A highway cut failure in Cretaceous sediments at Maymont

A highway cut failure in Cretaceous sediments
at Maymont, Saskatchewan
J. KRAHN
Departmer~tof Civil Er~gineerirzg,University of Saskatchewan, Saskatoott, Sask., Carmda S7N 0 WO
R. F. JOHNSON
Clifton Associates Ltd., Saskatootz, Sask., Carlada S7L 5 W5
D. G. FREDLUND
Departnwtt of Civil Erzgineering, Urziversily of Saskcitclze~vat~,
Saskatootz, Sask., Car~adaS7N 0 WO
AND
A. W . CLIFTON
Cliftort Associates Ltd., Regirla, Sask., Camda S4P OK5
Received March 26, 1979
Accepted July 4, 1979
In 1973 the Saskatchewan Department of Highways began construction of a crossing over
the North Saskatchewan River at Maymont, Saskatchewan. The south approach to the river
required a cut some 20 m in depth at the top edge of the valley and when the excavation reached
the design elevation a massive failure occurred on one of the backslopes. The major portion
of the slip surface followed a slickensided clay shale zone. An analysis of the failure indicates
residual angles of shearing resistance were being mobilized. The reason for mobilizing only
the residual strength is attributed to previous shearing arising from glacial ice-thrusting.
The sliding occurred entirely within the sediments of the nonmarine Upper Cretaceous
Judith River Formation, but the strengths mobilized were essentially the same as those mobilized by slides in the marine Upper Cretaceous Bearpaw and Lea Park Formations. Negative
water pressures arising from the stress change due to excavating did not appear to influence
the stability. Direct shear box tests on natural slickensided surfaces gave strengths higher than
required for a safety factor of unity. The testing of pxecut surfaces gave results that seem t o
correlate more closely with the field residual strengths. Furthermore, the Maymont case
history clearly illustrates the need for identifying geological details and demonstrates the
engineering significance of glacial ice-thrusting.
En 1973 le Ministkre de Ia Voirie de Ia Saskatchewan a commencC ?I construire un pont au
dessus de la rivikre North Saskatchewan B Maymont, Saskatchewan. L'approche sud ?I la rivikre
exigeait une excavation de quelques 20 m de profondeur en cr&tedu talus de la rive. Lorsque
l'excavation a atteint la profondeur projetke, une rupture de masse s'est produite d a m I'un des
talus. Une grande partie de la surface de glissement passait d a m une zone de schiste argileux B
surfaces miroitantes. Une analyse de la rupture indique que l'angle de frottement residue1 a ktk
mobilisk. La cause de la mobilisation seulement de la resistance rksiduelle est attribuke B un
cisaillement prialable rksultant des mouvements de glaciers.
Le glissement s'est produit entikrement dans les sCdiments non marins de la formation Judith
River du cretack supirieur, mais les rksistances mobilisCes Ctaient essentiellernent identiques B
celles mobilisCes d a m des glissements dans les formations marines Bearpaw et Lea Park du
crCtacC supCrieur. Des pressions interstitielles negatives resultant des variations de contraintes
produites par l'excavation ne semblent pas avoir affect6 la stabiliti. Des essais B la boite de cisaillement direct sur surfaces rniroitantes naturelles ont donnC des risistances supirieures B celles
requises pour obtenir un facteur de sCcuritC de 1.0. Les essais sur kchantillons pr6-decoupes
semblent conduire B une meilleure corrdation avec les rksistances rCsiduelles it1 situ. D'autre
part, le cas type de Maymont illustre clairernent le besoin d'identifier les dktails gCologiques et
dkmontre la signification pratique des effets des mouvements des glaciers.
[Traduit par la revue]
Can. Geotech. J., 16, 703-715 (1979)
Introduction
In 1973, the Saskatchewan Department of Highways undertook the construction of a bridge crossing
over the North Saskatchewan River approximately
5 km south of Maymont, Saskatchewan. Maymont is
located on the Yellowhead Highway 64 km northwest of Saskatoon, as shown in Fig. 1. The south
approach to the river required a cut of approximately
0008-3674/79/040703-13$01.00/0
@ 1979 National Research Council of Canada/Conseil national de recherches du Canada
CAN. GEOTECH. J. VOL. 16, 1979
ALBERT
TRANS
-
-
,
&
W
CANADA-HIGE
CURRENT
~ N I P A W ~ NBRlDGE
,-
YELLOWHEAD
,
- ,/HIGHWAY
.
