first two years, followed by
a 25-35% decline. An overall decline rate of
30-35% can be anticipated. Figures 8 and 9
show the distribution of first month
production rate and recoverable reserves.
4 RESERVOIR EVJILUATION OF HORIZONTAL BAKKEN WELL PERFORMANCE SPE 22389
Approximately 70% of Group "A" wells have a
stabili'zed ihitial rate of 200 BOPD [31.8
m
3
/d] or greater, and 48% have recoverable
reserves greater than 200 MBO [31.8 E+03 m
3
].
Group "A" reserves range from 35 MBO to 442
MBO [5.6-70.3 E+q3 m
3
] with an average of 223
MBO [35.5 E+03 m] .
Key Parameters
The range of ultimate recoveries for Bakken
drainholes dictates that a list of key
parameters be identified for early
evaluation. Unfortunately for horizontal
drainhole analysis, there is no single factor
that can accurately predict future
performance. Tables 2A-2C are key criteria
charts. Each table consists of a number of
parameters that should be considered in the
early evaluation of ultimate recoveries.
Zonal penetration 7 within the pay, drainhole
path, ori entat ion, product i vi ty, and a
fracture index are all key parameters. These
parameters were reviewed, ranked, and a value
assigned to arrive at a composite success
factor. The ranking scale (A through E)
utilized in this study is shown in Table 2A.
Table 2A shows the same wells listed in
Table 1. These wells have an average
recovery of 261 MBO [41.5 E+03 m
3
] Table
2B and Table 2C provide examples of other
combi nat ions of well s ba:sed upon the
selection criteria chosen in each table.
Meridian #14-27H in the Rough Rider area both
have calculated drainage areas (Table 1) in
the 400-500 acre [1619-2023 E+03 m
2
] range.
Both are expected to have recoveries in
excess of 400 MBO [63.6 E+03 m
3
]. Several
other wells in Table 1 have calculated
r i n ~ e areas in the 200-300 acre [809-1214
E+03 m] range. At this time, it is not
known whether these values are representative
drainage areas or are influenced by the Qt/Qi
ratio in the drainage area formula.
Additional work in this area is planned.
Joshi's paper on "Methods Calculate Area
Orai ned by Hori zonta1 Well s" 8, was very
helpful in calculating the drainage area of
Bakken wells. For optimum recoveries from
horizontal drainholes, industry must have a
good understanding of drainage areas and the
proper spacing of wells. The calculated
drainage areas in Table 1 were obtained using
equation 5.
The physical 3D model for a horizontal
drainhole and its drainage area is defined by
Figure 10 9. The drainhole length (L) is in
the "x" direction. This is the low
permeability direction of the reservoir, and
is perpendicular to the dominant natural
fractures. The sides of the drainage
rectangle are 2Xe and 2Ye. The thickness of
the reservoir (h) is in the "z" direction.
The Xw and Yw values represent the location
of drainhole along the "x" and "y" axes.
The effective permeability in the horizontal
plane of an anisotropic reservoir is equal to
the square root of Kx times Ky. The
rectangular drainage area of a vertical well
can be represented by the equations 8
Tables 2A-2C support and identify a broad
range of ult i mate recoveri es for a Bakken
drainhole. This approach allows for
identification of key parameters, and an
early analysis of successful wells. It also
provi des a method to quant'i fy the factors
that affect horizontal performance, and to
evaluate how critical a particular parameter
is in a given play. Recoverable reserves in
Tables 2A-2C support an ultimate recovery of
2.5 to 3 times a vertical well.
DRAINAGE AREA
Much work remains in determination of
drainage areas for horizontal drainholes.
Different reservoir conditions and geometries
will have an effect on actual drainage areas,
and must be considered in the evaluation.
The Slawson #1-7 in the Ash Coulee area and
12
(2Xe) (2Ye) = Area (43560)
and
2 Ye / 2Xe = ..jXy! Kx
in an anisotropic reservoir 8
can be rewritten as
2Ye = 2Xe..jXy! Kx
.... ( 6)
.... ( 7)
Equation 7
.... (8)
SPE 22389 M.R. REISZ 5
3.
