2005 Prigiarto Yonathan Hokkal Drenaje y Subdrenaje

8/10/2019 2005 Prigiarto Yonathan Hokkal Drenaje y Subdrenaje

1/106

UNIVERSITY GHENT

UNIVERSITEIT

GENT

INTERUNIVERSITY PROGRAMME

MASTER OF SCIENCE IN

PHYSICAL LAND RESOURCES

Universiteit GentVrije Universiteit Brussel

Belgium

Drained and undrained slope stability

analysis using GIS on a regional scale

September 2005

Promotor: Master dissertation in partial fulfilment

Prof. F. De Smedt of the requirements for the Degree of

Master of Science in


2/106

Most true it is, that I have looked on truth

Askance and strangely; but, by all above,

These blenches gave my heart another youth,

And worse essays proved thee my best of love.

Shakespeare CX

Het is zeker waar: ik zag oprechtheid, deugd

met een scheel oog, maar hemel, alsjeblieft,

dit dwalen bracht mijn hart een nieuwe jeugd,

en jij bleek op mijn pad mijn zoetste lief.

Shakespeare CX

Il est vrai que jai regard ce qui est vrai,

etrangement de travers, mais aprs tout,

ces faux regards ont donn une autre jeunesse

mon coeur; et les pires essays te montrent le meilleur.

Shakespeare CX (Pierre Jean Jouve)


3/106

Acknowledgements i

ACKNOWLEDGEMENTS

This thesis on Slope Stability Analysis Using GIS on a Regional Scale is the final output of

my advanced study in Physical Land Resources organized by Free University Brussels (VUB)

and University of Gent (RUG). I would like to express my deepest appreciation and thanks to

my promoter, Prof. Dr. Ir. F. De Smedt, for his encouragement, comments, suggestions andconstant support throughout my study period and research work. It has been a privilege and a

pleasure to be supervised by leading researcher in the department.

I would like to express my best appreciation to Prof. Marc Van Molle for his valued support

in giving direction for this thesis work. My sincere thanks also go to Mr. W. Solomon Tuccu,

Mr. Corluy Jan and Mr. Hung Le Quock for their valuable support, criticism, guidance and

help to make this manuscript finished. I have also been fortunate to have the support of Mr. Y.

P. Chandra especially for sending me information and materials needed for finishing my

thesis.

My gratitude also goes to Anja Cosemans for her valuable support during my study. She hasbeen a computer IT advisor, a good friend and also an advisor for many technical questions

related to my study.

This has been a wonderful year for me to have an experience studying in Belgium. This

experience has been more colourful with many friends that support me during my study. My

special thanks go to all my colleagues, especially Mr. Michael Ndemo Bogonko, for sharing

computer room and accompanying me during my thesis work. I would like also to express my

special gratitude to my best friend Mr. Pascal Nottet for encouragement, valuable support and

especially sharing good and bad time together. Live in Belgium has never been wonderful


4/106

Abstract ii

ABSTRACT

This study is the continuation of the previous study done by Ram Lakhan Ray, 2004, that

applied stability model on an area of 341 km2 of Dhading district, Nepal. In this study, a

spatial distributed physically based slope stability model was presented and applied on 84 km2

of cohesive soil, covered about 25% of the original study area. Two methods of analysis were

performed, i.e. total and effective stress analyses and Taylor and infinite slope methods were

applied on the analysis. Critical height and safety factor maps were produced based on those

analyses. Steady state and quasi dynamic conditions were considered for the present study

with varying soil thickness. For quasi dynamic conditions, wetness index was applied based

on direct rainfall infiltrations. Slope angle of 38and 17can be considered as the average

mean slope angle to cause instability and the lower most slope angle for stable conditions,

respectively. This value was derived from the analysis based on half saturated conditions. It

was also concluded that this case can serve as general conditions of safety factor map at the

site where this case also has a similar result with models based on different return periods.

Taylor method was not applicable for this study area since this method is only applicable for

assessing safety factor with high slope angle. For short term safety factor map, completely dry

conditions resulted from infinite slope method can be used as a short term applications. Half

saturated case can be considered as general and long term safety factor map as this condition

reveals similar result as given by various return periods. This study has proved that models

developed with infinite slope models have given the best result even with some assumption.

Keywords: stability, total stress analysis, effective stress analysis, Taylor method, infinite

slope method, critical height, safety factor, steady state condition, quasi dynamic condition,

short term safety factor map, long term safety factor map.


5/106

Table of Contents iii

Table of Contents

ACKNOWLEDGEMENTS ................................................................................................. i

ABSTRACT......................................................................................................................... ii

Table of Contents................................................................................................................ iii

List of Figures..................................................................................................................... vi

List of Tables ...................................................................................................................... ix

List of Abbreviations ............................................................................................................x

CHAPTER 1 : INTRODUCTION.......................................................................................1

1.1 General ..........................................................................................................................1

1.2 Introduction to Study Area.......................................................................................... ...2

1.3 Scope of the Study.........................................................................................................41.4 The Objective of the Study ............................................................................................4

CHAPTER 2 : LITERATURE REVIEW ...........................................................................5

2.1 General ..........................................................................................................................5

2.2 Slope Failure Mechanism...............................................................................................6

2.2.1 Internal Factors Effecting Slope Instability...................................................... ...8

2.2.1.1 Slope and Gravity Force ......................................................................9

2.2.1.2 Influence of Groundwater ................... .................................................9

2.2.2 External Triggering Events.................................................................................9

2.3 Fundamentals of Soil Parameters ................................................................................ .10

2.3.1 Principle of Effective Stress .............................................................................10

2.3.2 Failure Criterion...............................................................................................112.3.3 Drained and Undrained Strength.......................................................................11

2.3.3.1 Undrained Strength............................................................................12

2.3.3.2 Drained Strength................................................................................14

2 3 3 3 Residual Strength 15


6/106

Table of Contents iv

2.4.1.3 Cohesionless Material in Dry Condition..................... ........................21

2.4.1.4 Cohesionless Material with Groundwater Effect ................................22

2.4.2 Total Stress Analysis............ ............................................................................22

2.4.3 Wedge Analysis ...............................................................................................25

2.4.4 Non-Linear Methods ........................................................................................25

2.4.5 Model Based on Root Cohesion .......................................................................26

2.5 Landslide Hazard Analysis with GIS....................... .....................................................26

2.5.1 Model Concept.................................................................................................27

2.5.1.1 Using Infinite Slope with Total and Effective Stress ..........................28

2.5.1.2 Using Taylor Method.........................................................................28

2.5.1.3 Assessment of Stability Classes .........................................................29

2.5.2 Hydrological Model .........................................................................................30

CHAPTER 3 : MATERIALS AND METHOD.................................................................32

3.1 General ........................................................................................................................32

3.2 Data Availability..........................................................................................................33

3.2.1 Available DEM and Raster Maps .....................................................................33

3.2.2 Available Hydrological Data ............................................................................37

3.3 Applied Methodology ..................................................................................................37

3.3.1 Soil Parameters Determination .................... .....................................................38

3.3.2 Model Development.........................................................................................41

CHAPTER 4 : RESULT AND DISCUSSION...................................................................46

4.1 General ........................................................................................................................46

4.2 Ground Condition at the Study Area ............................................................................46

4.3 Critical Height Maps....................................................................................................48

4.3.1 Based on Total Stress Analysis (TSA)..............................................................48

4.3.1.1 Using Taylor Method.........................................................................48

4.3.1.2 Using Infinite Slope Method ..............................................................50


7/106

Table of Contents v

4.4.2 Effective Stress Analysis..................................................................................59

4.4.2.1 Completely Dry Condition.................................................................60

4.4.2.2 Half Saturated Condition............... .....................................................63

4.4.2.3 Fully Saturated Condition ................... ...............................................65

4.4.2.4 Based on Different Return Periods.....................................................68

4.5 Discussion ...................................................................................................................71

