Content uploaded by Md. Enayet Chowdhury
Author content
All content in this area was uploaded by Md. Enayet Chowdhury on Jan 29, 2019
Content may be subject to copyright.
Proceedings, International Conference on Disaster Risk Management,
Dhaka, Bangladesh, January 12-14, 2019
1Undergraduate Student, Dept. of Civil Engineering, Bangladesh University of Engineering and
Technology (BUET), Dhaka-1000, Bangladesh
2*Corresponding Author, Undergraduate Student, Dept. of Civil Engineering, Bangladesh University of
Engineering and Technology (BUET), Dhaka-1000, Bangladesh, E-mail: enayet108@gmail.com
3Professor, Dept. of Civil Engineering, Bangladesh University of Engineering and Technology (BUET),
Dhaka-1000, Bangladesh
EVALUATION OF THE EFFECT OF RAINFALL ON SLOPE STABILITY
ANALYSIS BY USING HYDRAULIC UNIT FLUX INFILTRATION PROPERTY
Ahmed Hossain1, Md. Enayet Chowdhury2*, Mohammad Shariful Islam3
ABSTRACT
Slope stability analysis for evaluating landslide potential excludes the effect of rainfall while determining
the Factor of Safety (FOS). This study involves the simulation of rainfall on slopes through hydraulic unit
flux boundary condition to evaluate the FOS. Infiltration data taken from perforated tray rainfall simulator
was used as the hydraulic unit flux infiltration in slope stability analysis, later the FOS was validated by a
conventional procedure. The rainfall simulator is capable of producing storms up to 188 mmh-1 within 10
minutes’ duration with a drop size distribution slightly greater than the natural rainfall (median diameter of
raindrops for simulated rainfall is 4.18 mm at 188 mmh-1, where that for natural rainfall is 3.25 mm). FOS
in real time condition was evaluated on the basis of surface runoff which is a common reason of rainfall
induced landslides. The result shows that FOS from software analysis excluding rainfall simulation is over
estimated with respect to real time FOS analysis.
Keywords: GEO-STUDIO®, Rainfall simulator, Hydraulic unit flux boundary condition
Introduction
Rainfall induced landslides are a very common occurrence in the hilly region specially when they are
composed of finer material like alluvial deposits. To ensure the stability of these slopes, determination of
factor of safety is of paramount importance. In a tropical region like Bangladesh, rainfall is considered as
the trigger to a landslide, causing the surface runoff of the top layer of soil in absence of vegetative layer or
any kind of soil reinforcing materials. Previously, Oh and Vanapalli (2010) have studied the impact of
rainfall infiltration on the shearing strength of soil mass and the stability of the slope where Chowdhury et
al. (2018) analyzed the variation of shearing strength with varying fines content. Islam (2018) has shown
the variation of factor of safety of a slope due to the variation of the degree of saturation. Chowdhury et al.
(2017) investigated a mathematical approach for river bank slope stability. Islam et al. (2013) analyzed the
effect of root reinforcement of vetiver on the stability of a slope.
In any kind of slope stability analysis, the result will not be a conservative one if the effect of rainfall on the
slope failure is not considered as a significant parameter. But the effect of rainfall is not directly considered
in contemporary methods like Limit Equilibrium or Strength Reduction method of slope stability analysis.
So, in this study, unit flux infiltration property is used to assign rainfall effect coupled with slope stability
analysis which follows Limit Equilibrium theorem to determine the rainfall induced factor of safety of
slopes.
Physical and Index Properties of Soil
Detailed laboratory experiments were carried out to determine the physical and index properties of the soil
samples which were collected from Chattogram hill tracks. Soil composition indicates sand 57.4%, silt
28.3% and clay 14.3%, co-efficient of permeability of the soil is 5.25×10-8 ms-1. The soil sample has a
specific gravity of 2.75, the bulk density of 16.8 kNm-3 and the moisture content at site condition is 15.46%.
Evaluation of the Effect of Rainfall on Slope Stability Analysis by using Hydraulic Unit Flux Infiltration
Property, Ahmed Hossain1, Md. Enayet Chowdhury2*, Mohammad Shariful Islam3
The soil has Liquid Limit, Plastic Limit and Plasticity Index of 24, 27 and 7 respectively. The soil is
categorized as SM-SC. The direct shear test indicates that the cohesion and angle of internal friction of the
corresponding soil sample is 1.8 kPa and 30.2° respectively.
Method of Analysis
A trapezoidal glass model was manufactured by (Islam, 2018) having a slope of 1:1 and dimension of
900mm×600mm×600mm. The glass model was filled with soil and compacted to the extent to attain the
desired density (16.8 kNm-3).
A drip-type artificial rainfall simulator (Chowdhury et al., 2017) of 1.11 m2 (1.22m×0.91m) having a
rectangular tray consisting of mild steel with a pipe supply-based water connection system was used to
artificially simulate rainfall on a physical model to determine the amount of water infiltrated in the soil and
run-off occurred on the slope. The corresponding rainfall simulator is capable of producing storms up to
188 mmh-1 rainfall intensity within 10 minutes’ duration with a drop size distribution slightly greater than
the natural rainfall (median diameter of raindrops for simulated rainfall is 4.18 mm at 188 mmh-1, where
that for natural rainfall is 3.25 mm). This greater dimension of the raindrop size with respect to natural
rainfall keeps the model as a conservative one because it will cause more surface erosion of the slope soil
than the practical condition.
