Regional distribution of in situ horizontal stresses in rocks of Southern Ontario

K . Y. Lo
Facdty of Engineering Science, University of Western Ontario, London, Ont., Canada N6A 3K7
Received May 11, 1978
Accepted May 12, 1978

Some structures built in different rock formations in Southern Ontario have been subjected to
various degrees of distress. These case histories include heaves of quarry bottoms, buckling of
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

concrete lining of canal floors, cracking of concrete lining of tunnels at the springline, and long
,term movement of the walls of unsupported excavations.
Inference from these case histories, together with direct measurements of in situ stresses,
indicate that high horizontal stressesexist in the Silurian and Ordovician rocks. The magnitude of
the maximum stress in the horizontal plane varies from 6-14 MPa depending on the depth and
rock type.
Excavations in rock relieve the in situ stresses. The stress relief serves as an initiating
mechanism for time-dependent deformation to occur leading to the process loosely termed as
'rock squeeze'. It appears, therefore, that due consideration must be given to this prevalent
phenomenon for the design of underground structures in rock in this region.

Des structures construites dans differentes formations rocheuses du Sud de I'Ontario ont subi
des dommages a des degris divers. Ces cas comprennent des soulevements de fonds de carritre,
des flambages de dallages de bCton dans les canaux, de la fissuration dans le garnissage de beton
de tunnels et un mouvement a long terme des parois d'excavation non retenues.
Les deductions a partir de ces cas, de mZme que les mesures directes des contraintes en place
For personal use only.

indiquent que des contraintes horizontales Blevees existent dans les roches siluriennes et or-
doviciennes. La grandeur des contraintes maximales dans le plan horizontal varie de 6 14MPa
dependant de la profondeur et du type de roche.
Les excavations dans le roc reliiche lescontraintes en place; ce relHchement de contraintes agit
comme mecanisme d'amor~aged'un deformation en fonction du temps qui conduit un
phinomene abusivement appele "etreinte rocheuse". En consequence, il semble que I'on doive
prendre en consideration ce phenomene lors de la conception de structures dans les roches dans
cette region.
Can. Geotech. J., 15,371-381 (1978) [Traduit par la revue]

Introduction One of the early attempts to measure in situ

It has generally been considered, until quite stresses in rocks was made by Hast in the 1950's
recently, that rocks in Southern Ontario are usually in Scandinavia, as described by Teriaghi ( 1962).
competent for engineering purposes. However, Subsequently, a large number of measurements
structures built in rocks in this area have exhibited were performed by Hast (1973). Tabulated values
distress in various degrees during or subsequent to of horizontal stresses measured in Fennoscandia in
construction, attributable to continuous deforma- granite, leptite, limestone, and quartzite in the
tions of the rock. This behaviour, sometimes period 1957-1 966 have been reported (Hast
loosely termed 'rock squeeze', appears to be a 1974) and the results are plotted in Fig. 1. In all
general phenomenon provided that the rock struc- cases, the horizontal stresses are much higher than
ture coincidentally meets certain loading and the vertical stress calculated from the overlying
boundary conditions. weight of the rock. It is clear from Fig. 1 that the
This problem appears to have been recognized, principal stresses in the horizontal plane, P and
or at least intuitively appreciated, as early as the Q (measured in megapascals) , increase linearly
beginning of this century, when engineers respon- with depth (measured in metres) giving the ap-
sible for the construction of the wheel pits of the proximate empirical equations:
Canadian Niagara and Toronto Power Plants in P = 12 0.06H +
Niagara Falls began to measure closure of the pit
walls in 1903 (Morison 1957; Hogg 1959). In a
study of the geological features and movements of Stresses that deviate from the above expressions
structures founded in rock in New York State, are attributable to nearby faults, which cause stress
Rose (1951) associated the rock squeeze with the anomaly. Several measurements performed by
possible existence of high horizontal in situ stress Herget (1974) in Northern Ontario conform
in the rock formations. approximately to the same relationships. From
 372 C A N . GEOTECH. J. VOL. 15, 1978

buckling of canal floors, (ii) movement of bridge

foundations, and (iii) cracking of tunnel walls
during or after construction.
(b) Excavations to moderate depth such as
wheel pit structures of the power plants in the
Niagara area (50 m ) .
(c) Tunnels at different depths such as the
Hydro tunnels in the Niagara area (up to 92 m ) .
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

