Australian Journal of Basic and Applied Sciences, 3(2): 395-406, 2009

ISSN 1991-8178

Mitigation of Savadkouh Landslide Using Non-woven Geotextiles

A.J. Choobbasti, A. Barari, F. Farrokhzal M. Safaei
Babol University of Technology, Department of civil Engineering, P.O. Box 484, Babol, Iran.
Abstract: Hazardous landslides are ordinary phenomena in north of Iran. In this paper, mitigation of
a landslide in Savadkouh by using non-woven geosynthetic was investigated. Reinforced slope was
designed using limit equilibrium analysis and numerical method such as those based on the FHW A
guidelines and finite element method and with in-situ residual soil. The main goal of this paper in
designing a reinforced slope instead of failed slope is finding number of reinforcement layers and
vertical distance of reinforcement layers. Using the non-woven fabric to repair failed slope is a good
option to prevent respective failures. W hen the non-woven fabric is used, it not only generates the
tension forces to enhance the overall stability of slopes, but also prevents the development of pore
water trapped in clayey soils by providing horizontal drainage through the fabric. Results of analysis
showed that positive effects of reinforcement are clear and lead to decrease of horizontal
displacements.
Key words: landslide; mitigation; geosynthetic; FHW A; finite element method
INTRODUCTION
Landslides are frequently responsible for considerable losses of both money and lives, and the severity of
the landslide problem worsens with increased urban development and change in land use. Given this
understanding it is not surprising that landslides are rapidly becoming the focus of major scientific research,
engineering study and practices, and land-use policy throughout the world.
In every slope there are forces which tend to promote down slope movement and opposing forces which
tend to resist movement. A general definition of the factor of safety, F, of a slope results from comparing the
down slope shear stress with the shear strength of the soil, along an assumed or known rupture surface.
Starting from this general definition, (Terzaghi, K. 1950), divided landslide causes into external causes which
result in an increase of the shearing stress (e.g. geometrical changes, unloading the slope toe, loading the slope
crest, shocks and vibrations, drawdown, changes in water regime) and internal causes which result in a
decrease of the shearing resistance (e.g. progressive failure, weathering, seepage erosion). (However, varnes,
D.J. 1978), pointed out there are a number of external or internal causes which may be operating either to
reduce the shearing resistance or to increase the shearing stress. There are also causes affecting simultaneously
both terms of the factor of safety ratio. Fig. (1) Shows an example of factor of safety variation as a function
of time, for a given slope and Table 1 and 2 shows brief lists of landslide casual factors and remedial
measures.

Fig. 1: An example of changes in the factor of safety with time.

Corresponding Author: A.J.Choobbasti, A. Barari, M.Safaei, Babol University of Technology, Department of civil
Engineering, P.O. Box 484, Babol, Iran.
395

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Table 1: A brief list of landslide casual factors.
1. Ground Conditions
(1) Plastic weak m aterial
(2) sensitive m aterial
(3) Collapsible m aterial
(4) weathered m aterial
(5) Sheared m aterial
(6) Jointed of fissured m aterial
(7) Contrast in perm eability and its effects on ground water contrast in stiffness (stiff, dense m aterial over plastic m aterial)
2. Geom orphological Processes
(1) Tectonic uplift
(2) Volcanic uplift
(3) Glacial rebound
(4) Fluvial erosion of the slope toe
(5) W ave erosion of the slope toe
(6) Glacial erosion of the slope toe
(7) Erosion of the lateral m argins
(8) Subterranean erosion (solution, piping)
(9) D eposition loading of the slope or its crest
(10) Vegetation rem oval (by erosion, forest fire, drought)
3. Physical Processes
(1) Intense, short period rainfall
(2) Rapid m elt of deep snow
(3) Prolonged high precipitation
(4) Rapid drawdown following floods, high tides or breaching of natural dam s
(5) Earthquake
(6) Volcanic eruption
(7) Breaching of crater lakes
(8) Freeze and thaw weathering
(9) Shrink and swell weathering of expansive soils
4. M an-M ade Processes
(1)Excavation of the slope or its toe
(2)Loading of the slope or its crest
(3)D efective m aintenance of drainage system s
(4)Irrigation
(5)W ater leakage from services (water supplies, sewers, storm water drains)
(6)Vegetation rem oval
(7)M ining and quarrying (open pits or underground galleries)
(8)Creation of dum ps of very loose waste
(9)Artificial vibration (including traffic, pile driving, heavy m achinery)