Ah (LzlOOO ft.)
= 2
(Av)
..... (10)
Ah (Lz2000 ft.)
= 3
(Av)
..... (11)
4.
Solving equations 6 and 7 simultaneously, one
obtains the drainage lengths for a vertical
well in an anisotropic reservoir, assuming
that drainage area and permeabil i ty values
are known.
One can calculate 2Xe drainage length for a
horizontal drainhole based upon the following
assumption; 2Xe is equal to the drainhole
length (L) in the "x" direction plus drainage
radius of the vertical well at each end of
the horizontal drainhole 8 This
relationship can be stated as
..... (9)
Figure 11 9 represents an idealized view of
drainage volume for both vertical and
hori zonta1 well s. Drai nage 1ength for a
horizontal drainhole along the high
permeability direction 2Ye (Figure 10) is
assumed to be the same as the vertical well.
Per Joshi 9, the two following relationships
give a comparison of horizontal drainage area
(Ah) to vertical drainage area (Av).
See Appendix A for an example problem
utilizing well and reservoir parameters from
the first horizontal drainhole in the Bakken
formation, Meridian #33-11H. (Note:
Production data and calculations have not
been updated. This example represents
initial results.) A summary of the well
parameters both known and assumed are
contained in Table 3. Reservoir parameters
are summarized in Table 4. In Appendix A
several different methods, including Joshi's
method 8, will be presented and results
compared.
In summary, horizontal drainage area (Ah) is
two to three time Av (vertical drainage
area). This means that a horizontal Bakken
drainhole should achieve a drainage area
between 320 - 480 acres [1295-1942 E+03 m
2
]
where a vertical Bakken well drains 160 acres
[647 E+03 m
2
]. For a vertical Bakken well
that drains 320 acres [1295 E+03 m
2
], the
horizontal well could drain 640 + acres [2590
E+03 m
2
]. Based upon an average vertical
well recoveringr 108 MBO [17.2 E+03 m
3
] and
the increased drainage area from horizontal
wells, a horizontal drainhole should recover
300 MBO [47.7 E+03 m
3
] for a drai nho1e 1ength
of 2000 feet [510 mJ. This is consistent
with other work in the report and offers a
high level of confidence in the ultimate
recovery expected from a hori zonta1 Bakken
drainhole.
PRODUCTIVITY PROBLEMS
A revi ew of performance data 10,11,12 and key
parameters in the planni ng and drill i ng of
Bakken horizontal drainholes reveals four
major reasons fl:>r 1ess than ideal results:
1. Formation damage
The inability to effectively remove
format i on damage wi 11 become worse
as the reservoir pressure decreases.
2. Overdrilling
Current well spacing in the
"Fairway" appears too dense in some
areas for horizontal drainholes.
Orientation
Ideally the drainhole is parallel to
the 10".' permeabil i ty direct i on of
the resl!rvoi r.
Porpois i ~
Problems at the well site can prevent
zonal penetration in the desired
horizontal segment of the drainhole.
For the majority of the prospective Bakken
acreage in the Williston Basin, it may not be
enough to encounter fractures. Initially the
overpressuring of the Bakken enhances the
effect i veness I:>f the fracture network.
However, as the! pressure in the reservoi r
(fractures) decreases, the effectiveness of
the fractures may also decrease. Therefore,
it is desirable to prop the fractures around
the wellbore open for optimum long-term
productivity.
Format ion damagl! is a major concern in the
Bakken. The invaded zone around the well bore
may reduce well productivity. Thus, a key
benefit in the drilling of horizontal
drainholes may be lost or reduced. This
condit i on wi 11 be more prevalent when
excessive mud weights are used to control
13
6 RESERVOIR EVJlILUATION OF HORIZONTAL BAKKEN WELL PERFORMANCE SPE 22389
problems of hole stability, or when a portion
of is partially depleted.
Undoubtedly, fracture collapse and near
well bore formation damage have occurred and
prevented some of the Bakken horizontal
drainholes from achieving their productive
potential.