WOSE
2AW
,
/
,
-
FIG. 1. Location of the Maymont bridge.
20 m at the top edge of the valley. The approach
crossed numerous old landslides along the valley
wall. The gradeline of this approach was set so that
the net effect of the construction over the old landslides was an unloading of the scarp and a loading of
the toe. When the excavation at the scarp was completed, a massive failure occurred rather suddenly in
the east backslope. Figure 2 shows an oblique aerial
view of the failure.
This failure is of interest for several reasons. First,
the shear strength mobilized at failure by this slide,
which occurred within Cretaceous nonmarine sediments of the Judith River Formation, was close to
strengths typically mobilized in the Upper Cretaceous
marine sediments of the Bearpaw and Lea Park
Formations. This is of interest because it suggests that
for engineering purposes it is not crucial to identify
the different Cretaceous formations, an identification
that is often difficult to make on the basis of engineering site investigations. Second, the slide only mobilized the residual strength at failure. Since this was a
newly constructed slope, it becomes important to
determine how the strength was reduced from its
peak to residual value. Third, there was no delay in
the failures as is often the case with cut slopes in
stiff fissured clays.
This paper presents the details and implications of
the Maymont case history.
Site Description
The location selected for the bridge crossing is
about 915 m upstream from an old ferry crossing as
shown in Fig. 3. At this point the river valley is about
2.4 km wide with a rise in elevation of 68 m from
river level to the upland. The river is flowing in a
southeasterly direction.
The north shore of the river is flanked by an
alluvial floodplain about 305 m wide, and north of
this floodplain, the landform is as an eroded till plain
that rises to the upland at a slope of about thirteen
horizontal to one vertical (4.4'). Land-slun~ping
topography extends southward from the south riverbank for about 610 m. For the first 280 m the natural
slope rises at about 3", for the next 320 m this increases to about 5.7', and for the last 10 m, just before
the scarp, the slope rises very sharply. South of the
scarp the upland landform is a ground moraine.
Immediately landward from the south valley crest
there are some linear features, evident on the vertical.
air photograph (Fig. 3), which are due to icethrusting.
Slope Design
During the design stages it was realized from the
site investigation data that the base of the proposed
approach cut would intercept a layer of slickensided,
brecciated clay shale. Because of the slickensiding,
relatively low strength parameters (i.e., +' between
10 and 15' with c' = 4.8 kPa) were selected for
design. With these relatively low strength parameters,
fairly flat slopes were required to achieve a safety
factor of 1.3. The Saskatchewan Department of
Highways analyzed the risk and economics involved
in a failure and concluded that there would be no
great risk involved to the road user if a failure were
to occur and that there would be little added cost to
flatten the slope after a failure. On the other hand,
substantial economic savings c o ~ ~ be
l d realized if the
failure did not occur. On the basis of this rationale
the backslopes were designed at a factor of safety of
unity with the lower portion at a slope of three
horizontal to one vertical and the upper portion at
four horizontal to one vertical.
Excavation was just completed on the south
approach when the east backslope failed, the result
of which is shown in Fig. 2. The failure occurred
sometime between 6:00 p.m. Friday evening and
Saturday morning. A block of soil 270 m long and
120 m wide broke out of the backslope and moved
toward the highway centerline. A crack developed
some 33 m landward from the crest of the backslope,
and it varied in width from a few centimeters to 6 m.
The photographs in Figs. 2, and 4-6 show these
features.
KRAHN ET AL.
FIG. 2. An oblique aerial view of the Maymont slide.
Slide Investigation
immediately following the backslope failure, a
detailed investigation was undertaken. Arrangements
were made for aerial photographic coverage from
which a contour map with intervals of 0.6 m (2 ft)
was prepared (Fig. 7). In addition, ground surveys
were performed to obtain cross sections of the
failed cut.
