Int. J. Rock Mech. Min. Sci. & Geomech. Abstr. Vol. 15, pp. 167-178. 0020-7624/78/0801-0167502.

00/0
© Pergamon Press Ltd 1978. Printed in Great Britain

The Hydrofracturing Stress Measuring

Method and Recent Field Results
B. C. HAIMSON*

The hydrofracturing technique is a recent development in the area of stress

measurements in rock. Unlike most other methods, it does not measure strain
at a point through the use of delicate instrumentation in the test-hole. Rather,
it directly determines average stresses over large areas by recording two
hydraulic pressures, one necessary to crack open a segment of the test-hole
and the other required to keep the fracture open. To do so, it uses simple
down-hole mechanical tools so that the method can be employed at any depth
from the surface. Elementary elastic relationships exist between recorded
pressures and in-situ stresses, and between fracture direction and stress orien-
tation. Our laboratory experiments have confirmed these relationships, and
in the last five years we have conducted a number of successful field measure-
ments throughout the United States and elsewhere, at depths between very
near the surface and 5000 m. The results of these measurements, details
of which are given in the paper, have been used in studying earthquake
control problems (Rangely, Colorado), underground power station design
(California, South Carolina), hard rock tunnel design (Wisconsin), and geology
and plate tectonics (all the measurements and particularly those in Michigan
and Iceland). A general uniformity of maximum principal stress direction
throughout the continental United States has been established. Stress magni-
tudes are affected by regional geologic conditions but generally show linear
increases with depth.

INTRODUCTION rock elastic parameters in order to convert measured

strain to stress; (2) the very small areas at which
The magnitude and direction of rock stresses are stresses are actually determined; and (3) the ineffective-
among the most important factors affecting the stability ness in zones of highly differential stresses where over-
of surface and underground structures, the driving coring could produce discing.
mechanisms of plate tectonics, the prediction and con- The only thing that hydrofracturing (or hydraulic
trol of earthquakes, and the in-situ extraction of coal fracturing) has in common with the other methods is
gas, leached copper, oil shale, and dry hot-rock geo- that it is conducted in boreholes. However, it does not
thermal energy. The challenge of reliably measuring require overcoring and is not limited in depth except
rock stresses has fascinated many rock mechanics by the length of the drilled borehole (stresses in a 5-km
scientists and field experimentalists. However, the per- well were recently measured, as described elsewhere in
fect method of determining in-situ stresses has thus far this paper). It determines stresses directly and is not
eluded their ingenuity. dependent on knowledge of precise values of the elastic
Most of the methods developed to date require a rock parameters. It estimates stresses over large areas
borehole drilled from the point of access to the depth providing a better average than point measurements.
where stresses are to be determined. With the exception It is not affected by highly differential stresses. Last,
of hydrofracturing, all these methods use strain or but not least, it does not employ sophisticated down-
deformation gages capable of sensing changes in diame- hole instrumentation, but rather rugged equipment,
tral, circumferential or bottom-hole dimensions when which is in part commercially available and ready to
existing stresses are relieved. Stress relief requires 'over- be rented or purchased.
coring', and it is this operation that is primarily respon- The hydrofracturing stress measuring technique con-
sible for the limit on the effective length of test-holes sists of sealing-off a section of a borehole at the
(50m or less). Other distinct disadvantages of these required depth by means of two inflatable rubber
methods are: (1) the requirement to precisely determine packers, and pressurizing the isolated segment using
a hydraulic fluid such as water. At some critical (also
* Professor of Rock Mechanics, Department of Metallurgical and called breakdown) pressure, the rock at the borehole
Mineral Engineering, University of Wisconsin. Madison. WI 53706,
U.S.A.
bursts and develops a tensile fracture. This fracture can
167
168 B . C . Haimson

be extended away from the hole by continuing the

f
where ~ is the rock weight gradient, and d is the
pumping. When the pump is shut off with the hydraulic depth.
circuit kept closed, a 'shut-in' pressure is recorded. This If av is the least principal compressive stress, a verti-
is the pressure necessary to keep the fracture open. The cal fracture will nonetheless initiate at the borehole
breakdown and shut-in pressures, which are carefully wail, yielding the first shut-in pressure (P~I). Often the
monitored during the test, can be related to the prevail- fracture will run into a nearly horizontal weakness
ing in-situ stresses. An impression packer or other bore- plane and open it up. If this plane crosses the pressur-
hole and surface geophysical devices can be used to ized interval a second shut-in pressure (P.,2) will be
determine the orientation and inclination of the hydro- recorded. Clearly Ps~ > Ps2 and
fracture which develops along a plane perpendicular
to the direction of the least principal stress. In this P s i = O'Hmin (3)
manner, both the magnitudes and the directions of the Ps2 = fly" (4)
principal stresses can be evaluated. The hydrofracturing
In this situation, both the least horizontal principal
theory, its limitations, laboratory results and selected
stress and the vertical stress will be directly determined
field experience to date are detailed in the following
by hydrofracturing pressures.
sections.
Often the shut-in pressure will not stay constant as
C A L C U L A T I O N O F STRESSES explained above. It is then advisable to obtain several
pressure plateaus for various flow-rates of fluid pumped
Theoretical expressions relating hydrofracturing
into the fracture. If pumping is very slow (say, in the
pressures to in-situ stresses have so far been developed
range of a few l/min) and if simultaneous time-base
only for linear elastic and isotropic rocks in which one
recording of both pressure and flow-rate (Q) is con-
of the principal stresses acts in a direction parallel to
ducted, a simple relationship resulting from equations
the axis of the test-hole. The parallelism prerequisite
formulated by Perkins and Kern [3] and suggested by
between hole axis and a principal stress direction is
Aamodt (R. L. Aamodt, personal written communica-
usually achieved by drilling vertical test-holes, since the
tion, 1975), can be used to obtain shut-in pressure (P~)
overwhelming field evidence is that one of the principal
values :
stresses is nearly vertical in most locations.
(P1 - P s ) / ( P 2 - Ps) = ( Q I / Q 2 ) 1/2 (5)
Two distinct situations exist. In one, the vertical
stress (av) is the least principal compressive stress, and where P1 and P2 a r e two pressure levels corresponding,
in the other the vertical stress is either the intermediate respectively, to the two flow-rates, Q1 and Q2.
or the largest principal stress. The hydraulic fracture In order to estimate the value of the major horizontal
will follow the path of least resistance, i.e. a direction principal stress (am.,x) the poro-elastic relationship
perpendicular to that of the least-principal stress. between the critical (breakdown) pressure (P,) necessary
Theoretical calculations [-1], as well as abundant labor- to induce a vertical hydrofracture and the two horizon-
atory [2] and field results (this paper) show that when tal principal stresses is used [-2,4,5]:
rubber packers are used to seal off an unprefractured
interval, the initial hydrofracture is always vertical and T + 3trUmin -- O'Umax -- 2Po
Pc - P0 = (6)
perpendicular to the least horizontal principal stress K
(am, i.), regardless of the magnitude of av. If av is the where compressive stresses are taken as positive and:
least principal stress, the hydrofracture will turn around P0 is the pore pressure in the rock at the tested
as it travels away from the stress field imposed by the depth;
pressurized test-hole, and will become horizontal. T is the hydrofracturing tensile strength and is equal
Alternatively, existing sub-horizontal weakness planes to Pc when cru,,i,, = ou ..... = Po = 0 and K = 1;
may open up, thus diluting the pressure in the hydro- K is a poro-elastic parameter which can be indcpen-
fracture and stopping its advance. If av is not the least dently determined in the laboratory. The range of K
principal stress, the induced vertical hydrofracture will is 1 < K < 2. K = 1 when the formation is imper-
extend along its initial direction. meable to the fracturing fluid. K = 2 when the rock
The shut-in pressure (Ps) needed to keep the hydro- matrix compressibility and the rock bulk compressibi-
fracture open when pumping is stopped is equal to the lity are equal, or when the Poisson's ratio equals 0.5
in-situ compressive stress perpendicular to the fracture [2,5].
plane. If Crv is not the least principal compressive stress, In practice, the values of T and K can be derived
P~ will stay constant provided there are no leakages from a plot of ( P c - Po) vs (3anmi,- 6u,nax- 2Pot.
through the pipe, past the packers, or into rock pores based on laboratory-simulated hydrofracturing tests in
and joints, and the relationship which the principal stresses are known since they are
P~ = crH~i. (1) the applied loads. The trend emerging from such series
of tests in 5 rock types [6] is presented in Fig. 1. Note
will determine the least horizontal principal stress that in these tests P0 was kept at zero. As expected,
(am, i,). The vertical principal stress (av) will then be K is not constant as it varies with the amount of pres-
calculated from the weight of the overlying rock:
sure applied to the rock, and therefore the observed
~,d = av (2) relationship between (Pc - P~) and (3am,,, , crn .......
 The Hydrofracturing Stress Measuring Method 169