ROCK AND FLUID PROPERTIES
Arange of rock and fluid property values for
the Bakken have been identified to assist in
this study and future analytical studies.
Although there are large amounts of data
available concerning rock and fluid
properties in the Bakken, there does not seem
to be a consensus about the critical
parameters. Core data, DST's, reservoir
fluid studies, pressure build-ups, etc. have
been obtained by numerous operators, but
never incorporated into a c:ommon database
available to the general industry. In 1990,
a group of companies began a cooperative
effort to gather Bakken data.
Tables 5 and 6 13 display a l"ange of values
for different parameters that were utilized
in this study. Core analysis on Union Texas
Petroleum's Federal #12-1 located in Sec. 12,
T144N-R102W, indicates an average porosity of
3%. Permeabilities are in the 0.02 - 0.05 md.
range in the Bakken and 0.25 md. in the
Lodgepole. The maximum permeability of 0.59
md. was recorded from a fractured sample.
This well has an EUR of 101 MBO [16 E+03 m
3
]
and is representat i ve of a typi cal Bakken
well.
RECOVERY FACTORS
Documented numbers are not available for
recovery factors in horizontal Bakken wells.
Joshi indicates a 16% recovery factor is a
good approximation and is consistent with the
15 to 20% range listed in Table 5. These
numbers were obtai ned from commi ss ion heari ng
exhibits. Joshi has indicated that increased
recovery factors from drainholes may be
2 - 5% higher than for vertic:al wells 8,9.
14
FUTURE WORK
Our understanding of fractured reservoirs
will increase dramatically with continued
application of horizontal drainhole
technology. Additional study of
rel ationships and plotting parameters will
improve early answers when limited data is
available.
Two major concerns in the Bakken shale are
the 1ocat i on and product i vi ty of unpropped
natural fractures. It is necessary to have
alternatives for wells that do not perform as
planned. Future work may include:
1. Additional refinements in both
completion and stimulation techniques.
A "wellbore mini frac".
2. Better definition of fracture areas.
3. Solutions to hole stability.
4. Improved analytical models.
5. Improved methods for early analysis.
CONCLUSIONS
1. Drainage area and recoverable reserves
for a horizontal well are 2.5 to 3.0
times that of a vertical well.
2. Original oil in place in the "Fairway"
ranges from 2.0 MMBO to 2.7 MMBO
429 E+03 m
3
] per square mile [2.6 km].
3. Recoverable reserves of 200-250 MBO
[32-40 E+03 m
3
] with better wells
recovering 300-500 MBO [48-79 E+03 m
3
].
4. Fracture network effect i veness determi nes
product i vi ty and dec1i ne characteri st i cs.
5. Average initial decline is 40-45% with a
final decline of 25-35%.
6. Bakken drainholes, usually recover 20-25%
of their reserves in the first year.
7. Formation damage ,can have significant
impact on productivity.
8. Effective H (thickness) of the reservoir
is larger than the actual thickness (h)
of the upper Bakken shale.
SPE 22389 M.R. REISZ 7
NOMENCLATURE Subscripts
ACKNOWLEDGMENTS
REFERENCES
The author wishes to thank the management of
Union Texas Petroleum for their support in
the publication of this study. The valuable
input and insights from D.T. Boyd and V.K.
Bangia are greatly appreciated. C.D. Crowe
and S. Housley did a great job in preparation
of manuscript and tables.
e = external
f = fracture
h = horizonta'i
n = net drainhole length
p = produced oil
v = vertical
W= wellbore
1. Frick, T.C., Petroleum Production
Handbook, P = 37-10 thru 37-12, Millet
The Printer, Inc., Dallas, Tx., 1962.
2. Agulleria, R., "The Uncertainty of
Evaluating Original Oil-In-Place in
Naturally Fractured Reservoirs", SPWLA
19th Annual Logging Symposium, June 13-
16, 1978.
3. ORYX, North Dakota Industrial Commission
Hearing, Case No. 5030, Exhibit 6.
