Semi-Quantitative Approaches For Landslide Assessment & Prediction PDF

Springer Natural Hazards
Sujit Mandal
Ramkrishna Maiti
Semi-quantitative
Approaches
for Landslide
Assessment and
Prediction
Springer Natural Hazards
More information about this series at http://www.springer.com/series/10179
Sujit Mandal Ramkrishna Maiti
•
Semi-quantitative
Approaches for Landslide
Assessment and Prediction
123
Sujit Mandal Ramkrishna Maiti
Department of Geography Department of Geography and Environment
Raja N.L. Khan Women’s College Management
Paschim Medinipur, West Bengal Vidyasagar University
India Paschim Medinipur, West Bengal
India
ISBN 978-981-287-145-9 ISBN 978-981-287-146-6 (eBook)

DOI 10.1007/978-981-287-146-6
Library of Congress Control Number: 2014951664
Springer Singapore Heidelberg New York Dordrecht London
© Springer Science+Business Media Singapore 2015

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of
the Copyright Law of the Publisher’s location, in its current version, and permission for use must always
be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright
Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Preface
In the present time, our natural environment is frequently associated with a number
of destructive hazards such as flood, earthquake, storm surge, drought, landslide,
volcanic eruption, etc., worldwide. All these environmental problems may bring
about a tremendous change in the geomorphic system over the land surface and
may also divide a system into a number of subsystems. It is the task of Physical
Geographers to study natural hazards and their destructive impacts over the earth’s
surface in a scientific way. Here, we studied in detail geologic, geomorphic, and
hydrologic processes responsible for landslides in a representative mountain
watershed, Shivkhola of Darjiling Himalaya, where landslides are very much active
and have destroyed communication lines, settlements, and tea estates. A quantita-
tive understanding of all the triggering landslide inducing parameters can provide a
clear vision of the changes and evolution of landforms. The aim of this book is to
provide an integrated knowledge and understanding of the application of semi-
quantitative approaches in evaluating various landslide triggering factors and
assessment of landslide susceptibility as well as landslide risk, which can help
planners and policymakers to check the magnitude of landsliding in mountain
terrains.
The concerned area, Shivkhola, is badly affected by slope instability jeopar-
dizing the economy and social systems; taking a toll of lives, lands, and properties;
promoting road blockage and hampering flow of tourists; destroying pipelines of
water; and causing subsidence of road and rail cut benches. The effects are con-
veyed over a long distance, both upslope and downslope, by disruption of transport
and hydrological systems and thus invites instability into both physical and social
systems. Most landslides of this trouble-torn district of Darjiling are concentrated in
this area where most of the torrential jhoras (Hill Streams) at their upper catchments
develop a potential sinking zone, and the Hill Cart Road and Narrow gauge Rail
when traversing through these zones invite catastrophic slides. The present work
tries to identify the zones of potential slides by investigating into the systematic
interaction among the number of prominent physical and human triggering factors.
The geotechnical attributes, viz., major principal stress, minor principal stress,
normal stress, shear stress, angle of internal friction, angle of rupture, cohesion,
v
vi Preface
shear strength, and safety factor of the Shivkhola watershed have been determined
using the Tri-axial Compression Test from the geotechnical laboratory, Geological
Survey of India (Kolkata). There is every possibility for the generation of geo-
morphic thresholds for initiation of slides due to hydrologic factors. Thus debris
slide occurs and the slope on scar face is reduced to that of repose angle to attain
temporary stability through internal feed back in a process of homeostatic adjust-
ment. Remote Sensing and GIS-based Landslide Susceptibility Zone Map reveals
that Paglajhora, Shiviter, Gayabaria, and Tindharia are more susceptible in terms of
slope instability. The prepared Landslide Hazard Risk Map also states that Tin-
dharia, Gayabari, Shiviter, and Paglajhora are prone to landslide risk because of
prevalence of risk elements such as road network and settlements. As most land-
slides occur along Hill Cart Road (NH-55), and a huge amount is spent on Post
Slide Management, an attempt toward pre-slide management of the susceptible
areas has to be introduced with immediate effects with less efforts and investment.
A sector-wise job assignment has to be made for regular supervision of the slope
stability and a serious drive for pre-slide management of potential slope failure
zones should be introduced; these should be considered as emergency as in post-
slide condition. In such a highly instable region the protection of slope and soil is a
great challenge through rational use of these resources for harnessing greater utility
over a long time.
Midnapore, India, August 2014 Sujit Mandal

Ramkrishna Maiti
Acknowledgments
It is a moment of jubilance and euphoria to me because of the candid endeavor so

far made, be it small or large in this book, in illuminating the unexplored domain of
knowledge in “Semi-quantitative Approaches for Landslide Assessment and Pre-
diction”. We would like to pay our humble regards, indebtedness, and immense
sense of gratitude to our respected Teacher, Professor Late Subhas Ranjan Basu,
Ex-Head of the Department Geography, University of Calcutta, who had shared
with us a lot of his experiences, suggestions, and encouragement till his last breath
so that we always bear his fragrance of affection and bliss showered on us. His
prudent direction, all sorts of support, and sleepless effort for preparation of this
book can never be forgotten.
We also acknowledge the financial support of University Grant Commission,
New Delhi, through granting a research project that helped in data collection and
subsequent analysis involved in this research work. We would like to express our
appreciation and thanks to Mr. Manik Das Adhikari, Project Fellow, IIT, Kharagpur
for his immense support throughout the work.
We wish to record our sincere gratitude to all the officials of all Departments,
including Geological Survey of India (GSI), Kolkata, Tea Research Centre,
Department of Agriculture, Department of Forest of Kurseong, Government of
West Bengal who had provided valuable information in the study area. Thanks are
due to Department of Geography and Environment Management, Vidyasagar
University, Paschim Medinipur, West Bengal, India, and the authority of Raja N.L.
Khan Women’s College, Paschim Medinipur, W.B., India and especially the
department of Geography and Zoology for offering logistic support and fulfilling all
sorts of lack occurred in that institution for involvement in this research work.
It is difficult for us to fully express our gratitude to our families, parents,
brothers, sisters, relatives, and students for their sacrifice, constant love, cordial
affection, and inspiration throughout the work.
Midnapore, India, August 2014 Sujit Mandal

Ramkrishna Maiti
vii
Contents
1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 1
1.1 Key Concepts on Landslide . . . . . . . . . . . . . . . . . . . . . . . ... 1
1.2 A Brief Account of Some Destructive Landslides
Occurrences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Relevance of the Landslide Study. . . . . . . . . . . . . . . . . . . . . . 3
1.4 Landsliding in Darjiling Himalaya . . . . . . . . . . . . . . . . . . . . . 17
1.5 Mechanism and Causes of Landslides . . . . . . . . . . . . . . . . . . . 23
1.6 Types and Classification of Landslide . . . . . . . . . . . . . . . . . . . 30
1.7 Objectives of the Present Work . . . . . . . . . . . . . . . . . . . . . . . 34
1.8 Applied Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
1.9 Data Used for Quantitative Study on Landslide . . . . . . . . . . . . 39
1.10 Existing Literature on Landslide Inducing Parameters . . . . . . . . 39
1.11 Slope Stability Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
1.11.1 Hydrological Models and Slope Stability. . . . . . . . . . . 44
1.11.2 Landslide Susceptibility Zonation Models Concept . . . . 44
1.12 Study on Landslide Management . . . . . . . . . . . . . . . . . . . . . . 48
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2 Geo-spatial Variability of Physiographic Parameters

and Landslide Potentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.2 Assessment of Geomorphic Attributes . . . . . . . . . . . . . . . . . . . 59
2.2.1 Analysis of Contour Orientation . . . . . . . . . . . . . . . . . 60
2.2.2 Digital Elevation Model (DEM) . . . . . . . . . . . . . . . . . 61
2.2.3 Hypsometric Analysis . . . . . . . . . . . . . . . . . . . . . . . . 63
2.2.4 Lithological Composition and Landslide Potentiality . . . 63
2.2.5 Analysis of Lineaments and Landslide Potentiality . . . . 67
2.2.6 Slope Angle and Landslide Potentiality . . . . . . . . . . . . 69
2.2.7 Slope Aspect and Landslide Potentiality . . . . . . . . . . . 70
2.2.8 The Surface Curvature and Landslide Potentiality. . . . . 71
ix
x Contents
2.3 The Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 74

2.3.1 Textural Characteristics of the Soil . . . . . . . . . . . .... 75
2.3.2 Saturated Soil Depth. . . . . . . . . . . . . . . . . . . . . .... 77
2.3.3 Water Holding Capacity, Pore Space and Volume
Expansion of the Soil (in %) . . . . . . . . . . . . . . . . . . . 80
2.3.4 Friction Angle (φ) and Cohesion (c) . . . . . . . . . . . . . . 86
2.3.5 Wet Soil Density (γs) . . . . . . . . . . . . . . . . . . . . . . . . 88
2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3 Impact Assessment of Hydrologic Attributes

and Slope Instability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.2 Climatic Attributes and Landslides . . . . . . . . . . . . . . . . . . . . . 97
3.2.1 Study on Rainfall . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.2.2 Evapotranspiration . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.2.3 Relationship Between Rainfall and Evapotranspiration
and Determination of Deficit and Excess Moisture
Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
3.3 Drainage Network Evolution and Landslide . . . . . . . . . . . . . . . 103
3.3.1 Drainage Network Analysis of 1972 . . . . . . . . . . . . . . 103
3.3.2 Drainage Network Analysis of 1987 . . . . . . . . . . . . . . 105
3.3.3 Drainage Concentration and Stream
Confluence Point . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
3.3.4 Drainage Network Orientation . . . . . . . . . . . . . . . . . . 110
3.3.5 Drainage Density and Landslide Potentiality . . . . . . . . 110
3.3.6 Upslope Contributing Area and Landslide
Potentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
3.4 Drainage Morphometry and Slope Instability . . . . . . . . . . . . . . 114
3.4.1 Determination of Composite Ranking Coefficient
Value and Instability Rank . . . . . . . . . . . . . . . . . . . . 116
3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4 Surface Run-off, Soil Erosion and Slope Instability . . . . . . . . . . . . 123

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4.2 Run-off Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
4.2.1 Determination of Runoff Curve Number (RCN) . . . . . . 127
4.2.2 Calculation for Antecedent Moisture Condition
or Cumulative Effects of Rainfall in Last Storm
of the Year . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 133
4.2.3 Estimation of Discharge (in m3) from Individual
Sub-watershed . . . . . . . . . . . . . . . . . . . . . . . . ..... 134
Contents xi
4.3 Relationship Between Land Use and Land Cover,

Hydrological Soil Group, Curve Number
(Under Average-AMC), Expected Run-off (m3) . . . . . . . . . ... 136
4.4 Sub-watershed Wise Analysis of Curve Number (AMC-III),
Potential Retention (S) and Run-off . . . . . . . . . . . . . . . . . ... 139
4.5 Periodic (Monthly) Relationship Between Surface Run-off
and Slope Susceptibility . . . . . . . . . . . . . . . . . . . . . . . . . ... 139
4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 143
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 143
5 Geomorphic Threshold and Landslide. . . . . . . . . . . . . . . . . . . . .. 145

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 145
5.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 147
5.2.1 Angle of Internal Friction/Angle of Repose
and Cohesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 147
5.2.2 Determination of Threshold Slope Angle
for Initiation of Slide . . . . . . . . . . . . . . . . . . . . . . .. 148
5.2.3 Determination of Threshold Rainfall to Initiate Slide. .. 150
5.2.4 Threshold Slope Height to Initiate Slide . . . . . . . . . .. 151
5.2.5 Application of the ‘Poisson’ and ‘Binomial’
Probability Distribution Models to Estimate
the Temporal Probability of Landslide Events . . . . . . . 151
5.3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
5.3.1 Geo-technical Properties and Critical Slope Angle . . . . 152
5.3.2 Calculated Critical Rainfall to Initiate Debris Slide . . . . 155
5.3.3 Calculated Critical Height to Initiate Slide . . . . . . . . . . 159
5.3.4 Temporal Probability of Landslide Events . . . . . . . . . . 159
5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6 Slope Stability Model and Landslide Susceptibility

Using Geo-technical Properties of Soil . . . . . . . . . . . . . . . . . . . . . . 167
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
6.2 Materials and Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
6.2.1 Slope Stability Model Concept. . . . . . . . . . . . . . . . . . 169
6.2.2 Cohesion (c) and Friction Angle (φ) . . . . . . . . . . . . . . 171
6.2.3 Surface Inclination/Slope (β) . . . . . . . . . . . . . . . . . . . 173
6.2.4 Soil Saturation Index/Wetness Index (m) . . . . . . . . . . . 173
6.2.5 Depth of Failure Surface/Depth (z) of Soil Below
the Terrain Surface . . . . . . . . . . . . . . . . . . . . . . . . .. 174
6.2.6 Soil Density (γs) and Density of Water (γw). . . . . . . .. 174
6.2.7 Identification of Major Landslide Location/Landslide
Inventory Map . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 174
xii Contents
6.3 Application of 1 Dimension Slope Stability Model

and Stability Analysis . . . . . . . . . . . . . . . . . . . . . . . . . ..... 175
6.4 Shear Stress, Shear Strength and Safety Factor
and Stability Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
6.4.1 Shear Stress. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
6.4.2 Shear Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
6.4.3 Safety Factor Based Stability Classes . . . . . . . . . . . . . 178
6.5 The Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 179
6.5.1 Analysis of Strength Parameters (Shear Stress
and Strength) for 10 Major Landslide Locations. ..... 179
6.5.2 Stability Analysis Based on Safety Factor
for 10 Major Landslide Locations. . . . . . . . . . . ..... 181
6.5.3 Comparative Stability Analysis Between Dry,
Semi-saturated and Saturated Soil Condition
Based on 1D Slope Stability Model . . . . . . . . . . . . . . 181
6.5.4 Accuracy Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
7 Application of Analytical Hierarchy Process (AHP)

and Frequency Ratio (FR) Model in Assessing Landslide
Susceptibility and Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
7.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
7.2.1 Landslide Susceptibility Assessment . . . . . . . . . . . . . . 196
7.2.2 Landslide Risk Assessment . . . . . . . . . . . . . . . . . . . . 203
7.3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
7.3.1 Relationship Between Landslide Susceptibility
and Landslide Triggering Factors . . . . . . . . . . . . . . . . 208
7.3.2 Accuracy Result of Landslide Susceptibility Map . . . . . 213
7.3.3 Analysis of Landslide Hazard Risk . . . . . . . . . . . . . . . 214
7.3.4 Accuracy Result of Landslide Hazard Risk Map. . . . . . 221
7.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
8 Landslide Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 227

8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 227
8.2 Problem Perception Over the Major Landslides Prone
Areas of the Shivkhola Watershed, Darjiling Himalaya . ...... 231
8.3 Recommended Landslide Mitigation Practices . . . . . . . ...... 232
8.3.1 Armouring the Catch-Water Drain Along
the Junction Between Road and the Hill Slope. ...... 232
8.3.2 Jhora Training . . . . . . . . . . . . . . . . . . . . . . . ...... 233
8.3.3 Retaining Wall. . . . . . . . . . . . . . . . . . . . . . . ...... 235
Contents xiii
8.3.4 Breast Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

8.3.5 Catchment Water Drainage . . . . . . . . . . . . . . . . . . . . 238
8.3.6 Continuous Monitoring of Sub-surface Water. . . . . . . . 240
8.3.7 Landslide Mitigation by Improving Soil Strength . . . . . 240
8.3.8 Geo-textile Method for Landslide Mitigation
(A Bio-engineering Approach) . . . . . . . . . . . . . . . . .. 242
8.3.9 Introduction of Vegetation and Slope Stability . . . . . .. 243
8.3.10 Maintenance and Continuous Monitoring of NH-55,
Hill Cart Road . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 247
8.3.11 Road Diversion to Avoid the Paglajhora-Sinking
Zone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
8.3.12 Construction and Maintenance of Buildings . . . . . . . . . 249
8.3.13 Landslide Warning System . . . . . . . . . . . . . . . . . . . . 251
8.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
Appendix A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Appendix B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
Appendix C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Appendix D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Appendix E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Appendix F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Appendix G. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Abbreviations
AHP Analytical hierarchy process

AMC Antecedent moisture condition
CI Consistency index
CR Consistency ratio
CRV Class rating value
DEM Digital elevation model
Dw Height of water
Fi Fraction of flow
FR Frequency ratio
FS Factor of safety
GANN Geomorphology-based artificial neural networks
GPS Global positioning system
GSI Geological Survey of India
HHCS Higher Himalayan Crystalline Sequence
HI Hypsometric Integral
Ia Initial abstraction
ICAR Indian Council of Agricultural Research
IRS Indian Remote Sensing
IS Indian Standard
ITSZ Indo-Tsangpo Suture Zone
L Effective contour length
LF Landslide frequency
LHEF Landslide hazard evaluation factor
LHRA Landslide hazard and risk assessment
LHS Lesser Himalayan Sequence
LISS Linear imaging self-scanning sensor
LPIV Landslide potentiality index value
LSC Landslide Susceptibility Coefficient
LULC Land use and land cover
MATLAB Matrix laboratory
xv
xvi Abbreviations
MBF Main boundary fault

MBT Main boundary thrust
MCT Main central thrust
NRSA National Remote Sensing Agency
PCRV Prioritized class rating value
PFRV Prioritized factor rating value
Pt Depth of rainfall
Qt The depth of runoff
RCA Road contributing area
RCL Road contributing length
RCW Road contributing width
RCN Run-off curve cumber
RI Random index
S The potential maximum retention of water by the soil
S Directional slope
SCS Soil Conservation Service
SINMAP Stability index mapping
SOI Survey of India
SPOT PAN French: Satellite Pour l’Observation de la Terre
SRTM Satellite Radar Topographic Mission
SVL Soil-vegetation-land
TOPMODEL Topographic index model
TT Tethyan thrust
UCA Upslope contributing area
USDA United State Department of Agriculture
USGS United State Geological Society
cr Critical rainfall
cs Critical slope angle
ch Critical height
φ Angle of internal friction or angle of repose
Ѳ Slope angle
α Threshold angle of Failure
u Pore water pressure on potential sliding surface
γ Bulk unit weight of the sliding materials
γs Soil density
γw Density of water
σ1 Major principal stress
σ3 Minor principal stress
σn Normal stress
T Transmissivity
ps Wet soil density
pw Density of water
c Cohesion
h Thickness of total soil
z Thickness of saturated soil
Abbreviations xvii
Sc Compressive strength of the rocks

W Unit weight of the rocks
λ Average rate of landslide occurrence
μ Mean recurrence interval between successive landslide events
m Soil saturation index
τ Shearing stress along plane
µ Coefficient of internal friction
β The terrain surface inclination
λ Shearing strength
α Rupture angle
1D One dimensional
2D Two dimensional
Chapter 1
Introduction
1.1 Key Concepts on Landslide
Landslides are simply defined as the movement of mass of dislodged rock, debris or
earth materials down a slope including a broad range of motions whereby falling,
sliding and flowing under the influence of gravity. A landslide is a type of ‘mass
wasting’ or down slope movement of soil and/or rock under the influence of gravity
(USGS). The word ‘landslide’ is actually a general term for several kinds of slope
movements and they can be classified in several ways. They can be classified by
rate of movement, type of materials and nature of movement and presence of
lubricating agent (slide, slump, flow or fall, Chorley et al. 1985). In Australia,
landslide is defined as the movement of rock, debris or earth down slope. They
result from the failure of the materials which make up the hill slope and are driven
by the force of gravity. Landslides are known also as landslips, slumps or slope
failure. The most common types of landslide in Australia are earth slides, rock falls
and debris flows. The movement of landslide materials can vary from abrupt col-
lapses to slow gradual slides and at rates which range from almost undetectable to
extremely rapid. Sudden and rapid events are the most dangerous because of a lack
of warning and the speed at which material can travel down the slope as well as the
force of its resulting impact. Extremely slow landslides might move only milli-
metres or centimetres a year and can be active over many years (creep).
Landslide can also be termed as ‘mass movement’. According to Chorley et al.
(1985) ‘Mass movement’ is the detachment and down slope movement of soil and
rock materials under the influence of gravity. The sliding or flowing of the materials
are caused due to their position and to gravitational forces. But this type of movement
is accelerated by the presence of water, ice and air. Basically, all kind of rock-waste
movements including soil and ice are collectively termed as landslide. The moving
distance covered by the earth materials involved in landsliding from the place of
origin to the place of destination, called as ‘run-out distance’. To define landslides,
we must define fall, slide, flow, lateral spreads and topple as all these are the major
© Springer Science+Business Media Singapore 2015 1

S. Mandal and R. Maiti, Semi-quantitative Approaches for Landslide Assessment
and Prediction, Springer Natural Hazards, DOI 10.1007/978-981-287-146-6_1
2 1 Introduction
slope movement types. The continuous downward movement of weathered rock

materials involving large blocks, earthen materials from steep slopes or cliffed valley
sides of streams under the impact of gravity, termed as fall. According to Selby
(1982) rocks falls are relatively small landslide caused by the removal of individual
and superficial blocks from cliff base and are promoted by hydrofracturing, stress
release, the wedging action of tree roots and other weathering process. Bloom (1978)
treated fall as one of “the distinct landslide process but it is rarely independent of
subsequent events”. This type of landslide is common in humid region. Slide is one
of the most common forms of failure and can be subdivided into translational and
rotational slides. Rotational slides are sometimes called slumps because they move
with rotation. Translational slides have a planar, or two dimensional surface of
rupture. Slides are most common when the toe of the slope is undercut. They have a
moderate rate of movement and the coherence of material is retained, moving largely
intact or in broken pieces. The accounts of various translational slides in the
Appalachians are cited by Jacobson et al. (1989), in Puerto Rico by Simon et al.
(1990), in New Zealand by Salter et al. (1981), and in California by Ellen and
Wieczorek (1988). Flow is the most destructive and turbulent form of landslide.
Flows have a high water content which causes the slope material to lose cohesion,
turning it into slurry. They are channeled by the landscape and move rapidly.
Hutchinson and Bhandari (1971) studied this type of landslide and suggested that
flows may be promoted by the collapse of soil from the surrounding cliffs, steep
slopes on to the upper part of the concave moving mass and raising pore-water
pressure in the debris. Spread is characterized by the gradual lateral displacement of
large volumes of distributed material over very gentle or flat terrain. Failure is caused
by liquefaction which is the process when saturated loose sediment with little or no
cohesion such as sands or silts are transformed into a liquid-like state. This process is
triggered by rapid ground motion most commonly during earthquakes. Nearly all
known examples come from southern Norway, the St. Lawrence lowlands of eastern
Canada and the Alaska Coast. Topple is characterized by the tilting of rock without
collapse, or by the forward rotation of rocks about a pivot point. Topples have a rapid
rate of movement and failure is generally influenced by the fracture pattern in rock.
Material descends by abrupt falling, sliding, bouncing and rolling. Varnes (1978)
observed various states of activities of landslide with time such as active, dormant,
reactivated, suspended, abandoned, stabilized and relict. All these characteristics of
landslide vary from one place to other with varying geological, morphological and
physical attribute of topography. A landslide should have some nomenclatures such
as crown, main scarp, top, head, main body, foot, toe, failure surface, toe of failure
surface, surface of separation, displaced mass, zone of depletion, zone of accumu-
lation, depletion, accumulation, flank and pre-failure topography. But the entire
landslide does not possess all of these nomenclatures. The failure of the slope
happens when gravity exceeds the strength of the earth materials. Although the
action of gravity is the primary driving force for a landslide to occur, there are other
contributing factors affecting the original slope stability. Typically, pre-conditional
factors build up specific sub-surface conditions that make the area/slope prone to
failure, whereas the actual landslide often requires a trigger before being released.
1.2 A Brief Account of Some Destructive Landslides Occurrences 3
1.2 A Brief Account of Some Destructive

Landslides Occurrences
The large scale landslides in populated areas made a disastrous effect on natural
environment and human structure. The scale of damage caused by earth-quake
triggered landslide is more destructive than any other type of causative factors. The
destructive slope movements may threaten single structure, villages or towns;
agricultural and forest land; communication lines and tunnel in use or under con-
struction; reservoirs and lakes. In 1584, a large scale landslide on the Rhone valley
side slopes wrecked the community of Yvorne and more than 300 lives were lost
(Heim 1932). A catastrophic slope failure (rock slide) took place in North America
which destroyed part of the town of Frank in Canada in 1903 and only within 2 min
about 30 million m3 of Carboniferous limestone wiped out from the mountain face
and buried long railway track and took 70 lives (McConnel and Brock 1904). The
disastrous landslide occurred near Vaerdalen, north of Trondheim in Norway in
1893 (Holmsen 1954) and that destroyed 22 farms and 11 persons were killed. In
1920, the Kansu Province of China got affected by great magnitude earthquake and
caused the sliding of thick loess deposits covering an area of 160 × 480 km and
killed about 200,000 people (Close and McCormick 1922). In 1935 a noteworthy
tunnel failure took place due to landslide in New Zealand (Benson 1940). In 1936 a
rock fall, 106 m3 in volume occurred near Loen and produced a 74 m swell in the
Nordfjord that resulted a loss of 73 lives (Bjerrum and Jorstad 1966). Sometimes
major landslides are indirectly responsible for catastrophic events that occur when
the slipped material blocks a river and hold back the water. One of the biggest
catastrophy of this kind occurred in the Southern Alps in the Sixteenth Century.
Extensive landslides of this kind were recorded in the Himalayan region and in this
way in 1893 a temporary 320 m high dam was created on the upper Ganges. One of
the largest slope failure of this type occurred on April 25, 1974 in the Mantaro river
valley in Peru. Kojan and Hutschinson (1978) treated the high rate of river erosion
as a significant contributing factor of landslip which resulted in a very deep valley
with steep slopes. Recently (2013), the Himalayan region faced destructive slope
failure events and that have taken unexpected causalities. Statistical records prove
that landslide is one of the most destructive natural hazards all over the World
(Tables 1.1 and 1.2).
1.3 Relevance of the Landslide Study
In the hilly terrains of India, landslides have been a major and widely spread natural
disaster that often strike life and property and occupy a position of major concern.
In India, two regions are most vulnerable to landslides i.e. the Himalayas and the
Western Ghats (Fig. 1.2). The Western Ghats and Nilgiris are geologically stable
4 1 Introduction
Table 1.1 Some destructive landslide events

Date Place Casualties Remark
25th November, Mont Granier, 1,000+ Destroyed five villages
1248 France
2nd September, Canton of Schwyz, 457 Destroyed four villages
1806 Switzerland
1843 Mt Ida, Troy, 15 Sediment slump and flow
New York
1881 Elm, Switzerland 115 Rock avalanche demolished
83 houses
1893 Trondheim, 111 Liquifaction flow in marine
Norway clays
1903 Frank, Canada 70 Rock avalanche demolished
most of the town
18th February, Usoy, Tajikistan 54 Triggered by M 7.4 earthquake
1911
19th May, 1919 Kelud, East Java 5110 Lahars caused 5,110 deaths and
destroyed 104 villages
16th December, Haiyuan County, >100,000 Loess flows and landslides
1920 Ningxia, China over an area of 50,000 km2
25th August, Diexi, Mao *3,100 The largest landslide formed a
1933 County, Sichuan, 255 m high dam on the river
China Min
5 July, 1938 Kwansai, Japan *1,000 1,30,000 homes damaged or
destroyed
1953 Minamiyamashiro, 336 dead or 5,122 homes destroyed
Kyoto, Japan missing
13 December, Huaraz, Peru 4,000–6,000 Caused by rupture of a moraine
1941 dam
10 January, Ranrahirca, Peru 4,000–5,000 An avalanche of ice and rock
1962 triggered by collapse of part of
a hanging glacier
9 July, 1967 Kure, Hiroshima 159 Landslide destroyed 352
Prefecture, Japan buildings
3–5 October, Darjiling, India ‘Thousands’ Landslide destroyed 60 km
1968 long highway
31 May, 1970 Yungay, Peru >22,000 Combined rock avalanches and
debris-flow buried two cities
4 May, 1971 St Jean Vianney, 31 Slab flows buried people and
Cannada houses
April, 1974 Junin Region, Peru 450 218 mm rain in 5 h triggered
many landslides
13 November, Armero, Colombia 23,000 Volcanic mudslides, that flo-
1985 wed at speeds of up to 50 km/h
down the slopes of the volcano.
These lahars moved into
valleys, merging to form larger
flows, one of which destroyed
the town of Armero
(continued)
1.3 Relevance of the Landslide Study 5
Table 1.1 (continued)

