Abstract

Various fracture interaction criteria were reviewed and a

systematic study is presented to analyze the effect of natural
fractures on hydraulic fracture propagation. From the literature
review, the approach of Warpinski and Teufel was adopted to
evaluat the fracture propagation that would occur after a
hydraulic fracture intersects an existing natural fracture. The
study determined the conditions of differential stress, angle of
intersection, and fracture toughness for which dilation of the
natural fracture would occur. Criteria for the subsequent
modes of fracture propagation are then presented.

Introduction
Hydraulic fracture propagation in the presence of natural
fractures is substantially different from fracture propagation in
reservoirs without natural fractures. Conventional hydraulic
fracture design is based on the assumption that the rock is
homogenous and the fracture propagates symmetrically in a
plane perpendicular to the minimum stress. In naturally
fractured reservoirs due to interaction with natural fractures,
the fracture may propagate asymmetrically or in multiple
strands or segments.
The presence of natural fractures alters the way the
induced fracture propagates through the rock. Experimental
investigations
1-3
have shown that the propagating fracture
crosses the natural fracture, turns into the natural fracture, or
in some cases, turns into the natural fracture for a short
distance, then breaks out again to propagate in a mechanically
more favorable direction, depending primarily on the direction
of the natural fracture relative to the stress field. Laboratory
studies suggest that hydraulic fractures tend to cross the
existing fractures at high differential stress and high angles of
approach and the direction of induced fracture propagation is
near perpendicular to the natural fracture. At low angles of
approach and low differential stress, the natural fracture opens,
diverting the fracturing fluid and preventing the induced
fracture from crossing, at least temporarily.
This paper reviews several experimental and theoretical
studies and discusses different fracture interaction criteria
developed in the past. A study is presented based on an
idealized interaction model to investigate the effect of a
natural fracture on hydraulic fracture geometry based
primarily on the differential stress and the angle of interaction.

Review of Theoretical and Experimental Studies
Several field and lab experimental studies have been done
in the past to investigate the effect of natural fractures on the
propagation of an induced hydraulic fracture. This section
reviews all the results and conclusions of those studies.
In 1963, Lammont and Jessen
1
conducted a series of
laboratory experiments on six different types of rocks. These
experiments were run under triaxial compression up to 1142
psi and with angles of approach, (Fig. 1), between the
hydraulic fracture and pre-existing fracture varying from 30
o

to 60
o
. They found that in all successful cases the induced
fracture was able to cross the existing fracture and the
orientation of the hydraulic fracture was such that it turned and
intersected the existing fracture at right angles. A similar
behavior was observed when the hydraulic fracture departed
from the existing fracture. It has been observed that the
location of the point of exit on the existing fracture was
random and it was not controlled by the stress concentration at
the end of the fracture but rather by some particular weakness
in the rock matrix.
Daneshy
2
reported the results of laboratory experiments
run on granite blocks which had three types of flaws in the
forms of crystal and matrix boundaries, small scale fractures
( in) and large natural fractures. From the results of the
experiments he concluded that the first two types of flaws had
little or no effect on the direction of the propagation of
hydraulic fracture whereas large flaws exhibited a mixed
behavior. In most cases the hydraulic fracture was able to
cross the larger fractures but in some cases the natural fracture
was able to successfully stop the fracture propagation within
the limits of the specimen size. He also studied the effect of
deviotoric stress on the fracture propagation and concluded
that in a in a hydrostatic state of stress the hydraulic fracture
can propagate in any direction and is likely to follow the
prevailing weakness in the rock.
Blanton
3
reported an experimental study in which he
conducted tests on naturally fractured Devonian shale and
hydrostone blocks to analyze the impact of angle of interaction
and differential stress on the hydraulic fracture propagation.

SPE 94568
Effect of Natural Fractures on Hydraulic Fracture Propagation
N. Potluri, D. Zhu, and A.D. Hill, Texas A&M U.
Copyright 2005, Society of PetroleumEngineers

This paper was prepared for presentation at the SPE European Formation Damage
Conference held in Scheveningen, The Netherlands, 25-27 May 2005.

This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal 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 a proposal of not more than 300
words; illustrations may not be copied. The proposal 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.