4.5.1 Total and Effective Stress Analyses................... ...............................................71

4.5.2 Influence of Depth............................................................................................73

4.5.3 Slope Angle .....................................................................................................73

4.5.4 Selection of Maps........................................ .....................................................74

4.5.4.1 Critical Height Map ...........................................................................74

4.5.4.2 Safety Factor Map..............................................................................76

4.5.5 Comparison with Root Cohesion Method .........................................................78

CHAPTER 5 : CONCLUSIONS AND RECOMMENDATIONS .......... ......... .......... .......82

5.1 Conclusions.................................................................................................................82

5.2 Recommendations........................................................................................................84

REFERENCES................................................................................................................... ix

APPENDICES .................................................................................................................. xiii


8/106

List of Figures vi

List of Figures

Figure 1: Sensitive Landslides Area (Ray, 2004) ...............................................................3

Figure 2 : Simplification Mass on Slope.............................................................................7

Figure 3 : Results of Undrained Triaxial Tests on Saturated Clay .....................................12

Figure 4 : Relationship between su/' and plasticity Index (Bjerrum and Simons, 1960) ..13Figure 5 : Relationship between the Natural Shear Strength of Undisturbed Clays and

Liquidity Index (Carter and Bentley, 1991) ......................................................14

Figure 6 : Correlation between Effective Friction Angle and Plasticity Index for Fine-

Grained Soils (NAVFAC DM-7)......................................................................15

Figure 7 : The Concept of Residual Shear Strength................................... ........................16

Figure 8 : Forces on element of infinite slope (Cernica, 1995)..........................................20

Figure 9 : Total Stress Analysis............ ............................................................................23

Figure 10 : Taylor's Stability Coefficients for u= 0 (after Craig, 2004).............................24

Figure 11 : Location of the Study Area (Ray, 2004) ...........................................................32

Figure 12 : Digital Elevation Model (DEM) of the Study Area (Ray, 2004)........................34

Figure 13 : Slope Map of the Study Area............................................................................34Figure 14 : Soil Map of the Study Area (Ray, 2004)...........................................................35

Figure 15 : Clayey Soil in the Study Area...........................................................................35

Figure 16 : Land Use Map of the Study Area (Ray, 2004) ................................................. .36

Figure 17 : Flow Chart for the Present Study....................... ...............................................42

Figure 18 : Previous and Present Study Assumption on Soil Thickness ..............................43

Figure 19 : Map Calculation for Stability Coefficient (Ns) .................................................43

Figure 20 : Map Calculation for Critical Height with Taylor Method................................. .44

Figure 21 : Map Calculation for Critical Height with Infinite Slope....................................44

Figure 22 : Map Calculation for Safety Factor with Infinite Slope and TSA .......................45


9/106

List of Figures vii

Figure 28 : Area of Critical Height for Each Soil Types Using Lower Undrained Shear

Strength............................................................................................................51

Figure 29 : Area of Critical Height for Each Land Use Types Using Lower Undrained Shear

Strength............................................................................................................51

Figure 30 : Critical Height Map with TSA..................... .....................................................52

Figure 31 : Area of Critical Height based on ESA ..............................................................53

Figure 32 : Area of Critical Height for Each Soil Types under Different Steady State

Conditions........................................................................................................54

Figure 33 : Area within Safety Factor Class with Taylor Methods......................................56

Figure 34 : Safety Factor Map of Taylor Method with H = 5 m ..........................................56

Figure 35 : Area of Stability Class under Different Soil Thickness for Infinite Slope Method

with TSA..........................................................................................................57

Figure 36 : Stability Area under Different Soil Types and Thickness with Infinite Slope and

TSA .................................................................................................................58

Figure 37 : Range of Slope Angle against Stability Class for Different Soil Thickness .......58

Figure 38 : Safety Factor Map with Infinite Slope Method (TSA) for H = 2 m ...................59

Figure 39 : Area of Stability Class for Dry Condition with ESA........................................ .60

Figure 40 : Relationship between Area Occupied by Stability Class and Soil Thickness.....61

Figure 41 : Stability Area under Different Soil Types and Thickness in Dry Condition ......61

Figure 42 : Range of Slope Angle against Stability Class under Different Soil Thickness

(Dry)................................................................................................................62

Figure 43 : Safety Factor Map of Completely Dry Condition for H = 4 m ..........................62

Figure 44 : Area of Stability Class for Full Saturated Condition with ESA .........................63

Figure 45 : Stability Area under Different Soil Types and Thickness in Half Saturated

Condition .........................................................................................................64


(Half) ...............................................................................................................65


10/106

List of Figures viii


(Full)................................................................................................................67

Figure 51 : Safety Factor Map of Full Saturated Condition for H = 6 m...... ........................68

Figure 52 : Wetness Index for Various Soil Thickness and Soil Types .............................. .69

Figure 53 : Rainfall Intensity with Various Return Periods.. ...............................................69

Figure 54 : Area of Safety Factor with Various Return Periods...........................................70

Figure 55 : Stable Area with Various Soil Types and Return Periods with Soil Thickness of

2m....................................................................................................................71

Figure 56 : Comparison between Various Method Results................................................. .72


11/106

List of Tables ix

List of Tables

Table 1 : Classification of Landslides (Varnes, 1975)........................................................7

Table 2: Consistency-Strength Relationship from Field Inspection (BS 8004: 1986) ......13

Table 3 : Methods of Analysis............. ............................................................................18

Table 4 : Stability Clases.................................................................................................30

Table 5 : Various Types of Soils and Corresponding Slope Angle........... ........................36

Table 6 : Rainfall Prediction of Study Area with SMADA 6 Software (Ray, 2004) .........37

Table 7 : Index Properties of Soil Based on Deoja et al. (1991).......................................39

Table 8 : Undrained Shear Strength from Various References........................................ .39

Table 9 : Effective Stress Parameters for the Study Area.................................................40

Table 10 : Soil Parameter Used for the Analysis .......... .....................................................41Table 11 : Tabulated Area of Soil Types for each Land Use Types ........... ........................46

Table 12 : Summary of Critical Height Using Taylor Method ................... ........................49

Table 13 : Summary of Critical Height using Infinite Slope Method .................................51

Table 14 : Range of Critical Height, Area and Slope Angle.............................................. .52

Table 15 : Critical Height and Slope Angle under Different Steady State Condition..........54

Table 16 : Range of Mean Slope Angle.............................................................................74

Table 17 : Slope Angle for Unstable and Stable Conditions ..............................................74

Table 18 : Summary of Critical Height..............................................................................75

Table 19 : Percentage of Total Area of Safety Factor for TSA Result................................77

Table 20 : Percentage of Total Area of Safety Factor for ESA Result................................78

Table 21 : Previous Study Assumption on Soil Thickness for Cohesive Soil .................... .78Table 22 : Lower Most Slope Angle Causing Instability for Previous and Present Study.. .80

Table 23 : Summary Comparison between Previous and Present Study.............................80


12/106

List of Abbreviations x

List of Abbreviations

DEM Digital Elevation Model

DoR Department of Roads

ESA Effective Stress Analysis

FS Safety FactorGIS Geographical Information System

Inf. Infinite Slope Method

Mod. Moderately

Mst. Moderately Stable

Qst. Quasi Stable

RCM Root Cohesion Method

St. Stable

TSA Total Stress Analysis

Ust. Unstable


13/106

Chapter 1 : Introduction 1

CHAPTER 1 : INTRODUCTION

1.1 General

Slope stability is a term used to explain the general immovability performance of a slope

under natural conditions or man-made slope. A slope may be laterally unsupported earth

mass, natural or man-made, whose surface forms an angle with the horizontal. Hills and

mountains, riverbanks and coastal formations, earth dams, highway cuts, trenches and the like

are examples of slopes. Every slope experiences gravitational forces and it may also possibly

be subjected to earthquakes, glacial forces or water pressures. In turn, these phenomena may

be direct influences on the stability of the slope.