The methods used to design reinforced slopes are mainly based on the limit equilibrium concept. Methods
such as (Reugger, 1986. Schmertmann, et al., 1987. Leshchinsky, 1989. Jewell, 1980. 1991 and Michalowski,
1997), all utilize limit equilibrium analysis or limit analysis in the design of reinforced slopes. Table 3 lists
general information related to the mentioned methods.
Extensive experimental studies have been devoted to the evaluation of effect of reinforcement on stability
of soil slopes. One of the most important studies in this field were conducted by (Zornberg et al., 1998. and
2003), who observed the behavior of reinforced slopes in the centrifuge. The effect of reinforcement on
stability of soil slopes, and the associated failure mechanisms were assessed in their study. Zornberg et al.,
showed that if a prototype of actual dimensions is modeled with a reduced scale of 1/N and subjected to an
acceleration field N times that of gravitational acceleration, a stress field similar to the prototype structure
would be reproduced within the reinforced slope model in the centrifuge. Other parameters such as density and
internal angle of friction are unchanged while tensile strength of geotextile layers in the model is reduced by
a factor of N. Some important findings of (Zornberg et al., 1998. and 2003), from centrifuge tests performed
on reinforced soil slopes may be summarized as follows:
1. Failure in the reinforced slope model was observed to pass through the slope toe, which is in good
agreement with the assumption of limit equilibrium methods.
2. Failure initiated from mid height of the reinforced model which contradicts the assumption made in limit
equilibrium method that failure develops through the toe of the slope.
3. Location of maximum reinforcement load and the associated maximum strain along the potential failure
surface depend on slope angle and overburden pressure.
4. Stability of the reinforced slope is governed by peak strength of the soil.
396

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Table 2: A brief list of landslide rem edial m easures.
1.M odification of slope geom etry
(1) Rem oving m aterial from the area driving the landslide
(2) Adding m aterial to the area m aintaining stability
(3) Reducing general slope angle
2. D rainage
(1) Surface drains to divert water from flowing onto the slide area (collecting ditches and pipes)
(2) Shallow or deep trench drains filled with free-draining geom aterials (coarse granular fills and geosynthetics)
(3) Buttress counter forts of coarse-grained m aterials
(4) Vertical(sm all diam eter) boreholes with pum ping or self draining
(5) Vertical (large diam eter) wells with gravity draining
(6) D rainage tunnels, galleries
(7) Vacuum dewatering
(8) D rainage by siphoning
(9) Electro- osm otic dewatering
(10) Vegetation planting (hydrological effects)
3. Retaining structures
(1) Gravity retaining walls
(2) Gabion walls
(3)Passive piles, piers and caissons
(4) Cast-in situ reinforced concrete walls
(5) Reinforced earth retaining structures with, strip, polym er, m etallic reinforcem ent
(6) Rock fall attenuation or stopping system s
4. Internal slope reinforcem ent
(1) Rock bolts
(2) M icro piles
(3) Geosynthetics
(4) Anchors
(5) Stone or lim e/cem ent colum ns
(6) H eat treatm ent
(7) Grouting
(8) Freezing
(9) Electroosm osic anchors
(10) Vegetation planting
Table 3: Specifications of lim it
M ichaloski
1997
Kinem atics
lim it analysis

equilibrium m ethods
Jewell
1991
Two-part
wedge

(c = 0, )
0,0.25,0.5
30- 90
Parallel to slope face
to slope face
20-50
0.5-0.8
Triangular or
Rectangular

