42 In situ stress

0 0

u2

0 0O3 I

(a) (b) (C)

Figure 4.1 (a) Principal stresses acting on a small cube. @) Principal stresses
expressed in matrix form. (c) Principal stress orientationsshown on a hemispherical
projection.

4.3 Methods of stress determination

Clearly, any system utilized for estimating the in situ stress state must
involve a minimum of six independent measurements. There are methods
of ’direct’ stress measurement and there are methods of estimating the
stresses via various ‘indirect’ or ‘indicator’ methods. In this book, we will
concentrate on the four main methods recommended by the International
Society for Rock Mechanics (ISRM), while recognizing that there are a
multitude of direct and indicator approaches available.
The four direct methods recommended by the I S M (Kim and Franklin,
1987) are:
(a) the flatjack test;
@) the hydraulic fracturing test;
(c) the United States Bureau of Mines (USBM) overcoring torpedo; and
(d) the Commonwealth Scientific and Industrial Research Organization
(CSIRO)overcoring gauge.
Use of the overcoring method is shown in Fig. 4.2.
Some of the indicator methods are:
(a) borehole breakouts-damage to a borehole indicating principal stress
orientations;
(b) fault plane solutions-back analysis of principal stresses causing faults;
(c) acoustic emission-the rock emits low-intensity ’noise’ when it is
stressed;
(d) anelastic strain relaxation--core exhibits expansiodcontraction on
removal from the borehole;
(e) differential strain analysis-pressurizing a piece of rock indicates its
previous stress state through differential strain effects;
(f) core discing-geometry of stress-induced core fracturing indicates
stress components;
(g) observations of discontinuity states, e.g. open discontinuities are not
transmitting stress across the gap.
The four direct ISRM recommended methods are described below; for
 Methods of stress determination 43

Figure 4.2 In situ stress determination in the Carmenellis granite.

a fuller description of the indirect methods, the reader is referred to Dyke

(1988).The key reference for the I S M methods is Suggested Methodsfor Rock
Stress Determination, produced by Kim and Franklin (1987).Here, we now
go on to explain these methods in the context of their ability to determine
the components of the stress tensor.
In Fig. 4.3, we have shown four stress tensors and indicated the ability
of each method to determine the six components of the stress tensor in one
application. For the flatjack and with the x-axis aligned perpendicular to
the flatjack, one normal component-in this case oxx-canbe determined.
It immediately follows that, to determine the complete state of stress, six

1. Flatjack 2. Hydraulic fracturing

Principal stresses
xy T~~ One normal stress
component axes i.e. plane of the
I ‘y\y Tyz determined, say
fracture, two deter-
mined, say u, and
, parallel to x-axis.
u3,one estimated,
say isz.

3. USBM overcoring torpedo 4. CSIRO overcoring gauge

R-
Three components
All six components
\ -- determined from six
(or more) measurements

change.
Figure 4.3 The four ISRM suggested methods for rock stress determination and
their ability to determine the components of the stress tensor with one application of
the particular method.
44 In situ stress