A distinction should be made between natural and man-made slopes where both of the slopes

might have different effect on the stability performance. Man-made slopes are usually under-

human controlled where dimensions, material characteristics and strength are controlled by

several site tests and designs to adapt favourable slope. Natural slopes, on the other hand, are

mainly natural occurrence of slopes where materials characteristics and strengths are

generally un-controlled. Thus, in man-made slopes, the slope is designed in such a way to

fulfil the characteristics and strengths of the materials, while for natural slopes, an attempt isused to maintain the slope from failure, which is caused by external triggering factor.

Basically, the performance of immovability of a slope, safety factor, for both man-made and

natural slopes is evaluated in relative terms of forces ratio that withstands the slope from

movements against that of causes failure. Among many internal and external forces,

gravitational and seepage forces are the internal factors that mainly cause imbalance forces in

soil or rock structures. Gravity is the force that acts everywhere on the earths surface, pulling

everything in a direction toward the centre of the earth. While seepage or pore water pressure

causes failure due to the rapid build up of pore water pressure.


14/106


called as Landslide Hazard Evaluation or Mapping. Complexity of the terrain and uncertainty

in factors affecting failure of the slope are more substantial compared to local slopes. Thus,

the need of evaluating landslide hazard has led to the use of Geographical Information

Systems (GIS), which are capable to analyze regional areas based on spatial distribution.

However, the principle used for the evaluation of landslide hazard remains the same as in

conventional local slope, which evaluates imbalance in forces. The different is that in spatial

analyzes the safety factor is evaluated in a pixel. Despite the difference, many deterministic

methods can be applied for evaluating landslide hazard and one of the most common methods

is so-called limit equilibrium approach. In this method, a slope may be divided into a number

of slices and the factor of safety is computed by solving the static equilibrium equations based

on a set of assumptions (Ray, 2004). The parameters required for analysis includes slope

geometry and conventional soil mechanics parameters. In most cases, the accuracy generally

depends on a proper estimation of soil parameters, hydrogeology conditions and geometric

conditions (Burton, 1998). However, consideration on the type of analysis either drained or

undrained cases should be carefully taken into account, because these cases determined the

chosen of parameters to be used in the analyses and the use of the outcome safety factor map.

As the type of analysis shows different effect on the stability result, a decision must be made

whether to use a total or an effective stress analysis especially, in clayey soils. The choice

generally follows from the classification of a stability problem as short or long term. Slope

failures generally result from a change of loading on the soil and if this occurs quickly, which

is the case in hilly or mountainous areas, the stability during and immediately after the change

may need to be assessed. This will be particularly important if the change of loading results in

a change of pore-water pressure in the soil mass and the change is rapid compared to the

consolidation time for the soil (Nash, 1987). Thus, in principle a total or an effective stress

approach could be used to analyze any slope, although, in practice, the short term stability


15/106


Resources in Vrije Universiteit Brussel. Thus, materials and data used for this study are

basically collected and re-used from the previous study done by Ram Lakhan Ray.

The study area is located at Dhading district, Nepal. Nepal is located in the heart of the

Himalayan arc and occupies nearly one third of the mountain range (Ray, 2004) with the

longitude of 8004 to 8812 easting and latitude of 2622 to 3027 northing. The previous

study is a part of a project called Slope Stability Analysis using GIS on a Regional Scale,

which lies in the Dhusa Village in Dhading district along the Prithvi Highway leading from

the Western and Eastern parts of the country to Kathmandu, the national capital of Nepal. The

study area itself is located in the mountainous district in Nepal where national road

connecting major towns in some parts of Gorkha and Chitwan districts lies within this

mountainous area with latitude of 2745 to 275230 northing and longitude of 843730 to

845230 easting. The latitude varies from about 242 to 1922m above sea level. Detail

explanation related to the study area can be found in Slope Stability Using GIS on a

Regional Scale by Ram Lakan Ray, 2004. Figure 1 presents the sensitive area where

landslides are frequently occurred.


16/106


season and, because of that, the major national road that connects other major districts is

closed for several weeks. Due to the frequently occurrence of landslides within this area, the

government has decided to develop mitigation plan for this area.

1.3 Scope of the Study

This study is mainly focused on to which extend the used of total stress analysis and effective

stress analysis applicable for the proposed study area. Since, the study area is covered both by

cohesive and cohesionless soil, while the total stress analysis is mainly applicable for

cohesive soil. Thus the study is conducted only on cohesive soil presented in the study area.

Two types of analysis was performed, i.e. total and effective stress analysis, using Taylor and

infinite slope method. Critical height and safety factor maps were produced based on those

analyses. Steady state and quasi dynamic conditions were considered for the present study

with varying soil thickness. For quasi dynamic conditions, wetness index was applied based

on direct rainfall infiltrations.

1.4 The Objective of the Study

Stability analysis on a regional scale have been investigated and studied by many researcher.

However, the methods and assumption used are not well explained. Therefore, the present

study aims to find a better approach for stability analysis over a regional area. The outcome of

the study will be helpful in planning, designing and implementing the development paradigms

of landslide area.

The landslide hazard as an outcome of this study could then be used as a guidance to assistsplanners and administrators in making decisions related to the landslide area. Furthermore, it

can be used as an indication of stability conditions over the study area. Risk assessment and

measurement can be interpreted based on the outcome. This will certainly provide useful

i f i f bili j i i h l h di l i


17/106

Chapter 2 : Literature Review 5

CHAPTER 2 : LITERATURE REVIEW

2.1 General

Slides may occur in almost every conceivable manner, slowly or suddenly and with or

without any apparent provocation. The term landslide is commonly used to denote the

downward and outward movements of slope-forming materials along surfaces of separation

by falling, sliding, and flowing at a faster rate. Although landslides are primarily associated

with mountainous regions they can also occur in areas of low relief, especially in surface

excavations for highways, buildings and open-pit mines. The geological history and human

activities often cause unstable conditions that lead to slope failure.

A quantitative assessment of the stability of a slope is clearly important when a judgement is

needed about whether the slope is stable or not, and decisions are to be made as a

consequence. The quantitative assessment of the stability is referred to safety factor, which is

calculated as a ratio between forces that withstand the structural soil mass from falling or

resisting forces and forces that causes the structural soil to failure or driving forces.

The safety factor evaluation is depended on a number of factors and the evaluation itself

depends on the types of analysis used. The factors affecting slope instability are generally

influenced by gravity forces and seepage forces (Craig, 2004), while type of analysis to be

used is depended on whether the safety factor is considered as short or long term applications.

According to Nash (1987) both of analysis type can be applied for any slopes, eventhough,

the consideration taken for short term application is much simpler and regardless the seepage

forces.

Deterministic, or physically based, models are based on physical laws of conservation of

mass, energy or momentum. The parameters used in these models can be determined in the


18/106


areas, however, has never seen a more extensive development, due to the regional variability

of geotechnical variables such as cohesion, angle of internal friction, thickness of layers, or

depth to groundwater. Furthermore, the calculation of safety factors over larger areas involves

an extremely large number of calculations, which could not be executed without the use of

GIS.

2.2 Slope Failure Mechanism

The slope failure occurs because of instability forces acting on a soil or rock mass. As all

masses on earths surface are affected by gravity forces, the slopes, which are geometrically

elevated above certain latitude and have a certain degree of slope, tends to slide to lower

latitude. Once the balance of the forces is disturbed by internal changes or external triggering

events, the mass structures are no longer able to withstand the forces that push the mass to a

lower position. The movements of the mass from the original positions due to imbalance

forces is called landslide.