1 fl i I the vicinity of the test-hole is determined from the

results of two operations: pressurization and impres-
1.0 sion.
$ K"l'-~J /
0.5 LABORATORY AND PRELIMINARY FIELD
RESULTS
Extensive laboratory-scale experiments have been
Od~ ~ 1 0.5~'' I
1.0
I
1.5
I
2.0 2.5 conducted to verify the relationships between hydro-
fracturing pressures and in-situ stresses and between
3~rHmin-- O'Hr.a x (kb)
fracture orientation and stress direction [2,6-10]. The
Fig. 1. Relationship between b r e a k d o w n pressure and horizontal in- highlights of the experimental results are:
situ stresses as obtained by averaging results in five hard rock types 1. All the induced hydrofractures are tensile ruptures
and normalizing them for T = 0. In these tests, P0 was held at zero.
The rocks tested were: Tennessee sandstone, Tennessee marble, and no shear fractures are observed.
Weber sandstone, Dresser basalt and Valders limestone[6]. 2. When rubber packers are used, all hydrofractures
were vertical and perpendicular to a m , i,.
2P0) is not linear. However, in the approximate range 3. The critical pressures for vertical fracture initiation
of 0 < Pc - Po - T / K < 250 bars, K is approximately is related to the horizontal principal stresses as shown
1.0. Beyond 250 bars. the value of K increases at in Fig. 1.
various rates and approaches asymptotically the value 4. Studies in foliated slate and pre-cracked quartzite
of 2.0. Roughly, between 250-500 bars K can be show that provided sufficient laboratory testing is con-
approximated at 1.5 and beyond that at 2.0. In all the ducted, such rocks are amenable to hydrofracturing
field tests undertaken by us to date, the hydrofracturing stress measurements within limit. In some cases, lining
pressures were such that K = 1.0 was always the best of the test-hole may be required to ensure pressuriza-
approximation of the poro-elastic parameter. tion and vertical fracture initiation.
Depending on the relative magnitude of av, equa- The first indications of the potential of hydraulic
tions 1, 2 and 6 or 3, 4 and 6 determine the values fracturing as a stress measuring technique came from
of all three principal stresses. calculations reported by Scheidegger [11] and Kehle
Based on the sound theoretical relationship, on the [12] based on routine oil-well hydraulic fracturing jobs
consistency of results when several hydrofracturing run for production stimulation purposes. Haimson and
tests are run at the same location or in the immediate Stahl [13] reported on three series of oil field hydro-
meighborhood, and on the close correlation between fracturing jobs (in the states of New York, Illinois and
the Ps value obtained by using equations (1), (3) or Ohio) in open holes, where impression packers and bot-
(4) and that obtained from equation (5), it can be firmly tom-hole pressure transducers were used to determine
stated that the smallest principal compressive stress is crack orientation and fracturing pressures, respectively.
accurately determined by the hydrofracturing method. The results within each group of tests were very consis-
The value of am, ax which is obtained by using equation tent with respect to critical and shut-in pressures, and
(6) is necessarily only an estimate because of the fracture directions. The significance of the consistency
approximated value of K and the assumption of linear of results was the strong indication that they were
elasticity. The value of gv as determined by hydrofrac- closely related to local in-situ stresses. Stress calcula-
turing is always found to be closely approximated by tions based on these hydrofracturing jobs were subse-
the weight of the overlying rock (see field results, this quently published by Haimson [8].
paper).
As stated above, the direction of hydrofracturing can
be determined by a number of methods. The most com-
MEASUREMENTS T H R O U G H O U T THE
mon and reliable way is to use an impression packer,
UNITED STATES
an inflatable sleeve covered with a sheet of very soft
rubber. When forced against the wall of the borehole Following the laboratory study and the initial field
the soft rubber takes an imprint of the rock face condi- results, all indications were that hydrofracturing had
tion and maintains a clear picture of it, long after the a great potential as a deep-hole stress measuring tech-
packer has been deflated and raised to the surface. nique. Our opportunity to test the method came in
Packer impressions are oriented by employing a bore- 1971 at Rangely, Colorado. The success of the Rangely
hole magnetic surveying tool or other techniques. Im- experiment led to more measurements sponsored either
pressions of successful hydrofractures in unprefractured by government agencies or private industry. We have
rock invariably yield vertical fractures, the directions been involved in a number of significant hydrofractur-
of which are consistent within the same area, and which ing stress measurements throughout the United States
determine the directions of a m , i, and aHmax. When a v and Iceland in relation to a variety of mining, civil
is the smallest compressive stress both a vertical and engineering, geological and geophysical projects. The
a horizontal hydrofracture are often traced by the im- more important of these measurements are described
pression packer. Thus. the complete state of stress in in the following sections.
170 B . C . Haimson