Date Place Casualties Remark
14–16 Vargas, Venezuela 30,000 Caused by a heavy storm that
December, 1999 deposited 911 mm of rain in a
few days
9 November, Amboori, Kerala, 40 Supposedly worst landslide in
2001 India Kerala state’s history
17 February, Southern Leyte, 1,126 Rock-debris avalanche
2006 Philippines triggered by ten-day period
of heavy rain
11 June, 2007 Chittagong, 123 Series of landslides caused by
Bangladesh illegal hillside cutting and
monsoon rains
6 September, Cairo, Egypt 119 Rock fall from cliffs, individual
2008 boulders up to 70 tonnes
8 August, 2010 Gansu, China 1,287 caused by heavy rainfall and
flooding in Gansu Province
16 June, 2013 Kedarnath, Utta- 5,700 High intensity rain (cloud
rakhand, India burst)
22 March, 2014 Oso, Washington, 41 Confirmed the flow of the landslide was
United States missing extreme because of the
extraordinary run-out of mud
and debris
2 May, 2014 Ab Barak, 2,000 death The landslides were triggered
Badakhshan, confirmed by heavy rains in Badakhshan
Afganistan province bordering Tajikistan,
where melting snow and sea-
sonal showers make the region
vulnerable to such calamities
Source [Wikipedia, the free encyclopedia (Retrieved from http://en.wikipedia.org/w/index.php)
and Coates (1977), Table 2, p. 19]
but have uplifted plateau margins influenced by neo-tectonic activity. The Hima-
layas mountain belt comprises of tectonically unstable younger geological forma-
tions subjected to severe seismic activity. According to Prasad (1986), Rao and
Chacko (1986), Thigale and Khandge (1996) that landslide activities are increasing
gradually in the Western Ghats, for over 1,200 km, covering parts of Maharashtra,
Goa, Karnataka, Tamilnadu and Kerala. Major landslide locations of Western Ghats
are Ghatkopar (Mumbai), Morbad (Thane), and Panchmukhi (Thiruananthpuram).
Landslide has become an annual phenomenon in the Iddukki and Wynad district of
Kerala as a result of deforestation and soil erosion. Compared to Western Ghats
region, the slides in the Himalayas region are huge and massive and in most cases
the overburden along with the underlying rocks are displaced during sliding par-
ticularly due to the seismic factor. Himalaya is the youngest and highest mountain
range on Earth, which extends over a length of about 2,400 km. It is one of the most
active and fragile mountain chains in the world; it is home to millions of people
6
Table 1.2 Location and character of the major landslide phenomena in the Shivkhola Watershed
Landslide Location Specific location of Elevation Lithology Area of the Triggering factors Type of the
No. the slides and slope slide slide
angle
1 Mahanadi-I 36.5 km. N of 57° Phyllite 390 m2 Solifluction and gravita- Debris slide
Siliguri 1,150 m tional pull
2 Mahanadi-II 36.55 km. N of 62° Quartzite 120 m2 Solifluction and seeping Debris slide
Siliguri 1,148 m of water
3 Mahanadi-II 36.6 km. N of 65° Quartzite and 300 m2 Supersaturation of the Debris slide
Siliguri 1,150 m phyllite materials
4 Tindharia 26 km north from 57° Phyllite, 56 m × 31 m High intensity rainfall Earth slide
(Rly. Station) Siliguri 927 m Quartzite-schist and saturation of
—September materials
2006
5 Tindharia 26.5 km north from >65° Phyllite, 6532.5 m2 Cohesiveless materials Earth fall
downslope Siliguri 925 m Quartzite-schist and steep slope
6 Paglajhora 39 km north from >65° Foliated gneiss Around 2 km Reduction of shearing Subsidence
sinking zone Siliguri 1,136 m and mica schist (both up and strength due to and sinking
(lower) down slope) supersaturation of
slope materials
7 Lower 38.5 km north from 68° Foliated gneiss 200 m length Reduction of shearing Subsidence
Paglajhora Siliguri 1,137 m and mica schist of road strength due to and sinking
road supersaturation
subsidence of slope materials
8 Lower 38 km north from 70° Foliated gneiss 125 m length Reduction of shearing Subsidence
Paglajhora Siliguri 1135 m and mica schist of road strength due to and sinking
1
Road supersaturation of
Abandonment slope materials
(continued)
Introduction
angle
9 Upper Pag- 15 km donwn from 67° Foliated gneiss 85 m of road Steep slope and Subsidence
lajhora road Kurseong 1,220 m and mica schist saturation of and sinking
subsidence unconsolidated
materials
10 14 miles 37.5 km North from >65° Foliated gneiss 68 m length Existence of Subsidence
bustee road Siliguri 1,130 m and mica schist of road non-cohesive materials and sinking
abandonment on steep slope
11 14 miles 37.25 km north 59° Calc-granulite, 2,250 m2 Steep slope, Earth slide
1.3 Relevance of the Landslide Study
bustee Slide from Siliguri 1,118 m Foliated gneiss unconsolidated

and mica schist materials, formation
of narrow nala
12 Shiviter T.E. 2.0 km north from 64° and Highly foliated 312 m2 Steep slope, Weathered
(along the Mahanadi 950 m and decom- supersaturation of debris
road) posed granite unconsolidated failure
materials
13 Shiviter tea 3.0 km north from >65° and Highly foliated 3 acres of tea Steep slope, supersatu- Debris fall
garden slide Mahanadi 1,050 m and decom- area ration of unconsolidated and earth
posed granite materials fall
14 Jogamaya Downslope between 50°–65° Phyllite, Large tea Soil-slip
T.E. Tindharia and 900–1,050 m Quartzite-schist area affected and debris
Gayabari by slide slide
15 Sepoydhura 2.00 km north from >65° Soft yellowish 66.6 m2 Steepening of slope by Debris slide
T.E.-I Chunabhati Rail 295 m mudstone, road cutting and less and earth
Station (Hill Cart shale and cohesive materials slide
Road) conglomerate
(continued)
7
8

angle
16 Sepoydhura 2.5 km north from 45° Soft yellowish 77 m2 Steepening of slope by Debris slide
T.E.-II Chunabhati Rail 295 m mudstone, road cutting and less and earth
Road) conglomerate
17 Sepoydhura 1.5 km. north from >65° Soft yellowish 102.6 m2 Steepening of slope by Debris slide
T.E.-III Chunabhati Rail 300 m mudstone, road cutting and less and earth
Road) conglomerate
18 Nurbong T.E. 4.5 km. north from >65° 375 m Soft yellowish 1,425 m2 Steep slope, loss of Subsidence
Chunabhati Rail mudstone, basal support by toe and sinking
Station (Hill Cart shale and erosion, unconsolidated
Road) conglomerate materials
19 Gitange T.E. 5.00 km north from >45° Soft yellowish Lower Bared slope surface, Rock-soil
Chunabhati Rail 300 m mudstone, section is saturation and Steep creep
Station (Hill Cart shale and affected by slope
Road) conglomerate failure
1
Introduction
living in northern India, northern Pakistan, Nepal, Bhutan, Tibet and parts of other
Asian nations. Owing to the rugged topography, the complex geological structures,
the fragile soil cover, occurrences of unconsolidated Quaternary sediments along
hill slopes, the high physical weathering, the high intensity monsoon rainfall, the
large temperature variations, and the occurrence of very large magnitude earthquake
events, landslides, debris flows, soil erosion, and other mass wasting processes are
very frequent in this region, which in fact are the primary cause of environmental
degradation in the region. Most of the landslides in the Himalayan region occur
along or very close to major thrust and fault like the Main Boundary Fault (MBF),
the Main Central Thrust (MCT), the Tethyan Thrust (TT), and the Indo-Tsangpo
Suture Zone (ITSZ) where rocks are heavily crushed due to regular tectonic activity
(Fig. 1.1). The region comprises lithological groups of Quaternary sediments,
Siwaliks, Lesser Himalayan Sequence, Greater Himalayan Sequence, Leucogra-
nites, Chekha Formation and Tethyan Sediments from south to north. There are six
thrust planes passing parallel to Himalayan range that have made the region most
vulnerable to landslips.
Himalayan Region can be grouped into four major sections with respect to the
distribution of landslides: i. Jammu and Kashmir Region ii. Himachal Pradesh
Region iii. Uttaranchal Region and iv. North-Eastern Region (Fig. 1.2). Major
landslide of Jammu and Kashmir region are Khuni Nala Slide (Before 1947), Nashri
Slide (1953) and Sansara Slide (1972). In Himachal Pradesh region, Malling slide
(1980, 1990), Ropagad Slide (1990), Thangi Slide (1985), Telangi Slide (1975),
Sapni Slide (1989) and Urni Rock Fall (1992) are well known landslide sectors.
Kaliasaur Slide (1969), Nandprayag Slide (1970), Patalganga Slide (1970), Debidhar
Slide (1979), Banswara Slide (1998), Bhiri Slide (1982), Guptkashi Slide (1984),
Barua-Bhenti Slide (1998) are the major landslide locations of Uttranchal Region
in Alakananda Valley (Fig. 1.3). In Uttaranchal, steep mountain slope, passing of
major thrust plane and existence of weak lithology promotes landslide activities.
Fig. 1.1 Lithology and various thrust plane passing through himalayan range (Source www.iisc.
ernet.in)
10 1 Introduction
Fig. 1.2 Location of major landslide prone area in India
Major landslides in North Eastern Region are Padamchen Slide (1968, 1971), Rishi
Slide (1964), Lepchajhora Slide (1968), Barrik Slide (1968), Rangpo Slide (1968,
1971), 9th Mile Slide (1968), Labha Slide (1968), Manul II Slide (1966), Myang
Slide (1967, 1973), Richu Slide (1966), Vong Slide (1962), Shivakhola Slide (1968,
1998, 2006, 2011), Keukhola Slide (1965, 1968) and Likibur Slide (1998, 2006)
(Source: Chandra 1975; Sinha et al. 1975; Sinha 1975; Engineer 1975; Krish-
naswamy and Jain 1975; Didwal 1988; Bartaya et al. 1996; Sah et al. 1996; Sah and
Bartaya 2002; Mandal and Maiti 2012, 2013).
Alakanand river valley of Uttaranchal region is one of the most important
landmarks with regard to landslide events at present. All the major thrust planes
(weaknesses planes) crossed the river valley region and made lthologically more
fragile segment of the Himalayan Range (Fig. 1.3a). In this region, Main River and
its tributaries are always engaged in down cutting processes. The active erosion
process over weak lithology has aggravated the mountain slope more susceptible to
landslide by steepening the valley side slope (Fig. 1.3b).
Fig. 1.3 Alakananda River Valley with various thrust planes and lithology (Source www.esc.cam.
ac.uk)
Every year, especially during the summer monsoon period, landslide and related
natural disaster events claim many lives and destroy property, infrastructure, and
the environment of the Himalayas. The economic loss in landslide alone in this
region is estimated at $1 billion per year. It is estimated that the loss of life due to
landslides and related earth flow phenomena in the Himalayan Region constitutes
about 30 % of the world’s total landslide-related damage value. The earthquake
induced landslide in Himalayan terrain is quite common phenomena. A large
number of landslips took place after Uttarkashi Erathquake in 1991 (SOI, Publi-
cation 30). The Durham Landslide Fatality Database suggests that over 1,000
people were killed in landslide events in the Himalayas in 2007 alone, which
represents almost 35 % of the global total. It was estimated that over 20,000 people
were killed by landslides during the 2005 Kashmir Earthquake in Pakistan and
India. These people generally live in widely-spread settlements in the fragile
Himalayan terrains, and suffer more from the landslides than any other types of
natural disaster. A large number of human settlements on the Himalayas are situated
either on old landslide masses or in landslide-prone areas. As a result, a great
number of people are affected by large- and small- scale landslides throughout the
Himalayas, especially during rainy seasons. In 1988, a huge landslide at Darbang,
about 200 km west of Kathmandu in Nepal, killed 109 people and temporarily
blocked Myagdi River. About 62 years before this incident, the same landslide had
buried Darbang area killing about 500 people. Likewise, one of the worst landslide
tragedies took place at Malpa Uttarkhand, India on 11 and 17 August 1998 resulting
in death of 380 people when massive landslides washed away the entire village
(www.hils.org.np/publication.hils.org.np).
Due to the 18th September 2011 Sikkim earthquake (M: 6.8), several “new” and
a few “reactivated” landslides occurred right from the Himalayan foot-hill region
(e.g. Dudhia in Kurseong Sub-division, Darjeeling District, West Bengal) up to the
12 1 Introduction
higher Himalayan range in the higher reaches of Sikkim-Darjeeling Himalayas

(Chakraborty et al. 2011). This earthquake invited cracks and subsidence of roads
along Nathula-Gangtok Highway (Fig. 1.4). The new landslides that occurred in the
lower elevations are mostly concentrated within the terrace deposits of trunk
streams such as Tista, Rangit and Balason etc. (Fig. 1.5) and within the old colluvial
deposits on the lower reaches of slope adjacent to trunk streams (e.g. Jorethang-
Rishi-Legship section along the right bank of Rangit river). In the slopes having
steep relief, the frequencies of rock/debris fall are more than to slides, whereas field
observation clearly indicated that proximal to epicentral region, frequencies of both
rock fall and rock slides are much higher than the debris/soil slides and the same are
more concentrated towards the crest of steep ridges. Recent field observation of the
Fig. 1.4 Cracks and subsidence, Nathula-Gangtok Road, Sikkim-Darjiling Himalaya (Source
GSI, Engineering Division, Gangtok, Sikkim)
Fig. 1.5 Some fresh landslips after Sikkim Earthquake, 2011 (Source GSI, Engineering Division,
Gangtok)
landslide-related damages further indicated that lithology and geomorphology also

played a significant role in causing these earthquake induced landslides. The debris-
laden slope which is made of loose unconsolidated material and the slope covered
by thin unconsolidated scree deposits have been more prone to failure by this
earthquake. Frequency of rock fall and rockslides are more in areas which are
generally vulnerable due to steep slope, weathered and fractured lithology and
unfavourably jointed and kinematically-unstable slopes (GSI Report October 19,
2011, Engineering Geology Division, Eastern Region, Kolkata).
From 14 to 17 June 2013, the Indian state of Uttarakhand and adjoining areas
received heavy rainfall, which was about 375 % more than the benchmark rainfall
during a normal monsoon. The upper Himalayan territories of Himachal Pradesh
and Uttarakhand are home to several major and historic Hindu and Sikh pilgrimage
sites besides several tourist spots and trekking trails. Heavy rainfall for four con-
secutive days as well as melting snow aggravated the floods. Torrential and unre-
lenting downpour, landslides and floods saw the raging waters sweep away more
than 180 people and hundreds of livestock to death. More than 70,000 pilgrims
have been stranded even as the Army and Air Force try to rescue them in a hostile
terrain. Of all the affected areas, Kedarnath was the worst hit. Extreme rains have
wreaked havoc in the region, with the tenuousness of the Himalayan soil stability
resulting in killer landslides. But environmentalists claim that widespread and
almost unregulated expansion of giant hydro-electric projects in the region, the
incessant construction of roads to serve the burgeoning tourist population, and the
adverse effect on the fragile ecosystem in the region due to growing human pres-
ence and pollution are the major causes for the devastation that Uttarakhand has
been subjected to.
However, according to eyewitnesses huge rocks broke away from Kedar Dome
after the flash floods caused by the cloudburst (Fig. 1.6a, b). Reports say that the
temple’s courtyard has been washed away and Nandi, Shiva’s mount at the gateway
to the temple, too was buried under over 6 feet of mud and rocks, like most of
Kedarnath. It however survived the onslaught of the mudslide. Even the 14-km
stretch of roadway from Gaurikund to Kedarnath has been totally submerged,
making entry or exit impossible. This is what led to pilgrims being stranded
(Fig. 1.10). Even Ram Bada, between Gaurikund and Kedarnath, is totally invisible
from the rescue helicopters. So while the two sides—the environmentalists and the
government—squabble over the root causes of the disaster, thousands of lives hang
by a thin thread. The massive slope failure destroyed human settlements (Figs. 1.7
and 1.8) and eliminated Ram Baba Road (Fig. 1.19) at Kedarnath and its adjoining
areas. Some fresh slope failure completely damaged mountain slope vegetation
cover (Figs. 1.9, 1.11 and 1.12) [Source: Disaster Mitigation and Management
Centre, Govt. of Uttarakhand, 2014].
Apart from such catastrophic landslides, many small-scale slope failures go
unreported, especially when they occur in remote areas of the Himalaya. Further-
more, the loss of productive lands in the hills due to landslides and related mass
erosion phenomena during every rainy season, which are seldom reported unless
they involve the loss of life, is so great that a quantified economic loss would
14 1 Introduction
Fig. 1.6 a, b Previous satellite image and settlement around Kedarnath Temple (a); and Post-
disaster satellite image and complete destruction of the town Kedarnath (b) (2013) (Source www.
dailymail.co.uk)
Fig. 1.7 Slope failure and

destruction of settlement at
Kedarnath (Source www.
dailymail.co.uk)
probably be in the same range as for a one-off natural disaster. National infra-
structures like roads, bridges, dams, hydropower stations, canals, buildings are also
repeatedly damaged by landslides in this region. A rapid rise in construction of
national infrastructures including roads, hydropower stations and dams, etc. with
inadequate or little consideration for the natural hazards has considerably contrib-
uted to triggering of landslides in the mountains of the Himalayas (Himalayan
Fig. 1.8 Slope failure and destruction of settlement at Kedarnath (www.dailymail.co.uk)
Fig. 1.9 Elimination of road by landslide, Kedarnath (Source www.srinistuff.com)
Fig. 1.10 Trapped people by landslide along the road, Gourikund to Kedarnat (Source www.
theaustralian.com.au)
16 1 Introduction
Fig. 1.11 Down ward movement of fragile slope materials (Source www.dailymail.co.uk)
Landslide Society, www.hils.org.np/publication.hils.org.np). Similarly, due to a

rapid increase in population over the Himalayan hills in the last three decades, the
trend of settling in comparatively hazardous areas is increasing. Thus, the rising
levels of risk from the landslides triggered by hydro-meteorological variability
invariably entail considerable loss of life and property losses and inflict significant
damage on the vital economic system of the Himalayan nations (Dahal 2009).
A further key issue is the deterioration of the ecological balance and environment of
the Himalayas, most notably through excessive deforestation, soil erosion and river
sedimentation. This is likely to be exacerbated by global warming, which is likely
to cause increased levels of extreme climatic events. Warming is also putting many
Himalayan glacial lakes in great danger of bursting, which may lead to complete
destruction of downstream human settlements and the habitat to world’s rare flora
and fauna. Unfortunately, however, despite the rapid climatic, geomorphological,
environmental and ecological changes taking place in the Himalayas, all of which
can be linked to landslide occurrences, systematic research on landslide processes
and environmental changes in the Himalayas is at best in its infancy. Although
some efforts have been made by the professionals and researchers from government
agencies in, for example, India, Nepal, Pakistan, and Bhutan as well as from
nongovernmental organizations, international agencies, and academic institutions,
the areas of investigation, the methodologies adopted, and the classification criteria
considered in the study differ considerably. Furthermore, there is a serious lack of
knowledge transfer and research output dissemination among the researchers.
Partly, this is because there are very few scientific gatherings among the geosci-
entists, environmentalists, and engineers who are involved in Himalayan landslide
and environmental research. To deal with all these issues in the Himalayan Region
and to foster investigations, collaborations, discussions, and integration among the
stakeholders in Himalayan landslide and environmental issues, a common forum of
Fig. 1.12 Location of earthquake induced landslide (2011) in Sikkim-Darjiling Himalaya (Source
Geological Survey of India, Eastern Region (Kolkata), darjeelingnew1blog)
geoscientists, environmentalists, engineers, and stakeholders needs to be estab-

lished immediately (Bhandary et al. 2012, 2013; Dahal and Bhandary 2013).
1.4 Landsliding in Darjiling Himalaya
The appalling landslip of 23rd September, 1899 which occurred in the eastern side of
Darjiling Town and confined to the soil-cap covering the gneisses that form the
Darjiling ridge. This landslip took away 72 lives and the damage to property was
18 1 Introduction
extensive. The widespread landslides occurred throughout the Darjiling District in

1950 as a result of heavy spell of rain for 3 days. The loss of life reported from the
district was 127 and out of which 100 was in Darjiling Sadar Sub-division alone. The
town Darjiling was cut-off for about 5 days and the Siliguri-Kalimpong Railway line
was washed away. Large sections of the Kurseong-Darjiling Railway track were
washed away. In 1950, K.K. Dutta, a Geologist, carried out a detailed investigation
and opined that while the pore-water pressure was the immediate causes of slope
instability, the stability of a slope depended primarily upon its inclination and the
nature of the structure of material adjoining it. During rains the pore spaces and the
voids are completely get filled with water and cause a rise of piezometric surface,
which, in turn increases the pore pressure and decrease the shearing resistance of the
materials and create slope failure. Due to incessant and heavy rain of 809 mm for
3 days between 3rd and 5th October 1968, there were numerous landslides in
Darjiling Himalaya. On October 5, the rainfall was 499 millimetres. The number of
deaths were estimated officially 677 on 12 October. The Rangpo Bridge on the
Sikkim border along the Tista-Gangtok Highway, the magnificent, one-span concrete
bridge on the Tists River near Tista Bazar and several other bridges at strategic points
were badly damaged. The Hill-Cart Road, the arterial route between Siliguri and
Darjiling was breached at 18 different places. The Indian Tea Association in a
statement claimed that between ten and fifteen percent of the total tea area in Darjiling
Himalaya has been destroyed. It also claimed that over 100 lives had been lost in the
tea estates and there had been widespread damage to factories buildings and other
installations. It is assumed that the problems of landslide cannot be prevented entirely
but they can be checked by proper preventive measures. Turfing and afforestation of
bare slopes, well-directed and efficient drainage, reduction of the steepness of hill
slopes by terracing, outward protection of the soil cap by means of revetments and
buttresses, protection of the harder rock outcrops, systematic quarrying in hillsides
and control of the erosive action of streams and waterfalls are some of the preventive
measures which may check landslides (Figs. 1.13, 1.14, 1.15, 1.16 and 1.17).
Fig. 1.13 IRS IC LISS III satellite image of Darjiling district (Source www.darjeeling.gov.in)
1.4 Landsliding in Darjiling Himalaya 19
Fig. 1.14 Location of concerned study area, the Shivkhola Watershed, Darjiling Himalaya
The concerned study area, the Shivkhola Watershed is badly affected from slope
instability jeopardizing the economy and social systems taking the toll of lives,
lands and properties. The effects are conveyed to a long distance propagating both
in upslope and down slope direction by the disruption of transport, hydrological
systems. Most of the landslides of this trouble torn district of Darjiling are
20 1 Introduction
Fig. 1.15 Road damage at

Tindharia
Fig. 1.16 Elimination of Hill

Cart Road
Fig. 1.17 Destruction of toy

train line
1.4 Landsliding in Darjiling Himalaya 21
concentrated to this area where most of the torrential jhoras (Hill Streams) at their
upper catchments develop a potential sinking zone and the Hill Cart Road and
Narrow gauge Rail, when traverse through these zones, invite catastrophic slide.
The present work tried to identify the zones of potential slides by investigating into
the systematic interaction among the numbers of prominent triggering factors.
The Shivkhola watershed, a landslide prone spatial unit in Darjiling Himalaya,
dominated by a large number of destructive landslide events. Most of the landslides
took place at Tindharia, Lower Paglajhora, Gayabari, Shiviter and Nurbong. Since
the first half of 20th century, the Watershed experienced following major landslip
events.
• The landslip in Tindharia on 15th Januay, 1934 due to Bihar-Nepal earthquake.
• Between the 11th and 13th June, 1950 a series of devastating landslide after a spell
of 834.10 mm rainfall around Kurseong, Darjiling Himalaya (Basu et al. 2003).
• Landslide at Giddapahar near Kurseong damaged over 175 m of road and
railway track and demolished many bustee hamlets between 3rd and 5th
October, 1968 due to heavy rainfall of 1121.40 mm (Basu et al. 2003).
• During 3rd and 4th September, 1980 again Tindharia was affected by severe
landslide due to heavy and concentrated rain of 299.1 mm (Basu et al. 2003).
• During 15th and 16th September, 1991 numerous landslide phenomena at
Paglajhora and Chunabhati (Basu et al. 2003) due to 462.5 mm heavy rain.
• During 11th and 13th July, 1993 innumerable landslide devastated Mahanadi,
Gayabari and Tindharia due to concentrated rainfall of 211.3 mm.
• The year of 1995, 1998, 2001, 2002, 2003, 2006 and 2007 are major cases of
landslips along the Hill Cart Road mainly at Paglajhora, Giddapahar, Tindharia
and Gayabari.
• A devastating landslide at Shiviter Tea Estate in 2005 (Fig. 1.18), which
destroyed a tea garden area of 1.5 acres (Field study by author, Tea Garden
Manager, 2006).
• On 10th September, 2006 a devastating landslide took place along the Hill Cart
Road (Fig. 1.17) at 14 miles Bustee and blocked the railway line for few days
due to sudden and catastrophic rainfall (Opinion Survey, 2007).
Fig. 1.18 Massive damage of