4. Joshi, S.D., "Production Forecasting
Methods for Horizontal Wells", SPE 17580,
SPE International Mtg. Tianjin, China,
Nov. 1-4, 1988.
5. Nelson, R.ll., Geologic Analysis of
Natura11 y Fractured Reservoi rs, Chapt. 1,
P = 72-102
t
Gulf Publishing Company,
Houston, Tx., 1985.
6. King, G.R., "Material Balance Techniques
for Coal Seam and Devon ian Shale Gas
Reservoirs", SPE 20730, Annual Mtg. in
New Orleans, La., Sept. 23-26, 1990.
7. Williams Petroleum Consulting and
Minerals Diversified Services, Inc.,
"Williston Basin Bakken Formation
Studies, Part I and II", Industry Report
in 1990.
8. Joshi, S.D., "Methods Calculate Area
Orai ned By Hori zonta1 Well s", Oi 1 and Gas
Journal (Sept. 17, 1990), P = 77-82.
9. Joshi, S.D., "Reservoir Aspects of
Horizontal Wells", SPE Short Course at
the
initial reservoir pressure, psia
[kPa]
bubble point pressure, psia
initial production rate, BOPD [mid]
= production rate at time (t),
BOPD[m
3
/d]
= drainage radius, ft. [m]
= drainage diameter, ft. [m]
= recovery factor, %
= initial oil saturation, %
= water saturation, %
= viscosity, cpo [Pa s]
= half the side of a drainage area
which is parallel to the horizontal
well, ft. [m]
= fracture half length, ft. [m]
= distance from drainage boundary to
the center of the drainhole, ft. [m]
= drainage length of rectangle
(diameter), ft. [m]
= half the side of the drainage area
which is perpendicular to the
horizontal well, ft. [m]
= distance from the x axis drainage
boundary to the drai nho1e, ft. [m]
= side of drainage rectangle in
horizontal plane, ft. [m]
= porosity, %
a
A
b
Bo
c
o
e
EUR
h
H
Pb
Qi
Qt
= half the major axis of drainage
ellipse, ft. [m]
= drainage area, acres [m
2
]
= half the minor axis of the drainage
ell ipse, ft. [m]
Boi = initial formation volume factor,
[dimensionless]
= formation volume factor at bubble
point, [dimensionless]
= compressibility, I/psi [I/Pa]
= fracture spacing, ft. [m]
= natural fracture width, [cm.]
= ultimate recoverable oil,
bbls. [m]
= reservoir thickness, ft. [m]
= effective thickness of
reservoir, ft. [m]
H.D. = horizontal displacement, ft. [m]
k = permeability, md.
Kh = horizontal permeability, md.
L = horizontal drainhole length, ft. [m]
L/2Xe = penetration ratio, [dimensionless]
N, = original oil in place, bbls. [m
3
]
OOIP
Pi
Ye
Yw
xf
Xw
r
2rev
RF
Soi
Sw
u
x
2Xe
2Ye
15
8 RESERVOIR EVtlLUATION OF HORIZONTAL BAKKEN WELL PERFORMANCE SPE 22389
Annual Mtg. in New Orleans, P = 46, Sept.
20; 1990:
10. Mukherjee, H., "A Parametric Comparison
of Hori zonta1 and 'Vert i cal Well
Performance", SPE 18303, Annual Mtg. in
Houston, Tx., Oct. 2-5, 1988.
11. Karcher, B.J., "Some Practical Formulas
To Predict Horizontal Well Behavior", SPE
15430, Annual Mtg. in New Orleans, La.,
Oct. 5-8, 1986.
12. Joshi, S.D., "Augmentation of Well
Product i vity With Sl ant and Hori zonta1
Wells", Journal of Petroleum Technology
(June 1988), P = 729-739.
13. Williams, LT., "Pressure Transient
Analysis of Horizontal Wells in a
Naturally Fractured Reservoir", SPE
20612, Annual Mtg. in New Orleans, Sept.
23-26, 1990.