~N boundary of the oil field, near the earthquake-prone

/'*'~ ~-- ~ " " " " . -.,.,,,~ RESERVOIRPRESSURE region (Fig. 2). The results obtained at the bottom of
I ,,""--~~"><" co~rrou~ the well (1900m below the surface), were: av = 435
I/. _ _ . . . _
bars (vertical), an,,~. = 315 bars (horizontal at N 20°W),
V i/ /i_ - - \ - - ' . , =~-'BDRY OF
"\ k \ (.~ ,'.-.~ k " , , ~NC-B_Y aH,,ax = 590 bars (horizontal at N 70°E).
...... ( 41 "Z The hydrofracturing test showed that the vertical
principal stress was intermediate in magnitude, strongly
EARTHQUAKE'%. ~ 7/7/ yb / \ indicating strike-slip faulting. The horizontal principal
stress directions were in accord with both the N 50~E
strike of the fault and its right lateral slip direction.
• I-NOR. FRAC. WF-.LL ... . . . . . . ....s The magnitudes of the measured principal stresses were
o EXPERIMENTAL WELL
used to predict a critical pore pressure of 240 bars for
e • | i i
0 4 km fault slip to occur (Fig. 3). This value was within 10','~
of the earthquake related threshold pressure monitored
Fig. 2. The Rangely Oil Field and the area where most earthquake
epicenters concentrated. Shown also are the pressure contours (in at the site (Fig. 4). The Rangely stress measurement
bars) in the oil-bearing Weber sandstone formation, the major fault, was considered very successful because of the close cor-
the 4 experimental wells through which reservoir pressure in the relation between the determined stresses and other field
earthquake zone was controlled and monitored, and the hydrofrac-
turing well.The short line crossing the latter is in the direction of observations.
the largest horizontal compressive stress.
~b) Ground stability under nuclear detonation-- Nevada
Test Site
(a) Earthquake research--the Rangely Experiment
Knowledge of the existing in-situ stress magnitudes
The first major utilization of hydrofracturing as a
and directions were required at the Nevada Test Site
deep-borehole stress measuring technique was in con-
for stability evaluation of tunnel complex U12n during
nection with the earthquake control research at
underground nuclear tests. Previous measurements
Rangely, Colorado [14-16]. Intense seismic activity
using an overcoring method were conducted within
centered in and around the Rangely Oil Field in the
only 5 m from the tunnel wall [18]. We performed
vicinity of a strike-slip fault had been recorded in the
hydrofracturing measurements both in a 250-m hole,
area for over a decade (Fig. 2). A primary objective
drilled from the slope of the mesa to the level of the
of the research program was to determine whether a
tunnel, and in vertical and horizontal 25-m holes drilled
correlation existed between earthquake triggering and
from the tunnel wall (Fig. 5). The horizontal holes were
formation pore pressure. The latter had been artificially
drilled in the general direction of the expected mini-
raised through water flooding of the field. A quantita-
mum stress and only shut-in pressures were obtained
tive solution relating pore pressure to slip initiation
because instead of fracturing, existing discontinuities
along the fault, which in turn could have triggered the
perpendicular to the hole axis were opened up. Were
earthquakes, required knowledge of the in-situ stress
the horizontal holes drilled in the direction of the larger
condition in or below the oil-bearing strata. Hydrofrac-
principal stress, more information might have been
turing was selected as the stress measurement method.
gained. All the other tests, both in the tunnel wall and
The borehole used was a newly drilled oil-well located
in the deep borehole, enabled a full evaluation of the
approximately l km west of the fault by the southern
stresses [19]. A linear stress-elevation relationship was
established for all principal stresses (Fig. 6). The in-situ
stresses around the tunnel at 380 m below the surface
I I
were determined at av = 70 bars (vertical), a,,,,~, - 35
RANGELY bars (horizontal at N 55'W), a,,,,x = 90 bars (horizon-
v tal at N 35°E). Both the magnitudes and the directions
to of the principal stresses are in agreement with the bore-
to
laJ
n," ~ , ~ MEASURED hole deformation measurements [18], and seismic in-
I--"
to
f-CRmC.,. / / 1 I
vestigation results [20], indicating again the reliability
of the method.

lad
"1"
(c) Pre-excavation design of underground powerhouses
tO Helms and Bad Creek projects
0 The determination of the in-situ principal stress mag-
0 .2 ~Hmin ~ ~Hmox ~ 6 nitudes and directions is of crucial importance in the
NORMAL STRESS (kb) design of large permanent openings such as power-
houses which possess one very long axis reaching some-
Fig. 3. Mohr diagram showing the measured stress regime (right times 150 m, and two considerably smaller dimensions
circle) at Rangely, the critical effective stress conditions (left circle) (within 50 m). The orientation of the chamber and the
that could induce slip along the existing fault (whose normal is at
70' to the direction of a.m..0, and the amount of pore pressure layout of the penstock and tunnels leading to and from
rcqnired to produce the critical effective stresses (240 barsj. it depend on the stress distribution.
 The Hydrofracturing Stress Measuring Method 171

EARTHQUAKE FREQUENCY AT RANGELY OIL FIELD, COLORADO

200

Ikes
,in I kilometer of bottom
total wells .3

150

i¢1
o-i- ~ E D E D TO
t- INITIATE EARTHQUAKES mc
I00 O

1
ua
at
-r
Z 7,
O
50

'~o .I

~,-- FI.UIDINJECTION -!- , o,o

WITHDRAWAL
U,O,NJECTION J
1

Fig. 4. Relationship between earthquake frequency and reservoir pore pressure at Rangely. The threshold pressure was
closely predicted by hydrofracturing stress determination, as shown in Fig. 3 (after reference [17]).