tea estate, Shiviter
22 1 Introduction
• A road crack was formed along NH-55, the road to Darjiling from Siliguri, at
Tindharia following heavy rain in the past few days on 4th August, 2007.
Another landslide incident happened in the same year on 16th July along the
Hill Cart Road and completely broken down the guard wall nearby Kurseong
town (Field Survey, 2008) (Fig. 1.15).
• A major landslides at 14 mile near Paglajhora disrupted traffic along the
National Highway-55 (Fig. 1.16) and toy train services between New Jalpaiguri
and Kurseong on 16th June, 2010 morning (Field Survey, August, 2010).
• Heavy showers and the vibration caused by Sikkim Earthquake invited
destructive landslide events at Tindharia (Fig. 1.12) and Lower Paglajhora on
10th May, 2011.
Recently, landslide risk analysis, like many other forms of management of either
natural or civil engineering hazards, is a relatively new discipline. Earlier attempts
to reduce landslide risk were largely the history of management of landslide terrain,
construction of protective structures or monitoring and warning systems, or the
ever-increasing sophisticated methods for mapping and delineating areas prone to
landslide (Dai et al. 2002). Risk of landslide is normally defined as the expected
number of lives lost, persons injured, property damaged and disrupted economic
activities due to particular landslide hazard for a given area and reference period
(Varnes 1984). To reduce the risk from the landslide events, the knowledge about
potentiality to landslide activity is crucially needed. But the information of land-
slide events is described in the form of landslide susceptibility map of the con-
cerned region and the preparation of this map depends on the complex sets of
knowledge of slope movement factors. The process of creating the maps involves
several qualitative or quantitative approaches (Soeters and Van Westen 1996;
Guzzetti et al. 1999a, b, c).
The Shivkhola, the Right Hand Tributary of the mighty Mahanadi is located at
the southern flanks of Darjiling Himalayan Range of Kurseong division of Darjiling
District in West Bengal between Kurseong to the north and Tindhria to the south
within 88°17′30″ E to 88°23′45″ E and 26°50′15″ N to 26°53′35″ N covering an
area of about 22.05 km2. It is famously known as Paglajhora, the most destructive
and torrential hill stream at its upper catchment (Fig. 1.14). The presence of a weak
and young set of rocks, a monsoonal climate with high intensity and long duration
rainfall and steep mountain escarpment slope are supposed to be the major prob-
lems associated with landslide in the study area become clear. Landslide in the
Himalayan region has been emerged as the dominant hill slope process. In the
Shivkhola Watershed, landslides at Jogmaya Tea Garden (Fig. 1.19), Tindharia Tea
Garden, and Paglajhora are destroying roads and others infrastructure, eliminating
the tea garden area and also disrupting transport and communication. This has
brought a serious attention of the local people and respective authorities to take care
the vulnerable communities living with hazards.
1.5 Mechanism and Causes of Landslides 23
Fig. 1.19 a–d Destruction in Jogmaya Tea Garden Area
1.5 Mechanism and Causes of Landslides
Recently the process studies are getting increasing importance in the field of hill
slope evolution though the obsession to trace cyclic landform elements and to fit
them in a global model of geomorphic cycle has ultimately retarded rather than
activated the development of empirical process research (Poesen and Poley 1987).
Slope failures are considered as the most active agent of soil erosion in steep
mountains. Proper understanding of the geomorphic evolution of mountain land-
scapes requires that they be given increased attention (Brardinoni and Church
2004). The evolution of hill slope is described as a complex and systematic
interaction between sets of physical and manmade factors governed by geologic,
hydrologic, climatic, geomorphic and land use attributes. Those factors in their
interactive combination yield more than the sum of individual factors (Wilson
1981). The geomorphic processes and elements, in their systematic relationship
show short-term equilibrium manifested by low rate of mass transfer and mainte-
nance of form characterized by slope and height. Although ‘true’ or absolute
equilibrium is a rare theoretical state towards which the system behaviour is tending
with greater rapidity by attempting to absorb the successive effects of a sequence of
process inputs of lower magnitude and frequency (Chorley et al. 1985). The
24 1 Introduction
geotectonic factors like angle of repose, static and dynamic friction of particles,
interlocking and sorting of grains (Brudsen 1979; Jumikis 1967; Carson 1977),
thickness of total soil and that of saturated soil (Borga et al. 1998) are considered as
important in the analysis of slope failure.
The hydrologic factors like daily rainfall threshold in connection with slope
angle and regolith thickness (Gabet et al. 2004), rainfall intensity, infiltration etc.
are practiced in the analysis of slope instability. The geographical factors including
the anthropogenic actions for slope instability in Himalayan slopes are widely
studied by Basu and Sarkar (1985, 1988), Basu and Ghatowar (1988). The slope
failure in the upper catchments and its subsequent effects on fluvial dynamics are
discussed considering hill and foot hill as an interactive whole and the operating
geomorphic processes being interconnected (Basu and Ghatowar 1986, 1988, 1990;
Basu 1989; Basu and Ghosh 1993). The anthropogenic processes like unscientific
use of slope for agriculture, mining (Basu and Ghatowar 1988; Basu and Maity
2001; Bhattacharya 1999) and deforestation (Bhattacharya 1996) etc. are widely
studied in relation to the slope instability. Numerous models in connection to the
slope stability, shallow and deep seated landslides are introduced and verified by
Hollingworth and Kovacs (1981), Burton and Bathrust (1994), Bradinoni and
Church (2004), Young (1963) and Montgomery et al. (1994). The stability of
natural slopes are examined and analyzed by Skempton and Hutchinson (1969).
The interaction among those processes are not always possible to express in
mathematical language and so an attempt is made for the establishment of empirical
and verbal models both to express the possible interactions which partially support
the governing processes of hill slope evolution following Ahnert (1970), Kirkby
(1980), Poesen (1985), and Deploey (1982). The understanding of probable inter-
action among the major factors helps in the assessment of processes and in their
management for the restoration of slope to ensure optimum utility from land and
mineral resources.
A body will not move unless a force is applied. In geomorphic processes, gravi-
tational force, water pressure force, expansion force and biological force play a
significant role in changing the shearing strength of slope materials and in moving the
earth materials down slope. We can say, the forces that drive sediments largely derive
from the gravity, the climatic effect and the action of plants and animals. Gravita-
tional force acts directly on rock bodies, sediments, water and ice and tending to make
them move. It acts world over at a nearly uniform magnitude of 9.81 m/s2, with slight
variation resulting from distance from the Earth’s centre and latitude. The presence of
water within rock-soil debris produces various forces that can drive them downward.
The forces developed by the presence of water are called as water-pressure force.
Attewell and Farmer (1976) pointed out that when the pore spaces are filled up, a
pore-water pressure is generated. This situation can reduce the pressure of contact of
the grain. The pore water pressure υ at depth h below the water table is given by
υ = γh where γ is the density of water. The pore water pressure acts in all direction and
it exerts uplift or buoyancy effect and that can produce slope instability. Sometimes
expansion forces expand and contract sediment, soil and rock body by changing
physical and chemical properties of the minerals and cause slope movement.
Biological force is being generated when plant root system push materials side ward
and if this occurs on a slope a down slope movement may take place. Gray (1970),
Brown and Sheu (1975), Greenway (1987) and Yim et al. (1988) studied the rela-
tionship between slope stability and vegetation.
Earth materials are always subject to stress and strain where a stress is a force
that tends to move materials down slope. All the mentioned forces produce stress
within a body. The stress of a body of soil on slope depends upon the mass of the
soil body, m and the slope angle, Ѳ (theta), expressed as, Stress = m sin Ѳ. Slope
materials possess an inherent resistance against downslope movement of the
materials. Friction force acts against gravity and resist movement. The downslope
movement of a soil body can happen when the applied stress is large enough to
overcome the maximum frictional resistance. Friction is expressed as co-efficient, μ
which is equivalent to the angle at which sliding starts and called as angle of sliding
friction (Hugget 2007). Frictional resistance is a function of both the inherent
frictional properties of slope materials and the normal stress acting upon them. As
the shear plane angle becomes steeper, the shear stress becomes larger and the
normal stress smaller. The slope angle attained after the slope failure is known as
the threshold angle of stability.
Cohesion is one of the most important factors of mass movement that affects the
internal strength of the slope materials. It includes the chemical bonding of rocks and
soil particles. The chemical bonding of rocks and soils occur with the presence of
cements composed of silica, carbonates or iron oxides. Cohesion also rises through
capillary suction of water in pores, compaction, and plant root systems. Soil particles
affect cohesion of a soil body and generate friction between one another, called as the
internal friction or shearing resistance which is determined by particle size and
shape and the degree to particles touch to each other. Mohr-Coulomb equation
defines the shear stress that a body of soil on a slope can withstand before it moves.
This equation shows the shear strength as effective normal stress and cohesion,
referred to as the Coulomb-Terzaghi shear strength equation (Summerfield 1991).
The total shear strength of slope materials (s) is given as,
s ¼ c þ r tanu; ð1:1Þ
where c is cohesion, σ is the effective normal stress and tan φ represents the
coefficient of plane sliding friction.
The slope stability of the hill slope can be expressed in terms of the relationship
between the forces tending to drive the slope materials and the forces tending to
resist driving stresses. It is clear that the movement will starts when driving force
exceeds the resisting force and this relationship is represented as the safety factor
(FS). The slope can exist in three states: i. where shear strength is larger than shear
stress, the slope is described as stable slope (FS > 1.3), ii. Where shear stress
exceeds the shear strength (FS < 1.00), the slope is described as actively unstable
slope. Sometimes the shear strength can vary over time with the interaction of
climatic phenomena and where the third stability category becomes prominent such
26 1 Introduction
as conditionally stable slope (FS = 1.00–1.30). Basically the triggering factors

transform the slope from a conditionally stable to an actively unstable slope.
The safety factor (FS) is defined as:
s
FS ¼ ð1:2Þ
s
where s is the total shear strength along a specific shear plane, and τ is the total
amount of shear stress developed along this plane.
For the shallow translational slide FS is described as:
ðcz cos2 b uÞ tan u

FS ¼ c þ ð1:3Þ
cz sin b cos b
where c is cohesion, γ is the unit weight of regolith, z is the vertical depth to the
shear plane, β is the angle of the shear plane, u is pore-water pressure and φ is the
angle of internal friction.
The stability condition of the hillslope varies from place to place as a result of
the variation in prevalence of landslide triggering factors. It is inferred that some
factors are responsible for increasing shearing stress and some factors for reducing
shearing strength. The changes of both shear stress and shear strength are major
contributing factors against slope instability. Varnes (1978), introduced several
factors contributing the increase in shear stress and the reduction of shear strength
that is stated below.
A. Factors contributing the increase in shear stress:
1. Removal of lateral support: erosion by rivers and glaciers, wave action,
faulting, previous rock fall or slide.
2. Removal of underlying support: undercutting by rivers, waves, sub-surface
solution, and loss of strength by extrusion of underlying sediments.
3. Loading of slope: weight of water, vegetation and accumulation of debris.
4. Lateral pressure: water in cracks, freezing in cracks, swelling and pressure
release.
5. Transient stress: earthquake and movement of tress.
B. Factors contributing the reduction of shear strength:
1. Weathering: disintegration and decomposition of granular rocks, hydration
of clay minerals, dissolution of cementing materials in rocks or soils.
2. Changes in pore-water pressure: saturation and softening of weathered
materials.
3. Changes of structure: creation of fissures in shales and clays, remoulding of
sands and sensitive clays.
4. Organic impact: burrowing of animals and decay of tree roots.
The causes of landslides are usually related to instabilities in slopes. It is

necessary to identify one or more causes for landslide and most important
landslide trigger. The difference between these two concepts is subtle but important.
The landslide causes are the reasons behind a landslide occurrence in a location and
at a time. Causes may be considered to be factors that make the slope vulnerable to
failure, that predispose the slope to becoming unstable. The trigger is the single
event that finally initiates the landslide. Thus, causes combine to make a slope
vulnerable to failure, and the trigger finally initiates the movement. The trigger is in
fact a slow but steady decrease in material strength associated with the weath-
ering of the rock—at some point the material becomes so weak that failure must
occur. Hence the trigger is the weathering process, but this is not detectable
externally. In most cases we think of a trigger as an external stimulus that induces
an immediate or near-immediate response in the slope and in this case in the form of
the movement of the landslide. Generally this movement is induced either because
the stresses in the slope are altered, perhaps by increasing shear stress or decreasing
the effective normal stress, or by reducing the resistance to the movement perhaps
by decreasing the shear strength of the slope materials forming.
In the majority of cases the main trigger of landslides is heavy or pro-
longed rainfall (Caine 1980). Generally this takes the form of either an exceptional
short lived event, such as the passage of a tropical cyclone or even the rainfall
associated with a particularly intense thunderstorm or of a long duration rainfall event
with lower intensity, such as the cumulative effect of monsoon rainfall in South Asia.
In the former case it is usually necessary to have very high rainfall intensities, whereas
in the latter the intensity of rainfall may be only moderate—it is the duration and
existing pore water pressure conditions that are important. The importance of rainfall
as a trigger for landslides cannot be underestimated. A global survey of landslide
occurrence in the 12 months to the end of September 2003 revealed that there were
210 damaging landslide events worldwide. Of these, over 90 % were triggered by
heavy rainfall. One rainfall event for example in Sri Lanka in May 2003 triggered
hundreds of landslides, killing 266 people and rendering over 300,000 people
temporarily homeless. In July 2003 an intense rain band associated with the annual
Asian monsoon tracked across central Nepal, triggering 14 fatal landslides that killed
85 people. The reinsurance company Swiss Re estimated that rainfall induced
landslides associated with the 1997–1998 El Nino event triggered landslides along
the west coast of North, Central and South America that resulted in over $5 billion in
losses. Finally, landslides triggered by Hurricane Mitch in 1998 killed an estimated
18,000 people in Honduras, Nicaragua, Guatemala and El Salvador. This is because
the rainfall drives an increase in pore water pressures within the soil. Movement is
driven by shear stress, which is generated by the mass of the block acting under
gravity down the slope. Resistance to movement is the result of the normal load.
When the slope fills with water, the fluid pressure provides the block with buoyancy,
reducing the resistance to movement. In addition, in some cases fluid pressures can
act down the slope as a result of groundwater flow to provide a hydraulic push to the
landslide that further decreases the stability.
28 1 Introduction
Considerable efforts have been made to understand the triggers for landsliding in
natural systems, with quite variable results. Simon et al. (1990) found
that storms with a total precipitation of 100–200 mm, about 14 mm of rain per hour
for several hours, or 2–3 mm of rain per hour for about 100 h can trigger landslides
in that environment. Corominas and Moya (1999) investigated the upper basin of
the Llobregat River, Eastern Pyrenees area and found that without antecedent
rainfall, high intensity and short duration rains triggered debris flows and shallow
slides developed in colluvium and weathered rocks. A rainfall threshold of around
190 mm in 24 h initiated failures whereas more than 300 mm in 24–48 h were
needed to cause widespread shallow landsliding. With antecedent rain, moderate
intensity precipitation of at least 40 mm in 24 h reactivated mudslides and both
rotational and translational slides affecting clayey and silty-clayey formations. In
this case, several weeks and 200 mm of precipitation were needed to cause land-
slide reactivation.
Rapid changes in the groundwater level along a slope can also trigger landslides.
This is often the case where a slope is adjacent to a water body or a river. When the
water level adjacent to the slope falls rapidly the groundwater level frequently
cannot dissipate quickly enough, leaving an artificially high water table. This
subjects the slope to higher than normal shear stresses, leading to potential insta-
bility. In some cases, failures are triggered as a result of undercutting of the slope by
a river, especially during a flood. This undercutting serves both to increase the
gradient of the slope, reducing stability, and to remove toe weighting, which also
decreases stability. For example, in Nepal this process is often seen after a glacial
lake outburst flood, when toe erosion occurs along the channel. Immediately after
the passage of flood waves extensive landsliding often occurs. This instability can
continue to occur for a long time afterwards, especially during subsequent periods
of heavy rain and flood events.
The second major factor in the triggering of landslides is seismicity. The passage
of the earthquake waves through the rock and soil produces a complex set
of accelerations that effectively act to change the gravitational load on the slope. So,
for example, vertical accelerations successively increase and decrease the normal
load acting on the slope. Similarly, horizontal accelerations induce a shearing force
due to the inertia of the landslide mass during the accelerations. These processes are
complex, but can be sufficient to induce failure of the slope. These processes can be
much more serious in mountainous areas in which the seismic waves interact with
the terrain to produce increases in the magnitude of the ground accelerations.
Some of the largest and most destructive landslides known have been associated
with volcanoes. There are two main types of volcanic landslide: lahars and debris
avalanches, the largest of which are sometimes termed as flank collapses. An
example of a lahar was seen at Mount St Helens during its catastrophic eruption on
May 18, 1980. The lahar killed more than 2,000 people as it swept over the towns
of El Porvenir and Rolando Rodriguez at the base of the mountain. Debris
avalanches commonly occur at the same time as an eruption, but occasionally they
may be triggered by other factors such as a seismic shock or heavy rainfall. They
are particularly common on strato volcanoes, which can be massively destructive
due to their large size. The most famous debris avalanche occurred at Mount St
Helens during the massive eruption in 1980. The debris avalanche had a volume of
about 1 km3 (0.24 cu mi), traveled at 50–80 m/s (110–180 mph), and covered an
area of 62 km2 (24 sq mi), killing 57 people (Table 1.3).
The major causes of the landslides that may induce the slope materials to move
downward in the representative mountain watershed of the Shivkhola Watershed of
Darjiling Himalaya are:
• Average inclination of rocks is generally steep towards the roadside (60–70°).
• Weathering due to diurnal and seasonal ranges of temperature causes rapid
disintegration and decomposition of the rock body.
• Due to intensity of local rainfall, the feldspar present in quartzite, decompose to
form kaolin which helps rock particle to slide down.
• Continuous and heavy rain during monsoon period reduces cohesion of slope
materials.
• Percolation of high intensity rain water through joints and cracks increases the
hydrostatic pressure within the soils and bring a change in the consistency of the
soil.
Table 1.3 Basic causes of landslide

Geological causes Morphological Physical causes Human intervention
cause of landslide
Weathered materials Slope angle Intense rainfall Excavation
e.g. heavy rainfall
Sheared materials Uplift Rapid snow Loading
melt
Jointed or fissured Rebound Prolonged Draw-down
materials precipitation
Adversely orientated Fluvial erosion Rapid Land use (e.g. construction
discontinuities drawdown of roads, houses etc.)
Permeability contrasts Wave erosion Earthquake Water management
Material contrasts Glacial erosion Volcanic Mining
eruption
Rainfall and snow fall Erosion of lat- Thawing Quarrying
eral margins
Earthquakes Subterranean Freeze-thaw Vibration
erosion
Working of Slope loading Ground water Water leakage
machinery changes
Vegetation Soil pore water Deforestation
change pressure
Erosion Surface runoff Land use pattern
Seismic activity
Soil erosion
30 1 Introduction
• Consequently the angle of internal friction is far less than the regional slope
(60°) that sets in instability. Cohesion is very less and that cannot bind the soil
effectively.
• With the onset of monsoon rill cutting as well as gully erosion starts and the run-
off, concentrating through these denuded portions, becomes the dominant agent
of soil erosion.
• The geological deformation due to the pushing effects of the Damuda Series of
rocks of lower Gondwana period against the Daling Series.
• Existence of very fragile rock-soil composition (rock waste of quartz-mica-
schist) over the slope surface.
• Coarse textured soil with very low amount of cohesive substances like salt,
calcium, and magnesium with high degree of percolation and leaching of upper
soil materials to the bottom leading to disintegration and displacement of soil.
• Unscientific coal dust drills and tunnels leading to rupturing of basal support.
• Depletion of vegetation covers which increases surface run-off and soil erosion.
• Concentration of settlement along the Hill Cart Road (NH-55) and their illegal
household activities over steep mountain slope and immense pressure over the
fragile slope materials by man-made concrete structure.
• Physiographic configuration (arcuate) of the Paglajhora sinking (major landslide
section) area provides favourable condition to produce hydrostatic pressure.
• Proximity to Main Central Thrust (MCT).
• Existence of intensely fractured and sheared nature of the bed rock.
• Toe cutting and headword erosion of debris covered slope by a fast flowing
tributaries which losses the basal support.
• Moderate to steep slope gradient which favours the concentration of water and
drainage.
• Improper drainage network orientation throughout the region.
• Accumulation of highly anisotrophic materials with a great thickness and low
shearing resistance.
• Disintegration of phyllite and its surcharge with water caused by heavy
downpour.
• Hill face, composed of phyllite and mica-schist, lost its shearing strength due to
saturation and tends to move downward in the name of failure.
• The entire tea garden area is dominated by steep slope (>60°) and toe cutting by
narrow channels.
1.6 Types and Classification of Landslide
The movement of the slope materials varies from place to place. Landslide can be
classified based on the rate of movement of slope materials, the shape of the slope
surface, the materials in the slope movements and other criteria involved in causing
slope movement. Considering all these many researchers such as Heim (1882),
1.6 Types and Classification of Landslide 31
Almagia (1910), Terzaghi (1925), Ladd (1935), Sharpe (1938), Savarenski (1939),
Emelyanova (1953), Varnes (1958), Hutchinson, developed varied schemes of
landslide classification. Terzaghi classified landslide based on the physical
properties involved in downward movement of materials. Sharpe (1938) classified
slide, on the basis of the materials displaced and the type and rate of movement.
Savarenski (1939) divided landslide into asequent, consequent and insequent
considering the shape of landslide surface. Asequent landslides develop in homo-
geneous cohesive soils along curved and approximately cylindrical surfaces.
Consequent landslides move along the bedding planes, joints or plane of schistosity
dipping downslope. Insequent landslides run transversely to bedding plane and are
generally of large dimensions.
Karson and Kirkby (1972) classified mass movement processes into three major
types such as, slide, flow and heave on the basis of type of movement. Slides take
place along clear-cut shear planes and are ten times longer than they are wide. Two
main types are translational slides and rotational slides. Translational slides occur
along planar shear planes and include debris slides, earth slides, earth block slides,
rock slides and rock block slides. Rotational slides are called as slumps which occur
along concave shear planes. This type of movement includes rock slumps, debris
slumps and earth slumps. Flow is the gravity induced mass movement (Selby
2005). It is one of the most destructive natural processes, causing hundreds of
deaths and losses of millions of dollars worth of property each year (Costa 1984).
Flows are generally categorized ad avalanches, debris flows, earth flows or mud
flows (Varnes 1978). Hsu (1975) suggested the term ‘sturzstroms’ to explain high
velocity mass movement that shows clear flowage behaviour. The very slowest type
of flow is termed as ‘solifluction’ which involves the oversaturation of earth
materials and their downward movement. Heave is caused by alternating phases of
expansion and contraction as a result of heating and cooling, wetting and drying
and burrowing activities of animals. Heave is classed as soil creep and frost creep.
Varnes (1978) proposed for five principal types of mass movements.
1. Falls: the free fall of loosened rock mass for the greater part of distance of
movement.
2. Topples: overturning of rock mass about a point below its centre of gravity.
3. Slides: the rock mass moves about a point above its center of gravity, known as
rotational slide. The rock mass when moves along more or less planar or gently
undulating surfaces, known as translational slide.
4. Lateral spreads: when the lateral extension movements occur in a fractured rock
mass.
5. Flows: it includes continuous deformation of the rocks and gradual downward
movement of the materials. In soil it represents flow like viscous fluid.
6. Complex Slides: a combination of two or more of the above stated types.
Zaruba (1969) classified slope movements considering the character of the rocks
and the type of movements:
32 1 Introduction
A. Slope movements of superficial deposits (slope detritus, weathering material),

produced mainly by subareal agents.
i. Talus creep that creates the terminal bending of strata
ii. Sheet Slides
iii. Earth Flows
iv. Debris Flows and liquefaction of sand.
B. Slides in Clays, Marls, Claystone, Clayey shales etc.
i. Along cylindrical surfaces
ii. Along composite sliding surfaces
iii. Caused by squeezing out of soft underlying rocks.
C. Slope Movements of solid rocks:
i. Rockslides on predisposed surfaces (bedding, schistosity, jointing and fault
planes)
ii. Long term deformation of slopes
iii. Rock falls.
D. Slope Movements include geological phenomena:
i. Solifluction
ii. Slides in sensitive clays
iii. Subaqueous slide.
Chorley et al. (1985) presented a very simplified classification of mass wasting-
mass movement phenomena (landslides) based on direction and type of movement:
i. Vertical movement:
(a) Fall—rock fall, earth fall and topple
(b) Subsidence—collapse and settlement.
ii. Lateral Movement:
(a) Slides—block slide
(b) Spreading—cambering (draping of sedimentary units) and Sackung
(lateral spreading away from the anticlinal crest).
iii. Diagonal Movement:
(a) Creeping—soil creep, rock creep and talus creep
(b) Slide—rock slide, debris slide, slumping
(c) Flows—earth flow, debris flow and mud flow.
Hutchinson in 1968 classified mass movements into three major types such as:
i. creep, ii. Frozen ground phenomena and iii. Landslides. His classification was
based on the rate of movement and the type of materials involved in failure. He
divided landslides into translational slide, rotational slips, falls, and subaqueous
slides. In 1988 Hutchinson classified landslides in the following manner based on
1.6 Types and Classification of Landslide 33
morphology of the moving mass with consideration of mechanism, material and

rate of movement.
1. Confined failures:
(a) in natural slopes and
(b) in excavated slopes.
2. Rotational slips:
(a) Single rotational slips
(b) Successive rotational slips
(c) Multiple rotational slips.
3. Compound Slides:
(a) In slide mass of low to moderate brittleness
(b) In slide mass of high brittleness.
4. Translational Slides:
(a) Sheet slides
(b) Slab slides; flake slides
(c) Peat slides
(d) Rock slides:
i. planar slides; block slides
ii. Stepped slides
iii. Wedge failures
(e) Slides of debris:
i. Debris-slides: debris avalanches (non-periglacial)
ii. Active layer slides (periglacial)
(f) Sudden spreading failures.
On the basis of Soil Fabric and Pore Water Pressure, Hutchinson (1988) also
classified landslides.
A. Soil Fabric [effects on c, φ]
1. First-time Slides in Previously Unsheared Ground: soil fabric tends to be
random and shear strength parameters are at peak or between peak and
residual values.
2. Slides of pre-existing shears associated with:
i. Re-activation of earlier landslides
ii. Initiation of landsliding on pre-existing shear produced by processes
other than earlier landsliding, i.e.:
(a) Tectonics
(b) Glacitectonic
34 1 Introduction
(c) Gelifluction of clays

(d) Other periglacial processes
(e) Rebound
(f) Non-uniform swelling.
B. Pore-water Pressure [effects on µ]
1. Short-term (undrained)- no equalization of excess pore water pressure set up
by the change in total stress.
2. Intermediate- partial equalization of excess pore-water pressures. Delayed
failures of cutting in stiff clays fall in this category.
3. Long-term (drained)- complete equalization of pore-water pressures to
steady seepage values.
The several types of landslides are being observed in the Darjiling Himalaya.
The fall of boulders from the steep slopes is the simplest form of landslide. Such
rock body is separated from the bedrock. Second type is the sliding of rock masses
which is quite frequent in the Tista Valley between Sivok and Kali Jhora, where the
hills consist of interbedded sandstones and shales inclined at high angles in the
same direction as the hill slopes. The scouring of underlying bands of soft shales by
rain-water causes the overlying sandstones to slip and slide down the hillsides. The
third type is the soil slip, caused by slow downward movement of soil or uncon-
solidated materials along unprotected hill slopes. This happen frequently in the
vicinity of the Hill Cart Road, between Mahanadi and Rangdong, where sections of
the road sink from a few inches to several feet. In the fourth type, slow downward
movements of soil sometimes causes violent landslip. In Darjiling Himalaya, the
soil slips are of small magnitude in regard to ‘length’ and ‘affected height’. But the
debris slips are of greater magnitude and are devastating in nature in Darjiling
Himalaya. Major rock slips were not observed in Darjiling, where true soil slip and
debris slips are common. The shearing resistance in slopes is made up of a com-
bination of internal friction and cohesion. In clay type soil, the angle of internal
friction is not more than 2° or 3° and with the internal friction alone, soil cannot be
stable. But due to cohesion, for some heights a soil slope is stable at very steep
angles. The angle of internal friction in rock debris which is sometimes 30°–40° is
known as angle of repose. A slope will safely stand to a great height with an
inclination slightly flatter than the angle of repose. In slopes formed of rock, slip
occurs along certain planes of weakness like joints, bedding planes etc.
1.7 Objectives of the Present Work
The main objective of the study is to estimate, quantify, and analyze various
landslide inducing parameters of land, water and soil and to prepare landslide
susceptibility map on a river basin scale which will be very much helpful in
planning, designing and implementing the development programmes. Without a
1.7 Objectives of the Present Work 35
proper knowledge and understanding of the local geological aspects, rainfall and
physical and chemical properties of soil, the construction of Hill Cart Road, trunk
road between Siliguri and Darjiling by outside planners and designers along the
steep escarpment slope has aggravated the problem of slope instability within the
Shivkhola watershed. After the completion of the project they leave the place
without assurance of sufficient training and maintenance of resources. The
responsible local government has also not allocated sufficient resources for regu-
lating land use conversion in the concerned area. Therefore the study on landslides
in relation to land, water and soil will be beneficial to the local and national
government in planning land use pattern and human construction in a rational
manner for the future prospect and well being of the people living in the study area.
For the fulfillment of three objectives mentioned above, the present work is carried
out under the following heads.
• Quantitative measurement and analysis of geomorphic and hydrologic param-
eters of slope instability in the Shivkhola Watershed of Darjiling Himalaya.
• Preparation of a landslide susceptible map and landslide risk map for under-
standing the spatial distribution of slope instability by analyzing all the landslide
triggering factors corresponding to the Shivkhola watershed.
• Preparation of a suitable management plan.
1.8 Applied Methodology
The deductive methodology is followed to investigate into the destructive impacts

of anthropogenic activities on local environment.
• Repeated and continued field studies for long duration (2006–2011) were made
for the proper cognition of the processes and their interaction. The topographic
factors mainly slope steepness, slope length, contributing area and concavity of
the slope were measured in field in consultation with SOI Topographical map
(RF-1:50,000). Slope length and upslope contributing area was measured by
locating the divide on the topographical map and measuring distance from it.
The concavity was measured by means of ratio between area (a) and contour
length (b) following Borga et al. (1998).
• The depth of the slide scar was measured by holding a measuring tape at both
the margins of scar and the other tape was allowed to hang, the reading is then
taken from the base of the hanging tape. The margin of the scars was surveyed
by prismatic compass. The intensive survey of the sliding scar was carried on by
Abney’s level at 0.5 m interval along radial lines originating from lower most
part of the scar. The altitude of the points at 0.5 m interval along the radial lines
was then estimated in reference to the central base point of known altitude
determined by GPS (Basu and Maiti 2001). The rate of road subsidence (along
36 1 Introduction
Hill Cart Road) was being monitored by Dumpy Level Survey and continuous
monitoring through GPS and measuring tape.
• The major hydrological parameter like rainfall data was collected during the
field investigation from Selim Hill Tea Estate situated within ½ km crow fly
distance from the major landslide location Paglajhora. Rainfall data were also
collected from Shiviter Tea Estate.
• The collection of soil samples from different locations of the Shivkhola
Watershed were being accomplished with GPS to analyze soil texture, water
holding capacity, pore space, cohesion, friction angle and others stress param-
eters of the soil.
• The total thickness of soil and that of saturated soil during monsoon was
measured from slope cutting. The soil here was mainly cohesion less with
coarser texture, having greater infiltration capacity and so even during the high
intensity rain Hortonian overland flow did not occur.
• The continuous changes in the land use character were studied during field
investigations with GPS and SOI (Survey of India Topo-Sheet, 1972 and 1987).
• The sufferings from the landslide phenomena and related perception studies
have been made through well structured questionnaire to make a suitable
management proposal for the unstable terrain, Shivkhola Watershed.
The analysis and interpretation of collected and measured secondary and primary
data were done with full care in order to make synthesis of all the studied parameters
in a holistic manner for understanding the process operating in the study area.
• Relief aspects i.e. percentage hypsometric curve, slope, and topographic index
of the shivkhola watershed were computed following the methods of Strahler
(1952), Wentworth (1930), Smith et al. (1979).
• Basin perimeter, basin shape, stream frequency, and drainage density were
measured after Horton (1932), Stoddart (1965), Miller (1953), and Schumm
(1956).
• The major hydrological parameter like rainfall data was collected from Selim Hill
Tea Estate situated within ½ km crow fly distance. The highly porous media helps
in the quick drainage of the subsurface water parallel to the slope and thus the
subsurface discharge through the porous media (permeable soil) in relation to the
hydraulic gradient and Darcian flow can be measured by the following equation.
h
Q ¼ b T sin h ð1:4Þ
z
h = Thickness of the saturated soil