A drilling program was carried out to augment the
subsurface information obtained during the initial
investigation. Drilling was done using a truckmounted rotary drill and the holes were advanced
using wet drilling techniques. Locations of the test
holes are shown in Fig. 7. Undisturbed samples were
collected during the drilling using thin-walled steel
tubes at selected locations and all the test holes were
electrically logged.
Open standpipe piezometers were installed in test
holes 8-75,9-75, and 201A-73. A slope indicator tube
was installed on the east slope to monitor any further
movements, and another slope indicator t ~ ~ bwas
e
installed near the crest of the west inf failed backslope
to detect any impending failure. The locations of
these are also shown in Fig. 7, as test holes S.I.10 and
S.I.11.
Geology and Stratigraphy
Except for a randomly located thin veneer of
glacial drift, the sediments consist of Cretaceous
bedrock materials belonging to the Judith River and
Lea Park Formations. The Judith River Formation
overlies the Lea Park Formation but the exact
boundary between them is difficult to ascertain at this
site. The landslide is interpreted as having occurred
CAN. GEOTECH. J. VOL. 16, 1979
FIG.3. A vertical airphoto of the Maymont bridge site prior to construction
entirely within the sediments belonging to the Judith
River Formation. The Upper Cretaceous Judith
River Formation comprises layers of clays, silts, and
sands, which were deposited in a nonmarine environment (McLean 1971).
At the south approach cut the sediments consist
primarily of a very dense silty sand overlying a zone
of alternating layers of silt, sand, and clay shale. The
clay shales are highly brecciated and slickensided. A
stratigraphic interpretation of the subsurface soils is
given by the cross section in Fig. 8. (The location of
the section is given in Fig. 7.) As may be seen in Fig.
8, the slickensided zone dips toward the base of the
cut beneath the east cut slope. Whether this dip does
or does not continue on beneath the west cut slope is
not certain. N o samples were taken during the drilling
of test hole S.I.10 and as a result no information is
available on the extent of the slickensiding in this
area. The stratigraphic sequence shown in Fig. 8 for
test hole S.l.10 is based on the electric log alone. It is
thought that beneath the west slope the slickensided
zone is horizontal or dips slightly into the slope.
The base of the slickensided zone within the landslide
was assumed to be at a depth below which slickensided surfaces were no longer evident in the undisturbed test hole samples.
It appears that the bedrock materials at this site
have been disturbed by glacial ice movement to a
depth of about 45 m. The jointing, brecciation, and
slickensiding noted in the samples are believed to be
due to these disturbances. The disturbance became
visibly evident in the backslopes as the excavation
proceeded. The bedding was folded and truncated as
illustrated in Figs. 9 and 10.
The piezometric levels were not high in the sliding
mass; however, there were minor amounts of ground-
KRAHN ET AL.
FIG. 4. A vertical airphoto of the Maymont bridge site at thc time of the backslope failure.
water seepage to the north of the northern extremity
of the landslide. The piezometers installed in test
holes 8-75 and 9-75 within the slickensided zone
indicated a piezometric level slightly above the top
of this zone. Three piezometers were installed in test
hole 20112-73, and the one with the highest tip elevation indicated the lowest water pressure. Generally,
these piezometers seem to indicate that this is an area
of groundwater recharge in a downward gradient and
that within the slickensided zone the groundwater
levels are close to its upper boundary.
Laboratory Testing
During the postslide investigation, soil samples
were obtained for the purpose of evaluating the shear
strength of the backslope material. Some thin-walled
tube samples were taken in the upper silty sand zone
but the majority of the samples were taken in the
slickensided brecciated zones.
Direct shear tests were performed on the lower
slickensided clay shale material. The tests were
performed by applying a normal load in one increment, submerging the sample in distilled water, and
CAN. GEOTECH. J. VOL. 16, 1979
FIG. 5. The headscarp towards the south end of the slide.