Stress measurements in the vicinity of future power- The Helms Pumped Storage Project will be located
houses have been almost routinely conducted using about 100 km east of Fresno, California in the Sierra
overcoring methods. Because of the distance limitations Nevada Mountains. Analyses of the local conditions
of these methods, exploratory holes could not be used led to a design calling for a 98 m long, 25 m wide and
and expensive pilot tunnels had to be driven into the 36 m high powerhouse to be constructed some 300 m
planned excavation area. In some cases, pilot tunneling deep in a granitic rock. Topographic difficulties and
was not feasible and stress measurements were only lack of excavation permit precluded the driving of a
conducted during the actual excavation, rendering any pilot tunnel. The only logical decision was to use
change in shape or orientation of the original design hydrofracturing for stress measurements in the existing
extremely expensive. The use of hydrofracturing in deep deep exploratory holes.
exploratory holes provides the needed information well Nine successful hydrofracturing stress measurements
ahead of the final design stage and without necessitat- were conducted in two of the existing coreholes, seven
ing additional drilling and tunneling expenses. in a vertical hole between the depths of 119 and 326 m,
and two in an inclined hole at depths of 239 and 271 m
[21]. All except for the shallowest test yielded nearly
vertical fractures oriented between N 8~'E and N 55"~E
and averaging N 25°E. This result indicated that the

2000 2:50 E

",
ml \l. max -- LLI
E o \
. ,,2o \ o
•S / " ,;t .,oooo Z
-- 0

oo \

,Tso, , , \ i i\ ~, 470
0 40 80 120
STRESS, bars

Fig. 6. V a r i a t i o n of m e a s u r e d stresses with e l e v a t i o n o r w i t h d e p t h

Fig. 5. Tunnel complex U12n, Nevada Test Site, and location of f r o m t h e t o p of the m e s a at t h e N e v a d a Test Site t u n n e l c o m p l e x
surface test-hole (UE7) and underground test-holes (in drifts 01 U 1 2 n . T h e v e r t i c a l stress s h o w n is for the p a r t o f the t u n n e l u n d e r
and 06). the mesa.
172 B.C. Haimson

STRESS (bors) (53 58 bars). This internal pressure would not be suffi-
150
50 |oo cient to fracture the wall of the penstock, but was high
I l
enough to open up existing joints or other discontinui-
• • ¢t HELMS 2200 ties which were oriented perpendicular to the direction
O" , | "~ E of am, i,. To prevent any communication to the power-
z house, the length of steel lining in the penstock would
O
22~ I-.- have to be sufficiently long to prevent joints intersect-
E
ing any of the penstock branches at the end of the
p- X-O'Hmo X lad
..I lining from reaching the powerhouse cavern. The ori-
bJ ~-O'.min 2100 ginal layout called for the bifurcation branch to be
/° \ ",, -

oriented at N 30°E, which subparallels the direction of

300
obtained hydrofractures. To avoid opening up existing
discontinuities in an axial direction, the decision was
made to rotate the bifurcation branch about 90 ° so
Fig. 7. Variation of measured stresses with depth at the site of the that the minimum principal stress acted parallel to its
Helms Pumped Storage underground powerhouse.
length. These and other changes in the original layout
were easy to make in the early stage of the design.
principal stresses acted approximately in the vertical Were the stresses not determined in the pre-excavation
direction and the horizontal plane with the direction stage, considerable expense and time delays would
of the maximum horizontal stress (an.,ox) at N25°E. have been encountered in making any of the above
The magnitudes of an,.~x ranged from 54 to 100 bars, corrections.
displaying a linear rate of increase with depth, almost No layout changes have yet been made in the power-
identical to that of the vertical principal stress (Ov) (Fig. house chamber itself but the stress measurement results,
7). The latter was calculated from the weight of the together with other rock properties determined, are
overlying rock. The least principal stress (art,.i.) ranged being incorporated in a finite-element analysis designed
between 45 and 55 bars in the direction N 65°W. The to determine stresses and displacements in the crown
uniqueness of these measurements was that for the first and wall of the chamber and establish the stability of
time an inclined hole was used for hydrofracturing tests the cavity, its most suitable orientation and the amount
in order to verify the orientation of the principal of required reinforcement and support.
stresses. The Bad Creek Pumped Storage Project will be
At the approximate depth of the future powerhouse located along the southeastern edge of the Blue Ridge
(300m), the stresses as derived from Fig. 7 were Escarpment, in the upper part of Oconee County,
~rv = 82 bars, an,,,i, = 54 bars at N 65°W and aH,,.~ = South Carolina. The powerhouse chamber will contain
95 bars at N 25°E. four 250 MW generating/pumping units and its tenta-
Some important changes in the original layout were tively planned size is 27 m by 137 m and 40m high.
made as a direct result of the pre-excavation hydrofrac- As preliminary layout and design work progressed, it
turing measurements (Fig. 8). It was discovered that was realized that additional information would be
the minimum principal stress was apparently lower required for the location, orientation and design of the
than the static pressure expected in the penstock area underground facilities. In order to obtain the needed

Iollo ( I I
~ -\ DRAJINAGE GALLERY ELEVATO4~ SH/kCT

~i~ I~..~ RA NGE OF L 150m ,;;:

~/'HYDRO FRACTURESI~t~C~[ /:;'" if-" .

EXCAVATION POWERHOUSE COMPLEX

Fig. 8. Original and modified layout of the underground powerhouse complex Helms Pumped Storage Project.
 The Hydrofracturing Stress Measuring Method 173

STRESS ( K b I underground annular tunnels. The tunnels will be sub-

O.I 0.2 0.3 jected to high mechanical pressures and are to be built
0.1 I t

%, x,, BAD (~REEK o - O" v in rock of unquestioned quality, strength and stiffness,
• %%
subjected to favorable in-situ stress conditions [22,23].
A
E \° /,,,o ..,,. O - O-H m i r a