z = Thickness of the total soil
b = contour length
T = wet soil transmissivity
θ = slope
1.8 Applied Methodology 37
• The hydrological factors mainly runoff depends on the upslope contributing

factor. The calculation of contributing area was made considering multiple flow
direction where the cumulative flow at a point (surrogated by the upslope area)
was distributed among more than one neighbouring down slope pixel (Borga
et al. 1998). The topographic irregularities at concave and convex slope are
responsible for the convergence and divergence of flow and thus necessitate the
implication of topographic index by Quinn et al. (1991). Another concept of
specific contributing area (total contributing area divided by the contour length)
was computed by distributing flow from a pixel among its entire lower elevation
neighbour pixel (Borga et al. 1998). Fraction of Flow (Fi) allocated to each
lower neighbour was calculated using Eq. 1.5 (Quinn et al. 1991).
Si Li
Fi ¼ ð1:5Þ
RSj Lj
where the summation is for the entire lower neighbour; S is the directional slope,
and L is an effective contour length that acts as the weighting factor. The value
of L used here is 10 m of the pixel size of the cardinal neighbour and 14.14 m of
the pixel diagonal for diagonal neighbour.
• The rain is the important factor for triggering slide by introducing lubrication
and increasing weight of wet soil in saturation condition. Once the stability
threshold is crossed, the slope remains unstable at greater rainfall rates. Thus it
is possible to determine the minimum steady state rainfall predicted to cause
instability, called critical rainfall (rcr) (Borga et al. 1998).
bps tan h
rcr ¼ Tsin h ½1 ð1:6Þ
apW tan u
• The amount of surface runoff from certain rain was calculated using USDA SCS
Curve Number Technique (1972) and Ministry Agriculture, Government of
India (1972).
• Angle of repose is considered to be the important stability factor (Van Burkalow
1945; Bloom 1991). The cohesion and angle of internal friction were measured
by tri-axial compression test following Mohr stress Diagram following (Brudsen
1979; Jumikis 1967; Carson 1977; Borga et al. 1998). Geo-technical parameters
of the collected soil and rock samples from different locations were measured at
Geotechnical Laboratory, G.S.I., Kolkata through tri-axial soil testing mecha-
nism. Besides, Keen Box Method and Sieve Method were used to measure bulk
density and texture. To assess the geotechnical parameters of the soil in the
Geological Survey of India (GSI) Laboratory (East Kolkata, Salt Lake) Indian
Standard IS: 2131 (1981) were applied. This is the Indian Standard (First
Revision), adopted by the Indian Standard Institution on 24th December, 1981,
after the draft finalized by the Soil Engineering and Rock Mechanics Sectional
Committee approved by the Building Division Council. Finally, using all the
38 1 Introduction
geo-technical parameters 1-D Slope stability model was used to prepare land-
slide susceptibility maps under dry, semi-saturated and saturated condition.
• The range of cohesion and friction angle of different soil was adopted from
Foundation of Engineering Geology (Waltham 2002) for the analysis of slope
stability. Similarly, saturated soil density of rock was adopted from the field
experiences done by Deoja et al. (Mountain Risk Engineering Handbook 1991)
and Specific Yield from Basic Ground-water Hydrology (Ralph C. Heath).
• The values
of the Safety factors were calculated dividing the value of shear
stress s ¼ r1 r
2
3
sin 2 a by the value of Shear strength ðkÞ ¼ ½rn tan u þ C.
• The angle that fractures (α) should theoretically make with the greatest principal
axis is:
a ¼ 45 u=2: ð1:7Þ
• Landslide susceptibility map of the Shivkhola watershed was prepared using

remotely-sensed data, field surveys and GIS tools applying Analytical Hierarchy
Approach (AHA)/Decision Support System after Saaty (1990, 1994), Saaty and
Vargas (2001) and Yang et al. (2006). In this method, landslide susceptibility
index (LSI) value for each pixel was computed by summation of each factor’s
weight multiplied by class weight (or rating) of each referred factor which is
expressed as under:
X
n
LSI ¼ ðWi Ri Þ ð1:8Þ
i¼1
where, Ri and Wi are class weight (rating value) and factor weight.
• Accuracy assessment of landslide hazard map was being performed after
Congalton (1991).
• Landslide hazard risk map of the Shivkhola Watershed was made with the help
of following equation.
Risk ¼ Hazard Exposure: ð1:9Þ
Risk exposure was calculated as a combination of three data layers i.e. weighted
land use/land cover, road contributing area and settlement density.
• Management proposal to mitigate landslide and also to avoid the landslide was
constructed following Royster (1979), Collison and Bhandari (1988), Valdiya
(1987), Carrara et al. (1995), Pla (1997), Montgomery et al. (1989) and Howell
(2001).
1.9 Data Used for Quantitative Study on Landslide 39
1.9 Data Used for Quantitative Study on Landslide
In the present work, atmospheric data was collected for Kurseong Tea Research
Centre (Darjiling), day wise rainfall data was collected from the Selim Hill and
Shiviter Tea Estate, historical landslide records and others relevant information
were obtained from research articles (Starkel 1972; Basu and Starkel 1985; Basu
and Ghatwar 1988; Basu and Maiti 2001; Basu and Sarkar 2003; Maiti 2007;
Ghosh 2009b), news papers (Ananda Bazar Patrika-10/09/06, 14/06/2007, 16/07/
2007 and 18/07/2007, Telegraph-04/08/2007, Statesman-August, 2006) and Geo-
logical Survey of India Technical Report, Topographical Map from Survey of India
(SOI) and Satellite data from NRSA and District Census of Darjiling for the year
1981, 1991 and 2001 and District Gazetteers’ of Darjiling (LSS O’Malley).
Mountain Risk Engineering Handbook provided information about population
growth, geology, soil, natural vegetation. Dow Hill Reserve Forest Office (Kurs-
eong), Soil Conservation Office and Agricultural Office (Kurseong) provided
information relating to land use and land cover and rainfall data. Satellite image
(IIRS P6/Sensor-LISS-III, Path-107, Row-052, date 18 March 2010), Modified
SRTM (Shuttle Radar Topographic Mission) data with scene size 10 latitude and 10
longitude (date 5 April 2008) and Google Earth Image (1 Sept. 2010) were also
used in the study. Various thematic data layers were generated using ERDAS
Imagine 8.5, Arc View and Arc GIS Software.
1.10 Existing Literature on Landslide Inducing Parameters
Among numerous literatures on slope stability, the following are of worth men-
tioning. These are grouped into four subheadings according to their major line of
contribution. The balance condition between shear strength and shear stress are
completely governed by both natural and man-made circumstances. The elements
that affect slope stability and landslides are numerous and varied, and interact in
complex and often subtle ways (Varnes 1984). Siddle and Ochiai (2006) summa-
rized and reviewed and divided natural factors influencing landslide into five major
groups: seismicity; strength, chemistry and mineralogy of soil; geology; geomor-
phology and hydrology.
Geological parameters play a pivotal role in landsliding. Siddle (1991) pointed
out that there exists an association between slope instability and different types of
regolith materials. Weathering alters the mechanical, mineralogic and hydrologic
properties of the regolith and hence acts as an important factor of slope instability in
many settings. Unstable bedding sequences are another important geological
attributes for landslide incidents. This situation occurs when mass movement on
bedding planes is triggered by either the increase in pore water pressures at the
interface between two different alternative strata, or the weakening of strength of
the clay deposit by water infiltration through the overlaying regolith layer.
40 1 Introduction
Landslide often occurs when there is heavy rainfall for a long period. Siddle and
Ochiai (2006) identified four common unstable bedding sequences: (1) alternating
bedding of hard and soft rocks (2) highly altered and permeable regolith overlaying
relatively low permeable sub-strata, (3) thin soils overlaying bedrock or till, and (4)
hard caprock overlaying deeply weathered rocks. Besides, faults, lineaments or both
are usually recognized as the most important triggering factors of slope instability.
The relative strength of the regolith is strongly influenced by past tectonic setting as
well as contemporary weathering (Julian and Anthony 1996; El Khattabi and
Carlier 2004). According to Ibetsberger (1996), and Pachauri et al. (1998) the
neotectonics contribute to slope instability by fracturing, faulting, jointing and
deforming foliation structures.
Engineering, chemical and mineralogical properties of soil determine shear
stress and shear strength of the soil. Shear strength of the soil is an important
engineering property which governs the stability of natural and constructed hill
slope. The shear strength is described as the function of normal stress, cohesion and
angle of internal friction. The relationship within these properties to other attributes
of the soil has been given by Terzaghi and Peck (1976), Wu and Sangrey (1978),
and Fredlund and Rahardjo (1993). Clay minerals are another important chemical
weathering product of the soil and regolith. Yatsu (1966), Duzgoren-Aydin et al.
(2002) studied soil chemical properties and concluded that landslide susceptibility
and slide type are closely associated with specific clay minerals. Clay accumulation
within relict joints is also associated with landslides. Matsuura (1985), Shuzui
(2001), Zheng et al. (2002) and Wen et al. (2004) mentioned that clay mineralogy
and chemistry can also provide indicators of potential sliding plane conditions.
Geomorphic Factors of Slope Instability include slope angle, slope aspect,
height of slope, slope curvature etc. Slope gradient is sometime treated as an index
of slope instability and due to availability of Digital Elevation Model (DEM) it can
be numerically evaluated and depicted spatially (O’Neill and Mark 1987; Gao
1993). Siddle and Ochiai (2006) suggested that slope aspect strongly affects the
hydrologic processes by influencing the evapotranspiration process and thus affects
weathering and vegetation and root development especially in drier environments.
Churchill (1982), Gao (1993), Hylland and Lowe (1997) and Lan et al. (2004)
studied the increase in slope failure in relation to slope aspect characteristics.
Altitude and landslide are intimately related by virtue of other factors such as slope,
lithology, weathering, rainfall and land use. The strong statistical relationship
between elevation and landslide has been studied by Pachuri and Pant (1992),
Linebak Gritzner et al. (2001) and Die and Lee (2002).
Spatial and temporal distribution of rainfall, water recharge into soil, lateral and
vertical movement of water within the regolith, evapotranspiration and interception
are the important hydrologic attributes for the initiation of landslide in hilly area.
Spatial distribution of rainfall are closely associated with landslide initiation by
means of their influence for the generation of pore water pressure in unstable hill
slope and it was studied by Campbell (1966), So (1971), Starkel (1976), Siddle and
Swanston (1982), Siddle (1984), Iverson and Major (1987). Generally, researchers
consider total amount of rainfall, short-term intensity, antecedent storm
1.10 Existing Literature on Landslide Inducing Parameters 41
precipitation and rainfall duration for the landslide susceptibility analysis. Some
researchers e.g. Siddle and Swanston (1982), Keefer et al. (1987) concluded that
short-term rainfall intensity is the most important determinant whereas others e.g.
Endo (1969) and Glade (1998) found a correlation of long-term precipitation with
landslide phenomena. Slope stability over the hill slope is also governed by the rate
of water movement into and through the regolith and the water holding capacity.
These two important landslide governing parameters are influenced by the structure,
density and orientation of fractures and intensities in bedrock and other substrata
that underlie the soil profile. On micro-scale, the rate of water movement in hill
slope soils is best understood by the hydraulic conductivity (K) and the sub-surface
flux of water per unit hydraulic gradient. Clayey soils and compact silty soils with
very small interstitial pores have much lower values of K than coarse textured soils.
Siddle et al. (1985) put forwarded that the hydraulic conductivity of a confining
layer underlying unstable landforms regulates long-term drainage and thus controls
the moisture content of the overlying soil mantle. Hardenbicker and Grunert (2001)
and Siddle and Ochiai (2006) studied pore water pressure induced slope failure on
steep slope with high porosity in moderately deep soils. Infiltration into the soil
increases pore-water pressure and make the slope materials unstable which is
controlled by soil physical properties (porosity, hydraulic conductivity, pore size
distribution, and preferential flow networks), vegetation cover, cultural practices,
freezing phenomena and macro and micro topography. Horton (1993) analyzed all
these properties and concluded that there exists an indirect relationship between the
rate of water infiltration and slope instability. The preferential flow of water both
within the soil and with the underlying bedrock was studied by Tsukamoto et al.
(1982), Siddle et al. (2000a, 2001), Montgomery et al. (1997), and Siddle and
Chigira (2004). Anderson and Burt (1978), Pierson (1980b), Tsukamoto et al.
(1982, 2000) analyzed pore water pressure and revealed that the development of
perched water table within the regolith is responsible for the initiation or acceler-
ation of landslide.
After analyzing all the geologic, geomorphic, hydrologic, engineering, chemical
and mineralogical factors of slope instability, ten landslide triggering factors have
been taken into account in the present work such as slope angle, slope aspect, slope
curvature, lithology, drainage, lineaments, upslope contributing area, land use and
land cover, road contributing area, and settlement density. Besides, relative relief,
ruggedness index, constant of channel maintenance, and drainage confluence were
also studied to understand the nature of slope instability.
The landslide mechanism and triggering process are key problems acknowl-
edged by the Ersmann (1979), Sassa (1988), Huang (2004). The study of the
processes and the triggering mechanism of landslide from the view point of rainfall
and hydro-dynamics is attacting more importance in the present day. The mecha-
nism of landslide is accomplished by sliding plane which results from loss of
cohesion and friction before the formation of a boundary during a heavy rainfall. It
is very much difficult to identify the potential landslide sites through investigation.
But there is a close relationship with rainfall where the landslide can be forecasted
and analyzed through statistical analysis of the rainfall data. The landslide is also
42 1 Introduction
controlled by the sliding mass resulting from large potentially weak surface around
the boundary of the sliding mass, eventually leading to rapid movement on the
plane with loss of cohesion and friction. This kind of landslide is associated with
long-term relief fracture, weathering disintegration, erosion and washout. Zaruba
et al. (1969) identified three main controlling factors of slope instability: (i) slope
gradient, (ii) slope consolidation and (iii) presence of water. There are four different
stages of landslides (Coppola et al. 2006; Chowdhury et al. 2000):
(a) Pre-failure stage: the soil mass is continuous and is mostly controlled by
progressive failure and creep.
(b) Failure stage: characterized by the formation of shear surface through the
entire rock-soil mass.
(c) Post-failure stage: includes the movement of the rock-soil mass involved in
landsliding.
(d) Reactivation stage: occurs when soil-rock mass slides along one or several
pre-existing shear surface.
The landslide in the mountainous slope is being governed by resisting force or
shear strength and driving force or shear stress which is explained by means of
ratio, i.e. resisting forces/driving forces, a dimensionless value, known as Safety
Factor (FS). If the safety factor value is less than or equal to 1, the slope will fail
because driving forces will equal or exceed the resisting force. During rainy reason,
the driving force becomes maximum and resisting force becomes minimum and so
landslides are quite common in this season in the concerned study area as a result of
reduction of soil cohesion in response to soil wetting, increase of soil weight
resulting from water absorption and decrease in effective stress derived from pore-
water pressure.
The character and mechanism of slope failure and the processes responsible for it
along with their control mechanism were studied in details by Terzaghi (1950),
Skempton and Hutchinson (1969), Eden (1970), Brudsen (1979), Zaruba and Mencl
(1982) etc. The mechanism of slope failure and related soil loss as a result of rain as
well as gradient was studied by Ghosh (1950), Dutta (1966), Starkel (1972),
Onodera et al. (1974), Weichmier and Smith (1978), Morgan (1986), Borga et al.
(1998) etc. Triggering mechanism of slope instability is caused due to rise of
ground water level that saturates the soil and increase the pore water pressure.
Simple models have been developed for estimating the soil saturation of the
mountainous region as the wetness index was defined in TOPMODEL by Beven
and Kirkby (1979). More acceptable soil saturation model was applied by Mont-
gomery and Dietrich (1994), Borga et al. (1998) and Pack and Tarboton (1998).
Such model considers the ground water condition and its flow and rainfall intensity.
In the present study, threshold rainfall and threshold slope angle have been
considered as triggering mechanism to illustrate site specific slope instability that is
described in detail in Chap. 6. An One Dimensional (1-D) slope stability model has
been summarized in Chap. 7 to explain the distribution of instability by estimating
the safety factor value (FS) from 50 locations considering cohesion, angle of
internal friction, slope angle, saturation index, soil-water density and depth of the
1.10 Existing Literature on Landslide Inducing Parameters 43
soil. Topographic Index Model is applied after Beven and Kirkby (1979) to
understand the soil saturation of the slope. Geo-technical parameters were estimated
through Tri-axial compression test for determining the safety factor values. The
study reveals that weak lithological composition with cumulative rain for few days
makes the slope more vulnerable and causes debris slide in the Shivkhola Water-
shed of Darjiling Himalaya.
1.11 Slope Stability Models
Among the various natural hazards, slope failure is the most widespread and dam-
aging hazard (De Smedt 2005). A sudden failure of the slope is caused by sliding,
rolling, falling or slumping. The constant pull of gravity makes all the hill slopes and
mountain cliffs susceptible to slope failure. When failure occurs, material is trans-
ported down slope until a stable slope condition is reestablished. The high suscep-
tibility to failure in the Darjiling Himalayan terrain is due to a complex geological
structure and an interaction among various processes acting upon the steep mountain
escarpment slopes. The interaction among the processes leading to hill slope evo-
lution was studied by Ahnert (1987), Kirkby (1980), Poesen (1985), and Deploey
(1982). The stability of mountain slope is generally viewed in relative terms. It is
apparent, that in analysis of slope, stability is not totally a descriptive term. Perhaps,
an appropriate term would be functional stability, which necessarily relates to a
specific need and specific governing criteria (Cernica 1995). Soeters and Westen
(1996) recommended infinite slope stability analysis but due to complication in
establishing vertical depth of failure plane in 3D mechanism Monte Carlo Method, a
simplified approach was considered by them reducing 3D depth to 2D equivalent
depth based slope stability model. Again it (2D MODEL) was converted to equiv-
alent translational depth of 1D slope stability model by Bhattarai et al. (2005).
Hollingworth and Kovacs (1981), Burton and Bathrust (1994), Bradinoni and
Church (2004), Young (1963), Montgomery and Dietrich (1994), Van Westen and
Terlien (1996), Burton and Bathurst (1998), Pack et al. (2001), Borga et al. (2002)
and Saha et al. (2002) introduced numerous models in connection to the slope
stability, shallow and deep seated landslides. The stability equation is applied for a
mass of loose, friable cohesion less debris after Jumikis (1967), Melnikov and
Chesnokov (1969). The most widely used landslide inventory techniques include
(Montgomery and Dietrich 1994): (i) field investigation using a check list to
identify landslide susceptibility sites; (ii) projection of future pattern of instability
from the landslide inventories; (iii) multivariate analysis of factors; (iv) stability
ranking based on criteria such as slope, lithology, landform; and (v) failure prob-
ability analysis based on slope stability models with stochastic hydrologic simu-
lations. Recently, the availability of GIS data has provided a lot of advantages to
quantify topographic attributes related to slope instability and landsliding. Mont-
gomery and Dietrich (1994) propounded a contour based steady state hydrologic
model with the infinite slope stability (simplified for cohesionless soils) to define
44 1 Introduction
stability classes based upon slope and specific catchment area. The infinite slope
stability model concept based on soil cohesion and root cohesion with the help of
determining the safety factor is put forwarded by De Smedt (2005).
1.11.1 Hydrological Models and Slope Stability
The hydrologic system over the mountainous area is seldom simple and rarely exist
in steady state. The hydrologic response of a hill slope to rainfall leads to a water
table rise that is very difficult to predict. The amount and duration of rainfall and lag
time between the rainfall events, and water table rise may vary widely depending on
the hill slope configuration, intensity of rainfall, initial saturation condition and the
saturated and un-saturated hydro-geologic properties of the hill slope materials
(De Bleeschauwer and De Smedt 2002). The hydrologic factors like daily rainfall
threshold, rainfall intensity, infiltration were studied by Guzetti et al. (2007), and
Gabet et al. (2004) etc. The critical rainfall for initiating landslide was studied (rcr)
by Borga et al. (1998). The recurrence interval of rains of certain intensity above
threshold value was analysed using log probability law by Chow (1951, 1954) and
Schwab et al. (2002). Hydrological model incorporating the saturation excess
run-off were constructed by Beven and Kirkby (1979), O’Loughlin (1986), Moore
et al. (1988), Moore and Grayson (1991). A geomorphology based artificial neural
networks (GANNs) was put forward for estimating direct runoff over watersheds by
Zhang and Govindaraju (2003). Mishra et al. (2003) presented a modified SCS-CN
Method. Saragni et al. (2007) evaluated three unit hydrograph models to predict the
surface runoff from a Canadian watershed. Recently, an integrated approach for
estimating surface run-off using Remote Sensing and GIS in the applied field of
hydrological research is applied by Durbude et al. (2001), Ambazhagan et al.
(2005), Jasrotia and Singh (2006), Tripathi et al. (2002) and Zade et al. (2005).
SCS Curve Number was used for estimation of runoff by correlating generalized
land cover with hydrologic soil groups and data were derived from the SCS table by
Chandra et al. (1984), Ragan and Jackson (1980), Tiwari et al. (1991) etc. Sing
(1975) has pointed out Antecedent Moisture Condition (AMC) as the water content
present in the soil at a given time. The build-up of soil pore water pressure at the
contact between the soil mantle or weathered regolith and the underlying lithology
during rainstorms and the period of rapid snow melt was well recognized as the
triggering mechanism of slope failure and it was studied by Pierson (1977),
Swanston (1982), and Megahan (1983).
1.11.2 Landslide Susceptibility Zonation Models Concept
Landslide hazard zonation consists the division of an area into several zones, which
indicates progressive levels of landslide hazard. To constitute the zonation map of
slope instability it is necessary to understand triggering mechanism of landslides.
1.11 Slope Stability Models 45
Generally the location of the landslide and its behavior is governed by the
hydrology of the sub-catchment in which it is located rather by the characteristics of
the catchment as a whole. Without a thorough mapping of the sub-catchment and
without assigning the weighting accordingly, the match between the inferred hazard
rating and the observed hazard rating will remain elusive (Bhandari 1987). Disaster
Management Support Group suggested a working scale for a slope instability
analysis is determined by the requirements of the user for whom the survey is
executed. Planners and engineers use the following examples of scales:
• National scale (<1:1,000,000) provides a general inventory of problem areas for
an entire country, which can be used to inform national policy makers and the
general public.
• Regional scale (1:100,000–1:500,000) is used in the early phases of regional
development projects to evaluate possible constraints, due to instability, in the
development of large engineering projects and regional development plans.
• Medium scale (1:25,000–1:50,000) is used for the determination of hazard
zones in areas affected by large engineering structures, roads and urbanization
plans.
• Large scale (1:5,000–1:15,000) is used at the level of site investigations prior to
the design phase of engineering works.
There are some basic principles proposed by Varnes (1984), Carrara et al.
(1991), Hutchinson and Chabdler (1991), Hutchinson (1995), Turner and Schuster
(1996), Guzzetti et al. (1999a, b, c, 2003) behind the landslide susceptibility
zonation, though there are some conflicting views among experts. These are:
1. Firstly, the main controlling factors for initiating landslide phenomena should be
identified and mapped. So, slope failure can be recognized, classified and
mapped in the field or through remote sensing, chiefly stereoscopeic aerial
photographs (Rib and Liang 1978; Varnes 1978; Hansen 1984; Hutchinson
1988; Cruden and Varnes 1996; Dikau et al. 1996; Guzzetti et al. 2003).
2. According to Varnes (1984), Carrara et al. (1991), Hutchinson (1995), for
landslides, “the past and present are keys to the future”. Under this assumption,
landslide in future is likely to occur under the same geologic, geomorphic and
hydrologic conditions as those that led to landslide in the past. That is why
Varnes (1984), Carrara et al. (1991), (1995), Hutchinson (1995), Guzzetti et al.
(1999a, b, c, 2003) expressed that the understanding of the past failure is
essential for the assessment of landslide hazard.
3. Researchers like Crozier (1986), Hutchinson (1988), Dietrich et al. (1995),
Guzzetti et al. (2003) assured that conditions that initiate landslides, or which
are directly or indirectly linked to slope failures, can be collected and used to
build predictive models of landslide occurrence, because landslides are con-
trolled by mathematical laws that can be determined empirically, statistically or
in deterministic fashion.
4. Landslide occurrence can be inferred from the heuristic investigations, com-
puted through the analysis of environmental information or inferred from
46 1 Introduction
physical models studied after Carrara et al. (1995), Soeters and Van Westen
(1996), Guzzetti et al. (1999a, b, c, 2003).
Essentially, the following two broad options are available at the start of the
mapping program.
1. Direct mapping of landslide hazards, based on geological, geomrophological,
and geotechnical investigation, and mapping by a single multidisciplinary team.
2. State-of-nature (factor) mapping and integration of factor maps into landslide
hazards map.
British geomorphologists have cited some of the best examples of landslide hazard
maps by using the direct hazard mapping approach. The second option is, however,
normally preferred because Factor Maps are, in themselves, a great value to the users
of the project output and provide the best assurance of independent evaluation, easy
revision, and constant upgradation and revalidation (Bhandari 1994). Guzzetti et al.
(2003) evaluated the requirements of indirect/state-of-nature method as: (i) the rec-
ognition and mapping of landslides over a target region which is obtained by pre-
paring a landslide inventory map; (ii) the identification and mapping of the physical
factors which are directly or indirectly correlated with slope instability; (iii) an
estimate of the relative contribution of the instability factors in initiating slope
instability; (iv) the classification of land surface into domains of different levels of
susceptibility; and (v) the assessment of the model performance.
Crrara et al. (1992, 1995), Van Westen (1993), Hutchinson (1995), Soeters and
Van Westen (1996), Van Westen et al. (1997a), Guzzetti et al. (1999a, b, c) and
Committee on the National Landslide Hazards Mitigation Strategy (2004), India
suggested the most common approaches in the literature review of landslide
susceptibility mapping procedure and which can be grouped into five major cate-
gories, namely: (i) direct geomorphological mapping; (ii) analysis of landslide
inventories; (iii) heuristic or index based methods; (iv) statistical methods,
including neural networks, fuzzy logic and expert systems; and (v) process based
conceptual models.
Bureau of Indian Standards (BIS-1998), proposed a guidelines for landslide
hazard zonation map on 1:25,000 or 50,000 scale. Bhandari (1987) proposed
landslide hazard zonation on the basis of the hazard rating values relating to the
responsible factors for landslide. The landslide zonation map at regional scale has
been attempted in different parts of the world for past two/three decades (Nilsen and
Brabb 1973; De Graf 1978; Varnes 1981, 1985). A landslide hazard map (macro)
on 1:25,000 scales has been prepared following the landslide hazard evaluation
factor (LHEF) rating scheme proposed by the Indian Bureau of Standards (1998)
for the area between Singtam and Ranipool (after S.R. Basu and B. Bera). Data for
LHEF have been computed from interpretation of 1:50,000 Survey of India topo-
graphical maps, 1:25,000 prints of panchromatic aerial photographs (stereoscopic
coverage), 1:50,000 geo-coded LISS III Satellite data, existing geological maps and
extensive field work. Bhandari (1987) proposed landslide hazard zonation on the
basis of the hazard rating values relating to the responsible factors for landslide.
1.11 Slope Stability Models 47
Vecchia (1978) proposed the terrain index for the stability of hill slope that includes
numerical rating for lithology, altitude, slope and friction along the potential failure
planes. Pachuri et al. (1998) carried out terrain classification and landslide
susceptibility mapping in Yamuna valley region of Garhwal Himalaya based on
aerial photographs (1:40,000), Landsat Images (1:2,50,000) and topographic maps
(1:50,000). In the Himalaya, landslide zonation map has been attempted by
Majumder (1980) Chatterjee (1986), Gupta (1998), Gairola and Shukla (1990),
Sharan (1992), Chandra (1992), Sharma et al. (1996). The Landslide Hazard
Mapping and Risk Assessment (LHRA) was attempted by Fiener in (1999). He
considered nine factors i.e. lithology, degree of weathering, structure, slope con-
dition, hydrology, erosion, physical properties, land use and land cover and slope
history for the purpose of hazard assessment. Champati Roy et al. (1997) prepared
landslide hazard zonation map on the basis of information value method of an area
in Chamoli district near Pipalkota town using aerial photographs and IRS LISS-II
data. The SINMAP (Stability Index Mapping) was studied by Hammond et al.
(1992), Montgomery and Dietrich (1994). Swanston and Dyrness (1973), Dietrich
and Dunne (1978), Sidle et al. (1985) concluded that in the forested steep slope
region, shallow landslides are dominant erosional processes. Bergin et al. (1995),
Sidle et al. (1985), Royster (1979) suggested the reforestation and improved road
drainage as the most important slope stabilization strategies and such strategies
received considerable attention from Cruz and Reyes (2000).
The landslide hazard and risk analysis with the help of the geotechnical model,
logistic regression model and the safety factor was done by Gokceoglu et al. (2000),
Romeo (2000), Carro et al. (2003), Shou and Wang (2003), Zhou et al. (2003),
Atkinson and Massari (1998), Dai et al. (2001), Dai and Lee (2002), and Ohlmacher
and Davis (2003). Rowbothan and Dudycha (1998), Jibson et al. (2000), Luzi et al.
(2000), Praise and Jibson (2000), Rautelal and Lakheraza (2000), Baeza and
Corominas (2001), Lee and Min (2001), Temesgen et al. (2001), Clerici et al.
(2002), Donati and Turrini (2002), Lee et al. (2002a, b), Rece and Capolongo
(2002), Zhou et al. (2002), and Lee and Choi (2003) have applied the probabilistic
model for landslide risk and hazard analysis.
Shasko and Keller (1991), Guzzetti et al. (1999a, b, c) discussed the possibility
and feasibility of integrating sophisticated slope stability modeling with GIS.
Ramakrishnan et al. (2002) has carried out extensive work on landslide analysis
using aerial photographs. Gorsevski et al. (2000a) used logistic regression model
for spatial prediction of landslide hazard. Recently, landslide hazard evaluation
using fuzzy logic, and artificial neural network models have been mentioned in the
various literature of Ercanoglu and Gokceoglu (2020), Pistocchi (2000) and Lee
et al. (2003a, b, 2004a, b). Rowbothan and Dudycha (1998), Baeza and Corominas
(2001), Lee and Min (2001), Temesgen et al. (2001), Clerici et al. (2002), Donati
and Turrini (2002), Lee et al. (2002a, b), Rece and Capolongo (2002), Lee and Choi
2003), Chung and Fabrri (2003), Lee and Pradhan (2006, 2007), Youssef et al.
(2009), and Pradhan and Lee (2010a, b, c) have studied and applied the probabi-
listic model for landslide susceptibility and risk evaluation. Joshi et al. (2002)
carried out landslide hazard zonation in part of Alakananda valley in collaboration
48 1 Introduction
with Space Application Centre (SAC), Ahmedabad in 1:50,000 scale. They used
IRS LISS-II and PAN data and integrated different thematic maps giving weigh-
tages and ranks in a GIS environment. Lee and Pradhan (2006) introduced landslide
hazard and risk mapping on Penang Island, Malaysia using Geographic Information
System and Remote Sensing Data. Barbieri and Cambuli (2009) presented statis-
tical method in landslide susceptibility mapping in 18th World IMACS/MODSIM
Congress, Australia. An integrated approach for landslide susceptibility mapping
using Remote Sensing and GIS was produced by Sarkar and Kanungo (2004),
Sharifikia (2007) Pande et al. (2008), and Nithya and Prasanna (2010). Recently,
Remote Sensing and GIS based Analytical Hierarchy Approach is one of the most
popular and widely applied decision support system and has been used to syn-
thesize various judgement comparisons to derive priorities among the criteria
responsible for landslide and to prepare landslide susceptibility map (Saaty 1990,
1994; Saaty and Vargas 2001; Yang et al. 2006).
1.12 Study on Landslide Management
Given the rapid demographic expansion found in much of the region, it simply
won’t be possible to set aside all vulnerable slopes for protection; therefore, some
strategies of slope stability on steep lands are required. Reforestation and improved
road drainage (e.g. Bergin et al. 1995; Siddle et al. 1985; Royster 1979; Cruz and
Reyes 2000; IADB 1999), have received considerable attention but there is still a
need to identify agricultural practices that reduce vulnerability to slope failure.
A soil conservation survey conducted in Honduras, Nicaragua, and Guatemala in
the wake of Hurricane Mitch failed to find any beneficial effect of “agro-ecological”
farming on landslide resistance during the storm (World Neighbors 2000). Use of
agro-forestry, contour cropping, physical and vegetative erosion barriers, or inte-
grated weed management were found to reduce soil degradation and surface erosion
on farms affected by Mitch, but these techniques did not correlate with a reduction
in landslides. While it is possible that soil conservation farming truly has no effect
on slope stability, the experience of geotechnical engineers and landslide modelers
indicates that land management targeted to soil or site-specific conditions can
reduce the probability of slope failure (Montgomery et al. 2000; Collison et al.
1995; Royster 1979).
The number of variables involved makes it difficult to characterize the influence
of land management on landslide susceptibility and even more difficult to predict
how changes in management will impact the current state. In recent years, the field
of physical landslide modeling has made excellent progress in dealing with the
complexities of slope failure. Applications of simplified slope-stability models have
proved effective as descriptive and predictive tools in temperate zones, allowing for
rapid stability assessment over a wide area (Jibson et al. 2000). Knowledge of a
stability index alone may be useful for planning timber harvests (e.g. Montgomery
et al. 2000) or for citing infrastructure (e.g. Carrara et al. 1995), but it is not helpful
1.12 Study on Landslide Management 49
for steep-land farmers for whom non-use and relocation are not viable options. If
physical landslide models are to contribute to slope stabilization in tropical,
agricultural watersheds, then a driving cause for instability must be identified, along
with the probability that a slope will fail. Importance of causal information is borne
out by physical slope-stability models that do consider the influence of dynamic
hydrology and plant growth. These models are computationally intensive and, to
our knowledge, have only been applied to engineered or “characteristic” slope
formations (Duncan 1996; Collison et al. 1995; Anderson et al. 1990). Nonetheless,
results of these modeling exercises are relevant to field-level management. Collison
et al. (1995) found that planting trees on an engineered embankment only enhances
the embankment’s stability if soil hydraulic conductivity is relatively high. For
deep-soiled embankments with low hydraulic conductivity, planting trees was, in
fact, detrimental to stability. Macro-pore flow along tree roots increased perme-
ability, leading to elevated pore pressures in and below the rooting zone, while roots
themselves did not penetrate deeply enough to anchor against deep-seated slope
failure. Under such conditions, it is preferable to vegetate the embankment with
grass, or some similar shallow-rooted ground cover, that sheds water off the
saturated portion of the slope. Water management can be more important than
physical reinforcement for weathered and frequently saturated soils. Drainage
ditches, or water-shedding ground cover, may offer better slope stabilization than
trees. Information on the soils and hydrology of a landslide-prone area is therefore
necessary for the development of appropriate management recommendations;
simply reporting the probability of failure is not enough (Collison et al. 1995).
Landslide mitigation and control measures depend on detailed investigations,
including identifying of the causative factors (Bhandari 1988). The landslide
mitigation work can be broadly classified into two categories: control work and
restriant work (Valdiya 2006). The control works involve modification of the
natural conditions such as topography, geology, ground water, and other conditions,
that indirectly promote landslide. Restraint works cover construction of structure
such as surface and sub-surface drainage works, removal of earth from the unstable
area, and building buttress walls, piles, anchors, and retaining walls. Montgomery
(1986) and Valdiya (1987) have suggested four-fold strategy of the control of
landslides. This includes—(i) reduction of the slope angle and placement of addi-
tional supporting material at the foot of the slope; (ii) reduction of the load on the
slope by removing the rock or soil situated high up on the slope; (iii) the utilization
of retention structure; and (iv) removal of fluid by various kind of drainage systems.
Vegetation is being widely used for erosion control, to achieve slope stabilization
along the transportation routes in countries like Japan, Korea and Hong Kong. The
new technique for slope stabilization with the help of vegetation has been adopted
by the Korean Highway Corporation (Sung-Hwan Kim et al. 1997). A list of shrubs
and trees that are useful for slope stabilization is given by Gupta (1979). Vegetation
turfing is the most effective and important corrective measures, particularly for the
freshly-exposed surfaces produced by road cutting and mining. Planting of grasses,
shrubs, trees and bamboos, followed by putting of jute net or vegetated stone
50 1 Introduction
pitching upon the freshly-turfed area would ensure rapid and undisturbed growth of
vegetation (Howell 2001).
All the studies, mentioned, are area case specific and highly specialized.
Understanding the uniqueness of the present study area was of high priority. Degree
of importance of individual landslide triggering factor varies from place to place.
Present research work was organized to identify relative importance of the landslide
triggering factors for the area under study. Again, within the Shivkhola basin, there
is also spatial variation in the contribution of the triggering factors. It is essential to
analyze spatial variation in the relative importance of the factors and also
site-specific variation in the triggering factors. Geomorphic, hydrologic, land use
and anthropogenic attributes and their spatial distribution were studied in details
through intensive field work. Runoff models, Digital Elevation Model (DEM), I-D
Slope Stability Model, Run-off Model (The United States Department of Agricul-
ture Soil Conservation Service (USDA SCS) Curve Number (CN) technique),
Geomorphic Threshold, Topographic Index Model, Land Use Index Model, RS &
GIS base Frequency Ratio and Analytical Hierarchy Process (AHP) etc. were used
to make a detailed spatial analysis and proper integration of factors for proper
synthesis. Rainfall characters were analyzed to determine the critical rain and return
period of the critical rain at various recurrence intervals. The cohesion, angle of
internal friction, and safety factor were derived applying IS: 2131 (1981) Standards
and Codes to realize the spatial distribution of slope instability. To identify the
potential landslide susceptible places in the study area, RS & GIS based Analytical
Hierarchy Approach was applied. Verification is made for accuracy judgment.
Most of the data were collected from intensive field study. In the present study
slope instability in the Shivkhola Watershed was analyzed considering both
physical and human factors taken together as a comprehensive whole in a cognitive
approach in an attempt to propose for better management and thus to achieve social
relevance as an active device for decision makers.
References
Ahnert F (1970) Functional relationship between denudation, relief and uplift in large mid-latitude
basins. Am J Sci 268:248–263
Ahnert F (1987) Process response models of denudation at different spatial scales. Catena Suppl
10:31–50
Anderson RS et al (1990) Interaction of weathering and transport processes in the evolution of arid
landscapes. In: Cross T (ed) Quantitative dynamic stratigraphy. Prentice Hall, Englewood
Cliffs, pp 349–361
Atkinson PM, Massari R (1998) Generalized linear modeling of susceptibility to landsliding in the
central Apennines, Italy. Comput Geosci 24:373–385
Attewell PB, Farmer IW (1976) Principles of engineering geology. Chapman & Hall, London
Barbieri G, Cambuli P (2009) The weight of evidence statistical method in landslide susceptibility
mapping of the Rio Pardu Valley (Sardinia, Italy). In: 18th World IMACS/MODSIM congress,
Cairns, Australia, 13–17 July 2009
References 51
Bartarya SK, Virdi NS, Sah MP (1996) Landslide hazards-some case studies from Sutlej Valley,
Himachal Pradesh. Himal Geol 17:193–207
Basu SR, Ghatwar L (1988) Landslide and soil erosion in the Gish basin of the Darjiling
Himalayas and their bearing on North Bengal Floods. Studia Geomorpholigica Carpatho-
Balcanica, vol 22. Krakow, Poland
Basu SR (1989) A study of the impact of landslides on the fluvial processes of the rivers Lish and
Gish of the Darjeeling Himalayas. In: Second international conference on geomorphology,
Georka Plus, Frankfurt, p 23
Basu SR, Maiti RK (2001) Unscientific mining and degradation of slopes in the Darjeeling
Himalayas. Changing environment. Scenerio of the Indian Subcontinent (Bd), pp 390–399
Benson WN (1940) Landslide and applied features in the Dunedin District in relation to geological
structure, topography and engineering. Trans Roy Soc NZ 70:249–263
Bjerrum L, Jordtad F (1966) Stability of rock slopes in Norway, vol 67. Nowegian
Geotechechnical Institute, Oslo, pp 50–78
Beven KJ, Kirkby MJ (1979) A physically based variable contributing area model of basin
hysrology. Hydrol Sci Bull 24(1):43–69
Bhandari RK (1987) Slope stability in the fragile Himalaya and strategy for development. J IGE
17:1–78
Bhandari RK (1994) Landslide hazard zonation mapping in Sri Lanka—a holistic approach. In:
Proceedings of national symposium on landslide in Sri Lanka, vol 1. National Building
Research Organisation, Columbo, p 271
Bhandary NP, Dahal RK, Okamura M (2012) Preliminary understanding of the Seti River Debris-
Flood in Pokhara, Nepal on May 5th 2012. A report based on a quick field visit program. In:
International society of soil mechanics and geotechnical engineering (ISSMGE), vol 6, issue 4.
Bulletin, pp 8–18
Bhandary NP, Yatabe R, Dahal RK, Hasegawa S, Inagaki H (2013) Areal distribution of large-
scale landslides along highway corridors in central Nepal. Georisk Assess Manage Risk Eng
Syst Geohazards doi:10.1080/17499518.2012.743377
Bhattarai P, Aoyama K (2001) Mass movement problems along Prithwi Highway Nepal. Annual
report of Research Institute for hazards in Snowy Areas, vol 23. Niigata University, pp 85–92
Bhattarai P et al (2005) Quantitative slope stability mapping with Arc GIS: prioritize high way
maintenance
Bergin DO, Kimberley MO, Marden M (1995) Protective values of regenerating tea tree stands on
erosion prone hill country. NZ J Foresty Sci 25:3–19
Bhattacharya SK (1999) A constructive approach to landslide through susceptibility zoning case
study in the Rakti Basin of Eastern Himalaya. Transaction 20–23:317–333
Bloom AL (1978) Geomorphology. Printice-Hall, New Delhi
Bloom AL (1991) Geomorphology, a systematic analysis of the Cenozoic Landforms. Prentice
Hall, New Delhi, pp 76–177
Borga M, Fontag DG, Ros DD, Marchi L (1998) Shallow landslide hazard assessment using a
physically based model and digital elevation data. Environ Geol 35(2–3):81–88. Springer
Brardinoni F, Church M (2004). Representing the landslide magnitude frequency relation,
Capilano river basin, British Colombia. In: Kirkby JM, Darby ES (eds) Earth surface processes
and landforms, vol. 29(1). pp 115–124
Brown CB, Sheu MS (1975) Effects of deforestation on slopes. J Geotech Eng Div Am Soc Civ
Eng GT2:147–65
Brudsen D (1979) Mass movement in process in geomorphology. Embelton C, Thornes J (eds)
Wiley, New York, pp 130–186
Burton A, Bathurst JC (1998) Physically based modeling of shallow landslide sediment yield at a
catchment scale. Environ Geol 35(2–3):89–99
Caine N (1980) The rainfall intensity-duration control of shallow landslides and debris flows.
Geogr Ann 62A:23–27
Carrara A et al (1991) GIS technique and statistical models in evaluating landslide hazard. Earth
Surf Process Land 16(5):427–445
52 1 Introduction
Carrara A, Cardinali M, Guzzetti F, Reichenbach P (1995) GIS-based techniques for mapping