GEO-STUDIO® was used to analyze the Factor of Safety (FOS) of the physical model using limit
equilibrium following Morgenstern-Price principle for both dry and after rainfall condition. To assign
rainfall effect in slope stability analysis, a boundary condition named unit flux hydraulic boundary
condition in SEEP/W module was introduced. This boundary condition is applicable when the precipitation
rate is greater than infiltration capacity of soil throughout the duration of rainfall. The ϕ-index value for the
situation is used as input for this hydraulic boundary conditions. In cases where the infiltration capacity of
the soil is lesser that the precipitation rate, hydraulic flux function needs to be defined for the use of
hydraulic boundary condition. Here, the amount of infiltrated water got from perforated tray simulation
(Islam, 2018) for 15 min and 30 min of rainfall (0.0014 ms-1 and 0.0025 ms-1 respectively) was used as the
input of unit flux in boundary condition where the hydraulic analysis will be coupled with SEEP/W module
which acts as the pore water pressure condition in SLOPE/W stability analysis to determine the rainfall
induced FOS of the slope. Throughout the duration of rainfall simulation, it was observed that surface run-
off on the slope was present which means the infiltration rate can be used as the input of the unit flux in
SEEP/W analysis.
Results
Figure 1: (a) Unit flux infiltration boundary condition input, (b) Seepage field for 15 minutes rainfall
(a)
(b)
Evaluation of the Effect of Rainfall on Slope Stability Analysis by using Hydraulic Unit Flux Infiltration
Property, Ahmed Hossain1, Md. Enayet Chowdhury2*, Mohammad Shariful Islam3
Figure 2: Comparison of the failure pattern of the slope between the physical model and coupled stability
analysis in GEO-STUDIO®; (a) & (d) when no rainfall is occurred, (b) & (e) when rainfall duration is 15
min, (c) & (f) when rainfall duration is 30 min.
In Figure 1(a), the assigned unit flux infiltration hydraulic condition is shown where in Figure 1(b) the
seepage field for rainfall duration of 15 minutes is provided. The seepage field shows that the vertical
percolation of rainwater was assumed and no lateral movement of water is considered. In Figure 1(b), the
lighter color in greater depth signifies more moisture content in soil whereas the darker colour in upper
portion indicates less moisture content in soil.
Rainfall duration
No rainfall
15 minutes
30 minutes
Factor of safety
1.69
1.06
0.87
Comments on slope stability
Ok
Ok but not
recommended
Not ok
Table: Comparison of the factor of safety of the slope in various rainfall condition
For dry condition, the model is directly assigned in SLOPE/W module and analyzed where the FOS for this
condition was 1.69 (>1.50). It means the slope was stable for this particular condition. For wet condition,
coupled analysis of SLOPE/W with SEEP/W module is required to assign the rainfall effect on the slope to
analyze rainfall induced FOS of the slope. At first, using hydraulic boundary condition in SEEP/W,
infiltration representing rainfall effect is assigned then in SLOPE/W the initial pore water pressure (PWP)
condition is set to parent analysis (Previous SEEP/W analysis) and stability analysis was done to attain FOS
of the slope for this condition. From the analysis, after 15 min rainfall, the FOS of the slope was reduced to
1.06 which is stable but not recommended. But after 30 min rainfall, the induced FOS lowers to 0.87 which
refers that the slope has become unstable.
Discussions and Conclusions
In this study, slope failure envelope analyzed from GEO-STUDIO® intersects the top surface of the slope
(for 30 min rainfall duration) which is representative to the practical condition (Figure 2(f)). So, the
reduction of FOS in the coupled analysis of SEEP/W and SLOPE/W is validated with the practical
condition. In the tropical region, this method can be used to determine the reduced FOS of the slope under
(a)
(b)
(c)
(d)
(e)
(f)
Evaluation of the Effect of Rainfall on Slope Stability Analysis by using Hydraulic Unit Flux Infiltration
Property, Ahmed Hossain1, Md. Enayet Chowdhury2*, Mohammad Shariful Islam3
the influence of rainfall where conventional methods fail to do so. To be conservative, the soil condition is
modeled here as bare surface, thereby neglecting the increase in FOS considering vegetation or
reinforcement effect on a slope. Further study can be conducted to observe the influence on coupled
analysis-based FOS in vegetated or reinforced slope.
Acknowledgement
The authors are acknowledging GEO-STUDIO® Student Edition License (free version) for conducting this
study. They are also grateful to Geotechnical Engineering Laboratory, Bangladesh University of
Engineering and Technology (BUET) for providing the opportunity to conduct lab tests.
References
Chowdhury, M.E., Hossain, A. and Muktadir, H.M. (2018) “Evaluation of the river training work of Padma
river through a mathematical approach”, Journal of Modern Science and Technology, 6(1):52-62
Chowdhury, M.E., Islam, M.A., Islam, M.S., Shahriar, M.S. and Alam, T. (2017) “Design, operation and
performance evaluation of a portable perforated steel tray rainfall simulator”, Proceedings of the 14th
Global Engineering and Technology Conference, 29-30 December, 2017, BIAM Foundation, 63 Eskaton,
Dhaka, Bangladesh
Chowdhury, M.E., Islam, M.A., Islam, T. and Khan, N. (2018) “Evaluation of shear strength of
cohesionless soil from maximum, minimum dry density and fines content using polynomial surface fitting
method”, Electronic Journal of Geotechnical Engineering, 23(1):31-56
Islam, M.A. (2018) “Measures for landslide prevention in Chittagong hill tracks of Bangladesh”, M.Sc.
Thesis, Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh
Islam, M.S., Arifuzzaman, Shahin, M.H. and Nasrin, S. (2013) “Effectiveness of vetiver root in
embankment slope protection: Bangladesh perspective”, International Journal of Geotechnical
Engineering, Vol. 7, Issue 2
Oh W.T. and Vanapalli, S.K. (2010) “Influence of rain infiltration on the stability of compacted soil
slopes[J]”, Computers and Geotechnics, 37(5): 649-65