X - 0" H rnox To test the applicability of a Pre-Cambrian granite site

\ O"Hmi n ~o x • • O"Hn')ax
at Montello, Wisconsin, a vertical corehole, 7.5 cm in
"I"
I-- 0.2
\ \, diameter was drilled and 7 hydrofracturating stress
0.
bJ \O'v(O,,e~u,~)~ "'.. x measurements were conducted between the depths of
75 and 188 m. Five of the borehole impressions after
\o "-
hydrofracturing yielded both vertical and horizontal (or
inclined along the planes of previously closed joints)
0.3 i i I
fracture traces, indicating that the least principal stress
Fig. 9. Variation of measured stresses with depth at the site of the acts in the vertical direction (av). All five vertical frac-
Bad Creek Pumped Storage underground powerhouse. tures were within -t-20 ° from the mean direction of
N 63°E. In these five tests, as in the Bad Creek Pumped
data, a preliminary location for the powerhouse was Storage case, two shut-in pressures were identified, the
established in the Toxaway gneiss and one NX size first corresponding to the vertical fracture and, hence,
hole was drilled vertically, approximately 275 m deep yielding the least horizontal principal stress (am, i.), and
into the zone of the future excavation. the second approximately equal to the overburden
Seven successful hydrofracturing stress measurements weight (av). The latter was determined in the laboratory
were conducted at different depths in the hole between as 0.25 bars/m × depth (m). The least horizontal stress
120 and 2 7 0 m [21]. Two levels of shut-in pressures appeared insensitive to the minor depth variations of
were recorded in most of the tests, the lower one always these tests and was limited between 62-82 bars. The
close to the calculated vertical stress based on the maximum horizontal stress (an,.,~) calculated from the
weight of the overburden. The impression packers recorded breakdown pressure, the first shut-in pressure
showed clear traces of both vertical and horizontal frac- and the laboratory determined hydrofracturing tensile
tures. The average orientation of the vertical fractures strength ( = 172 bars), varied between 140-200 bars (Fig.
in the powerhouse zone was N 60°E. The breakdown 10). For example, at 136 m depth the principal stresses
pressure and the first value of the shut-in pressure were were: or,, = 35 bars, an,.i. = 70 bars at N27°W,
used to estimate the magnitudes of the horizontal prin- trn,,, ~ = 160 bars at N 63°E. These results have been
cipal stresses. The results shown in Fig. 9 indicate a incorporated in a finite-element program in order to
linear increase in stress with depth; a remarkable co- determine the effect of the stress distribution on the
incidence in values of calculated and hydrofracturing magnet-tunnel stability.
determined vertical stresses; consistent least horizontal
principal stresses; and a rather wide scatter in the mag- (e) Crustal stress an ultra deep well in the Michigan
nitudes of the maximum principal stress. Based on Fig. Basin
9, at the depth of 230 m which is about the center of Cratonic basins, because of their subsidising tenden-
the planned powerhouse, the following in-situ stresses cies, preserve the most complete geologic record of con-
are expected: ~rv = 62 bars, am, i. = 159 bars at tinental interior histories. They are also the sites of
N 3ff~W, an,,,~ = 228 bars at N 60°E. large fossil fuel, salt, potash and oil shale deposits.
The stress results will be the basis for laying out Despite the importance of these basins, little is under-
a pilot tunnel into the powerhouse area. The pilot tun- stood about the mechanics of their subsidence, and,
nel will provide further geological and mechanical hence, no adequate models exist relating cratonic
details for the final design of the powerhouse, and in
particular, its orientation. Analysis of the underground STRESS (bars)
powerhouse by the finite-element method will be con- O IOO 200
ducted prior to the completion of the pilot tunnel using 1 ' I

the information obtained from the hydrofracturing tests MONTELLO

I
=
as a guide in selecting the appropriate ranges of t I
A IO(2
stresses. After the pilot tunnel is completed, the in-situ E ¢,_ O-Hmin t
stresses will be determined again using several methods o-o'- v
-r I
and the results will be compared to the values used I.-
rl J
hi
in the present analysis. O
Io '
I x
r O'Mmin i
(d) Underground storage caverns--Montello, Wisconsin ,, f'
I
The University of Wisconsin--Madison is currently I

studying the feasibility of constructing superconductive 2Oq

i
O
I
x
= i
I

magnets for electric energy storage in Wisconsin. The

doughnut-shaped magnets, 50-100min in radius and Fig. 10. Variation of measured stresses with depth in Montello,
about 5 × 10m in cross-section, will be mounted in Wisconsin.
174 B.C. Haimson

STRESS (kb) •

I0 0.5 1.0
DEVONIA~,,O ~ MICHIGAN
EW E L L

Z -SILURIAN ~
. . . . . ~-~x-- -- --~ - O'Hmift
ORDOVICIAN "~... N"~°'v
I
I1
....... -:.-\.
/ O-
%..
"r"
I,I
r~
.'AM..tA. ",~, \
/.: /: MICHIGAN ....
,~/ ,:: ~-

O BASIN
C..

~0
1 . . . . . . . . . . . . .

.5-LATE PRECAMBRIAN
-

b.
~-

\,
I Ix

" (bo~)
Fig. 12. Variation of measured an,,~, and of av with depth near the
center of the Michigan Basin.
/% DEEPI~
WIELLm .

O 3660m: c~n,,i,=645 bars, a H , , a ~ 2 8 8 5 bars. The

GRANDRAPIDS ..... . O
Prairie du Chien dolomite was tested at 2806m:
• " LANSING
am.i. = 400 bars, a n , , . ~ ~- 560 bars. The shallowest test,
DE'I'~OIT
run at 1230 m in the Devonian Amhertsburg formation,
,-I
gave a m , i, = 295 bars and anm,x - 505 bars. This a m . i ,
is considerably higher than expected from extending
the linear curve fitting the three deeper values, and is
very close to av (Fig. 12). This discrepancy could par-
tially be due to the discontinuity between the Devonian
and the Silurian (5--10 x 10 ~' years) and/or a result of
T~ OF TRENTON LIMESTONE (ORD~N)
stress relaxation in and above the Salina salt layers.
. . . . . . . .

A more detailed interpretation of the stress results can

be found in Haimson, 1978 [24]. Directions of the prin-
cipal stresses were not determined in these tests. How-
Fig. 11. Map of the Michigan Basin showing the contours on top ever, by interpolating k n o w n results in the Midwest
of the Trenton limestone and the location of the deep well where
hydrofracturing was carried out. and Southern Ontario, we speculate that o-n,.~ is
oriented at about N 6 0 ° E (Fig. 13).

events to global tectonic theory. A recently drilled, ultra

MEASUREMENTS IN ICELAND
deep oil well provided an excellent opportunity to con-
duct basic geophysical tests in the Michigan Basin in Iceland occupies a unique position in plate tectonics
an effort to find answers to some of the major puzzles theory because of its location at the boundary of the
related to continental interiors. One of the important N o r t h American Plate, on the axis of a mid-ocean ridge
tests undertaken was the evaluation of the stress regime underlain by shallow asthenosphere. The rift structure
in the basin. in Iceland and earthquake focal mechanism solutions
A series of five hydrofracturing tests were conducted along the mid-Atlantic Ridge support a crustal exten-
in the 5325-m deep well in Gratiot County, near the
axis of the Michigan Basin (Fig. 11). These were prob-
ably the deepest stress measurements ever undertaken
[24]. The top three tests were run through perforations
in the casing and were limited by the availability of
well-cemented cased zones. The controlled manner of
hydrofracturing and the repeatability of results during
repressurizations suggest that the perforation tests were
probably as reliable as open-hole ones, at least with
respect to the smallest horizontal principal stress
(~r,,,,i,,). In calculating an ..... an arbitrary tensile strength
of 100 bars was used since no core samples were avail-
able for testing. The vertical stress (O-v) was estimated
at 0.250 bars/m x depth (m). The two deepest tests
were run in open-hole. At 5320 m, the m a x i m u m allow-
able pressure (limited by pipe strength) of 1 kbar was
not sufficient to hydrofracturc the gabbro zone. At
5110 m, a late P r e - C a m b r i a n red silty shale was hydro-
Fig. 13. Maximum horizontal stress directions in the Midwest and
fractured yielding an,,i,, : 960 b a r s , a n d a n , , ~ x ± 1470 Ontario based on hydrofracturing (blocked squares) and strain relief
bars. The Mount Simon sandstone was tested at (open circles} measurements [24].
 The Hydrofracturing Stress Measuring Method 175