landslide hazard in geographical information system in assessing natural hazards. Academic
Publication, Dordrecht
Carson MA (1977) Angles of repose, angles of shearing resistance at angle of talus slopes. Earth
Surf Process 2:363–380
Chakraborty I, Ghosh S, Bhattacharya D, Bora A (2011) Earthquake induced landslides in the
Sikkim-Darjiling Himalaya—an aftermath of 18th September 2011 earthquake. Engineering
Geology Division, Geological Survey of India, Eastern Region, Kolkata
Chandra H (1975) Landslide in Teesta Valley. In: Proceeding on landslide and Toe Erosion
problems with special reference to Himalaya. ISEG, Kolkata, pp 147–164
Chowdhury G et al (2000) Numerical simulation of building performance under different low
energy cooling technologies. Int J Energy Environ 1(1):28–36
Chorley RJ, Schumm SA, Sugden DE (1985) Geomorphology. Methuen, London, New York
Close U, McCormick E (1922) Where the mountain walked. Nat Geogr Mag 41:445–464
Coppola L et al (2006) Reconstruction of the conditions that initiate landslide movement in
Weathered Silty Clay Terrain: effects on the historic and architectural heritage of Pietrapertosa,
Basilicata, Italy. Landslides 3(4):349–359
Coates DR (1977) Landslide perspectives. Geol Soc Am Rev Eng Geol 3:3–28
Congalton R (1991) A review of assessing the accuracy of the classification of remotely sensed
data. Remote Sens Environ 37:35–46
Crozier MJ (1986) Landslides: causes, consequences and environment. Croom Helm, London,
pp 252
Costa JE (1984) Physical geomorphology of debris flows’. In: Costa JE, Fleischer PJ (eds)
Developments and application of geomorphology. Springer, Berlin, pp 268–317
Dai FC, Lee CF (2002) Landslide characteristics and slope instability modeling using GIS, Lantau
Island, Hong Kong. Geomorphology 42:213–228
Dahal RK (2009) Representative rainfall threshold for landslide in the Nepal Himalaya.
Geomorphology 100(3–4):429–443
Dahal RK, Bhandary NP (2013) Geo-disaster and its mitigation in Nepal. In: Wang F et al (eds)
Progress of geo-disaster mitigation technology in Asia, environmental science and engineering.
Springer, Berlin, pp 123–156. doi:10.1007/978-3-642-29107-4_6
Deoja B et al (1991) Mountain risk engineering handbook. International Centre for Integrated
Mountain Development (ICIMOD), Kathmandu, 875 pp
De Vleeschauwer C, De Smedt F (2002) Modeling slope stability using GIS on a regional scale.
In: Proceedings of the first geological Belgica international meeting, vol 12. Aardkundige
Mededelingen, Leuven, 11–15 Sept 2002, pp 253–256
De Smedt F (2005) Slope stability analysis using GIS on a regional scale: a case study of
Narayanghat-Mungling highway section, Nepal. Dissertation of Master Degree, Universiteit
Gent, Belgium
Didwal RS (1988) Remedial methods employed in Jammu Province in India to control landslides
and their efficacy. In: Bonnard C (ed) Landslides, vol 2. A.A Balkema, Rotterdam, pp 897–902
Donati L, Turrini MC (2002) An objective and method to rank the importance of the factors
predisposing to landslides with the GIS methodology, application to an area of the Apennines
(Valnerina; Perugia, Italy). Eng Geol 63:277–289
Ellen SD, Wieczorek GF (1988) Landslides, floods, and marine effects of the storm of January
3–5, 1982, in the San Francisco Bay Region, California. US Geol Sur Prof Pap 1434:1–310
Engineer MN (1975) Landslide and erosion problems—a case study on road Kimin-Ziro in
Arunachal Pradesh. In: Proceedings landslide and Toe Erosion problems with special reference
to Himalaya. ISEG, Kolkata, pp 187–206
Ersmann TH (1979) Mechanisms of large landslides. Rock Mech 12:15–46
Gabet EJ, Burbank DW, Putkonen JK, Pratt-Sitaula BA, Oiha T (2004) Rainfall thresholds for
landsliding in the Himalayas of Nepal. Geomorphology 63:131–143
Glade T (1998) Establishing the frequency and magnitude of landslide-triggering rainstorm events
in New Zealand. Environ Geol 35:160–174
References 53
Gokceoglu C, Sonmez H, Ercanoglu M (2000) Discontinuity controlled probabilistic slope failure

risk map of the Altindag (settlement) region in Turkey. Eng Geol 55:277–296
Gray DH (1970) Effects of forest clear-cutting on the stability of natural slope. Bull Assoc Eng
Geol 7:45–66
Gupta PK (1979) Plants for environmental conservation. Bishan Singh, Mahendrapal Singh,
Dehradun, p 247
Guzzetti F, Carrara A, Cardinali M, Reichenbach P (1999a) Landslide hazard evaluation: a review
of current techniques and their application in a multi-scale study, Central Italy. J Geomorphol
31:181–216. Elsevier, London
Guzzetti F, Cardinali M, Reichenbach P, Carrara A (1999b) Comparing landslide maps; a case
study in the upper Tiber River Basin, central Italy. Environ Manage 18:623–633
Guzzetti F, Cardinali M, Reichenbach P, Carrara A (1999c) Landslide hazard evaluation: an aid to
a sustainable development. Geomorphology 31:181–216
Guzzetti F, Cardinali M, Reichenbach P, Ardizzone F, Galli M (2003) Impact of landslides in the
Umbria region, central Italy. Nat Hazards Earth Syst Sci 5:1–17
Guzzetti F, Peruccacci S, Rossi M, Stark CP (2007) Rainfall thresholds for the initiation of
landslides in central and southern Europe. Meteorol Atmos Phys. Online first version, doi:10.
1007/s00703-007-0262-7. Accessed 7 Aug 2007
Hammond C et al (1992) Level I stability analysis (LISA) documentation for version 2. General
technical report INT-285, USDA Forest Service, Intermountain Research Station, 121 pp
Heim A (1932) Bergstruz and Meanschenleben. Naturf Gesell 77:218. Vierteljahrsschrift, Zurich
Holmsen P (1953) Landslip in Norwegian quick-clays. Geotechniques 3:187–194
Horton RE (1932) Drainage basin characteristics. Trans Amer Geophys U 14:350–361
Howell JH (2001) Application of bio-engineering in slope stabilization: Experience from Nepal,
landslide hazard mitigation in the Hindukush Himalaya. ICIMOD, Kathmandu, pp 147–161
Hsu KJ (1975) Catastrophic debris stream (sturzstorms) generated by rockfalls. Bull Geol Soc Am
86:129–140
Huang R (2004) Mechanism of large scale landslide in Western China. Adv Earth Sci 19:443–450
Hugget RJ (2007) Fundamentals of geomorphology. Routledge, London
Hutchinson JN, Bhandari RK (1971) Undrained loading, a fundamental mechanism of mudfloes
and other mass movements. Geotechnique 21:353–358
Jacobson RB, Cron ED, McGeehin JP (1989) Slope movement triggered by heavy rainfall,
November 3–5, 1985, in Virginia and West Virginia, USA. Geol Soc Am 236:1–13 Special
Paper
Jibson WR, Edwin LH, John AM (2000) A method for producing digital probabilistic seismic
landslide hazard maps. Eng Geol 58:271–289
Kirkby MJ (1980) The stream head as a significant geomorphic threshold. In: Coates DR, Vitek JD
(eds) Threshold in Geomorphology. Allen and Unwin, Winchester, pp 53–73
Kojan E, Hutchinson JN (1978) Mayunmarca rockslide and debris flow, Peru. In: Voight B (ed)
Rockslides and abalanches, vol 1. Elsevier, Amsterdam. pp 315–361
Krishnaswamy VS, Jain MS (1975) A review of major landslides in north and north-western
Himalayas. In: Proceedings landslide and Toe Erosion problems with special reference to
Himalaya. ISEG, Kolkata, pp 3–40
Lee S, Choi U (2003) Development of GIS based geological hazard information system and its
application for landslide analysis in Korea. Geosci J 7:243–252
Lee S, Pradhan B (2006) Landslide hazard assessment at Cameron Highland Malaysia using
frequency ratio and logistic regression models. Geophy Res Abstr 8. SRef-ID: 1607-7962/gra/
EGU06-A-03241
Lee S, Pradhan B (2007) Landslide hazard mapping at Selangor, Malaysia using frequency ratio
and logistic regression models. Landslides 4(1):33–41
Maiti RK (2007) Critical analysis of slope instability on mining scars at Tindharia Cricket Colony,
Darjiling, West Bengal. In: Proceedings of eighteenth convention and national seminar on
“quarternary climatic changes and landforms. Organized at Manonmaniam Sundaranar
University, Tirunelveli 627012, Tamilnadu, pp 189–205
54 1 Introduction
McConnell RG, Brock RW (1904) Great landslides at Frank, Alberta. Canada Department of
interior annaul report, 17 pp
Melnikov M, Chensokov M (1969) Safety in open cast mining. Mir Publications, Moscow
Miller VC (1953) A quantitative geomorphic study of drainage basin characteristics in the Clinch
mountain area, Virginia and Tenneses, vol 3. Contract N60NR271-30, Technical Report,
Department of Geology, Columbia University, pp 1–30
Montgomery DR, Dietrich WE (1989) Source areas, drainage density and channel initiation. Water
Resour Res 25(8):1907–1918
Montgomery DR, Dietrich WE (1994) A physically based model for the topographic control on
shallow land sliding. Water Resour Res 30(4):1153–1171
Nithya ES, Prasanna RP (2010) An integrated approach with GIS and remote sensing technique for
landslide zonation. Int J Geomat Geosci 1(1):121–132
Pack RT, Tarboton DG, Goodwin CN (1998) Terrain stability mapping with SINMAP. Technical
description and users guide for version 1.00. http://www.engineering.usu.edu/dtarb/. Accessed
15 June 2007
Pandey A, Dabral PP, Chowdhary VM, Yadav NK (2008) Landslide hazard zonation using remote
sensing and GIS: a case study of Dikrong river basin, Arunachal Pradesh, India. Environ Geol
54:1517–1529
Parise M, Jibson WR (2000) A seismic landslide susceptibility rating of geologic units based on
analysis of characteristics of landslides triggered by the 17 January, 1994 Northridge,
California earthquake. Eng Geol 58:251–270
Poesen J, De Poley J (1987) Geomorphological models—theoritical and emperical aspects. Catena
Suppl 10:67–72
Pradhan B, Lee S (2010a) Delineation of landslide hazard areas on Penang Island, Malaysia, by
using frequency ratio, logistic regression, and artificial neural network models. Environ Earth
Sci 60:1037–1054
Pradhan B, Lee S (2010b) Landslide susceptibility assessment and factor effect analysis:
backpropagation artificial neural networks and their comparison with frequency ratio and
bivariate logistic regression modeling. Environ Modell Softw 25(6):747–759
Pradhan B, Lee S (2010c) Regional landslide susceptibility analysis using back-propagation neural
network model at Cameron Highland, Malaysia. Landslides 7(1):13–30
Prasad MK (1986) Kerala’s Western Ghats in the five year plans. In: Nair KSS, Ganaharan R,
Menon ARR, Ramachandran KK (eds) Proceedings of the seminar on Eco development of
Western Ghats. Kerala Forest Research Institute, Trivandrum, pp 288–292
Quinn P et al (1991) The prediction of hillslope flow paths for distributed hydrological modeling
using digital terrain models. Hydrol Process 5:59–79
Roy P, Chacko T (1986) The causes and control of landslides. In: Nair KSS, Ganaharan R, Menon
ARR, Ramachandran KK (eds) Proceedings of the seminar on Eco development of Western
Ghats. Kerala Forest Research Institute, Trivandrum, pp 244–246
Rautelal P, Lakhera RC (2000) Landslide risk analysis between Giri and Tons Rivers in Himachal
Himalaya (India). Int J Appl Earth Obs Geoinf 2:153–160
Rogan MR, Jackson JT (1980) Run-off synthesis using LANDSAT and SCS models. J Hydraulics
Div 106(5):667–678
Rowbotham D, Dudycha DN (1998) GIS Modelling of slope stability in Phewa Tal Watershed,
Nepal. Geomorphology 26:151–170
Sah MP, Bartarya SK (2002) The impach of March 29, 1999 Chamoli Earthquake on slope
stability and spring discharge in Chamoli and Rudraprayag districts of Garhwal Himalaya.
Himal Geol 23:121–133
Sah MP, Virdi NS, Bartarya SK (1996) Slope failure and damaging of river channels: some
examples from the Sutlej Valley, Himachal Pradesh. Himal Geol 17:183–191
Saha AK, Gupta RP, Arora MK (2002) GIS based landslide hazard zonation in the Bhagirathi
(Ganga) Valley, Himalayas. Int J Remote Sens 23(2):357–369
Sarkar S, Kanungo DP (2004) An integrated approach for landslide susceptibility mapping using
remote sensing and GIS. Photogram Eng Remote Sens 70(5):617–625
References 55
Saaty TL (1990) The analytical hierarchy process: planning, priority setting, resource allocation,
1st edn. RWS Publication, Pittsburgh, 502 pp
Saaty TL (1994) Fundamentals of decision making and priority theory with analytic hierarchy
process, 1st edn. RWS Publication, Pittsburgh, 527 pp
Saaty TL, Vargas LG (2001) Models, methods, concepts and applications of the analytic hierarchy
process, 1st edn. Kluwer, Boston, 333 pp
Salter RT, Crippen TF, Noble KE (1981) Storm damage assessment of Thames-Te Aroha area
following the storm of April 1981. Final report, Water and Soil Conservation Science Centre,
Aokautere, New Zealand. Ministry of Works and Development, Internal report, p 44
Sarangi A, Madramootoo CA, Enright P, Prasher SO (2007) Evaluation of three unit hydrograph
models to predict the surface run-off from a Canadian Watershed. Water Resour Manage
21:1127–1143
Sassa K (1988) Geotechnical model for the motion of landslides. In: Proceedings of the 5th
international symposium on landslides, vol 1. A.A. Balkema, Rotterdam, pp 37–56
Schumm SA (1956) Evolution of drainage systems and slopes in badlands at Perth Amboy, New
Jersy. Bull Geol Soc Am 67:597–646
Schwab GO, Fangmeier DD, Elliot WJ, Frevert RK (2002) Soil and water conservation
engineering, 4th edn. Wiley, New York, pp 34–35
Selby MJ (1982) Hillslope materials and processes. Oxford University Press, Oxford
Selby MJ (2005) Hillslope materials and processes, Oxford University Press, Oxford (Reprinted
volume)
Sharpe CFS (1938) Landslide and related phenomena. Columbia University Press, New York
Sidle RC (1991) A conceptual model of changes in root cohesion in response to vegetation
management. J Environ Qual 20(1):43–52
Simon A, Larsen MC, Hupp CR (1990) The role of soil processes in determining mechanisms of
slope failure and hillslope development in a humid tropical forest, eastern Puerto Rico.
Geomorphology 3:263–286
Sinha BN (1975) A engineering geologic approach to landslides and slope failure, In: Proceedings
landslide and Toe Erosion problems with special reference to Himalaya. ISEG, Kolkata,
pp 56–65
Sinha BN, Pradhan SR, Sinha P (1975) Geological analysis of Padamchen Slide, Sikkim. In:
Proceedings of landslide and Toe Erosion problems with special reference to Himalaya. ISEG,
Kolkata, pp 227–234
Skempton AW (1953) Soil mechanics in relation to geology. Proc Yorkshire Geol Soc 29:33–62
Skempton AW, Hutchinson JN (1969) Stability of natural slope and embankment section. In:
Proceedings of 7th international congress on soil mechanics and foundation engineering,
Mexicom, pp 291–340
Soeters R, Westen CJ (1996) Slope instability recognition, analysis and zonation. In: Turner AK,
Schuster RL (eds) Landslides: investigation and mitigation, vol 247. Transportation Research
Board Special Report, National Academy Press, Washington, pp 129–177
Starkel L (1972) The role of catastrophic rainfall in the shaping of the relief of the Lower Himalaya
(Darjeeling Hills). Geogr Polonica 21:103–147
Strahler AN (1952) Hypsometric analysis of erosional topography. Geol Soc Amer Bull
63:1117–1142
Summerfield MA (1991) Global geomorphology: an introduction to the study of landforms.
Pearson, Prentice Hall, Harlow
Kim S-H et al (1997) Slope stabilization using erosion control methods and its application. In:
Proceedings of international symposiun on landslide hazard assessment, Xian, pp 387–399
Terzaghi K (1950) Mechanism of landslides. In: Paige S (ed) Application of geology to
engineering practice, Berkey volume. Geological Society of America, Boulder, pp 83–123
Terzaghi K, Peck RB (1976) Soil mechanics in engineering practice, 2nd edn. Willey, New York
Thigale SS, Khandge AS (1996) Lnadslide in Western Ghats of Maharashtra. In: Pathan AM,
Thigale SS (eds) Contributions to environmental geoscience. Arvalli Books International, New
Delhi, pp 161–174
56 1 Introduction
U.S. Soil Conservation Service (SCS) (1972) Hydrology. In: Natural engineering handbook,
section 4. GPO, Washington
Valdiya KS (1987) Environmental geology: Indian context. Tata McGraw Hill, New Delhi,
pp 269–315
Van Burkalow A (1945) Angle of repose and angle of sliding friction: an experimental study. Geol
Soc Am Bull 56:669–707
Van Westen CJ, Castellanos-Abella E, Sekhar LK (2008) Spatial data for landslide susceptibility,
hazards and vulnerability assessment: an overview. Eng Geol 102(3–4):112–131
Varnes DJ (1958) Landslide types and processes. In: Eckel EB (ed) Landslides engineering
practice, vol 544. Highway Research Board, Special report 29, NAS-NRC Publication,
pp 20–47
Varnes DJ (1978) Slope movement types and processes. In: Schuster RL, Krizek RJ (eds)
Landslide analysis and control, vol 176. Special report, National Academy of Sciences,
Washington, pp 11–33
Varnes DJ (1984) Lanslide hazard zonation: a review of principles and practice. UNESCO Nat
Hazard 3:61
Waltham T (2002) Foundations of engineering geology
Wu TH, Sangery DA (1978) Strength properties and their measurement. In: Schuster RL, Krizek
RJ (eds) Landslides, vol 176. Transportation Research Board, Special Report, National
Academy of Sciences, pp 139–154
Yatsu E (1966) Rock control in geomorphology. Sozosha, Tokyo
Yim KP, Heung LK, Greenway RD (1988) Effects of root reinforcement on the stability of three
fill slopes in Hong Kong. In: Proceedings, 2nd international conference on geomechanics in
tropical soils (Singapore), vol 1. pp 293–299
Young A (1963) Deductive models of slope evolution. Rep Int Geogr Un Slopes Comm 3:45–66
Zaruba Q et al (1969) Landslides and their control. Elsevier, Amsterdam, p 205
Zaruba Q, Mencl V (1982) Landslides and control, vol 10017, issue 324. Elesvier, New York,
35 pp
Zhou CH, Lee CF, Li J, Xu ZW (2002) On the spatial relationship between landslide and causative
factors on Lantau Island, Hong Kong. Geomorphology 43:197–207
Chapter 2
Geo-spatial Variability of Physiographic
Parameters and Landslide Potentiality
Abstract The stability of mountain slope depends upon physical and chemical
properties of the soil. In the present work geomorphic properties such as slope
angle, slope aspect, slope curvature, lithological composition, and lineament as well
as behaviour of slope materials such as texture, cohesion (c), friction angle (φ),
water holding capacity, porosity, weight soil density and density of soil water were
assessed and found out the relationship with landslip in the Shivkhola Watershed.
Slope angle, slope aspect and slope curvature were derived from DEM on GIS
platform. Lineament map was prepared using Satellite Image LISS III (2010). Soil
samples were collected from 50 different locations and their laboratory test were
being carried out to assess cohesion, friction angle, water holding capacity, pore
space, and wet soil density. The spatial distribution of the mentioned characteristics
of slope forming materials is done using ARC GIS Software. To estimate evolu-
tionary stage through which the Shivkhola Watershed is passing, hypsometric
analysis was made. Integration between landslides inventory map and derived
thematic geomorphic maps was done to assess the spatial distribution of landslide
potentiality.