such flatjack measurements have to be conducted at six different orienta-

tions. Note that, in general, the reference axes will not be aligned with the
flatjack orientation and separate transformations will have to be used for
each flatjack measurement, because it is the normal stress perpendicular
to the plane of the flatjack that is being determined, rather than a specific
component of the stress tensor. In fact, it is interesting to note that whilst
a normal stress can be determined directly, there is no equivalent method
of determining a shear stress: the shear components in the tensor must be
calculated from the measurements of normal stresses in different directions;
they cannot be measured directly. It should also be remembered that this
technique determines the stress tensor in an excavation wall and therefore
determines the induced stress rather than thefield stress. (A glossary of terms
for in situ stress can be found in Section 4.10.)
With reference to the top right-hand diagram in Fig. 4.3, the basic
hydraulic fracturing method provides only two items of information-the
breakdown pressure and the shut-in pressure. Thus, only two components
of the stress tensor can be established by this technique: the shut-in
pressure is assumed to give the minor principal stress, g,whilst the major
principal stress, q,is given via the breakdown pressure, the value of o3and
the magnitude of the tensile strength of the rock.
We have seen that, in the case of the flatjack, the six components can be
determined by using the method at six different orientations. In general,
this is not possible with hydraulic fracturing, because the tests are
conducted deep in a borehole. The major advantage of hydraulic fractur-
ing is that it is the only method of determining part of the stress state more
than a few hundred metres from man-access, and, indeed, may be used up
to 5 or 6 km depth. However, the major disadvantage is that assumptions
have to be made in order to complete the stress tensor. These assumptions
are that the principal stresses are parallel and perpendicular to the borehole
axis, and that the vertical principal stress can be estimated from the depth
of overburden. As a result, in the hydraulic fracturing stress tensor in Fig.
4.3, the two circled components are determined but the three zero values
for the shear stresses are an assumption, as is the value (of what is taken
here to be) oz.
In the case of the USBM overcoring torpedo, a two-dimensional state of
stress is determined, i.e. the three circled components in the diagram in
Fig. 4.3, giving three components of the three-dimensional stress tensor.
Thus, two, and preferably three, non-parallel boreholes must be used to
determine the complete state of stress. It should be noted that in the cases
of the flatjack and hydraulic fracturing, the material properties of the rock
have not been used except for the tensile strength which is required in
hydraulic fracturing. For the flatjack, only the transformation equations are
required; for hydraulic fracturing, only the stress concentration factors for
a circular hole are required and these are independent of material
properties (assuming ideal elasticity); but, for the USBM overcoring
torpedo, in order to convert the measured displacements to stresses, the
elastic properties of the rock are required. This introduces a whole new
series of assumptions.
Finally, in the case of the CSIRO overcoring gauge, as we have shown
 Methods of stress determination 45
in Fig. 4.3, the complete state of stress can be determined from measure-
ments of strain in six or more different directions taken during one
application of the method. The material properties of the rock are also
required for this method: a device which is equipped with nine or 12 strain
gauges can determine the state of stress in a transversely isotropic rock with
five elastic parameters.
It is emphasized that the understanding of how the components of the
stress tensor are established by these four different methods is crucial to
the planning of an optimal strategy for stress measurement. There are other
complicating factors which we will be discussing later, but the key is to
understand the fundamental basis of the tests as described here. In this
context, none of the indicator methods, with the possible exception of
differential strain analysis, can estimate the complete stress tensor. It
follows that invariably our strategy for stress determination will comprise
of integrating all the information to hand.
In the following sub-sections, the four main ISRM methods are outlined
and discussed. The diagrams are those used in the ISRM Suggested
Methods document.

4.3.1 Flatjack
In Fig. 4.4, the basic principle of the flatjack test is shown. Two pins are
drilled and fixed into the excavation boundary. The distance, d, between
them is then measured accurately. A slot is cut into the rock between the
pins, as shown in the diagram. If the normal stress is compressive, the pins
will move together as the slot is cut. A flatjack, which is comprised of two
metal sheets placed together, welded around their periphery and
provided with a feeder tube, is then grouted into the slot. On pressurizing
the flatjack with oil or water, the pins will move apart. It is assumed that,
when the pin separation distance reaches the value it had before the slot
was cut, the force exerted by the flatjack on the walls of the slot is the same
as that exerted by the pre-existing normal stress. There will be some error
in this assumption, mainly due to jack edge effects, but these can be taken
into account if the jack is suitably calibrated. The test provides a good
estimate of the normal stress across the flatjack.
The major disadvantage with the system is that the necessary minimum
number of six tests, at different orientations, have to be conducted at six
different locations and it is therefore necessary to distribute these around
the boundary walls of an excavation. Invariably, these tests will be
conducted under circumstances where the actual stress state is different at
each measurement location. Hence, to interpret the results properly, it is
also necessary to know the likely stress distribution around the test
excavation.

4.3.2 Hydraulic fracturing

The hydraulic fracturing method of stress measurement basically provides
two pieces of information via the breakdown pressure and the shut-in
pressure (cf. the introductory text in Section 4.3 and part 2 of
46 In situ stress

r /, t t t
FI atj ack

Note identification of cancellation pressure, P,

Excavation time - ' Flatjack pressure -

Figure 4.4 The flatjack test (from SuggestedMethods for Rock Stress Determination, Kim
and Franklin, 1987). (a) Flatjack. @) Test configuration.(c) Pin separationversus slot
excavation time and flatjack pressure. (d) The flatjack tests in progress.

Fig. 4.3). A length of borehole is chosen for the stress measurements and
an interval, typically 1 m long, is located for the test and isolated using a
straddle packer system. The isolated zone is pressurized by water until a
fracture occurs in the rock. The two measurements taken are the water
pressure when the fracture occurs and the subsequent pressure required
to hold the fracture open, known, respectively, as the breakdown and shut-
in pressures.
In connection with this method, it is most important to realize the
following. First, the packed-off section should be free from fractures so that
a new fracture is in fact created: a method of establishing this is to use a
borehole television camera. Second, it is obviously best if the water
 Methods OF stress determination 47