-~.j
,2
J
J

IF
;R

Fig. 14. Map of Iceland showing the location of the mid-Atlantic Ridge boundaries (shaded area). Also shown are the
more reliable of Hast's overcoring locations (B), and the hydrofracturing sites (A). Straight lines at A and B indicate
direction of an,,,,~,. Representative focal mechanism solutions are shown by circles. Dark quadrants indicate direction
of (~Hrain.

sion condition perpendicular to the ridge [25]. How- cedures and calculations are given by Haimson and
ever, in-situ overcoring stress measurements by Hast Voight [27].
[26] conducted in shallow boreholes around Iceland The calculated stresses are plotted with respect to
indicate relatively high horizontal compressive stresses. depth in Fig. 15. The data suggest that each borehole
These results do not appear compatible with ocean site represents a distinct stress population. The quality
floor spreading processes, nor are they in accord with of fit achieved by linear regressions is excellent for ~rn,,i,
focal mechanism solutions for Icelandic earthquakes in both instances; the fit is good with respect to a n , . , x.
(Fig. 14). Since Hast's data were probably affected to The results for H32 are very consistent. Both an,.i,
some extent by near surface topographic and thermal and an,..x increase steadily with depth at about the
perturbations rather than by deep-seated tectonic forces same rates, which are substantially lower than the rate
it was considered essential that a set of deep measure- of increase of overburden pressure. The relative orien-
ments by a different technique be undertaken. tation of crn,.,x remains virtually constant with depth
During the summer of 1976 we conducted seven suc- at N 23°W 4- 3 °. The results of H18 are also relatively
cessful hydrofracturing experiments in two boreholes consistent: an,.~, and a v seem to increase steadily with
(H18 and H32) in Quaternary igneous rocks in Reykja- depth at about the same rate; an,.ox also increases with
vik, Iceland, on the flank of the Reykjanes Peninsula depth, but perhaps at a higher rate. Throughout the
continuation of the mid-Atlantic Ridge. The holes are depth range tested the vertical stress is the intermediate
located about 20-25 km northwest of the axis of the principal stress. The data from each series of tests dis-
active zone of rifting and volcanism (Fig. 14). These plays internal consistency, although different states of
holes are relatively small diameter observation wells stress are suggestive for boreholes H32 and H18 which
drilled 15-20 years ago near the periphery of the are only 2 km apart. On the whole, the data indicate
Laugarnes hydrothermal system. a dominant N N W to NW orientation for anm,x; at H18
The major problem with the hydrofracturing tests the stress state favors strike-slip faulting (for
was the pre-existing joints in the test-holes. In success- d > 150 m), whereas this regime seems restricted (say
ful tests these joints appeared to be sufficiently tight 100m < d < 250m) at H32. A normal fault regime
or healed so as not to interfere with pressurization. seems dominant at H32 (for d > 250 m).
In such cases a vertical hydrofracture was induced. To The hydrofracturing stress measurements at Reykja-
verify that hydrofractures were actually obtained, two vik suggest a dominant regional orientation of am.ax
packer impressions were taken in some tests, one before approximately perpendicular to the axial rift zone (Fig.
and one after pressurization. In each test both impres- 14). This orientation is furthermore supported by Hast's
sions showed pre-existing inclined cracks in the rock shallow overcoring measurements in southeast Iceland.
but only the impressions taken after hydrofracturing This state of stress is fundamentally different from that
showed also the existence of a vertical fracture travers- in the axial rift zones themselves (see focal mechanism
ing the length of the interval. More details on test pro- solutions. Fig. 14). In the rift zones, an,,,~,, is consistently
176 B.C. Haimson

STRESS (bors)
4O 8O 120 40 80 120
w !
i I i i •
o.t

H 18
%\-.
\\ ix ~k~,,
~ "0.2 I \ ". ".,..
E
\ -L 1
' HYOROFRACTURING N
\ \" N "",.-,~ I~Hmax
-I"
\
I s. ,c,.,.o \ \ "-.,
o.~ I" \Q%-. Ov ""-.,
~ 0
\ \.
0.4
\ \
_ i L i
i _ L

Fig. 15. Variation of measured stresses with depth in two boreholes in Reykjavik, Iceland.

aligned perpendicular to individual rift zone fissures seismicity in America. Although northeast is the major
and faults. For detailed interpretation of these results, direction trend, local deviations up to +45 ° do exist
see Haimson and Voight [27]. which necessitate measurements where more precise
stress orientation is required.
The ability to use hydrofracturing for stress deter-
mination at depths never attempted before is beginning
CRUSTAL STRESS IN THE UNITED STATES
to provide us with a more complete picture of the stress
The accumulating data on crustal stresses in the profile. Figure 17 shows the variation of horizontal and
United States, based on recent hydrofracturing vertical stresses with depth in the United States down
measurements, and corroborated by other stress deter- to 5 km, and is based only on hydrofracturing measure-
minations and geological evidence have a common de- ments. Similar profiles have been obtained before,
nominator in the direction of the major horizontal based on other stress determination methods
principal stress, which is consistently found to be ap- [26,29-31], but they were the result of measurements
proximately northeast (Fig. 16). Sbar and Sykes [28] conducted from underground openings and were
have related this phenomenon to plate tectonics and limited in the overall depth.

- ' - " o'Hrnox DIRECTION

Fig. 16, M a x i m u m horizontal principal stress directions in the continental United States based on the orientation of
vertical hydrofractures. Data obtained from measurements described in this paper and in references [l,8, 33-43] and s u m m a r -
ized by H a i m s o n [32].
 T h e H y d r o f r a c t u r i n g Stress M e a s u r i n g M e t h o d 177

s~ess (XK)Zbo.) in anm,x m a g n i t u d e s , w h i c h are also m o r e widely scat-

5 I0 t e r e d with r e s p e c t to the a v e r a g i n g curve. M o r e o v e r ,
~ *
Oo
Fig. 17 is n o t i n t e n d e d for use in lieu of m e a s u r e m e n t s .
L o c a l stresses c o u l d significantly vary from the p l o t t e d
a v e r a g e as c a n be verified f r o m the results of the
m e a s u r e m e n t s d e s c r i b e d in d e t a i l in this p a p e r .