Keywords Slope instability Geomorphic parameters Friction angle (φ) soil

texture Cohesion (c) porosity and water holding capacity Volumetric expan-

sion Wet soil density Remote sensing and GIS
2.1 Introduction
As the form of a slope is the end product of past geological processes, the
morphological history of the slope must also be understood. Each and every
spatial segment of the earth surface possesses some physiographic aspects and the
analysis of all the aspects enables us to predict an interrelationship between
physical and cultural phenomena and as a whole. The study area, Shivkhola
Watershed comprises a number of diversified physical aspects and there is a great

58 2 Geo-spatial Variability of Physiographic Parameters …
diversity of forms and the complexity of interrelationships. The practical rele-

vance of landslide can be recognized only by the systematic and thorough study
of geomorphic attributes such as relief, geology, and soil. A detailed and inte-
grated investigation of the geological structure of the area, the petrographical and
physical properties of the rocks and the local hydro-geological conditions with
changing slope of the Shivkhola watershed will help to prepare the corrective and
preventive measures in a reasonable scheme. The Shivkhola Watershed provides a
wide range of elevation from 2,040 m in north-west to 300 m in south-east. The
middle section of both north facing slope and south facing slope is attributed with
sudden and abrupt steepness. Geologically, the study area owns seven major
lithological units with varying degree of resistance and intensity of landslide
phenomena. The development of drainage network in the Shivkhola watershed is
the outcome of elevation and slope which are the ubiquitous elements of land-
scape, the structure and tectonic history of the area, and the existing humid
climate. There is continuous branching and headward extension as well as the
sharpening of the interfluves area caused by the present drainage network within
the basin. The soil up to the depth of 1.5 m is heavily disintegrated and
decomposed in the study area. The existence of finer to large size soil-rock
composition has aggravated the problem of soil erosion and soil slip in the
Shivkhola Watershed. Besides the size of the soil particles, the mineralogical
composition of the soil changes all the physical and chemical properties within
the soil. The amount of sand, silt and clay; porosity, water holding capacity and
bulk density; cohesion; and saturated depth of the soil are some of the significant
properties which continuously changing the actual nature of the soil-rock prop-
erties of the hill slope causing slope failure.
The study of various geomorphic attributes and their interrelations generally
offers a concrete accounts and evidences on the morphological characteristics and
landscape evolution. The analysis of relief, lithology, dissection, ruggedness,
topographic index, slope, aspect and curvature in relation to slope instability will
contribute an impression about the degree of importance and dimension of indi-
vidual geomorphic attribute. Landslide potentiality was estimated incorporating
landslide inventory map (Fig. 2.1) for all the geomorphic attributes by determining
class/ranges wise Landslide Potentiality Index Value (LPIV) of each factor by
means of a ratio (Eq. 2.1) between the number of cells/pixels disturbed by land-
slides and the total number of cells/pixels for that specific class. More details of
these procedures were obtained in other studies (Vieira et al. 1998; Guimaraes et al.
1999). Topographic Index (TI) Value was calculated in consultation with slope and
upslope contributing area. The effectiveness of all these parameters were being
influenced by hydrologic conditions and other atmospheric processes. Anderson
and Burt (1978) presented the role of topography in controlling through flow
generation and related landslips. GIS tools were applied for the identification of
topographic settings conducive to landslide occurrences by Gao (1993). Various
geomorphic models were being introduced for understanding slope instability by
Beven and Kirkby (1979), Ahnert (1987), Montgomery et al. (1994), and Dietrich
et al. (1998). Cruz (2000) studied in detail the role of geomorphic processes on
2.1 Introduction 59
Fig. 2.1 Landslide inventory map
mass movements. Zhou et al. (2002) investigated the spatial relationship between
landslides and causative factors.
LPIV ¼ ðF2 F1Þ 100 ð2:1Þ
where,
F1 number of pixels/cells or grid without landslide
F2 number of pixels/cells or grid with landslide.
2.2 Assessment of Geomorphic Attributes
The success of Geomorphologic research depends mostly on field study. The

emphasis has been laid on the field work in the present study, wherever possible.
All the problems relating to the study could not be solved because of the great
complexity of geomorphologic processes in nature. But the observation of geo-
morphic features such as the shape of the scarps, roughly the degree of slopes, the
development of drainage pattern, shape of the valley and ridges, nature of exposed
surface rock layers, were done by simple instruments, maps and imageries
(clinometer, abney’s level, GPS, SOI Topographical Map and LISS III Imageries)
and their photographs during the field work helped a lot for studying and analyzing
the topography and for preparing the various morphometric maps of the Shivkhola
Watershed.
2.2.1 Analysis of Contour Orientation
Shiv-khola watershed spreads over a region with varied relief character. Being the
hilly drainage system, it possesses a saucer shape having steeper slope almost
everywhere except at few localized section at the mid-central and lower part. The
contour map prepared in consultation with SOI Topographical Map (No78B/5) and
Satellite Imagery (LISS-III, 2003) shows a wide range (1,740 m) of altitude
between 2,040 and 300 m. The central middle portion and extreme right part
(confluence with the mighty Mahanadi) show the gentle slope. The other parts
depict steeper slope through closer contour spacing. The orientation of contours is
depicted in the Fig. 2.2. The valleys and the spurs in between those set the
remarkable character of the study area. The long profile along the main river, the
Shiv-khola, shows a waxing (convex summit) slope at the source region followed
down slope by a steep section near Paglajhora area and then by a gentle middle and
lower segment. The Hill Cart Road has to cross twice this steep section to harness
the facility along-contour extension.
This steeper (almost vertical) slope is more than 500 m high and extends over
2 km length along Hill Cart Road. After this steeper part the Shivakhola river
develops the gentle slope along the river and develops cut-and-fill terrace being
incapable of clearing all the dislodged material sliding down from steeper upslope.
After emerging at this stage river becomes sluggish and shows some spectacular
meandering on erosional slope which is quite unusual for a hill stream of such
shorter length. The wider valley floor (former) is mainly occupied by the Shivitar
Tea Estate and mostly covered with Tea Plants. The newly cut valley under this
former one is rather steeper and prone to slide along its entire length. After crossing
this, the river has to pass through the constricted valley between Tindharia Tea
Garden and Shivitar Tea Garden. Some flat land is again developed at the junction
with the mighty Mahanadi due to combined erosional effects of the Mahanadi and
the Shiv-khola. The upper catchment of the basin towards the water divide shows
gradual steepness and more dissection and efficiency in drainage. More tributaries
of lower orders are developed at the upper catchment due to swift drainage. Greater
rate of release of kinetic energy from potential is responsible for the dissection of
slope at the upper section. The land uses mainly the Hill Cart Road and North
Eastern Frontier Rail line (0.61 m gauge) are absolutely guided by the relief.
Almost at every part these run parallel to the contour running along the basin
boundary. Surprisingly Transport links crosses the main stream through the steepest
section of the basin at Upper and Lower Paglajhora section.
2.2 Assessment of Geomorphic Attributes 61
Fig. 2.2 Contour orientation of the Shivkhola watershed
2.2.2 Digital Elevation Model (DEM)
The prepared contour map at 20 m interval from the topographical map of 1987 at
the scale of 1:50000 was used for generating the Digital Elevation Model (DEM) by
using ARC GIS Software which helped to have an appreciation of the nature of
relief and the distribution of slope over the basin area (Fig. 2.3). The relief is mainly
guided by the position of the drainage lines and amount of dissection. The graded
state, characterized by gentle slope (graded slope) which is established at the
confluence at the flag end gradually creeps upward to the upper catchment
following the channels. The valley-side slopes are steeper enough showing absolute
Fig. 2.3 Digital elevation model (DEM)
instability. The steeper part of the slope gets adjustment through slides and thus
retreats backward setting a gentle and graded slope at its base which is sufficiently
steeper to release kinetic energy sufficient only to transport already eroded mate-
rials. Thus the graded condition gradually moves upslope from the confluence
point. The deposition at the middle part and erosion at the upper reach and again
erosion at the lower section through further down cutting execute the concept of
complex response following Chorley et al. (1985).
Table 2.1 Hypsometric analysis

Elevation (m) (h) Area above the elevation (km2) (a) a/A Cumulative a/A h/H
300 1.23 0.06 0.06 0.16
400 5.72 0.26 0.32 0.22
600 5.06 0.23 0.55 0.33
800 3.79 0.18 0.73 0.44
1,000 2.34 0.11 0.84 0.55
1,200 1.24 0.06 0.90 0.66
1,400 1.26 0.05 0.95 0.77
1,600 0.81 0.03 0.98 0.88
1,800 0.58 0.02 1.00 1.00
Total area 22.05 km2 , H maximum elevationfriction angle ranges betweenA total area
2.2.3 Hypsometric Analysis
In the present study to develop hypsometric curve the ratio between relative height
(h/H) and relative area (a/A) were plotted on the ordinate and abscissa respectively to
recognize the stages of cycle of erosion in the Shivkhola Watershed after Strahler
(1952). The hypsometric analysis of the said basin shows the distribution of area
against respective elevation zones (Table 2.1). The study shows that maximum basin
area lies within 400–600 m altitude and gradually the area under each successive zone
of 100 m is decreasing at a diminishing rate. The area below the curve is waiting for
erosion and so the ratio of the two represents the phase of erosion and thus the said
basin is approaching to mature stage where maximum of the areas are in sloppy
condition and no such flat surface is seen. The ridges are steep enough favouring easy
drainage and the absence of flat land is the basic hindrance in the land use and thus the
land resource cannot be used with full utility. Any attempt for reduction of slope at one
place for effective use increase the slope at other and thus introduces instability of
slope which sets instability in the other and the total system becomes instable.
Here, hypsometric integral (H.I.) has been accepted as an important morpho-
metric indicator of the stage of basin development which is the percentage of the
total volume of the basin area below the curve and thus it reveals the volume of area
unconsumed by the dynamic wheels of erosion (Strahler 1952). The derived
hypsometric integral value is 0.46 which depict the area is still passing through the
late youthful stage of landform development with moderate to high drainage density
which may invite havoc slope failure.
2.2.4 Lithological Composition and Landslide Potentiality
Mallet (1874) stated that in the Darjeeling territory the “Gondwana” rocks are
overlain by the metamorphic rocks, which are termed as “Darjeeling” (mainly mica-
gneisses and schists) and “Daling” (mainly slates and phyllites). The “Darjeeling”
structurally overlies the “Daling” and Mallet observed that there is a gradual pas-
sage from one metamorphic unit into another. Daling consists of garnetiferous
biotite schist, biotite schist and kyanite-sillimanite bearing garnetiferous gneiss,
indicating higher grade of metamorphism. Garnetiferous biotite-calc-granulite, calc-
gneiss, quartzite and muscovite-sillimanite-quartz-schist occur as inclusions in the
gneisses and represent the original calcareous, siliceous or other sediments.
According to Mallet, “the gneiss should be the older rock and their inverted on to
the slates and their turn on to the Damuda or that the boundaries should be faulted
one or finally that the relation of these formations to each other should resembles
those of the Tertiaries to Damuda”. The problem is that the-“Darjeeling”-“Daling”-
“Gondwana”-contacts are structural stratigraphic discontinuities.
Roy and Sensharma (1967) suggested that “Darjiling” and “Daling” strati-
graphically constitute one continuous unit called “Senchal Series”. According to
Ray the boundary between Gondwana and overlying Daling near Tindharia cannot
be drawn with certainty as there is a possibility that the two constitute one con-
tinuous formation in different grades of metamorphism with somewhat different
lithology. Folding has not been commonly recorded in “Gondwana” except in a few
cases in sandstone and shale. These folds are developed on bedding planes. A few
discontinuities fractures occur parallel to the axial planes of the folds, no appre-
ciable recrystalisation or neomineralisation has been observed along this plane.
Successive appearance of chlorite, biotite, garnet and kyanite which apparently
suggest the occurrence of progressive zones of regional metamorphism of Barro-
vian type can be recorded from south to north.
Gondwana sandstones and shales are mostly unaltered. Constructive metamor-
phism has already been observed except for some recrystalisation of small sericitic
micas and biotites; these occur along bedding (Laahiri 1973). The lithological map
of the concerned study area was collected from Geological Survey of India (GSI),
Kolkata (Eastern Region) and then necessary modifications were being incorporated
after thorough field investigation. Final lithological map was prepared with seven
rock groups and transformed into raster value domain on ARC GIS platform. Each
and every lithological group responds differently whenever it exposes to atmo-
spheric processes and also produces varying magnitude of landslide susceptibility/
Landslide Potentiality. Darjiling Gneiss (A), Reyang formation (E) and Swialik (G)
associated with highly foliated gneiss, mica-schists and occasional bands of flaggy
quartzites and granulitic rocks, slates phyllites with occasional quartzite, quartz-
schists and greywake schists, soft grayish sandstone, mudstone and shales and
conglomerate along with thin bands of marly shales and lignite cover more than
60 % area of the Shivkhola watershed. Chungtung Formation (B) with calc-gran-
ulitie, marble, quartz-granulite and mica-schist; Lingtse Granite (C) with foliated
granite or mylonitised granite with several close space sub-parallel thrust; Gorub-
athan Formation (D) with low grade phyllite and silvery mica-chlorite-schist, grey
sericite, and Damuda Formation (F) with coarse grained hard sandstones, quartzites,
carbonaceous shales and slates, thin seams of crushed and powdery coal share
almost same area (around 8 %) each (Fig. 2.4).
Fig. 2.4 Lithological map of the Shivkhola watershed
Some assessments have been presented for the study area after studying the
lithology in detail (Laahiri 1973):
1. The structural-cum stratigraphic succession can be observed in a traverse across
Tindharia-Kurseong region.
2. The “Daling”-“Darjiling” metamorphic boundary is gradational and correspond
to the passage between silvery mica-schist and garnetiferous mica-schist.
3. The metamorphites (Daling and Darjiling) disclose a well defined sequence of
deformation and recrystalisation.
4. The “Gondwana” rocks on the other hand, are mostly unaltered and these show
little metamorphism except some local cataclasis.
Landslide occurrences phenomena as well as landslide potentiality index value
(Table 2.2) are very high for the lithological composition of gneiss, mica-schist and
granulitic rocks (73.07), mylonitised granite with sub-parallel thrust (61.54),
phyllite, silvery-mica-chlorite-schist, grey sericite (69.23), and slate phyllite with
quartzite, quartz-schist and greywake schist (65.52) at Lower Paglajhora, 14 Miles
Bustee, Gayabari, Jogamaya, Tindharia, Shiviter and Mahanadi due to following
reasons:
Table 2.2 Lithological composition and landslide potentiality index value (LPIV)
Lithological Number Number Number of Number of Landslide Landslide
composition of cells of cells landslide landslide occurrence potential
(F1) (%) occurrence occurrences ratio index
(0.25 cells (F2) cell (%) (LPI) = F2/
km2) F1 × 100
Gneiss, mica- 26 18.57 19 22.09 0.73 73.07
schist and granu-
litic rocks (A)
Calc-granulie, 16 11.42 9 10.47 0.56 56.25
marble, quartz-
granulite and mica
schist (B)
Mylonitised gran- 13 9.28 8 9.30 0.61 61.54
ite with sub-paral-
lel thrust (C)
Phyllite, silvery- 13 9.28 9 10.47 0.69 69.23
mica-chlorite-
schist, grey seri-
cite (D)
Slate phyllite with 29 20.71 19 22.09 0.65 65.52
quartzite, quartz-
schist and grey-
wake schist (E)
Sandstone, quartz- 16 11.42 7 8.14 0.43 43.75
ites, shales, thin
seams of crushed
coal (F)
Soft sandstone, 27 19.29 15 17.44 0.55 55.56
mudstone, shales,
conglomerate and
marly shales and
lignite (G)
• Seepage through heavily disintegrated and decomposed materials and formation

of clay minerals, which induces slope instability.
• Rocks are traversed by quartz and quartzo-felspathic veins and the rocks are
often highly metamorphosed and jointed.
• Recrystallisation and cataclastic deformation have destroyed the clastic texture
with intense granulation along narrow zones of fracture.
• The apexes of the sliding zones are predominated with good amount of organic
matter which encourages high water holding capacity and volume expansion.
• The apexes of the sliding zones are deforested and are susceptible to both sheet
and gully erosion.
• Both Damuda and Swialik provide intensively deformed sandstone which
destroys the clastic texture and promotes slope instability.
Krishnaswamy (1982), Lahiri and Gangopadhyay (1974) studied structure and
stratigraphic pattern of rocks and their relation to landslide phenomena with par-
ticular reference to Himalayan region. Nautiyal (1951, 1966) presented some
Table 2.3 Rock properties

Rock types Strength (MN/m2) Bulk density (Mg/m3) Porosity (%)
Compressive Tensile Shear
Granite 100–250 7–25 14–40 2.6–2.9 0.15–1.5
Gneiss 50–200 5–20 – 2.8–3.0 0.5–1.5
Slate 100–200 7–20 15–30 2.6–2.7 0.1–0.5
Sandstone 20–170 4–25 8–40 2.0–2.6 5–25
Shale 100–200 7–20 15–30 2.0–2.4 10–30
Source Attewell and Farmer (1976)
geological report on the hill slope stability in and around Darjiling Himalaya.
Attewell and Farmer (1976) studied compressive, tensile and shear strength of
various rocks and concluded that slate and shale have low range of shear strength
(Table 2.3). Sandstone is characterized by high range of shearing strength. The bulk
density (Mg/m3) in most of the rocks varies from 2.00 to 2.9. In the Shivkhola
Watershed, Damuda and Swialik rock groups depict high range of shearing strength
where porosity ranges from 5 to 25 %.
2.2.5 Analysis of Lineaments and Landslide Potentiality
The lineaments exhibit the zone of weakness surface providing some linear to
curvilinear features such as fracture, joint, fault etc. in the geological structure
(Fig. 2.5). To extract lineaments of the Shiv-khola watershed PCI-GEOMATICA
Software of GIS was used and in the extraction process 3 bands of wavelength were
taken into account such as Near Infrared (Band-I; 0.7–1.3 µm), Red (Band-II;
0.6–0.7 µm) and Green (Band-III; 0.5–0.6 µm). The algorithm used to prepare
lineament map is ‘lineament extraction’. The lineaments at the places of Lower
Paglajhora, Gayabari lower, Tindharia, 14 Miles Bustee, Shiviter Lower slope,
Sepoydhura, and Norbong T.E. are closely spaced. The lineaments are absent at
extreme north, north-east and eastern marginal part of the Shivkhola Watershed.
The LPIV of each lineament class exhibits that the greater the distance from the
lineaments lesser is the probability of landslide phenomena. The distance of 125 m
from lineaments is dominated by high percentage of landslide affected pixels as
well as high landslide potentiality index value of more than 17. The distance of
more than 600 m from lineaments is less affected by landslide where LPIV ranges
from 0 to 10 (Table 2.4).
Fig. 2.5 Lineament distribution map of the Shivkhola watershed (distance from lineament in
meter)
Table 2.4 Lineament and landslide potentiality index

Class (dist. from Number of Number of landslide Landslide potentiality index
lineament in m) pixels (F1) affected pixels (F2) (LPI) = (F2/F1 × 100)
0.00–57.42 3,381 624 18.46
57.42–126.32 3,786 668 17.64
126.32–229.68 3,695 451 11.37
229.68–356.00 3,252 522 16.05
356.00–528.26 4,799 444 9.25
528.26–723.45 4,141 286 6.91
723.45–964.65 3,887 221 5.67
964.65–1251.75 3,921 120 3.06
1251.75–1642.20 1,419 37 2.61
1642.20–2925.40 850 0 0
2.2.6 Slope Angle and Landslide Potentiality
Firstly, the contour map at 20 m interval was prepared and digitized from the SOI
Topo-sheet (1987, 78B/5) at the scale of 1:50000 and was subsequently used for
generating Digital Elevation Model (DEM) using ARC GIS Software. Then slope
gradient, slope curvature and slope aspect maps were derived from DEM with
25 m grid cell size and classification was made following the earlier works of
Anbalangan (1992) and Dhakal et al. (2000). The slope is maximum near
Paglajhora area. Hill Cart Road and North Eastern Frontier Rail line (0.61 m gauge)
cross the entire river system twice through this steeper and unstable zone. The slope
is least at the central part where the river develops a cut and fill terrace. The slope
registers one of the minimum at the crest of water divide and towards the bottom
right of the basin. Slope gradient of the watershed varies from very gentle gradient
(around 10°) in the mid central and mid-lower part to that of high (more than 60°),
towards the marginal part/water divide. Most of the landslide phenomena were
found in the area of above 35° slope gradient.
In the Shivkhola Watershed high to very high landslide hazard risk is being
found at the place with slope angle between 24° and 40°. Low to moderate level of
risk is observed in the area pertaining to the slope angle of less than 24°. Small parts
Fig. 2.6 Spatial distribution of slope angle

Table 2.5 Slope angle and landslide potentiality

Classes Number of % of Landslide % of Landslide
(slope in pixels (F1) (Npix(Ni)) pixels (F2) (Npix(Si)) potentiality
degree) (Npix(Ni)) (Npix(Si)) index
0–7.17 3,353 10.12 190 5.63 5.67
7.17–14.34 3,238 9.77 202 5.99 6.24
14.34–19.92 3,587 10.83 211 6.26 5.88
19.92–24.97 2,445 7.38 201 5.96 8.22
24.97–29.75 3,555 10.73 311 9.22 8.75
29.75–34.53 2,776 8.38 329 9.75 11.85
34.53–39.57 3,854 11.63 413 12.24 10.72
39.57–45.95 3,276 9.89 427 12.65 13.03
45.95–54.71 3,557 10.74 543 16.09 15.27
54.71–67.73 3,490 10.53 646 19.15 18.51
of the Shivkhola Watershed with slope angle of more than 40° is registered with
high to very high landslide hazard risk (Fig. 2.6). It could be demonstrated that
50 % area of the Shivkhola basin is dominated by the high intensity of landslide
with 24° and 40°. The landslide potentiality index value increases with increasing
slope angle. It is inferred that there is a positive relationship between slope angle
and landsliding in the corresponding study area (Table 2.5).
2.2.7 Slope Aspect and Landslide Potentiality
Slope aspect of the Shivkhola Watershed was derived from the developed digital
elevation model (DEM) with the help of ARC GIS Software. In general, the
watershed shows the south facing slope with some local variation due to the
location of ridges, spurs and valleys. The concentrated local erosion and consequent
development of valleys, landslide scars and other topographic depressions are
responsible for the spatial distribution of slope aspect. This slope aspect is helpful
for the identification of slope segments which are required for the analysis of
potentiality of slope failure. The Fig. 2.6 shows the slope direction and most of the
places show the southward slope with some places of northward slope. The varied
direction of flow based on local slope and orientation of ridges, spurs and valleys
shows the pattern of concentration of surface water and the places of potential
surplus region. The availability of water increases with the increase in the distance
from water divide. Thus at the lower reach, of every stream, the availability
becomes more than the upper portion.
North, south, east, north east and south east facing slope are registered with the
landslide potentiality index value of 17.14, 13.10, 15.53, 11.67 and 12.54
Table 2.6 Slope aspect and landslide potentiality index (LPIV)