FIG. 6. The headscarp towards the north end of the slide. The fencepost gives an indication of the vertical drop.
then allowing the sample to consolidate. After consolidation (or swelling), the samples were sheared in
the forward direction at a rate of 2.5 X 10-3 cm/min
until a displacement of about 0.6 cm was reached.
Then the samples were manually sheared back to
their initial position. This procedure was repeated
until the shearing resistance had essentially reached
a constant value, which required about 5 cm displacement.
Grain-size analysis on the upper sedinlents indicated that the material can be classified as a nonplastic silty fine sand. The samples tested consisted
of about 50% sand size, 40% silt size, and 10% clay
size particles.
709
KRAHN ET AL.
are given in Table 1 and show that the samples were
very silty with clay contents varying between 10 and
24%. The liquid limits varied between 44 and 51%
and the plastic limits varied between 24 and 33%.
The natural water contents were slightly below the
plastic limits.
Results of the direct shear tests are shown in Fig.
11. Both the peak and residual values are shown and
the data points for each test are joined by a vertical
line. The samples (1, 2, 3) that were brecciated and
fissured showed a distinct drop-off from peak to
residual conditions. Sample 5, which was also brecciated and fissured, but softer than the other three,
exhibited little difference between peak and residual
conditions. Of the samples with slickensided surfaces (4, 6), one exhibited a significant drop from
peak t o residual and the other did not.
Three tests were also performed on remolded
samples. These tests were carried out in the same
manner as for the undisturbed specimens except they
were cut along the shear surface prior to testing.
The direct shear results show a large variation in
the residual strength between samples. If the cohesion
intercept is taken as zero, &.' for the natural samples
varies between a high of 28" and a low of 13".
Moreover, they are all higher than the residual +,'
of 8" obtained on the remolded precut samples. It is
possible that the samples should have been sheared to
larger strains even though the stress-strain curves
had essentially levelled off. To some extent, however,
a variation in residual strengths from laboratory
direct shear tests appears to be characteristic of these
overconsolidated sediments. Insley et al. (1977), for
example, obtained a variation of 8.5-15.0" in residual
angles of friction from tests on Upper Cretaceous
METRES
FIG. 7. A contour map of the slide area showing the
locations of the boreholes and section analyzed.
Six 5 cm X 5 cm direct shear box samples were
prepared from the thin-walled steel tube samples.
Three of these samples contained primarily brecciated and fissured material, whereas another similar
sample was much softer. Another two samples had
natural slickensided surfaces along the shear plane.
Atterberg limits, water contents, and grain-size
analyses performed on cuttings from these samples
TABLE1. Properties of samples tested in direct shear
Test
series
1
2
3
4
5
6
Liquid
limit
Plastic
limit
Water
Sand size
+0.074 mm
content
(beforelafter)
(%)
Silt size
Clay size
-0.002 mm
(%l
Description
(Cj,)
(5)
Brecciated
fissured
Brecciated
fissured
Brecciated
fissured
Natural
slickensided
Brecciated
Natural
slickensided
Remolded
and precut
44.3
31.7
26.5134.7
19.8
70.2
10.0
47.8
29.2
29.7133.9
21.5
64.0
14.5
47.4
30.5
30.0134.6
7.0
76.6
16.4
50.9
32.1
30.9134.6
3 .8
72.1
24.1
44.8
50.7
24.4
33.7
29.9/32.1
32.5/37.7
4.8
1.3
78.6
75.2
16.5
23.5
-
-
-
4.0
66.0
30.0
(96)
CAN. GEOTECH. J. VOL. 16, 1979
SL-E 310H IS31
KRAHN ET AL.
FIG.9. Contortions in the bedrock sedirnents in the west backslope (after Sauer 1978).
FIG. 10. Folding in the bedrock sedirnents in the east backslopes (after Sauer 1978).
sediments of the Lea Park Formation from the
Maidstone bridge site on the North Saskatchewan
River.
Analysis
A back-analysis of the cut slope failure was performed to ascertain the shear strength mobilized at
the time of failure. Most of the analyses were run
using the simplified Bishop method as extended for a
composite failure surface (Fredlund and Krahn
1977). A few checks were made using the Morgeostern-Price method but the computed factors of
safety were essentially unchanged from those obtained using the simplified Bishop method. The compu-
CAN. GEOTECH. .l.