The imbalance forces occurring on the soil or rock mass can be taken place due to internal

forces or external forces. The internal forces include strengths between particles and pore

water pressure, while external forces are the forces that act on the structural masses due to

triggering events such as earthquakes. The strengths between soil or rock particles are the

forces that generally withstand the soil mass from failure. Thus, in case of gravitation force

only that acts on the structural mass, the tangential components of gravity force to the slope

and the shear stress are the two forces that act inversely each other. Thus, if the shear stresses

are larger than the tangential gravity force, the structural mass will not move or deform as

illustrated in Figure 2.


19/106


Not MovedMoved

ggp

gt

ggp

gt

(a) Gentle Slope (b) Steep Slope

Figure 2 : Simplification Mass on Slope

Based on the type of mass movements, Varnes (1958) classified gravity-induced movements,

which was based on two variables, type of materials and type of movement. Movement types

are divided into falls, topples, slides, flows and a combination of those movements, while the

materials are divided into two classes, i.e. rocks and engineering soils, as listed in Table 1.

Table 1 : Classification of Landslides (Varnes, 1975)

Type of Material

Type of Movement Unconsolidated Sediment or Soil

BedrockCoarse Fine

Falls Rock Fall Debris Fall Earth Fall

Topples Rock Topples Debris Topples Earth Topples

Rotational Rock Slump Debris Slump Earth Slump


20/106


In fall movements, the movements occur by free fall or a series of leaps and bounds down the

steep slope. The movements are relatively free and lack of a slide plane. Depending upon the

type of slope materials involved, it may be a rock-fall, soil fall, debris fall, earth fall, boulder

fall, etc.

Slide type of movements occurs when the materials move as a block mass along the failure

plane. The failure plane is created as a result of imbalance forces that act in the plane in suchaway that the shear stresses of the particles are no longer capable to resist the soil or rock

mass. There are two types of slides as depicted in Table 1, i.e. rotational and translational

slides. The difference between those types is the type of the failure plane, translational slides

occur when the failure plane is a planar parallel to the surface, while rotational slides occur

when the failure plane is a circle.

The other two movements, topple and flow, are considered less sliding because the

movements are progressively. Topple type of movements occurs as a result of overturning of

the blocks rather than sliding, while flows are the movements of materials progressively

downward.

A distinction should be made between the factor that affects the slope stability and the

triggering factors that caused imbalances in forces. Both of the factors are explained in the

following sections.

2.2.1 Internal Factors Effecting Slope Instability

It is very important to recognize the factors that effect instability of a slope in order to know

the mechanism of failure and possible assessment of landslides. The factors that are those

which lead to a slide without any change in surface conditions, which involve unaltered

shearing stresses in the slope material (Ramiah and Chickanagappa, 1990) is called internal


21/106


2.2.1.1 Slope and Gravity Force

The angle at which material slopes is the major determining how much of the force of gravity

is directed downslope. If a block of rock or soil is placed on a flat surface, gravity acts

vertically and perpendicular to the flat surface and the full force of gravity is directed

downward ontothe surface. If the slope is rotated, some of the force of gravity is directed, or

resolved, perpendicular to the sloped surface, called normal force, and part is resolved parallel

to the surface, called shear force. As the angle of the sloped surface increases, the force of

gravity remains the same however the amount of that force resolved as shear force increases

and the amount resolved as normal force decreases as shown in Figure 2. At some point the

ratio of shear or normal force, called the coefficient of sliding friction, reaches a critical level

and the block begins to slide down the slope. Every material and slope type has an inherent

angle at which the material becomes unstable, called the angle of repose. Most unconsolidated

materials, such as soil or sediment, have angles of between 30 and 40 degrees. The angle of

repose for solid rock materials depends on the smoothness of the sloped surface and the nature

of the rock material, and can vary from 20 45 degrees.

2.2.1.2 Influence of Groundwater

Pore water is the water held within the void spaces, or pores, in the rock or sediment. Porefluid has two distinct effects on mass wasting risk. Pore water has a tendency to liquefy and

disaggregate unconsolidated materials, such as sediment or soil. Pore water tends to

destabilize rock layers on sloped surfaces. When pore water is under pressure it reducesthe

normal force holding rock layer stable on the sloped surface without reducing the shear force

that causes the downward motion of the rock.

2.2.2 External Triggering Events

External causes are those which produce an increase of the shearing stresses at unaltered

shearing resistance of the material. They include steepening of the slope, deposition of


22/106


consequently also shear strength may occur. In rock materials, breaking of cementation in

discontinuities or of intact rock may also occur.

Steepening of the slope is considered as human interaction rather than environmental effect. It

can be occurred when a mountainous area is cut for road, tunnel, aesthetic of residential, etc.

Modification of a slope causes changing in slope angle so that it is no longer at the angle of

repose. Then, the mass-wasting event happens in order to restore the slope to its angle ofrepose.

2.3 Fundamentals of Soil Parameters

A soil can be visualized as a skeleton of solid particles enclosing continuous voids which

contain water and or air. For the range of stresses usually encountered in practice the

individual solid particles and water can be considered incompressible; air, on the other hand,

is highly compressible. The volume of the soil skeleton as a whole can change due to

rearrangement of the soil particles into new positions, mainly by rolling and sliding, with a

corresponding change in the forces acting between particles. The actual compressibility of the

soil skeleton will depend on the structural arrangement of the solid particles. In a fully

saturated soil, since water is considered to be incompressible, a reduction in volume is

possible only if some of the water can escape from the voids. In a dry or a partially saturated

soil a reduction in volume is always possible due to compression of the air in the voids,

provided there is scope for particle rearrangement.

The stress-strain relationship for any material is used for analyzing the stability of structures,

slope, foundation, etc. Shear stress can be resisted only by the skeleton of solid particles, by

means of forces developed at the interparticle contacts. Normal stress may be resisted by the

soil skeleton through an increase in the interparticle forces. If the soil is fully saturated, the

water filling the voids can also withstand normal stress by an increase in pressure.


23/106


normal stresses and shear stresses through the interparticle contacts, but the pore fluid can

exert only all-round pressure. It is the stresses transmitted by the soil skeleton through the

inter particle contacts that control the strength and deformation of the soil. Where stresses

applied to the soil are wholly supported by the pore fluid pressure, they are not felt by the

contacts between particles and hence the soil behaviour is not affected. The effective stress

() acting on any plane is defined by the following equation :

= - u (1 )

in which is the total stress acting on the plane and u is the pore pressure.

2.3.2 Failure Criterion

Numerous failure criteria have been proposed for the stability analysis of soil mass, but most

of them are borrowed from basic mechanics. Since soil is a complicated material, some stress-

strain-time behaviour is highly non-linear. However, for practical uses the linear elastic model

and Mohr-Coulomb criterion and their shear equation are commonly used as expressed below:

= c + tan (2 )

where is the shear strength, c is the cohesion, is the total stress and is the angle of

internal friction. Depending on the type of analysis, total or effective stress analysis, the

parameters of c, and should be substitutes with c, and .

2.3.3 Drained and Undrained Strength

A distinction should be made between drained and undrained strength of cohesive materials.

As cohesive materials or clays generally posses less permeability compared to sand, thus, the

movement of water is restricted whenever there is change in volume. So, for clay, it needs


24/106


(a) Triaxial Undrained Test (b) Triaxial Drained Test

Figure 3 : Results of Undrained Triaxial Tests on Saturated Clay

2.3.3.1 Undrained Strength

It has been found empirically that the strength of a saturated soil is constant if its volume

remains unchanged. This description is given in Figure 3(a) which shows the result of testing

several identical specimens of saturated clay in a triaxial apparatus with different confining

pressures. If no drainage is allowed, the specimens have the same undrained shear strength

and it appears that the clay is purely cohesive. The different by an amount equal to the

difference in confining pressures, and hence the effective stresses are the same. This

behaviour is in contrast to what happens if the drainage is not restricted; the specimens would

have different drainage strengths as shown in Figure 3(b).