APPENDIX A
EXAMPLE PROBLEM: Drainage Area of the
Meridian #33-11H located
in Sec. 11-143N-102W
(Elkhorn Area)
vertical well (320 acre [1295 E+03 m
2
]
drainage case) is 2Xe = 3474'[1059 m] and 2Ye
= 4012' [1223 m].
For a horizontal drainhole 2Xe can be
calculated from equation 9
..... A-4
2Xe = 2082' + 2 (2106') = 6294 ft.
The rectangular anisotropic drainage lengths
for #33-11H is calculated as 2Xe = 6294'
[1918 m] and 2Ye = 4012' [1223 m]. The
area is equal to 580 acres [2347
E+03 m]. This number is consistent with the
drainage area of 646 acres [2614 E+03 m
2
],
which was calculated from equation 5
(material balance).
METHOD 2:
The drainage area of a conventional well in
an isotropic reservoir can be obtained by the
equation
METHOD 1:
Solving equations A-I and A-2 simultaneously,
we obtain
The calculation of the drainage area for a
horizontal drainhole in an anisotropic
reservoir, such as the Bakken, can be
obtained. Beginning with equations 6 and 7
and solving for a 320 acre [1295 E+03 m
2
]
vertical drainage case (assumed)
..... (A-5)
..... (A-6)
Area = 4Xe
2

Area = EUR (Bo) (5.615)
(RF) (h) (l-Sw)
From equation 6, the Area would be equal to
2Xe times 2Ye for a rectangular drainage
area. From equation 8, 2Ye can be written in
terms of 2Xe for an anisotropic reservoir.
Therefore, the area of a rectangul ar
anisotropic drainage area can also be
represented by the equation
..... (A-I)
..... (A-2)
2 Ye /2Xe = .jXy!Kx
2Ye / 2Xe = 1.155
(2Xe) (2Ye) = 320 x 43560
(1/2Xe) (2Xe) (2Ye) (2Ye)
= 320 x 43560 x 1.155
4Ye
2
= 16.1 x 10
6
2Ye = 4012'
Substituting equation A-6 into equation A-5
and solving, drainage length (2Xe) is
..... (A-3) obtained for a well in an anisotropic
reservoir where the drainage area is not
known.
Substituting 4012' [1223 m] into equation
A-2, 2Xe = 3474' [1059 m]. The rectangular
anisotropic drainage lengths for this
4Xe2 KKx Y = EUR (Eo) (5.615)
(RF) (Heff) (0) (l-SW)
..... (A-7)
16
SPE 22389 M.R. REISZ 9
From Table 3 and 4,
4xe
2
(1.155) = (427,000) (1.5) (5.615)
(.2) (20) (.04) (.85)
Xe = 2392'
2Xe = 4784'
Substituting 4784' [1458 m] into equation
A-2, 2Ye = 5526' [1684 m]. The drainage area
is equal to 607 acres [2456 E+03 m
2
]. This
number is consistent with the drainage area
of 646 acres [2614 E+03 m
2
], which was
calculated from equation 5.
METHOD 3:
The average of the area of an ellipse and the
area of a rectangle plus a circle will also
give a consistent answer (669 acres [2707
E+03 m
2
]).
The area of a rectangle plus a circle is
equal to
(2Xe) (2Ye) + pie (r
2
ev
) / 43560
= (6294') (4012') + pie (2106)2 / 43560
= 859 acres
The average of 478 acres [1934 E+03
m
2
](ellipse) and 859 acres [3476 E+03
m
2
](rectangl e pl us ci rcl e) is equal to 669
acres [2707 E+03 m
2
].
Area of Ellipse = pie (a) (b)
43560
a = L
n
/ 2 + rev
1/2 of major axis
1041' + 2106' = 3147'
..... (A-8)
..... (A-9)
1/2 of minor axis
..... (A-I0)
b = 2106'
Area of Ellipse = pie (3147') (2106') / 43560
= 478 acres
17
...
