1
24-038
Experimental and Numerical Analysis on Failure of Soil Slopes
Induced by Increasing Pore Water Pressure
Godfred Amponsah
New Mexico Institute of Mining and Technology,
Socorro, NM
Mehrdad Razavi
New Mexico Institute of Mining and Technology,
Socorro, NM
ABSTRACT
Slope failures very commonly occur during a long period
of heavy rainfall and groundwater rise. The failures pose
a substantial risk to people, infrastructure, and equipment
downslope. In the recent past, the mining industry has
faced an increase in tailings dam failure due to several rea-
sons including pore pressure increases in the dams. Many
failures have been observed to occur during times of water
level fluctuations but the critical factors that influence the
initiation of slope failures still need to be adequately clari-
fied. To investigate these factors, laboratory experiments
were conducted on model sandy slopes to determine pore
pressure-induced slope failure initiation.
This study also presents a method to examine water
seepage through soil slopes using an ultraviolet (UV) dye
and a UV flashlight. The small-scale model test simulation
demonstrated failures induced by either water percolation
from the upslope tank at a constant level or by raising the
water level at intervals. A bottom chamber fully filled with
water to replicate groundwater rise was also simulated. The
soil slope was monitored at every stage to study the defor-
mations and behavior until failure occurred.
Results gathered from the controlled laboratory condi-
tions were useful for the verification of the numerical mod-
eling method created in the computer program Slide. The
analysis showed that slope failure always occurred when
the toe was fully saturated. High pore pressures and seep-
age forces reduce the shear force at the toe to almost zero
causing it to slide. A comparison of slope height to water
level also indicated an average of 96% water level to cause
a failure. At this point, the soil slope is fully saturated and
has no matric suction. The findings in this study show that,
by monitoring the moisture content of slopes, failures can
be predicted.
INTRODUCTION
Soil has so many uses in our day-to-day life. Designing
structures in soil requires engineering judgments and cal-
culations. Soil particles naturally create friction between
grains, and when this bond is broken, there is instability.
Evaluating the causes of soil slope failure is essential in civil
and mining engineering, and a significant cause of this is
an increase in pore-water pressure. From this, it is necessary
to perform laboratory tests to understand better how such
failures occur. Several techniques are available to induce
slope movement and failure, such as rainfall, using a shake
table, and increasing pore-water pressure. All these meth-
ods require a laboratory set-up, which is time-consuming
and requires precision. These tools can prove useful if the
monitored slope is visible and not significantly affected by
massive external physical contact such as shaking of the
monitored slope. Any physical contact with an observed
24-038
Experimental and Numerical Analysis on Failure of Soil Slopes
Induced by Increasing Pore Water Pressure
Godfred Amponsah
New Mexico Institute of Mining and Technology,
Socorro, NM
Mehrdad Razavi
New Mexico Institute of Mining and Technology,
Socorro, NM
ABSTRACT
Slope failures very commonly occur during a long period
of heavy rainfall and groundwater rise. The failures pose
a substantial risk to people, infrastructure, and equipment
downslope. In the recent past, the mining industry has
faced an increase in tailings dam failure due to several rea-
sons including pore pressure increases in the dams. Many
failures have been observed to occur during times of water
level fluctuations but the critical factors that influence the
initiation of slope failures still need to be adequately clari-
fied. To investigate these factors, laboratory experiments
were conducted on model sandy slopes to determine pore
pressure-induced slope failure initiation.
This study also presents a method to examine water
seepage through soil slopes using an ultraviolet (UV) dye
and a UV flashlight. The small-scale model test simulation
demonstrated failures induced by either water percolation
from the upslope tank at a constant level or by raising the
water level at intervals. A bottom chamber fully filled with
water to replicate groundwater rise was also simulated. The
soil slope was monitored at every stage to study the defor-
mations and behavior until failure occurred.
Results gathered from the controlled laboratory condi-
tions were useful for the verification of the numerical mod-
eling method created in the computer program Slide. The
analysis showed that slope failure always occurred when
the toe was fully saturated. High pore pressures and seep-
age forces reduce the shear force at the toe to almost zero
causing it to slide. A comparison of slope height to water
level also indicated an average of 96% water level to cause
a failure. At this point, the soil slope is fully saturated and
has no matric suction. The findings in this study show that,
by monitoring the moisture content of slopes, failures can
be predicted.
INTRODUCTION
Soil has so many uses in our day-to-day life. Designing
structures in soil requires engineering judgments and cal-
culations. Soil particles naturally create friction between
grains, and when this bond is broken, there is instability.
Evaluating the causes of soil slope failure is essential in civil
and mining engineering, and a significant cause of this is
an increase in pore-water pressure. From this, it is necessary
to perform laboratory tests to understand better how such
failures occur. Several techniques are available to induce
slope movement and failure, such as rainfall, using a shake
table, and increasing pore-water pressure. All these meth-
ods require a laboratory set-up, which is time-consuming
and requires precision. These tools can prove useful if the
monitored slope is visible and not significantly affected by
massive external physical contact such as shaking of the
monitored slope. Any physical contact with an observed