2 NAGA POTLURI, D.ZHU AND A.D.HILL SPE 94568
He concluded that at low angles of interaction and low
differential stress the existing fracture opened, diverting the
fracturing fluid and preventing the induced fracture from
crossing, at least temporarily. At high angles of interaction and
high differential stress the induced fracture crossed the
existing fracture.
Warpinski and Teufel
4
and Warpinski
5
conducted
mineback experiments to study the effect of geologic
discontinuities on the hydraulic fracture propagation. They
observed three modes of induced fracture propagation:
crossing, arrest by opening and dilating the joint and arrest
by shear slippage of the joint with no dilation and fluid flow
along the joint. Crossing occurred at a very high differential
stress (1500 psi) which is in agreement with Blantons results.
At moderate to low differential stress the natural fracture
arrested the propagation of the hydraulic fracture. It was also
found that fractures were offset and multiple stranding was
predominant.
Beugelsdijk, de Pater and Sato
6
performed scaled
laboratory experiments on Portland cement blocks to analyze
the effect of discontinuities on the hydraulic fracture
propagation. The parametric study included the variation of
horizontal stress difference, flow rate, viscosity and stress
regime. From the experimental results it was found that at
large horizontal stress difference the hydraulic fracture
continued to propagate in the preferred fracture plane without
any interaction with discontinuities. A similar behavior was
observed at high flow rates and high viscosity. The stress
regime had a different impact on the fracture geometry. In
tectonic stress regime the hydraulic fracture was found to
interact more with the discontinuities even for high horizontal
stress difference. This is because in tectonic stress regimes the
vertical stress is at a low level and the number of open
discontinuities are more.
Many field studies
7-11
conducted in naturally fractured
formations reveal that the effects of natural fractures on
fracture propagation are
enhanced fluid leakoff
premature screen-out
arrest of the fracture propagation
formation of multiple fractures
fracture offsets
high net pressures

Fracture Interaction
Various authors have provided analytical equations and
numerical solutions for predicting the interaction between a
natural fracture and hydraulic fracture. Blanton
12
and
Warpinski and Teufel
4
derived fracture interaction criteria
relating differential stress and angle of approach. Renshaw
13

provided a criterion for crossing across unbonded interfaces.
In this section we review the above analytical criteria.

Blantons Criterion
Blanton presented a simple analytical fracture interaction
criterion relating differential stress and angle of interaction to
extrapolate the laboratory results to field simulations. The
criterion is based on the elastic solution of stresses in the
interaction zone and assumes that the initial interaction for the
hydraulic fracture is to be blunted by the pre-existing fracture.
According to him the criterion can be stated as follows,
opening will occur if the pressure at the intersection exceeds
the normal stress acting on the natural fracture and crossing
will occur when the pressure required for re-initiation is less
than the opening pressure. In mathematical form the criterion
for crossing can be written as,
0
T p
t
+ > (1)
and for opening,
0
n
p > (2)
where p is the pressure inside the fracture ,
n
is the normal
stress acting on the plane of the natural fracture,
t
is the stress
acting parallel to the natural fracture and T
o
is the tensile stress
of the rock. The expression for
t
depends on the far field
stress, pressure in the fracture, frictional slippage along the
natural fracture and the geometry of the interaction zone. The
final equation for crossing is given by Blanton as,


2 sin 2 cos
1
0
3 1
b T
>

(3)
where
( )
|
|
.
|

\
|

=
f
K
l x
x v
a
b
0
0
2
1
(4)
( ) ( )
( )
(
(

(
|
|
.
|

\
|


+
|
|
.
|

\
|

+
|
|
.
|

\
|
+
+ +
+ =
2
0
0
2
0
0
0
2
0
0
0 0
ln ln
ln
1
a l x
a l x
c
l x
a l x
l x
l x
a l x
l x x v

(5)
and
( )
2
1
2
2
2
0
1
1
|
|
|
|
|
.
|

\
|
+
+ +
=
f
f
K
K
e
e a
x

(6)
where
1
and
3
are the maximum and minimum principal
stress acting on the plane of the natural fracture as shown in
Fig. 2. a is the length of the zone of slippage, the region from
-l to + l is the open section of the natural fracture, is the
angle of interaction, and
f
is the coefficient of friction.
Beyond the open section to points ( ) l a + is the region
where the shear stress is increasing until it is equal to the far-
field shear stress.
The criterion was found to agree very well with the
experimental results and the angle of interaction was found to
be the most sensitive parameter in determining the fracture
propagation.
SPE 94568 EFFECT OF NATURAL FRACTURES ON HYDRAULIC FRACTURE PROPAGATION 3
Warpinski and Teufels Criterion
Warpinski and Teufel
4
derived a fracture interaction criterion
to predict whether the induced fracture causes a shear slippage
on the natural fracture plane leading to arrest of the
propagating fracture or dilates the natural fracture causing
excessive leakoff. Shear slippage occurs if the normal stress
acting on the plane of the natural fracture is not sufficient
enough to prevent the planes from sliding against each other.
Following a linear friction law
14
the relation between the shear
stress and the normal stress acting on the natural fracture plane
is given as
( ) p K
n f
+ =
0
(7)
where
0
is the inherent shear strength of the natural fracture
plane,
f
is the coefficient of friction and
n
p is the
effective normal stress acting on the natural fracture plane. If
( ) p K
n f
+ >
0
(8)
then shear slippage occurs on the natural fracture plane.
The shear stress and normal stress acting on the plane of
the natural fracture can be obtained from a 2D stress
resolution
14
(Fig. 3) and are given as
( )