Aspect Number of Number of landslide Landslide potentiality index
pixels (F1) affected pixels (F2) (LPI) = (F2/F1 × 100)
Flat 784 24 3.06
North 3,879 665 17.14
North east 3,797 443 11.67
East 4,346 675 15.53
South east 6,290 789 12.54
South 4,556 597 13.10
South west 3,332 35 1.05
West 2,870 69 2.40
North west 3,277 76 2.32
respectively (Table 2.6). South west, west, northwest and middle section is attributed
as minimum number of landslide occurrences phenomena as well as landslide
potentiality index (LPIV). According to the number of pixels affected by slope failure
south east facing slope rank first which is followed by east, north, south, north east,
north west, west, and south west. Upper halve of the watershed is dominated by south
east and south ward facing slope and lower halve is dominated by east and north ward
facing slope with LPIV >11 (Table 2.6). Such landslide prone facets are closely
associated with maximum slope and relief which is found at upper and lower
Paglajhora, Shiviter T.E., Gayabari Lower slope and Tindharia (Fig. 2.7).
2.2.8 The Surface Curvature and Landslide Potentiality
The retention of moisture in the soil depends mainly on the convexity (positive
curvature) or concavity (negative curvature) of surface. Slope curvature plays a
significant role in changing landform character (Gilbert 1909). The concave surface
holds moisture for long where the convex surface drains moisture immediately. The
curvature value represents the morphology of the topography. A positive curvature
represents the surface is upwardly convex at the pixel and a negative curvature
value represents the surface is upwardly concave at that pixel. A value of 0 indi-
cates that the surface is flat. The more positive and negative curvature value
indicates the surface is more susceptible to landslide occurrences. The reason for
this is that following heavy rainfall, a concave slope contains more water for a
longer period and saturates the soil perfectly and also reduces the cohesiveness
within soil. On the other hand a convex slope is generally more exposed to frequent
expansion and contraction processes which lead to the disintegration and decom-
position of rocks. The presence of decomposed or loosened materials along the
convex slope allows water particles to move downward, which is the triggering
mechanism of landslide phenomena.
Fig. 2.7 Slope aspect map
The slope curvature map was also extracted from Digital Elevation Model
(DEM) on ARC GIS. The map shows (Fig. 2.8) very high negative and positive
surface curvature along Hill Cart Road which passes through the steep escarpment
slope of the Shivkhola watershed. Paglajhora, Gayabari, Tindharia, and Shiviter are
dominated by the moderate to high levels of positive and negative surface curvature
with moderate levels of slope surface dissection (Fig. 2.8). Extreme middle and
eastern marginal part register minimum curvature value. The study depicts that
landslide Potentiality Index Value increases with increasing positive and negative
curvature values (Table 2.7). The positive curvature is common and that indicates
the tendency of immediate drainage of surface water causing ready washing and so
is detrimental to the stability of both soil and slope.
The study on slope angle states that there is rapid flowage of running water along
the steep slopes and formation of first order channel with active headward and
downward erosion. Such conditions introduce steep valley side slopes and devel-
opment of drainage network. The continuous development of drainage network
yield huge amount of surface water and their confluence at a particular location
generate more seepage and reduce cohesive strength of the soil and induce slope
Fig. 2.8 Slope curvature map
Table 2.7 Curvature and landslide potentiality index (LPIV)

Classes (value of Number of Number of landslide Landslide potentiality index
slope curvature) pixels (F1) affected pixels (F2) (LPI) = (F2/F1 × 100)
−25.87 to −11.41 995 221 22.21
−11.41 to −5.73 785 210 26.75
−5.73 to −2.33 2,111 486 23.02
−2.33 to −0.63 2,431 374 15.39
−0.63 to 0.50 1,0045 388 3.84
0.50–2.49 6,302 268 4.25
2.49–7.31 5,438 464 8.53
7.31–14.69 3,343 475 14.21
14.69–24.33 895 222 24.80
24.33–46.45 786 265 33.71
instability. Darjiling Himalaya is situated in the southern flank of Himalayan

Mountain Range, which receive frequent orographic rainfall during rainy period.
Most of the destructive landslides occur due to high intensity rainfall and it is
observed that the south facing slopes of Darjiling Himalaya are dominated by
higher amount of rainfall and large number of landslides. In Shivkhola Watershed,
concentration of drainage is one of the major causes of landslide. Slope charac-
terized by high positive (slope convexity) and high negative (slope concavity)
curvature value allows more concentration of water and drainage which on the other
hand decreases mineralogical bonding of soil and invite slope instability. The major
landslide prone areas in the Shivkhola Watershed are registered by such feature.
The occurrences of landslide phenomena are also aggravated because of the exis-
tence of high intensity lineaments at slope convex and slope concave section.
Lineaments permit seepage that helps to entrain the particles from vertical section
of the soil and introduce slope instability. It is to be concluded that the presence of
fragile lithological composition (>60 % area) and the continuous development of
drainage network over it, has made the Shivkhola Watershed more vulnerable to
landslides.
2.3 The Soil
Officially the soil of the Darjiling Himalaya is divided into two main groups (a) the
brown forest soil and (b) the terai soil. The first group of soil is found in the moun-
tainous region whereas the second group is found in the lower elevated zone. The
shivkhola watershed is falling under the brown forest soil as it covers hilly area of the
Kurseong Division. The inherent fertility of this brown forest soil is very high and has
free drainage through their profiles and also have very rich in well-distributed humus.
The percentage of organic matter in the top horizon increases with altitude but
gradually decreases down the profiles. The principal source of the organic matter
present in the top layer of the soil are leaves, stems, branches, roots, barks, fruits,
seeds, animals and micro-organism. The maturity of this forest soil depends on the
decomposition of the organic matter into humus. Accordingly, the decomposition is
slower at higher altitude due to climatic factors, but the decomposition is quicker at
lower altitude. Decomposition at higher altitude generally occurs only because of the
presence of fungi while at lower elevation it happens by climate, bacteria and animals.
In the Shivkhola watershed the major types of humus dominated soil found in the
forest soils are ‘Mull’ and ‘Mor’. The former is a porous, loose, crumble and friable
mass that develops under deciduous species. The ‘Mor’ is dominated by fungi and
mosses and associated to coniferous forest and also display a high degree of saturation
of A horizon and little accumulation of sesquioxide in the B horizon (Darjiling District
Gazetteers by LSS O’Malley 1999) (Table 2.8). Sarkar (1987a, b) introduced pedo-
geomorphic parameters and its impact on soil loss and landslip.
2.3 The Soil 75
Table 2.8 Percentage of organic matter, nitrogen, and pH of the AO horizon under varying
altitudinal extent
Altitude Mean Drainage Vegetation pH % of % of N2
(ft) annual condition type content organic in AO
rainfall matter horizon
(in.) (AO layer)
6,000 120 Well- Virgin high 7–7.5 22.32 1.14
drained forest
5,000 150 Well- Virgin high 6.5 19.23 0.97
drained forest
4,000 120 Well- Virgin high 6.5 18.76 0.77
drained forest
3,000 150 Well- Virgin high 6 24.9 1.05
drained forest
2,000 150 Well- Virgin high 6.5 10.29 0.35
drained forest
1,000 150 Well- Virgin high 6.5 7.05 0.31
drained forest
500 150 Not well- Virgin high 7–7.5 7.57 0.36
drained forest
Source Darjiling District Gazetteers, O’ Malley (1999)
2.3.1 Textural Characteristics of the Soil
Soil texture is defined by size distribution or mass fractions of primary particles in soil
(individual grains and particles). Primary mineral particles formed through physical
and chemical weathering of parent material and refractory organic substances make up
the solid phase. Particle size distribution and shape are the most important character-
istics affecting: pore geometry, total pore volume (porosity), and pore size distribution.
The presence of sand, silt and clay within a specific amount of soil are con-
sidered as the soil texture. On the basis of the amount of sand, silt and clay present
in the soil the porosity, water holding capacity of the soil varies and consequently
influence the cohesive nature of the soil. The higher amount of sand present in soil
is characterized by the higher rate of permeability and porosity and decreasing
tendency of the soil cohesiveness. The finer particles present in soil sometime
becomes mobile in presence of seepage water and disturbs the soil cohesiveness.
The soil becomes more compact in presence of higher amount of clay particles. All
these three attributes of soil texture generally control the seepage, permeability,
porosity, cohesiveness, water holding capacity etc. Hence, the stability condition of
the slope is governed by the existence of the sand, silt and clay. United State
Department of Agriculture (USDA) propounded a soil textural classes which states
that particles with <0.002 mm diameter in size is recognized as clay, the size class
≥0.002 and <0.05 mm as silt and Sizes between ≥0.05 and <2 mm and sand and
sizes of more ≥2 mm in diameter is treated as gravel. The present work is dealt with
USDA System and textural properties are determined applying sieving method.
Fig. 2.9 Spatial distribution of clay (%)
Lower part of study area is experiencing very minimum percentage of sand

particles whereas extreme south western part, small pockets in the south are
characterized by higher percentage of sand particles. Mid-section and most of the
marginal area are dominated by moderate to low percentage of sands (Table A.1,
Appendix A). Tindharia (26.5 km from Siliguri), Lower Paglajhora, Mahanadi and
Nurbong T.E. represents medium percentage of sand particles (Fig. 2.10) and this
situation provide numerous micro pore spaces within the sub-soil and consequently
increase the water holding capacity and reduce cohesive strength of the soil. In the
south eastern part mainly at Tindharia and Sepoydhura, Mahanadi, Upper Pag-
lajhora and Shiviter T.E. percentage of clay particles ranges from moderate to low
and so the bonding capacity is low. This leads to soil erosion and slope instability
condition. Small pocket in Giddapahar and Gayabari and lower eastern part register
with high percentage of clay particles (Fig. 2.9).
Giddapahar, Shiviter and Upper Paglajhora is experiencing fine sand, medium
sand, coarse sand and very coarse sand ranging between 10 and 30 % (Appendix
A). The percentage of coarse silt and fine silt are very low (below 5 %) and
percentage of very fine sand ranges between 5 and 10 %. The study on the
distribution of silt particle in the Shivkhola watershed depicts that yet the
2.3 The Soil 77
Fig. 2.10 Spatial distribution of sand (%)
percentage of silt is very low in the sub-surface soil, the eastern portions and few
other locations at north western part of the basin shows high percentage of silt
(Fig. 2.11). Middle section registers moderate to lower percentage of silt particles
and this site show moderate level of sands which helps easy percolation and
saturation of the soil and promote active soil loss at Tindharia, Gayabari, Lower
Paglajhora and Shiviter (Fig. 2.10). The most landslide prone section of the
Shivkhola watershed are registered with moderate to high percentage of granule
(Fig. 2.12). Such high percentage of granule in the sub-surface part of the
mountain slope helps to seepage and percolation of water and increases slope
materials saturation level and promote slope failure.
2.3.2 Saturated Soil Depth
Repeated and continued field studies for a period of 7 years were made for the
proper cognition of the processes and their interaction. The depth of the failure
surface was measured by holding a measuring tape at both the margins of scar and
Fig. 2.11 Spatial distribution of silt (%)
the other tape was allowed to hang, the reading was then taken from the base of the
hanging tape. The total thickness of soil and that of saturated soil from 50 different
locations during monsoon were measured from slope cuttings. After estimating the
approximate depth of all known points, a soil depth map (z/D) was made using Arc
GIS tool (Fig. 2.13). The maximum depth (>1.75 m) of the saturated soil is found in
the middle section (lower segment of Gayabari T.E., lower Paglajhora, Lower
segment of Shiviter and Tindharia T.E.) and lower section (mainly Nurbong and
Sepoydhura T.E.) of the basin. At marginal parts of the basin basically on the both
sides of the Hill Cart Road from Chunabhati to Gayabari, upslope parts from
Paglajhora proper, Lezzipur T.E. and Shiviter upslope, the saturated soil depth is
less than 1.75 m (Table A.1, Appendix A).
In the present study area, Shivkhola watershed the upper section of the soil at
major landslide locations are predominated by large percentage sand particle where
the tensile strength in dry and wet situation is low (Table 2.9). This situation
promotes shallow landslips. But the area dominated by large percentage of clay and
2.3 The Soil 79
Fig. 2.12 Spatial distribution of granule (%)
silt show the higher plasticity and high to very high tensile strength and low
landslide potentiality. We can say that the major size fraction of the soil controls
various physical properties of soil such as tensile strength, volumetric expansion,
plastic condition, porosity, permeability and water retention capacity as well as the
rate of liquefaction of slope materials.
Soil samples from 12 locations of Paglajhora Sinking Zone, one of the major
landslide prone areas of Darjiling Himalaya, were being collected and tested in the
laboratory (GSI) to understand physical properties of soils related with slope
instability. The study envisages that the saturated soil density varies from 2.00 to
2.60 and dry soil density varies from 1.20 to 1.96 (Table 2.10). The variation of
saturated soil density is less which helps to liquefy the soils chemical properties and
make all the places more conducive to slope failure. Percentage of sand, silt and
clay are also distributed in at this place uniformly. The cohesion in the Paglajhora
Sinking zone is very less (<0.7) and friction angle ranges from 22° to 37°.
Fig. 2.13 Spatial distribution of soil depth (in meter)
2.3.3 Water Holding Capacity, Pore Space and Volume

Expansion of the Soil (in %)
The capacity to contain water particles by a certain volume of soil mass is called
Water Holding Capacity. The water holding capacity of soil depends on the pres-
ence of impermeable layer below the permeable layer, pore space and solid space of
the soil, sources of water supply and shape of the underlying structure, and grav-
itational force. Pore space in soil consists of that portion of the soil volume not
occupied by solids. The volume of a given soil occupied by air is designated by air
space, air volume, air space porosity, non-capillary porosity and gaseous phase
2.3 The Soil 81
Table 2.9 Physical properties of soil according to major size fraction of the particles
Property Gravel Sand Silt Clay
Volume change None None Slight Large
Tensile strength Low Lower than Higher than Very high
(dry condition) when wet when wet
Tensile strength Low Low Intermediate High
(wet condition)
Plasticity (wet) None Slight Intermediate Very high
Plasticity (dry) None None None None, partial
cementation
Porosity Very High High Very high
high
Permeability Very High Intermediate to Very low
high low
Water retention Very Low High Very high
low
Size of voids Large Intermediate Capillary Subcapillary
Source Dapples (1959), US Department of the Interior, Bureau of Reclamation (1963)
Table 2.10 Study on the geo-technical parameters of soil at Paglajhora sinking zone
Sample Soil density (gm/cc) Grain size Direct shear test result
no. analysis (%)
Wet Dry Saturated Sand Silt Clay Cohesion (c) Friction (φ)
1 1.94 1.62 2.12 85 11 4 0.11 30
2 2.18 1.96 2.39 83 12 5 0.22 19
3 2.01 1.52 2.54 90 3 3 0.41 24
4 2.08 1.84 2.27 87 11 2 0.02 35
5 1.98 1.65 2.46 80 14 6 0.04 41
6 1.82 1.45 2.18 79 17 4 0.01 37
7 1.76 1.42 2.38 75 20 5 0.7 13
8 1.99 1.30 2.09 81 16 3 0.23 32
9 1.70 1.36 2.25 73 22 5 0.16 33
10 1.83 1.47 2.06 80 19 1 0.21 30
11 1.66 1.20 2.02 78 15 7 0.06 29
12 1.97 1.36 2.58 82 13 5 0.21 22
Source Mandal and Maiti (2012)
whereas volume of a soil occupied with water is known as liquid phase or capillary
porosity. These values are not constant but vary with the soil physical conditions
and moisture content. The amount of pore space is determined by the arrangement
of solid particles. If they tend to lie close together, the total porosity is low. If they
are arranged in porous aggregates, as is often the case in medium textured soils high
Table 2.11 Depth wise distribution of water holding capacity, pore space and volumetric
expansion (%)
Location Parameters Depth of the soil (cm)
0–10 10–20 20–30 30–40 40–60 60–80 80–100
26.5 km from Water 83.42 86.50 89.73 89.59 79.75 54.70 40.63
Siliguri near holding
Tindharia (1) capacity (%)
Pore 43.51 47.29 59.40 52.30 30.75 29.21 21.65
spaces (%)
Volume 31.40 27.10 27.21 28.93 25.23 19.75 13.21
expansion (%)
26.5 km from Water 73.04 74.06 75.03 71.93 60.51 46.44 34.65
Tindharia (2) capacity (%)
Pore 45.02 46.51 50.31 46.42 34.73 27.53 24.39
spaces (%)
Volume 21.09 21.09 23.20 24.15 17.46 12.03 9.95
expansion (%)
26.5 km from Water 70.21 71.39 74.45 75.09 62.43 51.03 37.51
Tindharia capacity (%)
(sample-A-3) Pore 52.40 59.31 50.43 42.36 37.05 29.00 26.74
spaces (%)
Volume 25.75 23.01 23.04 19.52 15.34 14.31 9.73
expansion (%)
26.5 km from Water 35.73 31.43 26.73 29.72 16.07 13.73 10.93
Tindharia capacity (%)
(sample-A-4) Pore 76.72 75.92 71.39 53.03 50.92 56.43 49.02
spaces (%)
Volume 14.71 14.92 13.75 9.78 7.02 5.03 4.01
expansion (%)
At Shiviter tea Water 45.45 41.25 35.50 32.55 27.30 22.53 10.93
estate holding
capacity (%)
Pore 69.54 63.75 60.50 49.85 34.75 26.65 22.55
spaces (%)
Volume 34.55 29.45 28.58 19.55 13.25 11.50 9.57
expansion (%)
At lower Pag- Water 29.22 25.54 22.45 18.50 12.34 9.08 5.45
lajhora holding
33.5 km from capacity (%)
Siliguri Pore 86.75 85.62 81.35 73.93 60.12 46.53 39.82
spaces (%)
Volume 21.11 19.21 15.42 12.22 9.75 6.45 5.40
expansion (%)
Source Basu and Sarkar (1985)
in organic matter, the pore space per unit volume will be high. In the present study a
comparison has been made between pore space, water holding capacity and volu-
metric expansion for few major landslide locational sites at different depth of soil
(Basu and Sarkar 1985) and their role in slope instability have also been studied
(Table 2.11).
2.3 The Soil 83
2.3.3.1 Volumetric Expansion
Soil Survey Standard Test Method was applied to measure volumetric expansion
of the soil. This method outlines the procedure for the determination of the free
swell of a disturbed soil on wetting. This test measures the free swelling of a
disturbed soil (ground and sieved finer than 0.425 mm) on wetting from air-dry to
saturation. The swell is calculated on a volumetric basis using a modification of
the Keen-Raczowski Test. The soil samples having at least 150 g of the material
passing a no 36 BS Sieve (0.425 mm) were collected from 50 different locations
of the Shivkhola watershed and prepared according to the procedure for testing
(Keen and Raczowski 1921). To calculate the percentage volumetric expansion
(VE), the following equation was applied.
W3 W2
VE ð%Þ ¼ 100 ð2:2Þ
W4 W1
where, W1 = weight of volume expansion box (g); W2 = weight of the weighing tin
(g); W3 = weight of wet expanded soil + tin (g); and W4 = wet of wet residual
soil + box (g).
2.3.3.2 Water Holding Capacity
Simply defined soil water holding capacity is the amount of water that a given soil
can hold for crop use. Soil texture and organic matter are the key components that
determine soil water holding capacity. In terms of soil texture, those made up of
smaller particle sizes, such as in the case of silt and clay, have larger surface area.
The larger the surface area the easier it is for the soil to hold onto water so it has a
higher water holding capacity. Sand in contrast has large particle sizes which results
in smaller surface area. The water holding capacity for sand is low. Soil organic
matter (SOM) is another factor that can help increase water holding capacity. Soil
organic matter has a natural magnetism to water. By using Keen-box and oven,
water holding capacity of soil was assessed applying the following equation.
total water in the wet soil

Water holding capacity of soil ¼ 100 ð2:3Þ
oven dry weight of the total soil
2.3.3.3 Porosity of the Soil
Porosity or pore space refers to the volume of soil voids that can be filled by water
or air. It is inversely related to bulk density. To calculate porosity, bulk density and
particle density were determined applying Keen-box method. The oven dry weight
of a unit volume of soil inclusive of pore spaces, called as bulk density. Generally
soil with low bulk density provides good physical condition. The bulk density of
sandy soil, loam soil, silt loam, and clay soil are 1.6, 1.4, 1.3 and 1.1 gm/cm3. On
the other hand, the weight per unit volume of the solid portion of the soil is called as
particle density. The particle density is the true density of soil. The particle density
of normal soils is 2.65 gm/cm3. The particle density of coarse sand, fine sand, silt
and clay are 2.655, 2.659, 2.798 and 2.837 gm/cm3 respectively. The particle
density of the soil increases with decreasing the size of particles. The porosity can
be calculated by using the following method.
bulk density
% solid space ¼ 100 ð2:4Þ
particle density
% of pore space ¼ 100 % % of solid space ð2:5Þ
In the present work, the total porosity in the soil has been derived using the
method applied by Brasher (1966).
Total porosity ð%Þ ¼ ðparticle density-bulk densityÞ particle density ð2:6Þ
Loose, porous soils have lower bulk densities and greater porosities than tightly
packed soils. Porosity varies depending on particle size and aggregation. It is
greater in clayey and organic soils than in sandy soils. A large number of small
particles in a volume of soil produce a large number of soil pores. Compaction
decreases porosity as bulk density increases. So, there exists a inverse relationship
between bulk density and porosity. If compaction increases bulk density from 1.3 to
1.5 g/cm3, porosity decreases from 50 to 43 %. Pores of all sizes and shapes
combine to make up the total porosity of a soil. Porosity, however, does not tell us
anything about the size of pores (http://www.agriinfo.in) (Table 2.12).
The collection and testing of soil samples from different locations and their
laboratory results shows that at 14 Miles Basti water holding capacity and volume
Table 2.12 Friction angle (φ) and landslide potentiality index (LPIV)
Classes Number of Number of landslide Landslide potentiality index
<18.00 3,500 626 17.88
18.00–19.486 3,547 523 14.74
19.486–20.971 3,266 417 12.77
20.971–22.456 3,864 413 10.69
22.456–23.940 2,786 329 10.81
23.940–25.425 3,545 311 8.77
25.425–26.910 2,435 201 8.25
26.910–28.395 3,597 211 5.86
28.395–29.880 3,248 202 6.21
29.880–32.848 3,343 190 5.68
2.3 The Soil 85
Table 2.13 Result of laboratory analysis (GSI Lab.) of collected soil samples from Tindharia
Sample Cohesion (C) Friction Dry soil Wet soil Water holding
angle (φ) density density capacity (%)
(gm/cm3) (gm/cm3)
I 0.64 22°30′ 2.20 2.43 35
II 0.25 19° 2.10 2.29 29
II 0.08 24° 1.99 2.21 28
expansion are low than the pore space. The pore space even at the depth of 60–80 cm
is prevailing above 35 %. The easy percolation of water through the subsurface soil
pore spaces creates the slope more vulnerable to soil erosion and slope instability. On
the other hand, near railway station and other places of Tindharia there is higher rate
of water holding capacity ranging between 40 and 55 % (Table 2.13) up to the depth
of 40 cm. where the percentage of pore space is experiencing 30–50 %. The water
holding capacity is reduced with decreasing the pore space beyond the depth of
60 cm. The volumetric expansion takes place in proportion to pore space and water
holding capacity and it is high (>10 %) above the depth of 40 cm and shows the
gradual decreasing tendency below the depth of 40 cm.
In the Shiviter T.E. the moderate to high percentage of pore space helps soil to
hold the moisture content and to expand moderately. The presence of sand, silt and
clay at moderate amount within the soil has helped to retain the moisture for a long
time without downward movement and has caused slope material more vulnerable
to soil loss. In the middle section of the watershed the water holding capacity as
well as the pore spaces is very high but there is a uniform rate of decreasing
tendency of both the parameters. The volumetric expansion is not as high as the
water holding capacity and pore spaces. The water holding capacity and pore space
is more than 50 % up to the depth of 40 cm when the volumetric expansion ranges
between 10 and 20 %. Beyond 40 cm water holding capacity and pore space are
experiencing 45–15 % where 0–10 % volumetric expansion takes place. The
decreasing of volume expansion, pore space and water holding capacity are in the
same rate at Upper Paglajhora and Tindharia T.E. The percentage of pore space is
very high up to the depth of 30 cm and that is why the water holding capacity is
reduced sharply and beyond 60 cm depth the pore space is being decreased
remarkably which causes easy saturation of surface soil and volumetric expansion
also takes place within the sub-surface soil. Such condition reduces the cohesion
and internal friction of the slope materials and makes the slope very much prone to
shallow soil slip. It is to be concluded from the present study that the average
percentage of sand and granules up to the depth of 100 cm are 35 and percentage of
silt is around 15 which promotes the soil layer to be saturated very easily and also
reduces the cohesion and shearing strength of the soil. Pore spaces, water holding
capacity and volumetric expansion decrease with increasing depth everywhere
which indicates that the near surface soil layer is saturated easily that reduced its
cohesion and shearing strength.
2.3.4 Friction Angle (φ) and Cohesion (c)
The shear strength of the soil is described as the function of normal stress on the
slip surface, cohesion, and angle of internal friction. The angle of internal friction
(φ) and cohesion are the two important physical properties of the soil which
determines angle of rupture, shearing strength, safety factor as well as stability
condition of the slope materials. A Mohr Stress Circle was developed to obtain
angle of internal friction and angle of rupture through confining pressure (σ3) and
compressive stress (σ1) with the centre on the horizontal axis; the centre of the
circle was obviously (σ1 + σ3)/2 and the radius was (σ1 – σ3)/2. The values of
confining pressure, σ3, and compressive stress, σ1 were plotted on horizontal axis
where stress difference is σ1 − σ3. On a plane parallel to the greatest principal stress
axis (2α = 0) the normal stress across the plane was σ3 and the shearing stress was
0. If the plane makes an angle of 45° with the greatest principal stress axis
(2α = 90), the shearing stress is at a maximum and the normal stress is (σ1 + σ3)/2. If
the plane makes an angle of 90° with the greatest principal stress axis (2σ = 180°),
the shearing stress is 0 and the normal stress is σ1.
Cohesion (C) is the attraction of particles to each other which is not directly
governed by a friction law but does provide a measure of strength of a material.
Thus sands do not exhibit cohesion, while soil which contains clay show cohesion.
It can be measured, as in soil mechanics, by the Mohr-Coulomb Equation.