Normally, the drained and undrained strength are derived by laboratory test by testing a

specimen on a triaxial compression test. Then, the drainage condition is applied on the

specimens whether drained or undrained, the strength result is comparable to drainage

condition.


25/106


p

u I0037.011.0'

c+=

(3 )

British Standard gives a rough guide of undrained shear strength in relationships with the

consistency as shown in Table 2. Bjerrum and Simons (1960) proposed the same correlation

as proposed by Skempton in the form of chart as shown in Figure 4. Another correlation

proposed by Carter and Bentley (1991) correlates natural undrained shear strength and

Liquidity Index (LI) as shown in Figure 5.

Table 2: Consistency-Strength Relationship from Field Inspection (BS 8004: 1986)

Consistency Field IndicationsUndrained ShearStrength (kPa)

Very Stiff Brittle or very tough > 150

Stiff Con not be moulded in thefingers

75 - 150

Firm Can be moulded in the fingers

by strong pressure

40 - 75

Soft Easily moulded in the fingers 20 - 40

Very Soft Exudes between the fingerswhen squeezed in the fist

< 20


26/106


The undrained shear strength usually uses when a total stress analysis is used. This correlation

explains that the relationships between undrained shear strength increases to the depth.

Figure 5 : Relationship between the Natural Shear Strength of Undisturbed Clays and Liquidity Index

(Carter and Bentley, 1991)

2.3.3.2 Drained Strength

When the water movement is not restricted, a specimen placed on triaxial compression test

will show different strengths for different confining pressures as shown in Figure 3. By


27/106


Figure 6 : Correlation between Effective Friction Angle and Plasticity Index for Fine-Grained Soils

(NAVFAC DM-7)

2.3.3.3 Residual Strength

For analysis of shear characteristics of overconsolidated soils relating to stability problems,

ordinary shear tests are not suitable because they give too high a shear value. Skempton

(1964) showed that the strength remaining in laboratory samples after large shearing

displacement corresponded closely with the computed strength from actual landslides;

therefore, he proposed a residual strength concept as shown in Figure 7. Because of the peak

or residual shear parameters are relatively time consuming and expensive, for practical uses

some simple experimental equations and correlations for estimating these strength parameters

have been proposed by numerous in investigators such as proposed by Jamiolkowski and


28/106


Figure 7 : The Concept of Residual Shear Strength

2.3.4 Choice Between Total and Effective StressA decision must be made when analysing slope stability whether to use a total or an effective

stress analysis. The choice generally follows from the classification of a stability problem as

short or long term. Slope failures generally result from a change of loading on the soil and if

this occurs quickly, the stability during and immediately after the change may need to be

assessed. This will be particularly important if the change of loading results in a change of

pore-water pressure in the soil mass and the change is rapid compared to consolidation time of

the soil or if the loading is a natural fluctuation of groundwater levels as occurs in natural

slopes the problem is considered to be long term.

Theoretically, both total and effective stress analyses could be applied to analyze any slope,

although since soils are predominantly frictional materials an effective stress analysis seems

inherently more logical especially for the analysis of long-term problems. In practice for

short-term stability problems a total stress analysis is often simpler and more convenient as

there is usually difficulty in predicting pore-pressure changes.


29/106


Another explanation related to total and effective stress is given by the permeability of the soil

structure. If the permeability of the soil is low, a considerable time will elapse before any

significant dissipation of excess pore water pressure will have taken place. At the end of

construction the soil will be virtually in the undrained condition and a total stress analysis will

be relevant. In principle an effective stress analysis is also possible for the end-of-construction

condition using the appropriate value of pore water pressure for this condition. However,

because of its greater simplicity, a total stress analysis is generally used. It should be realizedthat the same factor of safety will not generally be obtained from a total stress and an effective

stress analysis of the end-of-construction condition. In a total stress analysis it is implied that

the pore water pressures are those for a failure condition, while in an effective stress analysis

the pore water pressures used are those predicted for a non-failure condition.

2.4 Stability Analysis MethodsThe stability analysis methods are categorized into two basic approaches, i.e. (1) Limit

Equilibrium Analysis and (2) Deformation analysis, and It is also depended on the type of

analysis used, i.e. (1) Total Stress Analysis and (2) Effective Stress Analysis. So far, limit

equilibrium methods are the most common used for assessing slope stability, while the type of

analysis can be used both total and effective stress analysis.

Limit equilibrium approach postulates that the slope might fail by a mass of soil sliding on a

failure surface. When the failure occurs, the shear strength is fully mobilized all the way along

the failure plane, and the overall slope and each part of it are in static equilibrium. In the

analysis of stable slopes the shear strength mobilized under equilibrium conditions is less than

the available shear strength, and it is conventional to introduce a factor of safety F defined by:

stabilityforrequiredStrengthShear

StrengthShearAvailableFS = (5 )


30/106


curve or plane section, thus, it is necessary to consider the likely shape of the failure surface.

Table 3 presents the various method of limit equilibrium and their formed of failure planed.

The chosen of analysis type determines the shear strengths should be used for the analysis.

The shear strength of the soil is normally given by the Mohr-Coulomb failure criterion as

follow :

s = cu= su (for undrained total stress analyses) (6 )

s = c + tan (for drained effective stress analyses) (7 )

where, cuor suare the undrained shear strengths and c and are the effective cohesion and

the effective friction angle, respectively.

Table 3 : Methods of Analysis

Method Circular Non-CircularAssumption about

Interslice force

Infinite Slope * Parallel to Slope

Wedge Analysis * Defined Inclination

Total Stress Analysis *

Ordinary or SwedishMethod

*Resultant parallel tobase of each slice

Bishop's Method of Slices * (*) Horizontal

Janbu Simplified * * Horizontal

Spencer's Method * (*) Constant Inclination

Janbu Rigorous * * Define thrust line

As listed in Table 3, there are many limit equilibrium methods available; however, only linear


31/106


2.4.1 Infinite Slopes

Infinite slope is one of the simplest approaches for slope stability analysis. According to

Skempton and Delory, 1957, a landslides of a planar mass of soil occurs in slip surface which

is approximately parallel to the ground surface can be analyzed effectively using the infinite

slope analysis. The name infinite-slopes is given to earth masses of constant inclinations of

unlimited extent and uniform conditions at any given depth below the surface. Thus, in this

analysis the soil is assumed to slide on a plane slip surface which is parallel to the ground

surface and the slope is assumed to be infinite in extent at a certain inclination to the

horizontal (Nash, 1987). Even though, such assumptions adopted by infinite slopes are

realistically never taken place, infinite slope method provides a good general idea about the

stability of a slope. Based on the type of materials and groundwater occurrence, infinite slope

can be determined in several cases as elaborated below.

2.4.1.1 Cohesive Material in Dry Condition

As shown in Figure 8, a case of slope with slip failure parallel to the ground surface is applied

with the slope is infinite extent and no seepage is assumed. The gravity force (W) of a column

soil mass with thickness b is given by Hb. As a consequence of angle i, the weight of the

column mass can be divided into two components namely S, the force along the inclination of

the block and N, the force normal to the inclination of the block. Both of the force can be

expressed as follow, while forces acting parallel to the slip surface, F 1 and F2are assumed

equal and opposite, and are therefore ignored in the analysis.

Normal Force (N) = W cos i = Hb cos i (8 )

Shear Force (S) = W sin i = Hb sin i (9 )


32/106


Figure 8 : Forces on element of infinite slope (Cernica, 1995)

Resolving the two forces in Equation (8) and (9), the normal and shear stress can be derived

by dividing the two forces by the width of the soil mass on a plane failure, which is b/cos i.

Thus, the normal stress is given by :

icosHicosb

N 2== (10 )

and the shear stress is given by :

icosisinHicosb

S== (11 )

where, is the unit weight of soil.