= 90 2 sin
2
3 1
(9)
and
1 3 1 3
cos 2(90 )
2 2
n


+
= + (10)
When the hydraulic fracture intersects the natural fracture
the pressure at the intersection is given as

p
p
+
=
3
(11)
where p

is the treatment overpressure. The criterion can now

be mathematically written for shear slippage after substituting
Eqs. 9 and 10 into Eq. 8 as
( )
f f
f
K K
K p
+

>

2 cos 2 sin
2 2
0
3 1
(12)
and for natural fracture dilation as
n
p > (13)
Substituting Eqs. 10 and 11 in Eq. 13 gives
( )( )
2
2 cos 1
3 1

>

p (14)

Renshaws Criterion
Renshaw
13
provided a simple criterion for crossing by
applying a first order analysis of the stresses near a mode I
fracture impinging on a frictional interface oriented normal to
the growing fracture. According to this work, compressional
crossing will occur if the magnitude of the compression acting
perpendicular to the frictional interface is sufficient to prevent
slip along the interface at the moment when the stress ahead
of the fracture tip is sufficient to initiate a fracture on the
opposite side of the interface. Mathematically this is given
as,
06 . 1
35 . 0
35 . 0
1 0
3
f
K
T
+
>

(15)
The above criterion is restricted to orthogonal interactions and
assumes that the frictional interface does not alter the
propagation direction of the fracture.

Application of Interaction Criterion
When the hydraulic fracture intersects a natural fracture its tip
is blunted and the fracture propagation stops at least
momentarily. Following Warpinski and Teufel
4
, the pressure
at the intersection just when the tip is blunted is given by Eq.
11. Defining the pressure at the intersection of the hydraulic
fracture with the natural fracture as p
i
(t), with t = 0 being the
time that the hydraulic fracture reaches the natural fracture, we
consider all of the possible interactions between the hydraulic
fracture and the natural fracture.

I. p
i
(0)>
n

If the pressure in the propagating hydraulic fracture is
higher than the normal stress on the natural fracture, the
natural fracture will immediately dilate. Whether or not this
condition holds depends on the relative magnitudes of the
treatment overpressure, p

, in the propagating fracture, the

stress contrast,
1
-
3
, and the angle of interaction, , as given
by Eq. 14. The conditions for natural fracture dilation are
illustrated in Fig. 4 for a treatment overpressure of 100 psi at
the tip of the propagating fracture. For any intersection angle,
if
1
-
3
is below the curve shown, the fracturing pressure is
sufficient at the moment of intersection to dilate the natural
fracture.
As the natural fracture opens, leakoff into the natural
fracture will increase, and the net pressure, after leveling off
for a period of time, will increase
5
. As the pressure in the
natural fracture increases. Three possible modes of fracture
growth are possible as follows:

Ia. The pressure at the intersection point exceeds the
pressure necessary to initiate a fracture along the original
path of the propagating hydraulic fracture (Fig. 5), i. e.,

p
i
(t) >
3
+ T
o,i
(16)

where T
o,i
is the fracture toughness of the rock on the
natural fracture face opposite the blunted hydraulic
fracture. For this mode of fracture growth to occur, it
must be easier for the rock to crack on the natural fracture
wall opposite the blunted hydraulic fracture than to begin
propagating from one of the tips of the natural fracture.
This is favored by the fact that the pressure in the natural
fracture is highest at the point of intersection.

4 NAGA POTLURI, D.ZHU AND A.D.HILL SPE 94568
Ib. The pressure at one of the tips of the natural fracture
exceeds the net pressure required to begin propagating
from the natural fracture tip (Fig. 6). Presuming the
fracture extension will initially be in the direction of the
natural fracture, this occurs when

p
i
(t) >
n
+ T,
o,tip
+ p
nf
(17)

where T
o,tip
is the fracture toughness at the tip of the
natural fracture and p
nf
is the pressure drop in the natural
fracture between the intersection point and the nearest
fracture tip. A further requirement for this mode of
fracture propagation to occur is that the fracture toughness
must be sufficiently less than the toughness at the
intersection point to overcome the pressure drop in the
natural fracture and the higher normal stress. In other
words, for fracture propagation to begin from the tip of
the natural fracture, this should be a weak spot in the rock
satisfying the criterion

T
o,,tip
< T
o,i
(
n

3
) - p
nf
(18)

otherwise the fracture will initiate opposite the blunted
fracture.