(c = 0, )
0,0.25,0.5
30- 90
N ot parallel
to slope face
20-50
0.8
Triangular

for design of reinforced slopes

Reugger
Leshchinsky
1989
1989
slices
Internal stability:
Variational Lim it
equilibrium
(c = 0, )
(c = 0, )
0
0
40- 90
45- 90
Parallel
General case
to slope face
40-90
15-40
0.6-1
0.6-1
Rectangular
Triangular

Schm ertm an,

1987
Two-part
wedge
(c = 0, )
0
30- 80
N ot parallel
to slope face
15-35
0.9
Triangular

M ethod
M odel

Soil
(r 0 )
Slope Angle
Reinforcem ent
Arrangem ent

D istribution of
reinforcem en
force with height

Site characterization for Flourd landslide:

Site geography:
The site is located at Savadkouh Azad University grounds in the rural surroundings of Savadkouh, 5 Km
from Pol-Sefid city in the north of Iran. The elevation of the site is 350 m above sea level and like the
surrounding lands, a forest vegetation cover and a mountainous morphology is dominant. Fig. (2) Shows a
scarp of the landslide along with the tilted trees at indicating the extent of the slide.
The area investigated consists of an old slide mass which had several slides at the quaternary age,
associated with saturated conditions and dynamic loading caused by severe earthquakes. Peat, lignite remains
and coal remains from alder trees which were revealed from boreholes at the site indicate that the site had
several previous slides and that generally similar geographical conditions existed in the past.
397

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 2: Scarp of Flourd landslide with tilted trees

Geo-hydrology and underground water conditions:
Underground site investigations indicate that no stable underground water table exists. However,
underground seepage flow seems to exist within the direction of silt and gravel lenses which exist in the
stratification. Bore pits dug in the area show that seepage water exists at the depth of three to five meters
below ground level. W ater level was observed in several observation wells at depths ranging from 10 to 15
meters, at the interface between the slide mass and the underlying bedrock.
A part of the run-off flow outside the active slide area leaks into the slide area which is in addition to
the rainfall induced seepage from surface water absorption. The slide mass is therefore considered to be
saturated at the time of flow.
Geological characterization:
One of the important factors affecting the instability potential of a landslide hazard area is the engineering
geological parameters and characteristics of that site. In this context, the parameters considered usually consist
of site geometry, failure mechanisms observed, effect of slide on existing structures, assessment of the causes
and the risks for future occurrence of slides, and classification of the landslide and the existing soil
characteristics. All these have been studied in Flourd landslide and have revealed the high landslide hazard
of the site.
Site geom etry:
As previously mentioned, the slide mass at the university site is part of a sliding slope facing the south
eastern direction. The sliding mass extends to the calcareous sediments on the south west side, and to the river
at the foot. Steady state seepage passages through the medium and above the level of the rock foundation. A
volume of 15 to 20 liters per second were predicted for these seepage flows. The slide mass does not have
any visible surface flow paths and the existing paths are rather scattered in the whole region (Fig. (3)).

Fig. 3: Access road to the site and the high seepage flow of 10-20 liters per second observed.
From the lateral scarps, silty clay along with boulders is visible. The slide initiated at the point where little
vegetative growth existed. Surface water from rainfalls directly penetrated the underlying soils at theses
surfaces and was lead through the shear zone of the sliding mass. Springs visible at the foot of the sliding
mass is an indication of this process.
The upper surface of the slide is has a concave form which gathers rainfall water into the sliding mass.
Therefore, each period of heavy rainfall caused a reactivation of the slides. Tilting of the existing trees
indicates that a creep type of active slide is dominant in the area (Fig. (4)).