Q)
TABLE 1
BAKKEN RESERVE VARIABLES FROM FIELDS IN "FAIRWAY"
CALCULATED
--- --
DATE DRAINAGE CUM OOIP
LOCATION 1ST Qi Qt CUM OIL OOIP AREA EUR OIL 2
S-T-R FIELD WELL# PROD (BOPD) (BOPD) (MHO) (MHO) (ACRES) (MHO) (%EUR) (MBO/MI)
1. 11-143-102 ELKHORN 33-1lH 9-87 333 133 280 2738 664 442 63.3 2639
2. 07-142-101 ASH COULEE 1-7 2-90 636 221 131 1782 572 405 32.7 1994
3. 05-142-102 ROOSEVELT 23-5H 5-90 176 51 18 368 200 93 20 1178
4. 29-143-101 ELKHORN 34-29H 8-88 267 90 137 1242 293 262 52.3 2713
5. 33-143-101 ELKHORN 12-33H 4-89 441 139 167 1364 347 337 49.6 2516
6. 27-145-101 ROUGH RIDER 14-27H 9-89 412 216 157 1735 469 425 37.6 2368
7. 33-145-101 ROUGH RIDER 21-33H 2-89 218 60 111 1003 217 181 61.3 2958
8. 36-145-102 BUCKHORN 36-44H2 8-89 420 89 93 761 167 200 46.5 2916
9. 19-145-103 BICENTENNIAL 33-19H 6-88 485 22 ISS 822 230 183 84.7 2287
10. 29-145-103 BICENTENNIAL 31-29H 4-89 369 41 99 633 159 145 68.3 2548
11. 29-145-103 BICENTENNIAL 33-29H 10-89 291 89 72 S07 147 196 36.7 2207
NOTE: RECOVERY FACTOR =20%
TABLE2B
TABLE 2A
Parameters For Early Determination of Ultimate Recoveries Parameters For Early Determination of Ultimate Recoveries
KEY CRITERIA: HORIZONTAL DRAINHOLES IN TABLE 1
__ ___..... =-=== = ====... _ ._""_== ===== ===_: ......c ....
KEY CRITERIA: HORIZONTAL DRAINHOLES WITH WELL PATH RANKING OF" A"
SORTED BY EUR
Production Data
Well' location
Zonal
Penetration
(ft.)
Wellbore Orientation
Path (Azimuth)
Ranking (degrees)
Fracture
Index
Qi Ot CUM. PROtO EUR
(BOPO) (BOPO) (MBO) (MBO)
Drainage
Area
(acres)
Well' Locallon
Zonal
Penetration
(ft.)
Wellboro Orlentallon
Path (Azimuth) Fracture
Ranking (degrees) Index
ProducUon Data
Ot Ot
(BOPD) (BOPO)
Drainage
CUM. PROtO EUR Area
(MBO) (MBO) (acres)
Ideal Prefer> 2000' of pay In optimum direction
1. 33-11H 11-143-102 2082 A 161 A 333 133 290 442 664
3. 23-5H 05-142-102
1. 33-11H 11-143-102
636 221
333 133
347
170 392
167 337
123 408
139
179
180
364
639
441
88
177
164
A
A
A
2019
2858
2033
44-35H 35-144-102
44-7H 07-146-102
5. 12-33H 33-143-101
200
664
572
93
405
18
280 442
131
51 176
A
166
161
E
A
73
2082
07-142-101 2. 1-7
...