C. Direct Evidence
In situ stress measurement performed at ( a )
Thorold Tunnel in Lockport Dolomite and Lime-
stone, ( b ) a tunnel in Meaford-Dundas Shale,
(c) a tunnel in Collingwood Shale, ( d ) Wesley-
ville Generating Station in Trenton Limestone, and
( e ) Ontario Power Generating Station in Lockport
Dolomite and Limestone, and Rochester Shale.
It is clear that the records listed bear different
weights in assessing the in situ stresses. They vary
from category A, from which only the existence
of stresses may be inferred, to category B, from
which the magnitude of stresses may be calculated
very approximately, to category C, from which the
For personal use only.

stresses are directly measured in magnitude and

FIG. 1. Increase of major and minor principal stresses orientation. It is of interest to examine and evaluate
P and Q in the horizontal plane with depth in Fennoscandia these records in some detail.
(Hast 1974).
General Geology and Rock Conditions
these results, it would appear that the existence The area under consideration is shown in Fig. 2.
of high horizontal stresses is a global phenomenon The geology in this region has been described
not limited to Ontario alone. (e.g., Hewitt 1972; Morton et al. 1975). Generally,
In this paper, an attempt is made to assemble the rocks are of sedimentary origin and of the
all evidence, both direct and indirect, of high Ordovician, Silurian, and Devonian periods. The
horizontal stresses in different rock formations in area is virtually free from seismic activity and the
Southern Ontario. A preliminary map showing the only known fault is the Batavia (Clarendon-
geographical and geological distribution of these Linden) Fault in New York State.
stresses is clearly of engineering importance and Regional joint sets have been studied by
might also have important implications in the Brigham ( 1971 ) in Southwestern Ontario and
structural geology of the area. Rose ( 1951 ) in New York. Both reported joint
sets strike in approximately east-west directions.
Evaluation of Evidence However, the joints are usually widely spaced and
The sites at which stresses are either directly are limited in extent. The rocks are usually com-
measured by the overcoring method or at which petent by conventional engineering standards, with
the existence of horizontal stresses can be inferred RQD usually exceeding 90% and recovery 100%.
from observations is shown in Fig. 2. These records The Collingwood and the Meaford-Dundas forma-
may be divided into three categories. tions are softer and horizontally bedded and the
A. Indirect Evidence recovery is poorer and occasional rubble zones are
( a ) Natural geological phenomena that occurred encountered at depth (Robinsky and Morton
in post-glacial period: (i) folds and faults, and 1973; Morton et al. 1975). Typical values of the
(ii) natural 'pop-up'. uniaxial compression strength and elastic modulus
( b ) Heave of quarry floors. of some of the rock units are shown in Table 1.

B. Evidence Derived from Performance Records Geological Observations

of Engineering Structures Geological features, such as faulting, folding,
( a ) Shallow excavations (up to 20 m ) : (i) and pop-up or buckling of surface rock strata
 m o
OYWNGS POINT X FOLDS
0 MARMORA QUARRY FAULTS
0 NATURAL 'W-uV
0 F L O W HEAVE IN
OUARRIES

A BUCKLING ff CANAL
FLOW
v anloGE F W
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

opEN EXCAVATION
WHEEL PITS

+ TUNNELS
STRESS MEASUREMENT
LAKE ONTARIO I ~n MPo 1

BED ROCK INFORMATION

PERIOD I FORMATION W GROUP

@ UVLUA

SILURIAN @ WELW - LOCKWRT

@ CLlNTON 8 U T I R I C T
For personal use only.