33/106


where, c and are the cohesion and internal friction angle, respectively. Thus, substituting

Equation (11) and (12) into Equation (5), the safety factor for this condition becomes as

follow:

itan

tan

icosisinH

c

icosisinH

tancFS

+

=

+= (13 )

For clayey soil, it is interesting to defined a critical height (Hc) of the clay stratum, which can

be expressed by the formula :

=

tanitan

iseccH

2

c (14 )

2.4.1.2 Cohesive Material with Groundwater Effect

For a condition with groundwater effect, the pore pressure at a depth H equals wHwcos2i.

The effective pressure is (H - w Hw) cos2i, where w is the unit weight of water and Hwis the

height of water above the failure plane. Assuming that the thickness of water above the failure

plane equals to mH, then the shear resistance is given by :

s = c + (H - w Hw) cos2i tan

s = c + (H - w mH) cos2i tan = c + (- w m) H cos

2i tan (15 )

The factor m, in the above equation termed as the wetness index gives the condition of

saturation of the soil. If m equals to one, the soil is in a completely saturated condition while

the value zero indicates dry conditions of the soil. Similar to the procedure described above,

the safety factor in this condition is calculated by the following relationship.


34/106


of cohesionless soil in dry condition, the c and m in Equation (16) become zero and the safety

factor for this condition is given by :

itan

tanFS

= (17 )

Equation (17) expresses that for cohesionless soil the critical angle of the slope is equal to the

internal friction angle under dry condition.

2.4.1.4 Cohesionless Material with Groundwater Effect

Looking at Equation (16), for this condition, the wetness index, m, is no longer zero because

there is an effect of groundwater table. Thus, solving Equation (16) for this condition, the

safety factor becomes,

( )

itan

tanmFS w

= (18 )

2.4.2 Total Stress AnalysisThe permeability of clays is very much less than that of sands and this inhibits the movement

of water if there is tendency to change volume. As a result it may take years after a change of

surface loading on a deposit of clay for excess pore pressures to dissipate and for the effective

stresses to reach equilibrium. In this case, the condition of the soil is undrained where the

excess pore water pressures are unable to dissipate. However, the shear strength of a soil is

dependent on the effective stresses whatever the condition of drainage. Thus, when movement

of the pore water is restricted, the pore pressure increases in a soil which is trying to contract

and decreases in one trying to dilate. The change of pore pressure directly affects the effective

stresses and hence the shear strength


35/106


The stability analysis calculated by infinite slope for cohesive soil can be applied on total

stress analysis by assuming the internal friction angle () equals to zero. The explanation

about this analysis is given in Section 2.5.1.1. Another method for total stress analysis is

developed by Taylor (after Craig, 2004), which is assumed fully saturated clay under

undrained conditions as shown in Figure 9.

Figure 9 : Total Stress Analysis

As shown in Figure 9, only moment equilibrium is considered in the analysis and undrained

shear strength are used. In section, the potential failure surface is assumed to be a circular arc.

A trial failure surface (centre O, radius r and length La) is shown in Figure 9. Thus, the safetyfactor can be expressed as follow,

dW

rLcFS au= (19 )


36/106


Based on the principle of geometric similarity, Taylor (after Craig, 2004) published stability

coefficients for the analysis of homogeneous slopes in terms of total stress. For a slope of

height H the stability coefficient (Ns) for the failure surface along which the factor of safety is

a minimum is as follow,

HFS

cN us

= (20 )

and the safety factor can be expressed as follow:

HN

cFS

s

u

= (21 )

The coefficient Ns depends on the slope angle and the depth factor D, where DH is the

depth to a firm stratum. Figure 10 shows the Taylors stability charts.


37/106


pore water pressure. However, this method should be used with caution due to generalization

in pore water pressures. It might be possible to use this method with assumption that the

clayey soils are heavily impermeable and thus, the groundwater pressures are not easily

dissipated.

2.4.3 Wedge Analysis

There are situation in which the slip surface can be approximated by two or three straight

lines. This may occur when the slope is underlain by a strong stratum such as rock or there is

a weak stratum included within or beneath the slope. In these circumstances an accurate

assessment of the stability may be made by splitting the slope into several blocks of soil and

examining the equilibrium of each block.

In this method, the trial sliding mass is divided into two or three large sections or wedges. Theupper wedge is called the driving or active wedge, while the lower wedge is called the

resisting or passive wedge. In a three-wedge system, the middle segment is sometimes

referred to as the sliding block. The potential failure surface is simplified to a series of planes.

2.4.4 Non-Linear Methods

There are numerous non-linear methods, however, all of those non-linear methods has the

same assumption of failure plane that this method considers non-linear failure planes. One of

these methods is called as Method of Slices. There are also many methods of slices developed

by researcher such as General Formulation developed by Fredlund and Krahn, Bishops

Routine Method, Janbus Simplified Method, etc.

Despite the fact that there are many methods of slices, however, they share the same principle

that the slope being analyzed is divided into a number of slices. First of all, an assumed non-

linear failure plane is determined either circular or a combination between block and circular.

h h li d i d i hi h d f d h d fi d f il l h


38/106


based on an assumed central point of sliding. The safety factor is then determined as the

balance between forces that causing sliding against the central point and that of withstanding

the block against failure.

2.4.5 Model Based on Root Cohesion

This method is adapted by Montgomery and Dietrich (1994), Van Westen and Terlien (1996)

and de Vleeschauwer and De Smedt (2002), which combined the stability analysis with the

cover type of the land. Since, stability of a slope is not only depended on the internal factor

but also external factors, the method adopt the effect of external factor such as surcharge

pressure and root cohesion. By applying root cohesion, it means that the method also take into

account the possibility of translational failure because of land cover type. This method can be

expressed by the following formula:

itan

tanm1

isinD

CCFS

e

w

e

rs

+

+= (22 )

where, FS is the safety factor, Csand Crare the effective soil and root cohesion governed by

the vegetation type, respectively; D is the depth of the soil above failure plane; is the angle

of internal friction; i is slope angle; wis the unit weight of water and eis the effective unit

weight of soil as defined by Westen and Terlien (1996).

Actually, this method was developed based on infinite slope, however there are differences in

assumption and the philosophy behind the formula. First, the assumption of soil depth is taken

as the thickness of soil above the failure plane and it is perpendicular to the failure plane,

while in ordinary infinite slope the soil depth is the vertical depth against failure plane.

Secondly, there is a new parameter introduced in the formula that is root cohesion. By

introducing this parameter, the formula are no longer satisfy ordinary infinite slope equation,


39/106


infrastructure development (Joshi, 2002). The landslide potential mapping are becoming

useful for watershed management and they are proving themselves a good assistant to help

decision makers for careful development of hill slope which eventually can reduce the

economic and social losses, reducing the damage potential. Protection plans require the

description of scenarios that can be defined by means of simulation with mathematical

models, which incorporates the occurrence conditions of the failure including the triggering

mechanism

Regional landslide evaluation and mapping have been actively pursued by research

institutions and government agencies for a long time. Among different techniques of landslide

hazard model such as statistical approach, one widely used technique now a day is

deterministic approach. This approach seems to be superior because it has direct linkage to

physics. Evolution of fast processing computers and Geographic Information System (GIS)has enhanced its capacity of mapping. GIS technologies could provide a powerful tool to

model the landslide hazards for their spatial analysis and prediction. This is because the

collection, manipulation and analysis of the environmental data on landslide hazard can be

accomplished much more efficiently and cost effectively (Carrara and Guzzetti, 1999 and

Guzzetti et al., 1999). Many GIS-based analysis models and quantitative prediction models of

landslide hazard have been proposed since the beginning of GIS application in geohazards

research in the late 1980s (Carrara, 1983; Van Westen, 1994; Carrara et al., 1991; Carrara et

al., 1995; Carrara and Guzzetti, 1999; Jade and Sarkar, 1993; Chung et al., 1995; Chung and

Fabbri, 1998 and Chung and Fabbri, 1999).