398

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 4: A view of the active slide at the site tilting trees and visible roots indicate an active fault in
the area.
The effect of landslide on existing geotechnical structures in the area was generally in the form of slides
in the slopes, creep, tilt and structural cracks in the geotechnical structures. These defects were mainly
attributed to poor engineering characteristics of the structures, both in design and construction.
Modeling Flourd Landslide:
In order to better understand and assess Flourd landslide and the potential applicability of using geotextile
reinforced in-situ cohesive soil for mitigation of this landslide, initially, limit equilibrium method was utilized
in this study. The method suggested by FHW A was followed in the analyses. Limit equilibrium analysis code
Reinforced Soil Slopes (RSS) was effectively utilized to run limit equilibrium analyses in order to evaluate
the effect of reinforcement on the factor of safety against landslide. RSS calculates the factor of safety for an
existing non-reinforced slope and is able to calculate the required amount of reinforcement in order to reach
a desired factor of safety associated with equilibrium or stable conditions. RSS is able to design a reinforced
slope by one of the three following procedures:
1. Calculating reinforcement spacing for a given slope configuration in order to reach a desired factor of
safety.
2. Determination of the required reinforcement strength in order to reach a desired factor of safety.
3. Calculation of the factor of safety of a reinforced slope with a given configuration of reinforcements.
It is noteworthy that RSS follows all the design procedures suggested by FHW A.
In order to mitigate Flourd landslide using in-situ cohesive soils reinforced with horizontal geotextile layers,
an initial reinforcement configuration was first designed, and through limit equilibrium analysis using RSS, the
factor of safety of the reinforced slope was determined. Next, the resulting factor of safety was checked by
design suggestions from the literature in order to verify the adequacy of the design (Cornforth, D.H.2004).
After verifying the given initial reinforcement configuration, the design procedure suggested by FHW A was
followed in order to obtain the necessary reinforcement spacing and strength in RSS code.
The final design was obtained by following the above mentioned design procedure for three different slopes
of one, 1.5 and two horizontal to one vertical. The reinforced slope area is actually the slope which is
responsible for mitigation of the sliding backfill. Section L6 (Fig. (5)) was selected from the topography of
the landslide area for analysis. This section is shown in the topography in Fig. (5). Each analysis involves
careful determination of the initial unstable slope, and the evaluation of the necessary reinforcement
configuration for obtaining stable conditions. The initial design configuration involves a reinforcement vertical
spacing of 0.5 meters and a reinforcement length of 15 meters. Reinforcement length is automatically checked
for necessary factor of safety against pullout failure, by the program. The results of the initial analysis showed
that the slope geometry with a face slope of two horizontal to one vertical was stable under natural conditions
and no reinforcement design was required for this design. Therefore, only the slope angles of one and 1.5
horizontal to one vertical were considered in the consecutive analyses. A reinforced soil slope factor of safety
of 1.4 was chosen for design following the suggestions of (Cornforth, D.H.2004),. for a medium size landslide
with limited site characterization information. The results of the analyses are described in the following section.

399

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 5: Topography of the landslide area