AG'Hn~ CONCLUSIONS
• Oi.knox
H y d r o f r a c t u r i n g is the o n l y a v a i l a b l e m e t h o d of
stress d e t e r m i n a t i o n in d e e p holes. In the last decade,
t:: it has seen ever i n c r e a s i n g use in the U n i t e d States
a n d elsewhere. W e h a v e e m p l o y e d it in r e l a t i o n to
l-
t-
'g, e a r t h q u a k e c o n t r o l research, p r e - e x c a v a t i o n u n d e r -
fL
• • %\\ • g r o u n d o p e n i n g design, g r o u n d stability u n d e r n u c l e a r
d e t o n a t i o n , a n d s t u d y of t h e l i t h o s p h e r e in M i c h i g a n
\ and Iceland.

\. \ H y d r o f r a c t u r i n g results to d a t e i n d i c a t e t h a t s t r e s s
d i r e c t i o n s in the c o n t i n e n t a l U n i t e d States are r a t h e r

\ X~
uniform, while the increase of stress with d e p t h a p p e a r s
to b e l i n e a r for the r a n g e of d e p t h s tested.
\
\ ~k Acknowledgements--This contribution is based in part on a paper

\ \ delivered at the ISRM Symposium on Investigation of Stress in Rock,

held in Sydney, Australia in August 1976.
The work reported was supported in part by the United States
\ \ Geological Survey (grants 14-18-0001-12281 and 14-08-0001-6118),
United States Bureau of Mines (contract HO220080), National
Science Foundation (grants EAR 76-02952, EAR 76-03821, EAR
72-03564), Defense Nuclear Agency (contract DNA 001-73-CO212),
Fig. 17. Variation of principal stresses with depth in the continental American Petroleum Institute, Wisconsin Electric Utilities Research
United States based on hydrofraeturing results. Data obtained from Foundation, Pacific Gas and Electric Co., Duke Power Co., and
measurements described in this paper and in references [1,8,33~,3] the Wisconsin Alumni Research Foundation.
and summarized by Haimson [32]. Among the graduate students contributing to the success of our
measurements special thanks are due to J. Avasthi, L. Cheung, T.
Doe, J. Edl, S. Erbstoesser and K. Kim.
T h e h y d r o f r a c t u r i n g r e s u l t s have been fitted with
linear curves given by the f o l l o w i n g r e l a t i o n s h i p s : Receiced 19 Auoust 1977.

anmi, = 20 + 0.16 d REFERENCES

an,,,x = 75 + 0.24 d 1. Roegiers J. C. The development and evaluation of a field method
a v = 0.25 d for in-situ stress determination using hydraulic fracturing. Ph.D.
thesis, University of Minnesota (1975).
where stresses a r e in b a r s a n d d e p t h s (d) in m e t e r s 2. Haimson B. Hydraulic fracturing in porous and nonporous rock
and its potential for determining in-situ stresses at great depth.
(limited to 0 - 5 0 0 0 m). T h e v e r t i c a l stress is the result Ph.D. thesis, University of Minnesota (1968).
of b o t h o v e r b u r d e n weight a n d h y d r o f r a c t u r i n g 3. Perkins T. K. & Kern L. R. Width of hydraulic fractures. J.
m e a s u r e m e n t s . T h e r e l a t i o n s h i p s s h o w t h a t at s h a l l o w Petrol. Technol. 13, 937-944 (1961).
4. Blot M. A. & Willis D. G. The elastic coefficients of the theory
depths the mean horizontal stress given by of consolidation. J. appl. Mech. 594-601 (1957).
aHavo -= 50 -t- 0.20 d tends to b e higher t h a n the vertical 5. Haimson B. & Fairhurst C. Initiation and extension of hydraulic
stress, as p r e v i o u s l y n o t e d b y H a s t [26] a n d others. fractures in rock. Soc. Petrol. Enyr9 J. 7, 310-318 (1967).
6. Edl J. N., Jr. Mechanical instability of deep wells with particular
B e n e a t h 1 k m t h e m e a n h o r i z o n t a l stress is l o w e r t h a n reference to hydraulic fracturing. M.S. thesis, University of Wis-
the vertical stress and b e c o m e s 0.8 av at 5 k m . T h e consin (1973).
t r e n d is s i m i l a r to that o b s e r v e d by G a y [30] in S o u t h - 7. Haimson B. & Fairhurst C. ln-situ stress determination at great
depth by means of hydraulic fracturing. In Rock Mechanics--
ern Africa. It is to be r e m e m b e r e d t h a t m o s t of the Theory and Practice; Proc. l lth Syrup. Rock Mechanics (Edited
h y d r o f r a c t u r i n g m e a s u r e m e n t s b e n e a t h 0 . 5 k m were by Somerton W. H.), pp. 559-584. A.I.M.E., New York (1970).
t a k e n in s e d i m e n t a r y r o c k f o r m a t i o n s , a s s o c i a t e d with 8. Haimson B. C. A simple method for estimating in-situ stresses
at great depths. In Field Testing and Instrumentation of Rock.
g e o l o g i c a l l y s t a b l e areas o f little seismic activity. ASTM Special Tech. Publ. 554, pp. 156-182 (1974).
F u r t h e r d e t a i l s on crustal stresses in the c o n t i n e n t a l 9. Haimson B. C. & Avasthi J. M. Stress measurements in anistropic
U n i t e d States h a v e been p u b l i s h e d by H a i m s o n 1-32]. rock by hydraulic fracturing. In Application of Rock Mechanics;
Proc. 15th Syrup. Rock Mechanics (Edited by Hoskins, Jr.), pp.
W e e m p h a s i z e here t h a t Fig. 17 is p r e l i m i n a r y in 135-156. ASCE, New York (1975).
n a t u r e a n d m a n y m o r e m e a s u r e m e n t s are n e e d e d 10. Haimson B. C. Determination of in-situ stresses around under-
before the s t r e s s - d e p t h r e l a t i o n s h i p s can be u n e q u i v o - ground excavations by means of hydraulic fracturing. Final
report, ARPA H0229980, U.S. Bureau of Mines (1974).
cally e s t a b l i s h e d . W e also stress t h a t o u r c o n f i d e n c e 11. Scheidegger A. E. Stresses in earth's crust as determined from
in the a c c u r a c y of anmi~ values is s o m e w h a t h i g h e r t h a n hydraulic fracturing data. Geolo¢lie Bauwes 27, 45--50 (19621.
178 B.C. Haimson