2.5.1 Model Concept

The analysis of slope stability using GIS requires the overlying of various thematic maps such

as slope map derived from the Digital Elevation Model (DEM), land use map and soil map.

While for rainfall-triggered landslides, there are two main approaches for rainfall-triggered

l d lid di ti (1) t ti ti l l ti d f ti t h i t t bli h th


40/106


et al., 1995; Montgomery and Dietrich, 1994; Wu and Sidle, 1995 and Pack et al., 1998).

However, most models are valuable for certain applications and certain region.

The following sections discuss how the methods explained in Section 2.4 are applied for the

analysis of stability. The study mainly focuses on the stability for cohesive soil with emphasis

on Infinite Slope Method and Taylor Method by applying two stress cases, i.e. total and

effective stress.

2.5.1.1 Using Infinite Slope with Total and Effective Stress

The difference between total and effective stress analysis is the use of strength parameters and

the used of pore water pressures. For cohesive soil under effective stress analysis, the

cohesion should be replaced by effective cohesion (c) and if the cohesive soil is subjected to

internal friction angle, then it should be replaced by effective internal friction angle (). On

the other hand, for cohesive soil under total stress analysis, undrained shear strength (c u)

might be used and angle of internal friction () equals to zero (Nash, 1987) with pore pressure

being zero. Thus, the formulas for cohesive soil in dry condition (Total Stress Analysis)

becomes :

icosisinH

c

FS

u

= (23 )

and, the cohesive soil with groundwater influence (Effective Stress Analysis), the formula

becomes:

( )icosisinH

'tanicosHm'cFS2

w

+= (24 )

For effective stress analysis, m is the soil wetness index, which is defined the relative height


41/106


consideration is only for total stress analysis. Equation (21) can be used to estimate the safety

factor by applying stability coefficient as shown in Figure 10, which is depended on angle of

the slope and thickness of the stratum.

2.5.1.3 Assessment of Stability Classes

There is no general rule on how the safety factor should be classified. For instance, Van

Westen and Terlien, 1996, categorized safety factor into 3 classes, below one, which means

unstable, between 1 and 1.5, which means moderately stable, and above 1.5, which means

stable. SINMAP, Stability Index Mapping, an extension computed added modelling for slope

stability in ArcView, uses 6 classes for safety factor including division of safety factor below

1.

In the design of slopes, the factor of safety on shear strength traditionally has several

functions :

1. To take into account uncertainty of shear strength parameters due to soil variability, and

the relationship between the strength measured in the laboratory and the operational field

strength.

2. To take into account uncertainties in the loading on the slope such as surface loading, unit

weight, pore pressures, etc.

3. To take into account the uncertainties in the way the model represents the actual

conditions in the slope, which includes (a) the possibility that the critical failure

mechanism is slightly different from the one which has been identified, and (b) that the

model is not conservative.

4. To ensure deformation within the slope are acceptable.

Thus, a safety factor of 1 does not indicate that failure of a slope is necessarily imminent. The

real safety factor is strongly influenced by minor geological details, stress-strain


42/106


the stable conditions. Safety factor classes used by Westen and Terlien (1996) is strictly

categorized a slope being unstable, moderately stable or stable, however, for analysis, it is

necessary to quantify the area falls in safety factor between 1 to 1.5. Thus, it is convenient to

classify the safety factor in four classes as shown in Table 4.

Table 4 : Stability Clases

Safety Factor Slope Stability Class Remarks

FS >1.5 StableOnly major destabilising factors lead to

instability

1.25 < FS < 1.5 Moderately StableModerate destabilising factors lead to

instability

1 < FS < 1.25 Quasi Stable Minor destabilising factors can lead toinstability

FS < 1 Unstable Stabilising factors are needed for stability

2.5.2 Hydrological Model

One of the possible triggering mechanisms of slope failure is caused by the rapid increase of

ground water table, which finally affect the increasing pore water pressure. Beven and

Kirkby, 1979, developed soil saturation in function of hill slope topography as the wetness

index as follow,

=

tanalnm (25 )

where a is the contributing area per unit contour length and is the slope of the pixel.


43/106


D

SR

2

D

m

+

= (26 )

where, D is depth of soil [m], R is recharge or maximum daily rainfall [m], and S is Specific

Yield of soil [-].


44/106

Chapter 3 : Materials and Methods 32

CHAPTER 3 : MATERIALS AND METHOD

3.1 General

As this study is the continuation of the previous study done by Ram Lakan Ray, 2004, thus,

the necessary data for the analysis is collected by the previous analysis. In general, the study

area shown in Figure 11 has shown active landslides as reported by Ram Lakan Ray at

Krishna Bhir. It is covered not only by soil but also rocks (cliff), however, the existence of

rock is very small compared to soil. Besides, due to this study mainly focuses on clayey soils,

thus the existence of rock does not affect the result.

Figure 11 : Location of the Study Area (Ray, 2004)

The study mainly focuses on the applicability of total and effective stress analysis method by

applying infinite slope and Taylor methods on the study area. To be able to compare

objectively between the two analyses cases, the study is only conducted on a clayey soil. Even


45/106


3.2 Data Availability

Analyzing slope stability on a regional area requires two types of data, i.e. geotechnical

including topographical and hydrological data. Both of the data are equally important since

the geotechnical data represent the characteristics of the materials, while the hydrological data

represent the amount of rainfall in the area. However, sometimes it is difficult to collect such

information especially in rural area of a developing country, where information on earth

resources is always connected to the budget provided and development priority given by the

government. It is also the case that research and collection of data in a developing country are

not well organized.

Unfortunately, the situation is the same in Nepal for the study area. There is no soil map, land

use map, records of soil parameter and meteorological station inside the study area. The soil

map was then interpreted based on the Project Report prepared by Department of Roads

(DoR), Ministry of Works and Transport, Nepal. For land use map, it was produced by aerial

photographs prepared by Department of Survey. While for hydrological data, it was derived

from four meteorological stations around the study area, which is located at Dhading, Aru

Ghat, Gorkha and Rampur.

Since there is no actual measurement on soil parameters for this study area, thus, the soil

parameters were interpreted and adapted from various relevant books and papers. Even

though, the interpretation for soil parameters from various publications is quite useful to be

used for the analysis; however, the approach might not be accurate and involve a big

assumption due to the variation of soil parameters on the site.

For this study, the available data used from the previous study consist of four maps and a set

of hydrological data. The maps are DEM, slope map, land use map and soil map, while the

hydrological data has been calculated using statistical software as explained in Section 3 2 2


46/106


rows and 1237 columns, covering the area between 561524m to 586264 m Easting and

3070318 m to 3084338m Northing. The unit of the map is in meters.

Figure 12 : Digital Elevation Model (DEM) of the Study Area (Ray, 2004)

Ch 3 M i l d M h d 35


47/106


From the soil map, it was identified that the study area is covered by 11 soil types as shown in

Figure 14. There are three types of cohesive soil in the study area, i.e. Inorganic Silt, Organic

Silt and Sandy Clay as shown in Figure 15, covering a total area about 84.057 km2.

Figure 14 : Soil Map of the Study Area (Ray, 2004)

Ch t 3 M t i l d M th d 36


48/106


Table 5 : Various Types of Soils and Corresponding Slope Angle

Angle (degree)Soil-Code Soil Type Count Area (km2)

Min Max

1 Clayey Sand 241513 96.6052 0.0591 49.9658

2 Poorly G. Sand 82881 33.1524 0.3526 51.0144

3 Silty Gravel 107882 43.1528 0.1908 59.7071

4 Gravelly Sand 20053 8.0212 0.0106 50.925

5 Sandy Clay 117980 47.192 0.2843 60.945

6 Rock 1408 0.5632 1.7308 46.4667

7 Inorganic Silt 85819 34.3276 0.0193 57.7121

8 Poorly G. Gravel 35238 14.0952 0.244 57.958

9 Organic Silt 6344 2.5376 1.2014 43.4273

10 Silty Sand 84749 33.8996 0.1215 52.2431

11 Clayey Gravel 68815 27.526 0.3208 60.4487

Total 852682 341.0728

From the land use map, it was identified that the study area is covered by 9 types of land use

as shown in Figure 16. The study area is covered majority by three types of land cover,

agricultural land, bush land and forest with percentage of 48 %, 29 % and 20 %, respectively.