RESULTS AND DISCUSSION
Prior to numerical analysis of a landslide, it is necessary to select a critical section of the landslide site
in order to perform plane strain calculations on. In this context, Flourd landslide was observed by the
topography maps available, and a critical section as depicted in Fig. 5 by line L6 was chosen for analysis. This
section is believed to be representative of the behavior of the landslide since a relatively wide area is covered
by the slide and therefore the assumption of plane strain conditions is not irrelevant for this section. Moreover,
the majority of site investigation was conducted in the vicinity of this section and therefore, more thorough
information is available in this vicinity.
Limit equilibrium analysis was first performed on the section chosen, in order to observe the possible
failure mechanisms of the slide, as well as to suggest a design for the reinforced soil slope intended to mitigate
the landslide. An initial analysis was performed in each case in order to evaluate the factor of safety of the
slide section. Then, the design section of RSS was used to give a suitable design for a reinforced soil slope
in order to mitigate the landslide. A factor of safety of 1.4 was considered adequate for conditions of
equilibrium. The reinforced slope was extended into the bedrock in order to prevent failure surface from
developing through the interface between the reinforced mass and bedrock. Figs. (69) show several RSS
model analysis results, which show design outputs for the reinforcement based on FHW A guidelines. It is clear
from the results that a reinforced slope is adequately capable of mitigating the landslide. Tables (4) and (5)
list the parameters used for the RSS calculations and finite element model. Also, the results of the limit
equilibrium analyses are tabulated in table (6).
Limit equilibrium analysis gives a good initial design for the reinforced slope area in the landslide.
However, no information is provided by the method regarding the deformations. In order to verify the
suitability of the given design for mitigation of Flourd landslide, FE analysis was performed on one of the
models in order to assess the deformation characteristics of the model. PLAXIS finite element code was used
for the analysis. 15 node elements were used to create a mesh for the problem. The resulting finite element
mesh is depicted in Fig. (10). A reinforcement vertical spacing of 0.5 m was chosen for the model, since only
the deformation characteristics were intended from this analysis and no design output was to be arrived at in
this stage. Lateral boundaries were fixed in the horizontal direction while the bottom horizontal boundary was
fixed in both directions. Geotextile elements available in PLAXIS code was used to model geotextile layers.
An interaction index of 0.85 was chosen for the interface elements.
Maximum displacements in the reinforced slope were obtained as 11.4 cm, which is equivalent to % 0.9
of the slope height. Total displacement vectors are shown in Fig. (11) which shows a maximum displacement
of 0.3 cm.

400

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 6: Initial design of reinforced slope for 1.5:1 slope

Fig. 7: RSS design output for 1.5: 1 slope

Fig. 8: Initial design for 1:1 slope

401

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 9: RSS design for 1:1 slope

Table 4: Properties of soils used in lim it equilibrium and finite
Kind of Soil
M oist U nit
SaturatedU nit
W eight(kN /m 3 )
W eight(kN /m 3 )
Bed Rock
20
20.5
U nreinforced Soil
17.4
18.5
Reinforced Soil
18
19.2

elem ent analysis

Internal Friction
Angle
45
28
15

Cohesion
(kPa)
100
5
10

M odulus of
Elasticity (M Pa)
100
7
15

Poisson's
Ratio
0.25
0.35
0.3

Table 5: Properties of geotextile used in lim it equilibrium and finite elem ent analysis
Geotextile Param eters
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------U ltim ate Strength (KN /m )
Geotextile Stiffness (EA)
15
45
Table 6: RSS results for different slope angles
Reinforced Slope
------------------------------------------------------------------------------------------------------------------RSS D esign
Initial D esign
---------------------------------------------------------------------------------------------------------Reinforcem ent
Required
Sim plified Bishop
Sliding Block
layout
Reinforcem ent
Factor of Safety
Factor of Safety
7@ 0.54m
180.62 m /m
1.395
1.423
5@ 0.75m
3@ 1.25m
4@ 1.06m
153.1 m /m
1.443
1.812
3@ 1.1 8m
3@ 1.43m
----1.549
2.329

402

U nreinforced slope

Slope Angle

Factor of Safety

----------

1.260

1:1

1.306

1.5:1

1.485

2:1

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 10: FE mesh for the reinforced slope.

Fig. 11: Extreme total displacement vectors in the reinforced slope mass
Incremental shear strains as depicted in Fig. (12) reveals that strain localization occurs mainly in the
interface area between soil-geotextile layers. It can be seen that maximum shear strain increment is measured
as % 0.93. These low values prove that geotextile layers adequately mitigate the strains within the slide mass.
Fig. (13) Shows the plastic points of the slope, once again showing that maximum displacements occur in
the soil-geotextile interface, and that no clear failure mechanism is forming in the slope. Moreover, comparison
between plastic points and incremental strains show that plastic points are attributed not to the base soil of the
embankment, but to the interface area of soil-geotextile layers. This is due to low interface properties of nonwoven-geotextile layers. Although woven geotextiles have higher interface properties, however, drainage
characteristics of non-woven geotextiles result in an increase of shear strength in the interface area, thus
increasing interface properties compared to woven geotextiles. Therefore, the observed behavior in the
reinforced soil is inevitable and shear strength properties may only be increased through other methods such
as the sandwich technique.