12. Kehle R. O. Determination of tectonic stresses through analysis 27. Haimson B. C. & Voight B. Crustal stress in Iceland. Pure appl.
of hydraulic well fracturing. J. geophys. Res. 69, 259-266 (1964). Geophys. 115, 1-38 (1977).
13. Haimson B, and Stahl E. Hydraulic fracturing and the extraction 28. Sbar M. L. & Sykes L. R. Contemporary compressive stress and
of minerals through wells. In Proc. 3rd Symp. on Salt, pp. seismicity in Eastern North America: an example of intra-platc
421~,32. Northern Ohio Geological Society, Cleveland, Ohio tectonics. Geol. Soc. Am. Bull. 84, 1861-1882 (1973).
(1970). 29. Herget G. Ground stress determination in Canada. Rock Mech.
14. Haimson B. C. Stress measurements in the Weber Sandstone at 6, 53 64 (1974).
Rangely, Colorado. EOS Trans. Am. Geophys. U. 53, 524 (1972). 30. Gay N. C. In-situ stress measurements in Southern Africa. Tec-
15, Haimson B. C. Earthquake related stresses at Rangely, Colorado. tonophys. 29, 447-459 (1975).
In New Horizons in Rock Mechanics; Proc. 14th Syrup. Rock 31. Ranalli G. & Chandler T. E. The stress field in the upper crust
Mech. (Edited by Hardy H. & Stefanko R.), pp. 689- 708. ASCE, as determined from in-situ measurements. Geol. Rdsch. 64,
New York (1973). 653-674 (1975).
16. Raleigh C. B., Healy J. H. & Bredehoeft ,I. D. Faulting and crus- 32. Haimson B. C. Crustal stress in the continental United States
tal stress at Rangely, Colorado. In Flow and Fracture of Rocks, as derived from hydrofracturing tests. In The Earth's Crust, pp.
pp. 275-284. Geophysical Monograph 16, Am. Geophys. U. 576 592. Geophysical Monograph 20. Am. Geophys. U. (1977).
(1972). 33, vc;n Schonfeldt H. A.. Kehle R. O. & Gray K. E. Mapping of
17. Wallace R. E. Goals, Strategy and Tasks of the Earthquake stress field in the upper earth's crust of the U.S. Final Technical
Hazard Reduction Program. Geological Survey Circular 701 Report to U.S.G.S. (14-08-0001-122278) (1973).
(1974). 34. Zoback M. D., Healy J. H. & Roller J. C. Preliminary stress
18. Hooker V. E., Aggson T. R. & Bickel D. C. ln-situ Determination measurements in Central California using the hydrofrac tech-
of Stresses in Rainier Mesa, Nevada Test Site. U.S. Bureau of nique. Pure appl. Geophys. 115 (1977).
Mines report (1971). 35. Bredehoeft J. D., Wolff R. G., Keys W. S. & Shutter E. Hydraulic
19. Haimon B. C., Lacomb J., Jones A. H. & Green S. J. Deep fracturing to determine the regional in-situ stress field, Piceance
stress measurements in tuff at the Nevada Test Site. In Advances Basin, Colorado. Geol. Soc. Am. Bull. 87, 250-258 (1976).
in Rock Mechanics, pp. 557-561. National Academy of Science, 36, Haimson B. C. Stress measurements in faults and their vicinities.
Washington, D.C. (1974). Final Report to U.S.G.S. (grant 14-08-0001-G118) (19771.
20. Hamilton R. M. & Healy J. H. Aftershocks of the Benham 37. Power D. V., Schuster C. L., Hay R. & Twombly J. Detection
nuclear explosion, Bull. seism. Soc. Am. 59, 2271-2281 (1969). of hydraulic fracture orientation and dimensions in cased wells.
21. Haimson B. C. Design of underground powerhouses and the im- J. Petrol. Technol. 28, 1116-1124 (1976).
portance of pre-excavation stress measurements. In Design 38. Strubher M. K., Fitch J. L. & Glenn E. E., Jr. Multiple vertical
Methods in Rock Mechanics; Proc. 16th Syrup. Rock Mech. fractures from an inclined wellbore--a field experiment. J. Petrol.
(Edited by Fairhurst C. & Crouch S. L.), pp. 197-204. ASCE, Technol. 27, 641 -647 (1975).
New York (1977). 39. Tyler L. D. & Vollendorf W. C. Physical observations and map-
22. Haimson B. C., Doe T. W., Erbstoesser S. R. & Fuh G. F. Site ping of cracks resulting from hydraulic fracturing in-situ stress
characterization for tunnels housing energy storage magnets. In measurements. Paper No. SPE 5542 presented at Soc. of Petrol.
Site characterization; Proc. 17th U.S. Syrup. Rock Mech., pp. Engrg. of A1ME, Fall Mtg (1975).
4B41-4B49. Utah Engrg Exper. Station, Salt Lake City, Utah 40. Haimson B. C. A stress measurement in West Virginia and the
(1976). state of stress in the Southern Appalachians. EOS Trans. Am.
23. Haimson B. C., Doe T. & Fuh G. F. Geotechnical investigation Geophys. U. 58, 493 (1977).
and design of annular tunnels for energy storage. In Preprint 4l. von Schonfeldt H. A. An experimental study of open-hole
Proc. of First Ira. Syrup. on Storage in Excavated Rock Caverns, hydraulic fracturing as a stress measurement method with par-
Stockholm. 2, 59-66 (1977). ticular emphasis on field tests. Ph.D. thesis. University of Minne-
24. Haimson B. C. Crustal stress in the Michigan Basin. J. geophys, sota (t970).
Res. (1978) In press. 42. Parsons R. C. & Dahl H. D. A study of thc cause of roof it~stabi-
25. Klein F. W., Einarsson P. & Wyss M. Microearthquakes on the lit 3' in the Pittsburgh coal seam. In Proc. 7th ('amsdia~7 R~'t,
mid-Atlantic plate botmdary on the Reykjanes Peninsula in Ice- Mechanics Syrup., Mines Branch Dept. of Energy, Mines and
land. J. geophys. Res. 78, 5084-5099 (1973). Resources, pp. 79- 87 (1972).
26. Hast N. Global measurements of absoh, te stress. Phil. Trans. 43. Zoback M. D. and Healy J. H. In-situ stress measurcments ncar
R. Soc. 274, 409-419 {1973). Charleston, South Carolina. EOS Trans. Am. Geophy. U. 58, 493
11977).