FIG.2. Locations of geological observations, case histories, and stress measurements in Southern Ontario.
TABLE
1. Representative values of strengths and elastic moduli from uniaxial compression tests

Uniaxial
compressive Modulus of
strength deformation
Rock unit Rock type (MPa) ( x 104 MPa) Source of data
- - - - --

Lockport Dolomite and limestone 100-200 3-7 Lo et a/. 1975

Rochester Interbedded shale and dolomite 20-40 2 Unpublished data
1975 of K.Y. Lo
Meaford-Dundas Interbedded shale, siltstone, 30 (weak layer) 2-4 Lo and Morton 1976
mudstone, and limestone 190 (strong layer) Morton et a/. 1975
Collingwood Interbedded shale and mudstone 20-70 0.7-2 (vertical) Lo and Morton 1976
1.4-3.5 (horizontal) Morton et a/. 1975

without any apparent change of external loading, strike north-south, similar in direction to the
may be interpreted as evidence of existence of Batavia Fault.
high horizontal stresses. The basis of such an inter- Many cases of heave of rock outcrops without
pretation is the fact that the in situ stress system any intervening activity of man in this area have
satisfies the appropriate failure criteria of the rock been observed by geologists. These are actually
mass. The relevance of these observations to the strength failure of the rock mass by buckling.
present problem can only be established when the The rock breaks up into blocks and slabs and
time scale of occurrence is considered. Therefore, forms a ridge running usually for a few hundred
only geological features that can be clearly estab- feet. Winder (C. G. Winder, private communica-
lished as post-glacial in origin are of interest. tion, 1973) recorded two recent cases of these pop-
White et al. (1973) reported four folds and two ups in limestone at Young's Point, north of Peter-
faults in the locations shown in Fig. 2. All these borough (Fig. 3 ) . One of these strikes southeasterly
are minor geological structures, but appear to be in the same direction as the heave at Marmora
post-glacial in origin. There appears to be no pre- Quarry (Fig. 4 ) . Similar observations have been
ferred orientation of the fold axes, but both faults reported in western New York State and in Ogdens-
 C A N . GEOTECH. J. VOL. 15. 1978
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

Post-glacial pop-up, Young's Point (courtesy of Geological Survey of Canada, taken by C. G . Winder,

burg, New York (Sbar and Sykes 1973). Unfor-

For personal use only.

tunately, information of the exact geometry, depth

of rock affected, and time of occurrence of these
pop-ups are rarely recorded. However, the mechan-
ism of failure does suggest that horizontal stresses
are present, approaching a magnitude sufficient to
cause buckling failure. The final stage of failure is
probably triggered by some mechanisms such as
freezing and thawing, thermal stress, and reduction
of mass strength of rock due to weathering.
The above geological phenomena, though in-
teresting in themselves, cannot be taken in isolation
as conclusive evidence of existence of high in situ
stresses. However, interpreted together with the
case histories of performance of structures de-
scribed in the next section, they form useful supple-
mentary information to define the regional pattern.

Case Records
A number of records of performance of different
types of engineering structures or excavations in
rocks are available. For a better perception of the
phenomenon, 'these will be discussed under three
categories by which the high horizontal stresses
manifest themselves, involving ( a ) strength failure
of the rock stratum, ( b ) short term deformation,
and (c) long term movements.
(a) Strength Failure of Rock Strata
FIG.4. Heave of quarry floor, Marmora Quarry (cour- (i) Buckling Failure o f Quarry Floor
tesy of Geological Survey of Canada, taken by C. G. Heave or buckling of the bottoms of several
Winder, 1951). quarries in this region is said .to have occurred,
 TABLE
2. Failure of bottom of open excavations in rocks