403

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 12: Incremental shear strains within the reinforced mass

Fig. 13: Plastic points from the FE analysis

Stress-strain analysis was also performed by PLAXIS on non-reinforced model of the natural state of the
slide area in order to observe actual displacement characteristics of the landslide prior to mitigation. Natural
state prior to the slide was predicted from the slide geometry. Displacement vectors revealed that maximum
displacement of 52.7 cm occurs in the slope which is equivalent to % 4.25 of the slope height. These high
displacement values reveal the circular failure mechanism that is forming within the slope. Figs. (14) and (15)
show plastic points and shear strains within the model, better revealing the circular failure mechanism.

Fig. 14: Plastic points within the unreinforced embankment

404

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Fig. 15: Shear strains within the unreinforced embankment revealing circular failure surface.
Conclusions:
Limit equilibrium analysis and stress-strain analysis performed both on unreinforced and reinforced models
of Flourd landslide were performed in this study. General conclusions were drawn as follows:
1. Finite element analysis on the slide mass revealed that circular failure surface developed in the landslide,
therefore approving the circular failure mechanism assumed in the LE analysis.
2. The analyses showed that using in-situ cohesive soil reinforced with geotextile layers is adequately able
to mitigate shallow to medium landslides.
3. In order to make full use of beneficial effects of reinforcement, the reinforced area should extend into the
harder medium.
4. Reinforcing the slope reduces both horizontal and vertical displacements substantially.
5. Non-woven geotextile layers are more suitable for reinforcing cohesive soils than woven geotextiles, due
to proper drainage properties.
REFERENCES
Terzaghi, K., 1950. Mechanisms of Landslides, Geological Society of America, Berkley, 83-123.
Varnes, D.J., 19780 Slope Movements And Types And Processes, Landslides Analysis and Control.
Transportation Research Board Special Report, 11-33.
Reugger, R., 1986. Geotextile Reinforced Soil Structures, Proc. Third International Conference on
Geotextiles, Vienna, Austria, 453-458.
Schmertmann, G.R., V.E. Chourey-Curtis, R.D. Johnson and R. Bonaparte, 1987. Design Charts for
Geogrid Reinforced Soil Slopes, Proc. Geosynthetics. New Orleans, 108-120.
Leshchinsky, D., R.H. Boedcker, 1989. Geosynthetic Reinforced Soil Structure, J.Geotch.Engrg. ASCE,
1459-1478.
Jewell, R.A., 1980. Some Effects of Reinforcement in the Mechanical Behavior of Soils, PhD Thesis.
University of Cambridge.
Jewell, R.A., 1991. Revised Design Charts for Steep Reinforced Slopes, Reinforced Embankment. Theory
and Practice. Thomas Telford. London, 1-30.
Michalowski, R.L., 1997. Stability of Uniformly Reinforced Slopes, J.Geotech.Engrg. ASCE, 546-556.
Zornberg, j.g., N. Sitar and j.k. Mitchell, 1998. Performance of Geosynthetic- reinforced slopes at
Failure, J. Geotech.Geoenviron. ENG, 670-683.

405

Aust. J. Basic & Applied Sciences, 3(2): 395-406, 2009

Zornberg, j.g., N. Sitar and j.k. Mitchell, 1998. Limit Equilibrium as Basis for Design of GeosyntheticReinforced slopes, J. Geotech.Geoenviron. ENG, 684-698.
Zornberg, j.g. and F. Arriaga, 2003. Strain Distribution W ithin Geosynthetic Reinforced Slopes, Journal
of Geotechnical and Geoenvironmental Engineering, 32-45.
Zornberg, J.G., 2003. Peak Versus Residual Shear Strength in Geosynthetic Reinforced Soil Design,
Geosynthetics International, 234-237.
Cornforth, D.H., 2004. Landslides in practice, John wiley & sons, 1-596.

406