pressures are measured at the test section, i.e. downhole rather than at
the surface, if possible. Third, it is necessary to use an impression packer
or equivalent system to establish the orientation and location of fracture
initiation. Finally, it should be remembered that, using the basic
technique, it has to be assumed that the borehole is parallel to a principal
stress direction.
A schematic representation of the test equipment (consistingof a straddle
packer and an impression packer) is shown in Fig. 4.5, together with the
interpretative calculations. In Fig. 4.6, an early stage in the hydraulic
fracturing stress measurement procedure is shown.
There are several problems inherent in the use of this equipment to
measure the stress state. With reference to the four points mentioned
earlier, it can often be difficult, if not impossible, to identify a 1 m length
of borehole which is fracture free. Furthermore, there can be difficulties in
measuring water pressures accurately, and in correctly identifying the
breakdown and shut-in pressures. There is the question of whether the
crack initiating at the borehole wall in fact propagates in the same direction
(e.g. it may curl into the plane normal to the borehole axis). Lastly, it is
often a completely unjustified assumption that the borehole is indeed
parallel to a principal stress. Against all these points, however, is the fact
that the hydraulic fracturing method is the only direct method available
for stress measurement at any significant distance from the observer (i.e.
distances greater than 100 m), and it has been used to depths of several
kilometres.

--
(a1 (b)
Breakdown
Tn pump, flowmeter. I Pressure -P, pressure
pressure transducers

Shut-in P

housing P, = uh
fm P, = 3Uh - uH+u,
-1 I

4-
cture
a1 -0 9 rn Sti addle
lmprer
packer
-r
-1 2 ni
+M Rnre v
a,= tensile
strength of

i packer
J
Drillhole
Drillh

"Ft t \OH

Figure 4.5 (a)The hydraulic fracturing system and (b) associated calculations (from
Suggested Methods for Rock Stvess Determinution, Kim and Franklin, 1987).
48 In situ stress

Figure 4.6 Hydraulic fracturing straddle packer system being lowered into a
borehole during stress measurement tests in Cornwall, UK.

In the calculation method shown in Fig. 4.5(b), it is assumed that the

stress concentration of a principal stress component around the borehole
in the horizontal plane shown has extreme values of -1 and 3. As shown,
the shut-in pressure, Ps, is assumed to be equal to the minor horizontal
principal stress, oh.The major horizontal principal stress, oH,is then found
from the breakdown pressure. In the formula in Fig. 4.5, the breakdown
pressure, PB, has to overcome the minor horizontal principal stress
(concentrated three times by the presence of the borehole) and overcome
the in situ tensile strength of the borehole rock; it is assisted by the tensile
component of the major principal horizontal stress. Note that when a
borehole is pressurized with water at a given pressure, a tensile stress
component of the same value is induced in the borehole periphery.
Moreover, we have also assumed that the crack has propagated in a
direction perpendicular to the minor principal stress.
All of these factors contain further tacit assumptions, in particular
that the elasticity theory is valid. For this to be true, and the stress
concentration factors of -1 and 3 around the circular borehole to be valid,
the material of the borehole wall must be continuous, homogeneous,
isotropic and linearly elastic. Furthermore, we have also assumed in this
basic analysis that the rock is impermeable, so that borehole water has not
penetrated the rock and affected the stress distribution.
 Methods OF stress determination 49

If the elasticity assumptions are made, we find that the stress

concentration factors do not depend on the elastic constants of the rock nor
the diameter of the borehole. We do, however, need to know the tensile
strength of the rock and this is a subject fraught with controversy. Suffice
it to say, the best way to measure the tensile strength is under the
conditions for which it is required, i.e. by pressurizing a hollow cylinder
of rock. This is because the tensile strength (i.e. the stress causing tensile
failure) is not a material property. A material property does not depend on
the specimen geometry and loading conditions of the test: the tensile
strength does.
Against this background of many potential difficulties, a great deal of
research effort is being expended on improving confidence in stress
measurements made using this technique. There are ways of overcoming
all the difficulties mentioned above, see Cuisiat and Haimson (1992).

4.3.3 The USBM borehole deforrnution guuge

As indicated in part 3 of Fig. 4.3, the USBM technique allows the complete
stress state in a plane to be determined from three measurements of
the change in different diameters of a borehole when the stresses are
released by overcoring the borehole. The instrument is shown in Fig. 4.7.
When the torpedo is inserted in a borehole, six 'buttons' press against
the borehole wall and their diametral position is measured by strain gauges
bonded to the sprung steel cantilevers supporting the buttons. When this
borehole is overcored by a larger drill, the stress state in the resulting
hollow cylinder is reduced to zero, the diameter of the hole changes, the
buttons move, and hence different strains are induced in the strain gauges.
From previous calibration exercises, the actual diametral changes are
deduced. From these changes, and with the use of elasticity theory, the
biaxial stress state in the plane perpendicular to the borehole axis is
deduced.
In this test, as in hydraulic fracturing, we are determining far-field
stresses which have been concentrated around the measurement borehole.
A useful aspect of the USBM technique is that it produces an annular core

Figure 4.7 The USBM borehole deformation gauge.