Ic. The pressure somewhere in the natural fracture is high
enough to overcome the local fracture toughness and the
fracture breaks out of the natural fracture somewhere
between the intersection point and the natural fracture tip
(Fig. 7). For this mode of propagation to occur, local
flaws in the surface of the natural fracture exist so that the
pressure at position l along the natural fracture is high
enough to initiate a hydraulic fracture at this point. This
means that

T
o,l
< T
o,i
- p
l
(19)

where p
l
is the pressure drop in the natural fracture
between the intersection point and position l.
Additionally, the pressure drop at the natural fracture tip
must be below the pressure needed to propagate the
fracture at that point.

II. p
i
(0) <
n

If the pressure in the fracture is initially below the normal
stress acting on the natural fracture, dilation of the natural
fracture will not occur. If the fracture toughness at the
intersection point, Toi, is the same as the treatment
overpressure, the fracture will continue to propagate
unaffected by the natural fracture. However, if the fracture
toughness is higher because of the discontinuity at the natural
fracture, the fracture will be arrested, at least momentarily,
until the fracture breaks out on the other side of the natural
fracture or the natural fracture begins to dilate. Which of these
events occurs depends on the relative magnitudes of the
fracture toughness and the normal stress. Thus, if

T
oi
>
n

3
(20)

The natural fracture will inflate when p
i
(t) exceeds
n
,
while if
To,i <
n

3
(21)
the fracture will propagate from the intersection point when
p
i
(t) >
3
+ T
oi
. (22)

Conclusions

The approach of warpinski and Teufel was applied to predict
the manner in which a hydraulic fracture will propagate when
it intersects a natural fracture. The possible modes of
propagation are:
1. Crossing of the natural fracture, which occurs when
the normal stress on the natural fracture is high
relative to the fracture toughness of the rock.
2. Dilation of the natural fracture thane propagation
from the tip of the natural fracture.
3. Dilation of the natural fracture, then breakout of a
fracture from along the natural fracture.
Criteria for which when each of these modes of fracture
propagation will occur based on the horizontal stress contrast,
the angle of intersection of the propagating fracture and the
natural fracture, fracture toughness, and pressure drop within
the natural fractures are presented.
If the pressure in the fracture is below the normal stress
acting on the natural fracture, the propagating fracture will
likely be arrested, at least momentarily. For this case, criteria
for subsequent fracture propagation or natural fracture dilation
are presented.

Nomenclature
a = length of the zone of slippage, ft
b = dimensionless factor
G = Shear Modulus, psi
h
f
= height of the fracture, ft
K
f
= coefficient of friction, dimensionless
l = length of the natural fracture, ft
p = pressure, psi
p
i
(t) = pressure in the fracture at the intersection with the
natural fracture (psi)
p

= treatment over pressure, psi

q = flow rate, bbl/min
T
0
= tensile strength of the rock, psi
T
0i
= tensile strength of the rock at the intersection, psi
T
0,tip
= tensile strength of the rock at the natural fracture
tip, psi
T
0,l
= tensile strength of the rock at location l, psi
w = width of the induced fracture, in
x
o
= parameter in Blanton model (ft)
p
nf
= pressure drop in the natural fracture (psi)
p
l
= pressure drop in the natural fracture between
intersection and position l (psi)
= angle of interaction (approach), deg
= viscosity, cp
= Poisson ratio, dimensionless

1
= maximum in-situ principal stress, psi

3
= minimum in-situ principal stress, psi
SPE 94568 EFFECT OF NATURAL FRACTURES ON HYDRAULIC FRACTURE PROPAGATION 5

n
= stress acting perpendicular to the natural fracture
plane, psi

t
= stress acting parallel to the natural fracture
plane, psi
= shear stress acting on the plane of the natural
fracture, psi

0
= inherent shear strength of the material, psi

Acknowledgements
The authors thank the sponsors of the Improved Well
Performance Research Program of the Center for Petroleum &
Geosystems Engineering at the University of Texas at Austin
for providing the financial support for this study.