Depth of
excavation Length of
in rock Heave heave
Site (m) Rock type (m) (m) Remarks Reference
Dufferin Quarry, 18-21 Dolomite (Lockport) 0.6-0.9 < 76 Buckled in dolomite Golder Associates
Milton by sliding along Ltd. 1972
dolomite-shale
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

interface
Marmora Quarry, 15 Pre-Cambrian 2.4 % 150 Coates 1964
Ont. limestone
Barge Canal, 9 15 ft Lockport 0.6 > 90 Heave in shale Rose 1951
Lockport, NY. dolomite
15 ft Rochester
Shale
Queenston- 18 (max.) Lockport dolomitic 0.9 < 1000 Buckling of 15 cm Morison 1957 .
Chippawa Canal limestone thick floor slabs,
at least 3 cases

but only two incidences of this are recorded in

some detail. Information on these two quarries is
summarized in Table 2. The Dufferin Quarry in
Milton (Golder Associates Ltd. 1972) is of par-
For personal use only.

ticular interest and the heaves that occurred will

be discussed in some detail below.
The Dufferin Quarry is located on the Niagara
Escarpment (Fig. 2 ) . The rock formations at the
site consist of 23 m of Lockport Dolomite under-
lain by shale of the Queenston Formation. At the
time of field investigation (1971), the quarry had
been developed to an irregular-shaped rectangle
366 m wide and 732 m long. The depth of excava-
tion varied from 18 to 21 m into the Lockport
Formation. Thus a thickness of the cap rock vary-
ing from some 1.5 to 3.0 m was retained over the FIG. 5. Triangular void at base of heave, Dufferin
bottom of the quarry. Quarry (courtesy of Golder Associates Ltd.).
During the site investigation, at least seven
ridges of heaves were observed. These varied in tively intact (core recovery usually 100% ) brittle
length from 15 to 76 m and apparently had no dolomite is subjected to the in situ horizontal stress
preferred orientation. in a strained state. Thus a large amount of strain
At one of the heaves, a test trench was ex- energy is stored in the rock. As excavation pro-
cavated down to the shale. It was found that the ceeds, some stress concentration at the bottom of
cap rock is intensely fractured into blocks and the excavation will result. This effect, however, is
slabs, and the failure is confined to the dolomite relatively insignificant since the horizontal extent
rock. At the interface between dolomite and is usually large compared to the depth of excava-
weathered shale, a triangle void, 0.5-0.6 m high tion and, from field observation, heave is not con-
and about 2.4 m wide, was uncovered running fined to any particular area in the quarry (e.g.,
along the axis of the heave (Fig. 5 ) . At a recent near quarry wall). However, the removal of the
site visit (October 1974), new heaves were ob- overlying rock deprives the layer under considera-
served in the area recently excavated. A typical tion of its vertical stabilizing support, providing
example of these heaves is shown in Fig. 6. therefore a triggering mechanism of the failure.
From these and other similar observations, a The layer fails by buckling upwards and sliding
mechanism of failure of this phenomenon may be along the weakest horizontal plane, which is
deduced. In its natural environment, the rela- usually provided by a change in rock properties
 CAN. GEOTECH. J. VOL. 15, 1978
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

FIG.6. Heave of quarry floor, Dufferin Quarry, taken October 1974.

For personal use only.

FIG.7. Heave of floor during construction, Queenston-Chippawa Canal, October 1921 (courtesy of Ontario
Hydro).

at the interface of two rock strata or materials. swelling pressures due to the presence of small
The release of strain energy accompanying the amounts of swelling minerals may serve as supple-
buckling failure produces the substantial heave mentary triggering mechanisms but none of these
above the surrounding ground surface and intense factors alone can account for the magnitude of
fracturing of the layer results. The presence of horizontal stresses required or the mechanism of
minor geological features such as warping of the failure involved.
bedding planes enhances the possibility of occur-
rence of failure and often dictates the location of (ii) Buckling of Canal Floors
heaves. Other factors such as thermal stress due A similar phenomenon to that described above
to temperature changes and frost action, and occurred in the form of buckling of concrete floor
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