VOL. 16, 1979
I
1,2,3,5
4,6
NATURAL
l
l
FISSURED
AND
I
' I
BRECCIATED
SLICKENSIOEO
/
/
N O R M A L STRESS - k P o
FIG. 11. Shear strength envelopes from direct shear tests on clay shale samples.
tations were performed using the SLOPE computer
program at the University of Saskatchewan (Fredlund 1975).
For analysis purposes, the sliding mass was divided
into two soil types as shown in Fig. 12. No conclusive
evidence became available during the investigation of
the exact location of the slip surface except near the
crest and at the toe. It is reasonable, however, that
the slip surface was located within the slickensided
zone and for this reason the trial slip surfaces were
not permitted to penetrate below this zone. The
piezometric level was taken to be slightly above the
slickensided clay shale zone as indicated in Fig. 12.
A large number of slip surfaces were analyzed, each
one passing through the two known end points. The
one found to have the lowest safety factor was the
composite failure surface illustrated in Fig. 12.
In the analysis it was considered inappropriate to
use a cohesion intercept for the nonplastic upper
silty sand and therefore the cohesion was set equal to
zero. The influence of varying +' was investigated,
and the results indicated that the factor of safety was
insensitive to changes in the angle of internal friction.
Subsequently, all analyses were performed with
c' = 0 and +' = 30" for the upper unit consisting of
the very dense silty sand.
Several combinations of c' and +' were computed
that resulted in a factor of safety of unity. These are
given in Fig. 13 and indicate that when the cohesion
is equal to zero the friction angle required for a
factor of safety of unity is only 8.2". Inclusion of a
cohesion value reduces the friction angle required
for a factor of safety of unity according to the
relationship shown in Fig. 13. These low friction
angles indicate that the residual strength was mobilized at failure.
Raising or lowering the piezometric level by about
3 m altered the factor of safety by only about 10%.
With a lowering of the water level, the friction angle
required, with cohesion equal to zero and for a
safety factor of unity, is 7.5" and with a 3 m rise in
water level the friction angle required for the point
of limiting equilibrium is still within the range of
residual strength values.
Analysis of the postslide geometry, using a similar
slip surface and similar strength parameters as for
the back-analysis, results in a safety factor of 1.07.
This 7% increase in the factor of safety from unity
is thought to be the result of the rapid failure: its
momentum probably carried the sliding mass into a
stable position. The slope indicator installed in the
slide (S.I.ll in Fig. 7) confirmed that the sliding mass
reached a stable position. Remedial work at the slide
consisted of raising the gradeline of the roadway to
load the toe and landscaping the slope to unload the
crest and to provide drainage off the slope. All
crevices and cracks were filled and compacted to
prevent surface runoff from infiltrating the sliding
mass. The final slope geometry is shown in Fig. 12
and an analysis of this configuration indicates a
factor of safety of 1.36.
Discussion
The mobilized angle of friction (i.e., 8.2") at the
CENTER
OF
3
FAILED
FINAL
CROSS
X
Y
499.7
12
13
14
15
16
321.6
347.6
396.3
321.6
347.6
498.7
498.2
491.5
494.2
492.7
499.2
493.3
17
18
396.3
457.3
485.4
485.4
X
3
4
5
6
350.6
393.9
426.2
4573
517.4
506.4
495.7
495.7
7
321.6
8
9
347.6
396.3
ROTATION
X = 393.3
Y = 582.9
R = 99.4
',
FOlNT
Y
-
POINT
CROSS
SECTION
NOTE
SECTION
:
DIMENSION
ARE
IN
METRES
\
\
7
c--=-\
\
8
7 =l95 k ~ / r n ~
c' = 0
12
15
18
17
FIG. 12. The section analyzed in the stability analyses.