References
1 Lamont, N and Jessen, F.: The Effects of Existing
Fractures in Rocks on the Extension of Hydraulic
Fractures, JPT, 203-209, February 1963.
2 Daneshy, A.A.: Hydraulic Fracture Propagation in the
Presence of Planes of Weakness, SPE 4852, presented at
SPE-European Spring Meeting, Amsterdam, 29-30 May
1974.
3 Blanton, T.L. (May 1982), An Experimental Study of
Interaction Between Hydraulically Induced and Pre-
Existing Fractures, SPE 10847, presented at the
SPE/DOE unconventional Gas Recovery Symposium,
Pittsburg, Pennsylvania, 16-18 May 1982.
4 Warpinski, N.R and Teufel, L.W.: Influence of Geologic
Discontinuities on Hydraulic Fracture Propagation, JPT
209-220, February 1987.
5 Warpinski, N.R.: Hydraulic Fracturing in Tight, Fissured
Media, JPT 146-152,208-209, February 1991.
6 Beugelsdijk, L.J.L., de Pater, C.J., Sato, K.:
Experimental Hydraulic Fracture Propagation in Multi-
Fractured Medium, SPE 59419, presented at the SPE
Asia Pacific Conference on Integrated Modeling,
Yokohoma, Japan, 25-26 April 2000.
7 Britt, L.K. and Hager, C.J.: Hydraulic Fracturing in a
Naturally Fractured Reservoir, SPE 28717, presented at
the SPE International Petroleum Conference and
Exhibition, Veracruz, Mexico, 10-13 October 1994.
8 Rodgerson, J.l.: Impact of Natural Fractures in Hydraulic
Fracturing of Tight Gas Sands, SPE 59540, presented at
the SPE Permian Basin Oil and Gas Recovery Conference
held in Midland, Texas, 21-23 March 2000.
9 Vinod, P.S., Flindt, M.L., Card, R.J., Mitchell, J.P.:
Dynamics Fluid-Loss Studies in Low-permeability
Formations With Natural Fractures, SPE 37486,
presented at the SPE Production Operations Symposium,
Oklahoma, 9-11 March 1997.
10 Azeemuddin, M., Ghori, S.G., Saner, S., Khan, M.N.:
Injection-Induced Hydraulic Fracturing in a Naturally
Fractured Carbonate Reservoir: A case study from Saudi
Arabia, SPE 73784, presented at the SPE International
Symposium & Exhibition on Formation Damage Control,
Lafayette, 20-21 February 2002.
11 Murphy, H.D and Fehler, M.C.: Hydraulic Fracturing of
Jointed Formations, SPE 14088, presented at the SPE
International Meeting on Pertoleum Engineering, Beijing,
China, March 17-20 1986.
12 Blanton, T.L.: Propagation of Hydraulically and
Dynamically Induced Fractures in Naturally Fractured
Reservoirs, SPE 15261, presented at the SPE
Unconventional Gas Technology Symposium, Louisville,
KY, 18-21 May 1986.
13 Renshaw, C.E. and Pollard, D.D.: An Experimentally
Verified Criterion for Propagation across Unbonded
Frictional Interfaces in Brittle, Linear Elastic Materials,
International Journal of Rock Mechanics Mining Science
and Geomechanics, 32, 3, 237-249, 1995.
14 Jaeger, J.C and Cook, N.G.W.: Fundamentals of Rock
Mechanics, Halsted Press, New York, 1976.





Hydraulic fracture
Natural fracture

1


Fig. 1 Hydraulic fracture intersecting a natural fracture.




-l-a
3
1
-l
+l
l+a
Natural fracture

Zone of slippage


Fig. 2 Zone of slippage for natural fracture (from Blanton
12
).
6 NAGA POTLURI, D.ZHU AND A.D.HILL SPE 94568

Natural fracture

Natural fracture


Fig. 3 Stress resolution on the natural fracture plane.



0
100
200
300
400
500
600
700
800
900
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90

p
i
()>
n

Fig. 4 Conditions for natural fracture dilation,
treatment overpressure=100 psi.




dilated
natural fracture
propagating
hydraulic fracture
p
i
(t)>
3
+T
o


Fig. 5 Hydraulic fracture crosses natural fracture.



propagating
hydraulic fracture
dilated
natural fracture
propagating from
natural fracture
p
tip
>
n
+T
o,tip





Fig. 6 Hydraulic fracture propagates from the tip of
natural fracture.





propagating
hydraulic fracture
dilated
natural fracture
propagating from
natural fracture
T
o,e
<T
o,i
-p
e




Fig. 7 Hydraulic fractrure propagates from weak point along
natural fracture.