FIG.8. Heave of floor subsequent to construction, Qu eenston-Chippawa Canal (courtesy of Ontario Hydro).

slabs in canals. Several cases of buckling have been By treating the concrete slabs as a beam acted
recorded in the Queenston-Chippawa Canal. In upon by the time-dependent build-up of stresses,
For personal use only.

this case the rock was excavated to a depth of 18.3 the buckling load computed would be 2.9 and 11.6
m partly through overburden into the Lockport MPa if free and fixed end conditions are assumed,
Limestone, the bottom was lined with two 4 m respectively. The results of these calculations,
wide slabs at each side, and a 5.5 m slab was though approximate, are consistent with the in situ
poured later with no expansion joint provided. stresses measured in the area. Similar rock squeeze
All slabs were 15 cm in thickness. problems have not been reported for the Welland
The mechanism of failure is similar to that of Canal in this area, probably due to the much
the quarries. One of the failures extending for at thicker slab used ( 1 m ) .
least several hundred feet was known to occur in
October 192 1 during construction and was located ( b ) Short Term Movements in Open Excavations
at station 146 south of Lundys Lane (Fig. 7 ) . In During the construction of SABNGS No. 2 in
1964, when the canal was dewatered, a stretch of 1952, measurements of inward movements of the
at least 915 m along the bottom was found to have rock walls in the open excavation for the canal
heaved (Fig. 8 ) . The locations of these are shown were taken. These were generaliy measured at the
in Fig. 9. Similar heave of the bottom of the New rock surface or 1.5 m below the rock surface. The
York State Barge Canal was reported at Lockport, excavations penetrate about 6 m of overburden
and bridges founded on rock (Rochester Shale) and 18 m into rock. Thus approximately 15 m of
in the cities of Lockport and Rochester have been the Lockport Dolomite and 3 m of Gasport Lime-
subjected to substantial movements (Rose 1951 ) . stone were removed. The width of the excavation
It is clear, therefore, that both short and long was 58 m.
term failure of concrete floors poured at the The closure due to elastic movement at the top
bottom of open excavations could occur. The time of the rock slope amounted to 1-1.3 cm when the
of failure will depend on the movements of the first lift of 6 m of rock was removed. An additional
rock walls in relation to the displacements allowed closure of 1.3-1.8 cm occurred immediately
for in the design. When the provision for displace- after the bottom of the excavation was reached.
ment is exhausted and the ends of the floor are in The total elastic movement was therefore about
direct contact with the rock wall, stresses begin to 2.5 cm followed by small creep movements vary-
build up with time in the floor slabs. These stresses ing from 0.076 to 0.33 cm/year for the first two
will increase towards the in situ value with time. years during which measurements were taken.
If the thickness of the floor is insufficient to with- Analysis of deformation using elastic parameters
stand the stresses at a particular time, buckling measured for these rock formations showed that
failure will occur. if the in situ horizontal stresses were of the same
 378 CAN. GEOTECH. J. VOL. 15. 1978
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14
For personal use only.

SIR ADAM BECK NIAGARA G.S.

(///////// ) BUCKLING OF CANAL FLOOR

FIG.9. Locations of structures in Niagara Falls.