CAN. GEOTECH. J. VOL. 16, 1979
12.5
I
75
'
l"\
100 -
I
"\
-
"\
5 0 FACTOR
OF
SAFETY
2.5 -
0
4
EFFECTIVE
= 1.0
+
I
l
l
5
6
7
ANGLE
OF
FRICTION
FIG. 13. The r e q u i r e d effective s t r e n g t h
for a factor o f safety o f unity.
-
-
8
-
DEGREES
parameters re-
quired
Maymont slide is close to that obtained from the
back-analysis of other slides in Upper Cretaceous
sediments. A landslide on the Qu'Appelle River
Valley near Lumsden, Saskatchewan, in the Upper
Cretaceous clay shales of the Bearpaw Formation,
mobilized a friction angle of 9" when cohesion was
taken as zero (Johnson 1978). At the Maidstone
bridge site on the North Saskatchewan River, Insley
et al. (1977) found that the stability of the landslide
situated in the Lea Park Formation was governed by
residual strength parameters of +,' = 10 and c,' = 0.
Back-analyses of failures in the Bearpaw Formation
at the Gardiner Dam in Saskatchewan have shown
that with c,' = 0, the 4,' mobilized was between 7.0
and 10.5" (Ringheim 1964).
The reason for the similarity in the shear strengths
mobilized by landslides is that the three Cretaceous
formations all contain layers of low-strength material
such as bentonitic zones. These are usually associated with the marine Bearpaw and Lea Park
Formations but the nonmarine Judith River Formations also contain local bentonitic zones (Whitaker
and Pearson 1972) because of the tremendous interfingering of the sediments of the two depositional
environments.
This is significant because it means that for
engineering purposes it is not crucial to identify the
different formations, a task often difficult to perform
on the basis of engineering investigations. What is
crucial, however, is the identification of geological
details, such as bentonitic zones, regardless of which
formation they belong to.
The strength mobilized by the Maymont slide was
essentially equal to the residual strengths measured
in the laboratory on the remolded precut samples.
With cohesion equal to zero, the friction angle
mobilized by the slide was 8.2", whereas the residual
friction angle measured in the laboratory on precut
and remolded samples was 8.0". Considerably higher
and variable residual strengths were exhibited by the
direct shear tests on the natural samples. Insley et al.
(1977), Weisner (1969), and Thomson and Hayley
(1975) also measured the lowest residual strength
parameters for Cretaceous clay shales on surfaces
that had been precut. The close agreement between
the strength mobilized by the Maymont slide and
strength obtained in the laboratory from precut
samples suggests that testing precut samples gives
the most representative value of the residual field
strength.
whenever an excavated slope fails at a shear
strength corresponding to the residual friction conditions, it is important to ascertain how the strength
was reduced from peak to residual. For the Maymont
slide a logical explanation for the shear strength
having been reduced to residual is that it occurred
during the shearing resulting from glacial ice-thrusting. The linear surface features on the vertical air
photograph in Fig. 3 and the folds, brecciations, and
slickensides all indicate that the material has been
moved about through glaciation. It has been recently
shown by Sauer (1978) that this type of icethrusting
is a fairly widespread phenomenon in glaciated
regions.
The suddenness with which the failure took place
is of interest. Generally there is a time delay, which is
attributed to the time necessary for equalization of
negative pore pressures resulting from the unloading
(Vaughan and Walbancke 1973; Eigenbrod 1975 ;
Skempton 1977). It does not appear that the Maymont slide was influenced by negative pore water
pressures since this would mean that the shear
strength parameters would be even lower.
It should be noted that the west cut, which was as
high as the east cut, did not fail. The slope indicator
installed at the crest of this slope (S.I. 10 in Fig. 7) has
not shown any movement to date, which indicates
that the slope is still stable. There may be two possible explanations. First, the slickensided clay shale
zone beneath the west slope does not appear to dip
toward the base of the cut as it does beneath the'
failed slope. This stratigraphic difference has a
stabilizing effect. Second, there may very well be
negative pore water pressures acting in the more
intact soils below the west cut. If this is the case, it is
possible that this slope could fail as the negative
water pressures relax. Regardless of the reason, the
KRAHN ET AL.