co
4. 34-29H 29-143-101 1063 A 46 267 90 137 262 293
4. 34-29H 29-143-101 1063 A 46 267 90 137 262 293
5. 12-33H 33-143-101 2033 A 164 441 139 167 337 347
31-35H 35-146-104 1100 A 70 278 96 114 232
6. 14-27H 27-145-101 1754 B 178 412 216 157 425 469
11. 33-29H 29-145-101 1905 A 107 291 69 72 196 147
7. 21-33H 33-145-101 1280 B 90 218 80 111 181 217
36-44H 36-144-103 2088 A 166 135 58 37 156
8. 36-44H 36-145-102 2242 90 420 89 93 200 167
10. 31-29H 29-145-101 1761 A 71 369 41 99 146 159
9. 33-19H 19-145-103
AVERAGE EUR =
196 147
261
92 36
57 103 41
49
218
101
73 A
A
2065
1950 42-31H 31-145-103
43-3H 03-143-102
230
145 159
183
99
72
155
89
41
22
369
291
485
71
94
107
A
A
C 718
1761
1905 29-145-101
29-145-101
1
'0. 31-29H
11. 33-29H
AVERAGE EUR - 251
TABLE 2C
Parameters For Early Determination of illtimate Recoveries
KEY CRITERIA: HORIZONTAl DRAINHOLES WITH >2OOO'IN ZONE (NET PAY)
SORTED BY EUR
-- -_........."" _........----- ----...._-- ------ -------- ------ ----- ------ ;:::===::::=== ===== ======
Production Data
Zonal WeUbora Orientation ------- -------- Drainage
Penetration Path (Azimuth) Fracture QI at CUM. PROD EUR Area
Well # location (ft.) Ranking (degrees) Index (BOPD) (BOPD) (MBO) (MBO) (acres)
______... ......... == _liZ==... ___ ______
-===== ====== --_._-'"'-- ----- -_.....=..
1. 33-11H 11-143-102 2082 A 161 A 333 133 260 442 664
44-7H 07-146-102 2858 A 88 364 179 123 408
44-35H 35-144-102 2019 A 177 639 180 170 392
5. 12-33H 33-143-101 2033 A 164 441 139 167 337 347
'"
I C>
14-35H 35-143-102 2579 77 185 96 53 202
8. 36-44H2 36-145-102 2242 90 420 89 93 200 167
36-44Hl 36-144-103 2086 A 188 135 58 37 156
33-35H 35-146-104 2000 88 220 98 64 146
43-36H 36-147-103 2450 85 171 41 48 109
43-3H 03-143-102 2085 A 73 218 41 57 103
14-33H 33-143-102 2050 27 134 22 22 39
44-23H 23-143-102 2607 92 101 3 31 35
========= -- ------== ====== ."".........._- ------ ..".. ""..... -=..=== ======== ="""".- ._----
AVERAGE EUR 214
TABLE 3
Meridian ~ l l H
Well Parameters for Example Problem (Appendix A)
PARAMETER SYMBOL VALUE SOURCE
====== ====== ====== ====== ====== ====== ======
Horizontal Displacement H.D. 3132 ft. MDS& Williams
Net Drainhole .Ln 2082 ft. MDS &. Williams
length in zone
Azimuth Azm 161 degrees MDS data
Formation thickness h 8 ft. Lng.
Effective thickness Heff 20 ft. In house
Drainage radius of circle ,
1489 ft. Calculated
(160 acre case) ev (160)
Drainage radius of circle , 2106 ft. Calculated
(320 acre case) ev (320)
Drainage radius of circle ,
2979 ft. Calculated
(640 acre case) ev (640)
FOR 320 ACRE VERTICAL CASE:
---------
Penetration ratio Lnl2Xe 0.331 Calculated
(For 160 acre case) (0.411)
Drainage length along the 2Xe 6294 .ft. Calculated
drainhole in the x direction
Drainage length along high 2Ye 4012 ft. Calculated,
permeability direction
Drainage radius of ellipse . 3147 ft. CaleuWed
(major axis)
Drainage radius of ellipse b 2106 ft. Calculated
(minor axis)
Average Drainage area - A(b) 669 acres Calculated
horizontal drainhole
TABLES
TABLE 4
BAKKEN PARAMETERS
Meridian #33-IIH
FORMATION PROPERTIES

Reservoir Parameters for Example Problem (Appendix A)


PARAMETER SYMBOL VALUE SOURCE
====== ====== ====== ====== ====== ====== ======
PARAMETER SYMBOL VALUE SOURCE
(t) Total porosity 4.0 % Core
======= = ---
PBU Analysis
Initial Reservoir pressure Pi 4500 psi MDS data Initial Reservoir pressure Pi 6000 psi MDS data
Bubble Point pressure Pb 2500 psi MDS data
I
rUbble Point pressure Pb 2500 psi MDS data
Oil gravity 45.3 degrees MDS data
Connate water saturation Swc IS % Assumed
API
I
Iwater compressibility c(w) 3.0E--Q6 Assumed
Reservoir temperature T 240 degrees MDS data
F I IPermeability range K 0.2 - 1.0 md. Core
Ky .014 -.1 md. PBU Analysis
Gas oil ratio Rs 850 scf/stb MDS data
I I
Kx .01 -.09 md.