order of magnitude as the vertical pressure re- that the stresses built up due to the inward move-
moved by the excavation, the decrease in width ment reached the compressive strength of the con-
near the top of rock excavation is of the order of crete. Severe spalling of the brick lining of the
0.25 cm only. To be consistent with the magnitude walls also occurred.
of measured horizontal closure of the order of 2.5 Measurements of the closure of the walls were
cm, an in situ horizontal stress of about 10 MPa soon started at various locations and a most com-
is implied. plete and comprehensive record was kept at the
Canadian Niagara Wheel Pit (Morison 1957).
(c) Long Term Deformation Time-dependent deformations at four different
Along the Canadian side of the Niagara River, elevations are shown in Fig. 10. Superimposed on
five hydro-electric power plants were constructed these mean curves are seasonal fluctuations with
in the period from 1902 to 1952. The locations of the minimum occurring in the March-April period
these structures are shown in Fig. 9. The design and the maximum occurring in the September-
for the generating units is similar, consisting of a October period, but these fluctuations are no more
narrow slot cut into rock to house the penstocks than e0.13 cm. The peaks and troughs of these
and turbines. The wheel pit is 5.5 m wide, 50.3 m fluctuations are so uniformly repeatable that they
deep, and 174 m long for the Canadian Niagara are most likely due to corresponding variations in
Generating Station, and the rock units penetrated rock temperature. The maximum time-dependent
are shown in Fig. 10. deformation occurs at the turbine deck located
Since the completion of construction, inward near the boundary of the DeCew Dolomite and
movements of the walls resulted in various struc- Rochester Shale and amounts to 7.6 cm. At higher
tural distress, particularly at the turbine deck. elevations both the magnitude and time rate of
These took the form of jamming of structural units, deformation are substantially less. Observations in
bending and buckling of steel beams, and crushing at least two other wheel pits of similar design and
and spalling of the arch floors. Measurement of constructed in essentially identical geological con-
stress in the concrete arches in one plant showed ditions are similar and clearly indicate that the
 5.5 rn

GOAT ISLAND

FROM ONTARIO HYDRO R E P O R T

MORISON ( 1 9 5 7 )
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

DECK FLOODED

1905 1907 1909 1911 1913 1915 1917 1919 1921 1923 1925 1927 1929 1931 1933 1935 1937 1939 1941 1943 I945
DATE - YEAR

Typical results of closure measurements of wheel pit walls, Canadian Niagara Power Company.
phenomenon occurring in these rock formations is Y 7
HORIZONTAL
10.5 14
STRESS ( MPa )
1715 fl
of a general nature.
A complete interpretation of the pattern of de-
THOROLD SITE
formation requires a knowledge of the in situ
stresses and the elastic and time-dependent be-
For personal use only.

KEY

haviour of each rock unit. A finite element analysis Ic--.

0 P
taking into account these factors has been per-
ROCK SURFACE
formed (Lee and Lo 1976). For the present pur- *,'E
pose, it is sufficient to realize that the relief of
horizontal in situ stress due to the excavation serves
as an initiating mechanism, and the time-dependent
deformation behaviour must be taken into account
in the design of rigid or semi-rigid structures in
these rock formations. Similar time-dependent de-
formations have been observed at the lock walls at
the New York State Barge Canal in Lockport,
where closure of 15 cm has been noted (Rose
195 1) . The excavations penetrated Rochester
Shale to various depths.
In Situ Stress Measurements
The results of stress measurement at the Thorold
site have been described in detail (Palmer and Lo
1976). A summary plot is shown in Fig. 11. It
may be seen that the stresses vary with depth in
the range of 8-14 MPa, with the highest value FIG. 11. Results of stress measurements at Thorold
Tunnel.
measured in the 'shaly limestone' layer of the
Gasport member. The orientation of the major strength .= 34 NIPa) of these shale formations,
principal stress P is surprisingly consistent, being amounting to 6 MPa. For the strong layers, the
about N60°E. horizontal stresses are substantially less. The
At two tunnel sites, one in Meaford-Dundas orientation of the maximum stress is not, however,
Shale and one in Collingwood Shale, in the vicinity well-defined.
of Toronto, stress measurements were also per- In connection with the rock mechanics investiga-
formed at 9.2-15.3 m below rock surface (Lo and tion for the proposed Wesleyville Generating Sta-
Morton 1976). It was found that the highest stresses tion west of Port Hope, stress measurements at the
occur at the weaker constituent layers (uniaxial lakeshore were carried out in July 1975 to a depth
 C A N . GEOTECH
'i

