6
slope slide. After failure, the slope angle was measured at 9°
from the initial 35°. The experiment was terminated after
20 minutes when the failure occurred. The final shape of
the slope showed the toe slid, resulting in a collapse to the
head of the crest.
The second experiment was a gradual rise in water level
to induce slope failure. The water level was kept constant
for 3 minutes at each level before rising to the next level.
Failure occurred at the last stage when the soil sample was
fully saturated.
Experiments 1 and 2 produced similar results, although
they had different initial conditions. The results and
sequence of events for experiment 2 are tabulated below in
Table 2.
A comparison of the slope geometry before and after is
shown in Figure 10).
The chronology of events for this experiment is simi-
lar to that of Jia et al. (2009) where two phenomena were
observed, that is settlement of the crest and gradual collapse
of the toe induced by wetting.
Firstly, the settlement of the crest of the slope is attrib-
uted to the wetting-induced collapse of the saturated area
close to the toe. Along with Jia et al. (2009), other research
by Tadepalli et al. (1992) also observed a similar wetting-
induced collapse where a volumetric strain of 1.1% was
measured for the settlement. In this experiment, a 9% set-
tlement was measured. The original slope angle of 35° was
reduced to 20° after the failure occurred.
4.3 Finite element modeling for slope failure
During this experiment for slope failure, water was raised
at a regular interval. At each level, the water was kept con-
stant for 3 minutes and raised to the next level. At the last
level, the slope failed with failure initializing on the slope.
The noticeable high hydraulic gradient when water levels
rise produces high seepage forces through the soil. These
forces cause wetting to be formed early at the toe of the
slope. Again, the flow causes pressure ridges at the toe, ulti-
mately resulting in bulging and reduction of shear force at
the toe. Failure in the slope occurred primarily due to high
hydraulic gradients that resulted in large seepage forces
and the decrease of shear strength resulting from high
pore pressures.
The factor of safety reduces as the water level rises
from an initial of 1.26 to 0.841 at failure. Both Janbu’s and
Bishop’s Methods resulted in a factor of safety less than 1
at failure (Figure 11). This scenario is particularly true to a
(a)
(b)
Figure 9. Steady-state failure experiment (a) before and (b)
after failure
Table 2. Correlation of time, observation, and slope
deformation
Time
(min)
Upslope
Tank
water
level
(mm)
Visual Observation
(Saturation zone)
Visual
Observation
(Deformation)
3 70 Wetting front
gradually moving
No deformation
of slope
6 140 Wetting at the toe of
the slope.
Bulging at the toe.
Settlement begins
at the crest of the
slope
7 140 Lateral movement
at toe
Minimum
settlement at the
crest
9 215 Failure Slope failure.
Settlement
measured at
10 mm
slope slide. After failure, the slope angle was measured at 9°
from the initial 35°. The experiment was terminated after
20 minutes when the failure occurred. The final shape of
the slope showed the toe slid, resulting in a collapse to the
head of the crest.
The second experiment was a gradual rise in water level
to induce slope failure. The water level was kept constant
for 3 minutes at each level before rising to the next level.
Failure occurred at the last stage when the soil sample was
fully saturated.
Experiments 1 and 2 produced similar results, although
they had different initial conditions. The results and
sequence of events for experiment 2 are tabulated below in
Table 2.
A comparison of the slope geometry before and after is
shown in Figure 10).
The chronology of events for this experiment is simi-
lar to that of Jia et al. (2009) where two phenomena were
observed, that is settlement of the crest and gradual collapse
of the toe induced by wetting.
Firstly, the settlement of the crest of the slope is attrib-
uted to the wetting-induced collapse of the saturated area
close to the toe. Along with Jia et al. (2009), other research
by Tadepalli et al. (1992) also observed a similar wetting-
induced collapse where a volumetric strain of 1.1% was
measured for the settlement. In this experiment, a 9% set-
tlement was measured. The original slope angle of 35° was
reduced to 20° after the failure occurred.
4.3 Finite element modeling for slope failure
During this experiment for slope failure, water was raised
at a regular interval. At each level, the water was kept con-
stant for 3 minutes and raised to the next level. At the last
level, the slope failed with failure initializing on the slope.
The noticeable high hydraulic gradient when water levels
rise produces high seepage forces through the soil. These
forces cause wetting to be formed early at the toe of the
slope. Again, the flow causes pressure ridges at the toe, ulti-
mately resulting in bulging and reduction of shear force at
the toe. Failure in the slope occurred primarily due to high
hydraulic gradients that resulted in large seepage forces
and the decrease of shear strength resulting from high
pore pressures.
The factor of safety reduces as the water level rises
from an initial of 1.26 to 0.841 at failure. Both Janbu’s and
Bishop’s Methods resulted in a factor of safety less than 1
at failure (Figure 11). This scenario is particularly true to a
(a)
(b)
Figure 9. Steady-state failure experiment (a) before and (b)
after failure
Table 2. Correlation of time, observation, and slope
deformation
Time
(min)
Upslope
Tank
water
level
(mm)
Visual Observation
(Saturation zone)
Visual
Observation
(Deformation)
3 70 Wetting front
gradually moving
No deformation
of slope
6 140 Wetting at the toe of
the slope.
Bulging at the toe.
Settlement begins
at the crest of the
slope
7 140 Lateral movement
at toe
Minimum
settlement at the
crest
9 215 Failure Slope failure.
Settlement
measured at
10 mm