50 In situ stress

which may be used in the laboratory to determine the elastic properties

directly at the site where the test was conducted. Given the validity of the
assumptions, the USBM gauge and its homologues are efficaciousbecause
they are reusable, permit measurements to be made many times within a
borehole and are relatively cheap and robust. In Fig. 4.8, the raw data
recorded during a USBM overcoring stress measurement test are shown.
It can be seen that the effect of removing the pre-existing stress
components has been to cause an expansion along all three diameters, with
one of the deformations, u3, in the figure, being more than the other two
deformations.
The measurement of a diametral displacement is analogous to the use
of a flatjack for measuring the normal stress component. In a similar way
to the flatjack, each measurement of displacement effectively permits cal-
culation of one normal strain. Through the use of the stress transformation
equations, it is possible to calculate the principal components of the biaxial
stress state and their orientations. There is, however, the added com-
plication of the presence of the borehole, which perturbs the stress state
from its natural in situ state.

4.3.4 The CSIRO overcoring gauge

This device operates on a similar principle to the USBM torpedo except that
it is a gauge which is glued into the borehole and can measure normal
strains at a variety of orientations and locations around the borehole wall.
The gauge is glued into position within the pilot hole, initial readings of
strain are taken and the gauge is then overcored. This destresses the
resulting hollow cylinder and final strain gauge readings are taken. The

The traces are the electrical output from the device

plotted against time during overcoring and hence
illustrate the evolution of diametral change during
overcoring.

-O.OO~ r

measurement
for each curve
u3
0.020
\.-.-.-
0 100 200 300 400 500
Depth of overcoring bit (mm)

Figure 4.8 Data obtained during a USBM overcoring test.

 Methods OF stress determination 51

Figure 4.9 The CSIRO overcoring gauge. (a) The CSIRO gauge. (b) Installation of
the gauge. (c) A sectioned hollow cylinder core containing a CSIRO gauge.
52 In situ stress
gauge has either 9 or 12 separate strain gauges, in rosettes of three, so there
is some redundancy in the measurements-thus permitting statistical
analysis of the data. Alternatively, if the rock is assumed to be transversely
isotropicrather than completely isotropic, then the extra readings allow the
stress state to be calculated incorporating the rock anisotropy. For a fuller
discussion of anisotropy and the numbers of associated elastic constants,
the reader is referred to Chapters 5 and 10.
One major advantage of this and similar gauges is that the resulting
hollow cylinder is retrieved from the borehole and can be subjected to
laboratory testing under controlled conditions in order to determine both
the functionality of the system (e.g. whether any strain gauges have
debonded, whether the cylinder is composed of intact rock, etc.) and the
necessary elastic constants.
As with all the methods discussed, this technique has its limitations and
disadvantages. One major problem is the environment within the
borehole: prior to gluing the gauge in place, the surface of the wall can
easily become smeared with material deleterious to adhesion; if the drilling
fluid is at a different temperature to the rock, then thermal expansion or
contraction of the hollow cylinder can lead to misleading strains being
induced; and the long-term stability of the glue may not be compatible with
the installed life of the gauge. Against this are the factors that the gauge is
relatively cheap, it contains built-in redundancy (both electrical and
mathematical) and, uniquely of the four methods described here, the
complete state of stress can be established with one installation.

4.4 Statistical a n a l y s i s of stress s t a t e data

With repeated measurements of a variable, it is customary scientific practice
to apply some form of statistical treatment for the purpose of establishing
the accuracy and precision of the measurement system. Thus, when a scalar
quantity is being measured, the mean and standard deviation are
conventionallyused as measures of the value and its variability. However,
a scalar is defined by only one value, whereas, in the case of the stress
tensor, there are six independent values. This has crucial ramifications for
averaging a number of stress tensors and for specifying the variability of
the stress state.
We have explained that the stress state is normally specified via the
magnitudes and orientations of the principal stresses. So, if a number of
stress measurements have been made in a particular region, it is very
tempting to estimate the average stress field by averaging the principal
stresses and their orientations separately, as demonstrated in Fig. 4.10@).
This is incorrect:it is wrong to take the average of the major principal stresses
in a number of stress tensors-because they may well all have different
orientations. The correct procedure is to find all the stress components with
reference to a common reference system, average these components, and then
calculate the principal stresses from the six values of averaged components,
as demonstrated in Fig. 4.10@)and the box in the text. Note also that each
of the six independent components of the stress tensor has its own mean
and standard deviation: these will generally be different for each of the six