Maymont slide shows how localized geological details can influence stability. Had the cut been shifted
perhaps 30 n~ either direction, the situation could
have been considerably different.
It is becoming more apparent that stability problems in complex sedimentary bedrock deposits are
invariably controlled by geological details. A thin
presheared bentonitic layer controlled the stability
of a slide in Cretaceous bedrock at Devon, Alberta
(Eigenbrod and M-orgenstern 1972). The same feature
controlled the stability of landslides at the Nipawin
bridge in Saskatchewan, at the Maidstone bridge in
Saskatchewan, and at the Smoky 2 bridge in Alberta
(Insley et d.1977). The Maymont failure has once
again illustrated the importance of the geological
details in slope stability.
Acknowledgements
The authors are grateful to Mr. W. A. Sheard,
Chief Engineer, and to Mr. D. G . Metz, Principal
Geotechnical Engineer, of the Department of Transportation, Government of Saskatchewan, for permission to publish this case history. Special thanks
are also due to Mr. H. Eley and his Materials Testing
Branch for the field investigations and instrumentation installations.
This work was in part supported by financial
assistance received from the National Research
Council of Canada and the Saskatchewan Research
Council.
EIGENBROD,
K. D. 1975. Analysis of the pore pressure changes
following the excavation of a slope. Canadian Geotechnical
Journal, 12, pp. 429-440.
715
K. D., and MORGENSTERN,
N. R . 1972. A slide in
EIGENBROD,
Cretaceous bedrock, Devon, Alberta. Itr Geotechnical
~ r a c t i c efor stability in open pit mining. Edited by C. 0.
Brawner and V. Milligan. American Institute of Mining
Engineers, pp. 223-238.
FREDLUND,
D. G. 1975. A comprehensive and flexible slope
stability program. Presented at the Roads and Transportation Association of Canada Meeting, Calgary, Alta.
J. 1977. Comparison of slope
FREDLUND,
D. G., and KRAHN,
stability methods of analysis. Canadian Geotechnical Journal, 14, pp. 429-439.
P. K., and SMITH,L. B. 1977. Use of
INSLEY,
A. E., CHATTERJI,
residual strength for stability analyses of embankment
foundations containing preexisting failure surfaces. Canadian Geotechnical Journal, 14, pp. 408-428.
JOHNSON,
R. F. 1978. The Maymont bridge landslides: A case
history study. M S c . thesis, Dept. of Civil Engineering,
University of Saskatchewan. Saskatoon, Sask. 220 p.
MCLEAN,J. R. 1971. Stratigraphy of the Upper Cretaceous
Judith River Formation in the Canadian Great Plains.
Geology Division, Saskatchewan Research Council,- Report
NO. 11.96 p.
RINGHEIM.
A. S. 1964. Ex~erienceswith the B e a r ~ a wshales a t
the ~ o i t h~ a s k a t c h e d a n River Dam. 8th international
Congress on Large Dams, Vol. 7, p. 529.
SAUER,E. K. 1978. The engineering significance of glacier
ice-thrusting. Canadian Geotechnical Journal, 15, pp.
457-472.
SKEMPTON,
A. W. 1977. Slope stability of cuttings in brown
London clay. 9th International Congress of Soil Mechanics
and Foundation Engineering,
Lectures Vol.,
- Tokyo,
. Special
pp. 25-34.
THOMSON.
S.. and HAYLEY.D. W. 1975. The Little Smoky
~andslide.'canadianGeotechnical Journal, 12, pp. 379-392.
H. J. 1973. Pore pressure
VAUGHAN,
P. R., and WALBANCKE,
changes and the delayed failure of cutting slopes in overconsolidated clay. Geotechnique, 23(4), pp. 531-539.
WEISNER,W. R. 1969. Residual shear strength of overconsolidation clay shales. M.Sc. thesis, Department of Civil
Engineering, University of Saskatchewan, Saskatoon, Sask.
168 p.
D. E. 1972. Geological map of
WHITAKER,
S. H., and PEARSON,
Saskatchewan. Saskatchewan Research and Department of
Mineral Resources.