49/106


3.2.2 Available Hydrological Data

There are four meteorological stations surrounding the study area. Of the four meteorological

stations surrounding the study area, only three of them were analyzed for developing

hydrograph. The data derived from meteorological stations at Rampur was not considered

because it is located in a plain area where climate and rainfall patterns are completely

different than the study area. However, only the closest rainfall stations to the study area were

considered, i.e. Dhading. The rainfall data collected from the Department of Hydrology,

HMG, Nepal, consists of yearly rainfall data from 1956 to 1996. The rainfall frequency

analysis is developed using SMADA 6.0 software with a Log Pearson Type III distribution.

Table 6 presents the results for the study area.

Table 6 : Rainfall Prediction of Study Area with SMADA 6 Software (Ray, 2004)

Exceedence Return Period Daily Rainfall Standard

Probability (years) (mm) Deviation (mm)

0.995 200 370 102

0.990 100 322 74

0.980 50 277 52

0.960 25 235 35

0.900 10 185 20

0.800 5 150 130.667 3 124 9

0.500 2 103 8

3.3 Applied Methodology

As explained in the previous chapters, the safety factor for a regional area can be derived with

the use of GIS where the information related to the spatial data is stored in various map such

as topography, soil and land use map. The spatial information of a map in GIS is stored in

attribute tables of the respective map. Then the calculation of the safety factor for every grid

cell is done by applying the method in every grid.



50/106


As explained in Chapter 2, the Taylor method is a total stress analysis where the safety factor

is calculated based on a stability coefficient (Ns) expressed in Equation (21). The stability

coefficient developed by Taylor (1948) is expressed in terms of slope angle with different

thickness of soil. For this study, the thickness of the soil is assumed to be infinite and thus

only one line of the stability coefficient developed by Taylor is used, i.e. the line with D = .

To be able to calculate the stability coefficient in spatial analysis, the stability coefficient for

D = was first digitized. The data were then correlated using polynomial regression to be

able to derive the mathematical equations. As shown in Figure 10, the stability coefficient for

D = can be divided into two parts, i.e. constant and a polynomial function, for slope angle

of 0 to 52.8 and above 52.8, respectively. The mathematical equations derived from the

polynomial regression for stability coefficient of D = are as follow:

Ns= 0.183 for 0 < 52.8 (27 )

Ns= 6.10-7

3 10-42+ 0.0079 - 0.0263 for > 52.8 (28 )

For infinite slope methods, both total and effective stress analysis are applied with different

soil parameters. Total stress analysis applied on infinite slope uses undrained cohesion and

= 0, while effective stress analysis applied on infinite slope uses effective shear strengths.

3.3.1 Soil Parameters Determination

As explained previously, soil parameters for the study area were not available, thus, to

estimate the soil parameters, published references were used. There are many references

related to soil parameters of cohesive soil. Three strength parameters should be determined

using the available references, i.e. undrained shear strength (su), effective cohesion (c) and

effective angle of internal friction (), for three cohesive soil types identified in the study

area. The three cohesive soils identified in the study area are Sandy Clay (CL), Inorganic Silts

(ML or MH) and Organic Silts (ML)



51/106


ranging from 17 % to 38 % and thus, the plasticity index (PI) ranges from 4 % to 30 % as

shown in Table 7.

Table 7 : Index Properties of Soil Based on Deoja et al. (1991)

Unit Weight Atterberg Limit (%)

SoilCode

Soil Type ClassificationWater

ContentTotal Dry

LiquidLimit(LL)

PlasticLimit(PL)

PlasticityIndex(PI)

LiquidityIndex(LI)

(%) (kN/m3) (kN/m3) (%) (%) (%) (-)

5Sandy

ClayCL 19 18.50 15.55 33 17 16 0.1 - 0.4

ML 27 18.50 14.60 30 26 4 0.2 - 0.47

Inorganic

Silts MH 48 17.00 11.49 68 38 30 0.3 - 0.5

9

Organic

Silts OL 24 13.50 10.89 42 29 13 0.4 - 0.7

Table 8 : Undrained Shear Strength from Various References

Undrained Shear Strength (su)

Soil

CodeSoil Type Classification

Plasticity

Index

(PI)

Liquidity

Index

(LI)

Soil

Thickness Skempton

Bjerrum

&

Simons(1953)

Carter &

Bentley

(1959)

(%) (%) (m) (kN/m2) (kN/m2) (kN/m2)

1 3.13 3.24

2 6.26 6.48

3 9.39 9.715

SandyClay

CL 16 0.1-0.4

4 12.52 12.95

20 - 60

1 2.31 1.85

2 4.62 3.70

3 6.93 5.55ML 4 0.2-0.4

4 9.24 7.40

20 - 40

1 1.59 1.66

2 3 18 3 31

7Inorganic

Silts



52/106

p

Based on the atterberg limit derived from Deoja et al. (1991), the undrained shear strengths

were determined using the available correlations. Undrained shear strengths given by Bjerrum

and Simons (1960) and Skempton (1957) share the same correlation based on the effective

overburden pressures. However, those correlations show very low value as shown in Table 8

compared to the one given by Carter and Bentley (1959). This is caused by the fact that both

of correlations are mainly applicable only for normally consolidated clay or marine clay,

which is not applicable for this mountainous area. Thus, the correlations developed by Carter

and Bentley (1991) are more appropriate to be used.

The effective internal friction angle () was determined from the correlation chart explained

in Chapter 2 and compared to the one given by Deoja et al. (1991). Again, the correlations

given by NAVFAC DM7 are higher compared to the one given by Deoja et al. (1991).

However, the effective internal friction angle given by Deoja et al. (1991) seems to be at the

lower bound of the correlations given by NAVFAC DM7. Thus, the average values of the

correlations between both are used for further analysis. Table 10 presents the parameters of

the soil used for the analysis of safety factor.

Table 9 : Effective Stress Parameters for the Study Area

Effective Strength

from Deoja, et. al

(1991)SoilCode

SoilType

Classification

PlasticityIndex

(PI)Cohesion

Friction

Angle

Effective

Friction

Angle *

Used

Effective

Friction

Angle

(%) (kN/m2) () () ()

5

Sandy

Clay CL 16 20 28 32 30

ML 4 7 32 357

InorganicSilts MH 30 10 25 28

30



53/106

Table 10 : Soil Parameter Used for the Analysis

Effective Strength

SoilCode

SoilType

Classification

Total

UnitWeight

Undrained

Shear

Strength(su)

CohesionFriction

Angle

Specific

YieldConductivity

(kN/m3) (kN/m2) (kN/m2) () (m/day)

5Sandy

ClayCL 18.5 20 - 60 20 30 0.12 1.E-08

ML7

InorganicSilts MH

18.5 20 - 40 10 30 0.18 1.E-05

9OrganicSilts

OL 13.5 10 - 20 10 28 0 1.E-06

3.3.2 Model Development

The current study follows the flow chart illustrated in Figure 17. Basically, there are 2 groups

of map produced, i.e. the critical height (Hc) maps and safety factor (FS) maps. The critical

height maps are determined based on total and effective stress analysis by applying either

taylor method on total stress analysis (TSA) or infinite slope method on total and effective

stress analysis (ESA). However, the critical height maps for ESA are only calculated for

conditions of dry, half saturated and fully

2005 Prigiarto Yonathan Hokkal Drenaje y Subdrenaje

Documents