Kv 0.9124 md.
N IFormation Volume Factor
I I Kh 0.05 md.
...
Initial Boi 1.3 rb/stb MDS data
@Bubble point Bo 1.5 rb/stb MDS data
I
IAverage thickness
h 10 ft. MDS data
Heff 20 ft. Inhouse
Initial prod. rate Qi 333 BOPD MDS data
I
IReservoir temperature T 240 degrees MDS data
Current prod. rate Qt 151 BOPD MDS data F
Cumulative production Np 249 MBO MDS data
I
Ultimate recoverable EUR 427 MBO In house
FLUID PROPERTIES
reserves

Recovery factor RF 20.0 % MDS data
Formation Volume Factor
Initial Boi lAO rb/stb MDS data
(t)
@Bubble point Bo 1.54 rb/stb MDS data
Total porosity 4.0 % Core
Oil gravity 42 degrees MDS data
API
Initial hydrocarbon Soi 85 % Assumed
saturation
Oil viscosity at reservoir 0.3 cp MDS data
Effective permeability K 0.2 - 1. md Core
Average Gas oil ratio Rs 850 scf/stb MDS data
Ky .014 - .17
I
IRecovery factor RF 15.0 % MDS data
Kx .01 -.09
to 20.0 %
BAKKEN PLAY
BAKKEN PARAMETERS
TABLE 6
*.**** *** **********
FRACTURE NETWORK PROPERTIES
PARAMETER SYMBOL VALUE SOURCE
-= = =--= =---.
HORIZONTAL WELL
,/>(f)
I.P. 200-400 BOPD
Fracture porosity 0.05 to Empirical correlations
200-250 MBO RECOVERABLE
...
I
0.30 %
...
Fracture compressibility e(f) 81.8E-06 Williams Petroleum
I
SURFACE
I --
I
Area of
Development to
Fracture spacing D 3-4 ft. From Industry
I
Date
100 em. I
WYOMING
em. Empirical correlations I
I I
0 50
Natural Fracture width e IOE-03 to
t------l
IOE-02
MILES
Minimum width e IOE-OS em. Literature
Fracture permeability k(f) 0.2 md.
"""W,;Z ",,,,1.. ,,< ",M,b.,
Figure 1 Williston Basin
Figure 7 - Group A : 21 Wells
Horizontal Bakken Performance
Rate Va Montha From 'Iat Prod
Figure 8 - Bakken Play
Group A Wells
1st Month Production Rate

7
e
II
4
8
2
1
0.L.-_-
800
lbtal we 21
_a.r'",
0-100 100-200 200-800 800-400 400-100 100-100
BOPD
811 211 80
-MIN. RATE I
II 10 111 20
MONTHa OF PROD
I...... Me. PROD --+- MAX. RATE
1L-_----L__---L-__..l..-_----l ---L.__....l.-_--'
o
Figure 9 - Bakken Play
Group A W'ells
Recoverable Reuerves
Figure 10
THE PHYSICAL MODEL
II O::'..:.W'eI::::I.=-- -.,
4
8
2
o
0-100 100-10 1110-200 200-801) 800-400 400-1100
RESERVES (YBO)
lbtal welle 21
_a.I.. ,
h
..-- L
o
___ Xw --..
... 2Xe
Figure 11 - Drainage Areas
Vertical well Drainage Volume Schematic
OIL
__ r
Horizontal Well Drainage Volume Schematic
24

Related Interests