of 36.6 m in Trenton Limestone. The rock is strong parallel to the gorge face and low stress normal to
and competent, with the elastic modulus varying the face. Both the orientation and magnitude of
from 4 x l o 4 to 7 x l o 4 MPa and horizontal the stresses are, in general, consistent with the
uniaxial strength ranging from 83 to 131 MPa. regional pattern.
The results of the stress measurements are shown
in Fig. 12. The maximum stress P varies from 8
to 13 MPa. The orientation of P is about N15"W. Conclusions
An excellent opportunity of verifying the validity All the recorded evidence described in the
of the method of stress measurements arises with preceding paragraphs of high horizontal stress has
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

the investigation of the stability of the rock cliff been plotted in Fig. 2. It will be seen that the belt
overhanging the Ontario Power Generating Station of horizontal stresses stretches from Rochester
at Niagara Falls (consulting engineers: Acres Con- (New York State), westward through Niagara
sulting Services Ltd.). In one of the drill holes for Falls, turning northeast around Lake Ontario foI-
stress measurements about 25 m from the face of lowing the lake shoreline and extending at least as
the gorge, tests were performed in Rochester Shale far east as Wesleyville. The width of this belt could
at a depth of 35 m. The results confirm the exis- be as much as 30 km inland, and might be more.
tence of high horizontal stress in this formation, Direct measurement and (or) indirect evidence in-
the maximum stress P amounting to 10 MPa. In dicated the high in situ stresses exist in the Lock-
this location, the local in situ stress distribution port, Clinton and Cataract groups of the Silurian
has been modified by the geometry of the Falls Period, the Meaford-Dundas and Collingwood
and the Niagara River in plan and the deep gorge formations, as well as the Trenton - Black River
in elevation. The results of tests in each of the Formation of the Ordovician Period. No data are
successive rock units (Eramosa, Goat Island, yet available for Devonian rocks.
For personal use only.

Gasport, De Cew, and Rochester shales, respec- It is therefore clear that the existence of high

, ,
tively) showed high compressive stress aligned

3.,5
HORIZONTAL
7p IOi5
STRESS ( MPa )
I4p 1s: up , stresses is not confined to any member of the rock
formations; nor to any particular age of the forma-
tions. Similar phenomena of high horizontal stress
exist outside this studied area and elsewhere in the
North American continent and in the world (Hast
1973, 1974; Herget 1974). It is therefore unlikely
that these stresses can be solely ascribed to the
past overburden load during glaciation as suggested
by White et al. (1973). More probably, these
stresses could be due to the current movements of
continental drift according to plate tectonic theory.
Whatever the cause of these stresses, it is necessary
for engineering purposes to realize the possibility
of their existence in a given site unless evidence
indicates otherwise. Steps of investigation may then
be taken to assess ( a ) the magnitude and direction,
and ( b ) the influence of the stresses on the design
of the given project.
It must be emphasized that the existence of high
horizontal stresses alone does not necessarily lead
to adverse problems in every case. In fact, in some
instances, it is easy to conceive that its effect is
actually beneficial. However, in cases where ( a )
the geometric and boundary conditions of the
structure and rock, (b) the strength and time-
dependent deformation of the rock units, and (c)
the rigidity of the structure combine to form the
necessary and sufficient conditions, serious con-
FIG. 12. Results of stress measurements at Wesleyville sideration must then be given to design for the
Generating Station. phenomenon described.
 Acknowled~ements D
HERGET,G. 1974. Ground stress determinations in Canada.
Rock Mechanics, 6, pp. 53-64.
The research performed was supported by the H ~ w r r r D.
, F. 1972. Paleozoic geology of Southern Ontario.
National Research Council of Canada under Geological Report 105, Ontario Division of Mines. 18 p.
Special Project Grant No. 7307. HOGG,A. D. 1959. Some engineeringstudies on rock movement
The author gratefully acknowledges the helpful in the Niagara area. In Engineeringgeology case histories No.
3. Geological Society of America, New York, NY.
discussions with Mr. J. I. Adams and Mr. G. LEE,C. F., and Lo, K. Y. 1976. Rock squeeze study oftwodeep
Gorman, both of Ontario Hydro, on the perfor- excavations at Niagara Falls. ASCE Specialty Conference on
mance of the Hydro structures in the Niagara area Rock Engineering, Boulder, CO, Vol. I, pp. 116-140.
Lo, K. Y., and MORTON,J. D. 1976. Tunnels in bedded rock
Can. Geotech. J. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/20/14

and the permission to publish the data quoted in

with high horizontal stresses. Canadian Geotechnical Jour-
the paper. nal, 13, pp. 216-230.
The stress measurements at Ontario Power Lo, K. Y., LEE,C. F., PALMER, J. H. L., and QUIGLEY, R. M.
Generating Station were performed by Mr. C. 1975. Stress relief and time dependent deformation of rocks.
Yuen and Dr. M. Hori, the University of Western National Research Council of Canada Special Project Report.
Ontario. The work was carried out for Acres Con- MORISON,W. G. 1957. Rock squeeze investigation, Toronto
Power G.S. Ontario Hydro Research Division Report No.
sulting Services Ltd. Their permission to quote 57-13.
part of the data obtained is appreciated. MORTON,J . D., LO, K. Y., and BELSHAW,D. J. 1975. Rock
The generous permission by Golder Associates performance consideration for shallow tunnels in bedded
Ltd. to use the information from the Dufferin shales with high lateral stresses. Proceedings, 12th Canadian
Rock Mechanics Symposium, Kingston, Ont.
Quarry is also gratefully acknowledged. PALMER, J . H. L., and LO, K. Y. 1976. 111 siru stress measure-
ments in some near-surface rock formations - Thorold, On-
tario. Canadian Geotechnical Journal, 13, pp. 1-7.
BRIGHAM, R. J. 1971. Structural geology of southwestern On- ROBINSKY, E. I., and MORTON,J. D. 1973. Foundation investi-
For personal use only.

tario and southeastern Michigan. Mines and Northern Affairs gation for CN Tower, Toronto. 26th Canadian Geotechnical
Petroleum Resources Section, Ontario. Conference, Toronto, Ont.
COATES,D. F. 1964. Some cases of residual stress effects in ROSE, C. W. 1951. Preliminary report - rock squeeze studies,
engineering work. In State of stress in the Earth's crust. Niagara River Development. Report, U.S. Army Corps of
Edited by W. R. Judd. Elsevier Co. Inc., New York, NY. Engineers, Buffalo District, NY.
GOLDERASSOCIATES LTD. 1972. Foundation investigation for SBAR,M. L., and SYKES,L. R. 1973. Contemporary compres-
proposed plant re-location, Dufferin Materials and Construc- sive stress and seismicity in eastern North America: an exam-
tion Ltd., Milton, Ontario. Report to Holderbank Technical ple of intraplate tectonics. Bulletin of the Geological Society
Services Ltd. (private circulation). of America, 84, pp. 1861-1881.
HAST, N. 1973. Global measurements of absolute stress. TERZAGHI,K. 1962. Measurement of stresses in rock.
Philosophical Transactions of the Royal Society of London, Geotechnique, l2(2). pp. 105-124.
Series A, 274, pp. 409-419. WHITE,0. L., KARROW,P. F., and MACDONALD, J. R. 1973.
1974. The state of stress in the upper part of the Earth's Residual stress relief phenomena in Southern Ontario. 9th
crust as determined by measurements of absolute rock stress- Canadian Symposium on Rock Mechanics, Montreal, P.Q.,
es. Naturwissenschaften, 61, pp. 468-475. pp. 323-341.