r1 r3 tan2 45 þ u2
C ðcohesionÞ ¼ ð2:7Þ
2 þ tan 45 þ u2
The cohesion of the soil varies from place to place due to variation in the
presence of cementing materials which helps to combine soil particles tightly. This
is the bonding of the particles with each other. The natural bonding of the soil
particles are influenced and loosened by the presence of lubricating agent (water
and ice particles) and ensure the materials to collapse. The friction angle of sand-
stone under dry condition varies from 26° to 35° and under wet condition 25°–34°.
Fine-grained granite provides the friction angle of 31°–35° and 29°–31° for dry and
wet condition respectively. In case of gneiss, friction angle is 26°–29° for dry and
23°–26° for wet condition (Barton and Choubey 1977). The above mentioned
lithological compositions are available in the Shivkhola Watershed and laboratory
test of 50 soil samples shows that the friction angle ranges between 18° and 32°
(Appendix C).
At Lower Paglajhora and 14 Miles Bustee, the geo-technical properties of soil
are very much conducive to soil slip (Table 2.13). In the present study the friction
angle for the concerned material varies from 18° to 32° (Fig. 2.14). Around
Tindharia and Lower Paglajhora friction angle ranges between 18° and 22°. A steep
slope will decline by slope failure to an angle of repose slope to attain short term
stability. This concept leads to the concept of limiting or Threshold slope angle. It is
clearly observed from the figure that middle section, extreme lower most part,
2.3 The Soil 87
Fig. 2.14 Spatial distribution of friction angle (φ)
Sepoydhura and some northern marginal parts are registering the friction angle of
more than 24°. Large parts of Tindharia (Table 2.12), 14 Miles Bustee, and Lower
Paglajhora are facing the friction angle of less than 24° and also are considering as
the most vulnerable part of the Shivkhola Watershed. The friction angle at Shiviter
varies from 21° to 25° (Table 2.15). It is observed that there is an increase in the
volume of the landslide area and mass since 1986–2010.
Hill Cart Road is passing through the locations of Tindharia. 14 Miles Bustee
and Paglajaora where friction angle is quiet less. Analysis also reveals that more
than 50 % area of the Shivkhola watershed is below the friction angle of 24°. The
derived landslide potentiality index value reveals that the area having friction
angle of less than 20° are registered with high LPIV (Table 2.12). The low LPIV
is observed at the places where the friction angle is greater than 25°. The spatial
distribution of cohesion in the Shivkhola Watershed reveals that Paglajhora, 14
Miles Bustee (Table 2.14), Tindharia (Table 2.12), Shiviter (Table 2.14) are
characterized by very low cohesive strength of soil. The range of cohesion is
between 0.01 and 0.90 (Fig. 2.15). The places of Gayabari and its adjoining areas,
Sepoydhura, middle section of the watershed and extreme north-eastern part are
dominated by moderate to high cohesive strength of the soil, varying from 0.35 to
0.90. The estimated cohesion of all the 50 locations shows that the cohesion in
the Shivkhola watershed is very less, that is less than 0.90. The study indicates
that there is an inverse relationship between cohesion and LPIV. The region of
low cohesion of less than 0.29, showed the LPIV of more than 15 (Tables 2.15
and 2.16).
Table 2.14 Result of the laboratory test of soil samples from lower Paglajhora and 14 Miles
Bustee
Sample Lithology Cohesion Angle of Wet Dry Water
no. (kg/cm2) internal density density holding
friction (φ) (gm/cm3) (gm/cm3) capacity (%)
I Foliated 0.12 22°30′ 2.14 1.86 32
gneiss and
mica-schist
II Foliated 0.25 19° 2.08 1.79 28
gneiss and
mica-schist
III Foliated 0.18 17°30′ 2.02 1.74 23
gneiss and
mica-schist
Fig. 2.15 Spatial distribution of cohesion (c)
2.3.5 Wet Soil Density (γs)
Specific unit weight of water and unit weight of the soil were estimated by
examining the soil samples collected from 50 landslide locations during field
investigation from the GSI (Geological Survey of India, East Kolkata) laboratory.
The density of soil and water varies from place to place due to in situ
geo-hydrologic condition. The saturated soil density of rock was also consulted and
2.3 The Soil 89
Table 2.15 Volumetric expansion of the landslide with strength properties from Shiviter
Landslide Volume Volume Volume Slope Cohesion Friction Water
number (1986) (June, 2006) (August, angle (c) angle holding
(m3) (m3) 3
2010) (m ) (Ѳ) (°) (φ) (°) capacity
(%)
I 390 450 540 57 0.05 25 24
II 1,200 1,350 1,400 62 0.13 18 33
III 300 500 620 59 0.52 21 35
Source Laboratory test result
Table 2.16 Cohesion (c) and landslide potentiality index (LPIV)

Classes Number of No. of landslide Landslide potentiality index
<0.01 3,381 691 20.44
0.01–0.11 3,786 668 17.64
0.11–0.20 3,695 451 12.20
0.20–0.29 3,352 522 15.57
0.29–0.38 3,899 344 8.82
0.38–0.47 3,741 286 7.64
0.47–0.55 2,450 110 4.48
0.55–0.64 3,987 221 5.54
0.64–0.73 4,021 120 2.98
0.73–0.90 1,519 60 3.94
adopted from the field experiences done by Deoja (Mountain Risk Engineering
Handbook, 1991) and Specific Yield from Basic Ground-water Hydrology. Here,
the wet soil density was derived after Brasher (1966). Higher the density, greater is
the propensity of landslide occurrences. Wet soils help to liquify the mineralogical
properties present in the soil and reduce the cohesive strength.
The wet soil density plays a significant role in changing chemical properties,
cohesion (c) and friction angle (φ) of the soil particles. It is proved that greater is the
wet soil density, lesser is the cohesive strength of the soil because after wetting the
soil, it loses internal bonding capacity and becomes more susceptible to landslide.
In Shivkhola watershed, wet soil density is higher (>2.00 KN/m3) in the areas of
Tindharia, 14 Miles Bustee, Paglajhora, Sepoydhura, Gitingia and Upper Paglajh-
ora (Fig. 2.16). All these places are dominated by drainage concentration. Such
drainage concentration helped to increase the wet soil density and to reduce the
cohesive strength. The study on landslide potentiality shows that soils having high
wet soil density are very much prone to landslide phenomena (Table 2.17).
Fig. 2.16 Spatial distribution of wet soil density
Table 2.17 Wet soil density and landslide potentiality index (LPIV)
Classes Number of No. of landslide Landslide potentiality index
<1.79 3,303 180 5.45
1.79–1.88 3,138 212 6.76
1.88–1.96 3,687 201 5.45
1.96–2.03 2,495 211 8.45
2.03–2.11 3,505 321 9.15
2.11–2.19 2,726 319 11.70
2.19–2.27 3,954 423 10.69
2.27–2.35 3,176 417 13.13
2.35–2.43 3,657 553 15.12
2.43–2.57 3,490 636 18.22
2.4 Conclusion
The study area Shivkhola watershed possesses a wide range of elevation between
300 m in the south-east and 2,040 m in the north. A large part of the watershed is
lying between the altitude of 400 and 600 m. The steepness of the slope varies
significantly from place to place and its characteristics mostly depend on the
drainage density. The left hand side of the river Shivkhola is less steep in
2.4 Conclusion 91
comparison to right hand side. The analysis of hypsometric curve reveals that the
potential dissection is more at Tindharia and for the stretch along the Hill Cart Road
between Gayabari to Mahanadi including Paglajhora. The northeastern part of the
basin near the water divide is extensively dissected and shows absolute instability.
The study area is composed mainly of the Darjiling Gneiss, Daling formation
composed of Chungtung formation, Lingtse Granite, Garubathan formation and
Ryang formation. Gondwana formation, the most fragile one due to the presence of
carboniferous rocks is located along a narrow belt being sandwiched between
Daling to the north and Siwalik to the south. The structural-cum stratigraphic
succession can be observed as a traverse across Tindharia-Kurseong region. The
concerned study area is structurally instable as most of the unconformities are lying
across the drainage lines and so subsidence zones are developed at the junctions of
the drainage lines with the structural discontinuities and lineaments. The Paglajh-
ora, the biggest subsidence zone of the study area, is situated along the Darjiling-
Daling boundary. The mica schist in the Daling series is also a factor of instability
as most of the slides are located on the mica schist due to its less resistance. The
slope is maximum near Paglajhora area and the Hill Cart Road and North Eastern
Frontier Rail line (0.61 m gauge) cross the entire river system twice through this
steeper and unstable zone. The slope is least at the central part where the river
develops a cut and fill terrace. The study shows that the LPI for the slope categories
are increasing at a steady rate as the steepness increases and it is the indicator of the
direct control of slope on the slope failure. All the cells having steepness of 19°–23°
and above are affected by landslide. The positive curvature is more common
indicating the tendency of immediate drainage of surface water which is detrimental
to the stability of both soil and slope. Paglajhora, Gayabari, Tindharia, and Shiviter
are dominated by the moderate to high levels of positive and negative surface
curvature with moderate levels of slope surface dissection. Landslide prone north,
south, east, north east and south east facets are closely associated with maximum
slope and relief which is found at upper and lower Paglajhora, Shiviter T.E.,
Gayabari Lower slope and Tindharia where landslide potentiality is high. Sub-
surface soil over the steep slope at the places of Tindharia T.E. and Lower Pag-
lajhora is dominated by humus which is very loose, crumble and friable. Such
humus dominated soil gets saturated very easily due to moderate amount of rain and
reduces cohesion by increasing pore-water pressure and make the slope surface
more vulnerable to soil slip. At marginal parts of the basin basically on the both
sides of the Hill Cart Road from Tindharia to Gayabari, upslope parts from Pag-
lajhora proper, T.E. and Shiviter upslope where the saturated soil depth is less than
1.75 m. At all these places sub-surface soil gets saturated quickly and promotes
suitable condition for shallow soil slip. At the sub-surface layer of the soil per-
centage of pore space is high but at greater depth pore space decreases because of
the existence of large percentage of finer particles. The reduction of pore space at
greater depth results in the increase of water holding capacity and volumetric
expansion at the sub-surface soil which increases the pore-water pressure and
reduces cohesion and finally invites slope soil failure at most of the places of the
Shivkhola Watershed.
References
Ahnert F (1987) Process response models of denudation at different spatial scales. Catena Suppl
10:31–50
Anabalagan R (1992) Landslide hazard evaluation and zonation mapping in mountainous terrain.
Eng Geol 32:269–277
Anderson MG, Burt TP (1978) The role of topography in controlling through flow generation.
Earth Surf Proc 3:331–344
Attewell PB, Farmer IW (1976) Principles of engineering geology. Chapman and Hall, London
Barton N, Choubey V (1977) The shear strength of rock joints in theory and practice. Rock Mech
10:1–54
Basu SR, Sarkar S (1985) Some consideration on recent landslides at Tindharia and their control.
Indian J Power River Val Dev 1985:190–194
Beven KJ, Kirkby MJ (1979) A physically based variable contributing area model of basin
hydrology. Hydrol Sci Bull 24(1):43–69
Brasher BR (1966) Use of Saran resin to coat natural solid clods for bulk-density and water
retention measurement. Soil Sci 101:108
Chorley RJ, Schumm SA, Sugden DE (1985) Geomorphology. Methuen and Co Ltd, New York
Cruz O (2000) Studies on the geomorphic processes of overland flow and mass movements in the
Brazilian geomorphology. Revista Brasileira de Geosciencias 30:500–503
Dapples EC (1959) Basic geology for science and engineering. Wiley, New York
Deoja et al (1991) Mountain risk engineering handbook. International Centre for Integrated
Mountain Development (ICIMOD), Kathmandu, pp 875
Dhakal AS, Amada T, Aniya M (2000) Landslide hazard mapping and its evaluation using GIS: an
investigations of sampling schemes for a grid-cell based quantitative method. Photogramm Eng
Remote Sens 66(8):981–989
Dietrich et al (1998) SHALSTAB: a digital terrain model for mapping shallow landslide potential.
National council of the paper industry for air and stream improvement. Technical report
Gao J (1993) Identification of topographic settings conducive to landsliding from DEM in Nelson
County, VA, USA. Earth Surf Proc Land 18:579–591
Gilbert GK (1909) The convexity of hill tops. J Geol 17:344–350
Guimaraes RF, Montgomery DR, Greenberg HM, Gomes RAT, Fernandes NF (1999) Applicatioa
of a model for the topographic control on shallow landslides to catchments near Rio de Janeiro.
In: Lipard SJ, Naess A, Sinding-Larsen R (eds) IAMG99—annual conference of the
international association of mathematical geology. IAMG, Trondheim, Noruega, pp 349–354
Krishnaswamy VS (1982) Geological aspects of landslides with particular reference to the
Himalayan region. In: Proceedings of the international symposium on landslides, New Delhi,
pp 171–185
Keen BA, Raczkowski H (1921) Relation between the clay content and certain physical properties
of a soil. J Agric Sci 11:441–449
Laahiri S (1973) Some observations on structure and metamorphims of the rocks of Kurseong,
Tindharia region, Darjeeling district WB. In: Jhingran AG, Valdia KS (eds) Himalayan
geology, vol 3. Wadia Institute of Himalayan Geology, Delhi, pp 365–371
Lahiri S, Gangopadhway PK (1974) Structure pattern in rocks in Pankhabari—Tindharia region
Darjeeling district WB with special reference its bearing on stratigraphy. In: Jhingran AG (ed)
Himalayan geology, vol 4. Wadia Institute of Himalayan Geology, Delhi, pp 151–170
LSS O’Malley (1999) Darjiling district gazetteers. Logos Press, New Delhi
Mandal S, Maiti R (2012) Application of RS and GIS based semi-quantitative approach in
landslide hazard risk assessment of the Shivkhola watershed, Darjiling Himalaya. Georisk
Assess Manag Risk Eng Syst Geohazards 6(4):203–220
Mallet FR (1874) On the geology and mineral resources of the Darjeeling district and Western
Duars. Mem Geol Surv India 2:1–72
References 93
Montgomery et al (1994) A physically based model for the topographic control on shallow
landsliding. Water Resour Res 30:1153–1171
Nautiyal SP (1951) A geological report on the hill slope stability in and around Darjeeling, WB.
Unpublished report of the geological survey of India
Nautiiyal SP (1966) On the stability of certain hill slopes in and around Darjeeling, WB. Bull Geol
Surv India Ser B 15(1):31–48
Roy S, Sensharma SB (1967) Geological report on the stability of hill slopes in and around
Darjeeling town, Darjeeling district SW. Unpublished geological survey of India report
Sarkar S (1987a) Pedogeomorphic parameters in environmental management—a case study in the
upper Panchanai basin of Darjeeling Himalayas. In: Proceedings of the seminar on applied
geography in the perspective of planning the environment, the urbanscape and the regional
pollution. North Bengal University, pp 128–139
Sarkar S (1987b) Soil loss in the upper Mahananda basin in the Darjeeling Himalaya. Geog Rev
India 49(2):47–56
Strahler AN (1952) Hypsometric (area altitude) analysis of erosional topography. Bull Geol Soc
Am 63:1117–1142
US Department of the Interior, Bureau of Reclamation (1963) Earth manual. Government Printing
Office, Washington, DC
Vieria et al (1998) Controles fito-morphologicos dos escorregamentos da baisa de Quintic (RJ).
Revista GEOSUL 27:324–328
Zhou CH, Lee CF, Li J, Xu ZW (2002) On the spatial relationship between landslides and
causative factors on Lantau Island, Hong Kong. Geomorphology 43:197–207
Chapter 3
Impact Assessment of Hydrologic
Attributes and Slope Instability
Abstract Quantitative geomorphology provides a systematic approach to the

analysis of a complex landscape of any size. The stability of the mountain slope
depends upon the prevalence of various hydrologic variables. In the present work,
the excess and deficit moisture period in a year and its role in slope instability were
assessed studying rainfall and evapotranspiration. Study envisages that July and
August are the most consistent rainfall months of the year where the values of
co-efficient of variation are very low. The distribution of drainage and its evolution
has been studied to determine the drainage concentration over the slope surface and
their role in slope steepening and instability. To assume the slope saturation of
materials saturation, stream confluence points/junction points were studied for
individual sub-watersheds. The length of drainage per unit area and upslope con-
tributing area were analyzed spatially in connection to the landslide potentiality.
The existence of moderate drainage density may invite havoc slope failure on
convex slope segment. Greater the upslope contributing area, maximum is the slope
saturation and slope instability in the Shivkhola watershed. Some important
drainage basin parameters such as basin shape, form factor, circularity ratio,
elongation ratio, compactness factor, and elipticity index of the sub-watersheds
were considered to develop the priority scale on slope instability. The sub-basin I
and IV are more efficient in drainage and are more erosion and landslide prone
followed by sub-watershed V, II, III and VI.
Keywords Rainfall Drainage morphometry Landslide potentiality Instability

rank Watershed
3.1 Introduction
One of the most fundamental geomorphic unit is the ‘drainage basin’ in the study of
geo-hydrological parameters where drainage network characteristics and its
evolution strongly influences the whole hydrological parameters through initiating,

96 3 Impact Assessment of Hydrologic Attributes and Slope Instability
encouraging and changing others sub-areal processes such as surface run-off,

seepage, and sub-surface flow. But, for the development of the drainage network
and for changing actual nature of the slope materials rainfall must plays an
important role. The history of evolution of a drainage basin and its morphology are
closely related to rainfall characteristics and geology, i.e. the type of rocks and their
structure. Depending upon the susceptibility to erosion and structure of the rocks
the fluvial processes operate and the landform evolution proceeds through
successive stages with drainage network branching and head ward extension.
Hydrological attributes mainly rainfall and drainage are the two important factors of
slide and soil erosion as the material moves down slope with the help of water and it
may be either surface or sub-surface flow. The water when flows on the surface
with huge amount can perform remarkable erosion as the potential energy it pos-
sesses is readily transfers to kinetic energy which takes part in the mass transfer.
The soil water saturates the soil and increases the pore water pressure and the
weight of the mass which then becomes more prone to slide or subsidence. The
occurrence of few days’ continuous rainfall and their presence within the subsoil
may act as the lubricating agent and which makes the down slope movement easier
by reducing the friction, cohesion and increasing shearing stress in the slope
material. Rainfall as a trigger has been extensively studied by number of authors i.e.
Larson (1995), Polloni et al. (1996), Glade (2000), Wieczorek and Guzzetti (2000),
Polemio and Petrucci (2000), Toll (2001) and Zezere (2000).
One of the most important geomorphic properties is the degree of dissection of
the topography, sometimes expressed in terms of drainage density. Gradual
extension of drainage network and their regular branching increase the slope
steepness by increasing slope concavity and convexity. Gilbert (1909) argued that
convex-concave forms reflect a gradual transition in process dominance from creep
to wash with increasing distance from a drainage divide. Gilbert’s model was
quantified in terms of a linear stability analysis by Smith and Bretherton (1972).
Drainage density and landscape structure may alternatively be controlled, by
thresholds for run-off generation (Kirkby 1980; Ijjasz-Vasquez et al. 1992) or by
thresholds of slope stability (Montgomery and Dietrich 1989). Drainage density in
particular may be controlled to varying degrees by any of these thresholds, and each
different threshold may produce a different functional relationship between drainage
density and factors related to climate geology and relief. Howard (1997) repre-
sented a detachment-limited model in which the relationship between drainage
density and mean erosion rates depends on (i) the dominant hill slope transport
process (creep/landsliding) and (ii) the presence or absence of a threshold for run-
off erosion.
The Shivkhola basin is under the humid climatic condition receiving excessive
orographic monsoon rainfall, which enhances the erosion and denudation of the
surface within the basin. The nature of terrain, evolution of landscape, amount of
soil erosion and their removal, in the Shivkhola watershed could be well acquainted
by studying various hydrological factors/attributes i.e. climatic attributes (rainfall
and evaporation), drainage confluence, drainage density, upslope contributing area
and others. To understand the probability or chances of landslide occurrences
3.1 Introduction 97
phenomena to each class of all the hydrological factors responsible for slope
instability, landslide potentiality index value (LPIV) is determined by means of
ratio between number of landslide location to each class and total number of
landslide location in the basin. So, the study of all these attributes is very much
significant in analyzing the nature of slope susceptibility of the basin in term of
geo-hydrology.
3.2 Climatic Attributes and Landslides
The climate of the Darjiling district is especially unique because of its position in
relation to the Tibetan land mass, the wide differences in altitude, the powerful
effects of the monsoon against the Himalayan barrier and the peculiar configuration
of the neighboring mountains which deflects winds and affect local climate (rainfall
and temperature) On account of the hilly nature of the terrain of Shivkhola
watershed, there are sharp variations in rainfall even between nearby sections of the
mountain ranges. The precipitation during south west monsoon constitutes about
80 % of the annual rainfall, July being the wettest month. The variation in the total
rainfall from year to year is not much. On an average, there are about 120 rainy
days in a year. Temperature generally touches the highest level in May. January is
the coldest month of the year. Over the northern parts of the watershed the atmo-
sphere is highly humid throughout the year. In the north-western hilly tract the
relative humidity between 90 and 95 % during the rainy season. But in the low
lying tracts to the east and south-east, the relative humidity is slightly less. The
driest months are March and April when the relative humidity varies between 45
and 60 % (Table 3.1). The present study involves the analysis of some basic
information available either from field work or some reliable secondary sources.
The climatological information are collected from Tea Research Center (Kurseong),
Shiviter Tea Garden and Agriculture and Soil Conservation Office (Kurseong,
Darjiling).
3.2.1 Study on Rainfall
3.2.1.1 30 Years’ Monthly Average Distribution of Rainfall

and Assumption of Catastrophic Rainfall (mm) Year
Drainage basin is a proper spatial scale for analyzing hydrological parameters like
input of rainfall and resultant output of discharge in a systematic interactive
combination with other topographic and geometric attributes (Chorley 1969;
Strahler 1957). Amount of rainfall is one of the triggering factors for slope insta-
bility because it affects surface run-off, infiltration, depth of the saturated soil and
thus influences soil-moisture condition. Infiltration and evapotranspiration are
Table 3.1 Meteorological observation since 2001–2010 (10 years mean)

Months Mean max Mean mini Average rainfall Mean wind Mean % of Evaporation Mean dew
temperature temperature for 32 years speed (m/s) possible (mm/day) point
in mm (P) sunshine temperature
Jan 12.83 6.16 12.57 1.00 14 0.4 9.00
Feb 13.77 7.83 16.67 1.05 31 1.3 9.00
Mar 17.22 10.89 43.76 1.26 53 2.2 9.00
Apr 21.49 14.26 94.85 1.58 54 4.7 9.00
May 21.57 15.53 302.80 1.48 27 2.6 9.00
Jun 22.34 17.27 728.35 1.52 28 4.4 9.00
Jul 22.84 17.22 995.54 1.39 07 3.9 9.00
Aug 23.05 17.60 783.92 1.26 14 4.0 9.00
Sep 21.85 16.47 587.95 1.11 15 4.1 9.00
Oct 20.89 14.41 183.29 1.11 36 1.6 9.00
Nov 18.21 11.58 12.30 1.01 63 1.3 9.00
Dec 14.57 8.55 21.90 1.19 37 0.7 9.00
considered as the important hydrological parameters that determine slope insta-

bility. The hydrologic factors like daily rainfall threshold in connection with slope
angle and regolith thickness (Gabet et al. 2004), rain fall intensity, infiltration
(Schumm 1956) etc. are given due importance in the analysis of slope instability. In
the Shivkhola watershed the amount of rainfall increases from the month of May
and it reaches peak in July then it starts decreasing and reaches minimum in
January.
Analysis of monthly average rainfall since 1979–2009 also reveals that the
months of June, July, August and September are experiencing rainfall of more than
(between third quartile-Q3 and first quartile-Q1) average. On the other hand, below
average rainfall is being found mostly in the months of October, November,
December, January, February and March. July and August (Table 3.2) where the
values of co-efficient of variation are very low (37 and 30). These monsoon months
are also characterized by the catastrophic rainfall months because of the frequent
occurrences of landslide events due to few days’ continuous rainfall. The pictur-
esque slope failure took place in the Shivkhola watershed due to catastrophic
rainfall in the month of July of 1985, 1989, 1992, 1993, 1998, 2002, 2003, 2004,
2006 and 2007. Rainfall induced slope failure also occurred in September 1980 and
2006, August 2007 and 2010.
Table 3.2 Statistical analysis of monthly average rainfall since, 1979–2010

Months J F M A M J J A S O N D
Mean 13 17 43 95 302 728 996 784 588 183 12 23
S.D. 23 20 48 60 173 304 364 236 170 154 18 61
C.V. 176 117 111 63 57 41 37 30 28 84 150 97
3.2 Climatic Attributes and Landslides 99
3.2.1.2 Yearly Distribution of Rainfall with 3 Years Moving Average

Value and Its Relation with Landslide Events
Starkel (1972) for the first time, observed the geomorphic effects of an extreme rainfall
event in the eastern Himalaya (Darjiling). Froehlich et al. (1990) investigated the same
area (Darjiling Himalaya) and found that shallow slides and slumps on steep slope
segments occur when 24 h rainfall reaches 130–150 mm or continuous three days
rainfall totals 180–200 mm Researchers (Ghosh 1950, 2009; Nautiyal 1951, 1966;
Dutta et al. 1966; Basu and Sarkar 1985; Basu and Maiti 2001; Paul 1973; Sengupta
1995; Basu and De 2003; Pal 2006; Maiti 2007; Sarkar 2011) carried out a demand
oriented studies in Darjiling Himalaya and identified the role of rainfall in causing
landslide occurrences phenomena. The rainfall data since 1979–2009 was collected
from Selim Hill Tea Estate, Tindharia Tea Estate and Shiviter Tea Estate. Analysis of
yearly average rainfall since 1979–2009 (Appendix B, Table B.1) also suggests that
the year 1980, 1981, 1984, 1985, 1989, 1990, 1991, 1998, 1999, 2000, 2003, 2005,
2006 and 2007 are the year of above average rainfall (317.75 mm). The devastating
landslide occurrence in the 1984 and 1985, 1987, 1988, 1999, 2000, 2003, 2005,
2006, and 2009 are associated with the preceding year increasing trend average
rainfall. So, the landslide phenomena in the Shivkhola watershed are related with the
cumulative effects of precipitation and the changing nature of ground-water condition.
3.2.2 Evapotranspiration
Evapotranspiration, the leakage from immediate system, regulates water balance

and moisture availability. To understand the water loss from the corresponding
watershed monthly evapotranspiration is calculated following Jensen et al. (1990).
The equation for well-watered grass or reference ET0 (Penman 1963) and
converted to SI units (Jensen et al. 1990) is:
D c
To ¼ ðRn GÞ þ 6:43ð 1:0 þ 0:53v2 Þðes ed Þ ð3:1Þ
Dþc Dþc
where,
λETo reference ET for a well-watered grass expressed as latent heat flux density,
MJ m−2 day−1,
Δ slope of the saturation vapor pressure curve in kPa/°C,
γ psychometrics constant in kPa/°C,
Rn net radiation in MJ m−2 day−1,
G heat flux density to the soil in MJ m−2 day−1,
v2 average wind speed at a height of 2 m in m/s,
es saturated vapor pressure at mean air temperature in kPa,
ed saturated vapor pressure at mean dew point temperature also es × mean
relative humidity).
The following equations and constants were summarized from the Jensen et al.
(1990)
i. Slope of the saturation vapor pressure curve:
Values for the Δ can be obtained from,
D ¼ 0:20ð0:00738T þ 0:8072Þ7 0:000116 ð3:2Þ
where,
T temperature in °C.
ii. Determination of the psychrometric constant (in kPa/°C):
c ¼ 0:00163 P=k ð3:3Þ
iii. Determination of estimated atmospheric pressure:
P ¼ 101.3 0.01055 (EL) ð3:4Þ
iv. Determination of the latent heat of vaporization:
k ¼ 2:501 0:002361 T ð3:5Þ
where,
P estimated atmospheric pressure in kPa,
EL elevation in m,
λ latent heat of vaporization of water in MJ/kg.
v. The Calculation of net radiation (Rn):

The net radiation (Rn) can be calculated from,
h i
Rn ¼ ð1 aÞRs rT 4a 0:34 0:139ðed Þ0:5 ð0:1 þ 0:9 n/NÞ ð3:6Þ
where,
Rn the solar radiation received at the earth’s surface in MJ m−2 day−1,
α the radiation reflection coefficient or albedo with values near 0.25 for
green crops,
σ Stefan Boltzmann Constant (4.903 × 10−9 MJ m−2 day−1 °k−4).
Ta absolute air temperature in °k (°C +273),
n/N ratio of actual to possible hours of sunshine.
3.2 Climatic Attributes and Landslides 101
vi. Determination of the mean cloudless solar radiation:

If the solar radiation is not measured, it can be obtained from
Rs ¼ð0:35 þ 0:61n=NÞRso ð3:7Þ
where,
Rso is the mean solar radiation for cloudless skies in MJ m−2 day−1 from the
table made by Jensen et al. (1990).
vii. Determination of the saturation vapor pressure:

The saturation vapor pressure is calculated from the following formula,
h i
es ¼ 3:38639 ð0:00738T þ 0:8072Þ8 0:00019j1:8T þ 48j þ 0:001316
ð3:8Þ
where,
T the mean air temperature in °C.
The Eq. 1.8 can be used to determine the ed by substituting the mean
dew point temperature for T.
3.2.3 Relationship Between Rainfall and Evapotranspiration

and Determination of Deficit and Excess Moisture
Period
Rainfall ranging between 300 and 700 mm occurs in the month of May and Sep-
tember. The rainfall amounting below 150 mm is considered as deficit moisture
period as evapotranspiration exceeds rainfall (Appendix B, Table B.1) that occurs
in January, February, March, April (pre-monsoon) and October, November and
December (post-monsoon). On the other hand in the months of May, June, July,
August, September and October the precipitation exceeds the evaporation and sub-
surface gets saturated. This excess moisture is spatially distributed over the basin
unit. The maximum concentration is observed near the source region, where
overland flow and sub-surface flow are collected to initiate surface channel fol-
lowing Hortons’ (1932, 1945) rule. The excess moisture period (July, August, and
September) experiences significant run-off (Table 3.3) from the whole basin and
makes the slope surface more vulnerable. Thus soil and slope instability becomes
maximum at the upper catchment near Upper Paglajhora and the rational man-
agement of the excess moisture during monsoon is earnestly needed to avoid
inconvenience and loss of land and property by frequent landslide there.
102
Table 3.3 Monthly evapotranspiration (after Penman 1963 and Jensen et al. 1990)
Month parameters Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
T (in °C) 9.50 10.80 14.06 17.88 18.45 19.81 20.03 20.33 19.16 17.65 14.90 11.56
Δ (kPa/°C) 0.08 0.09 0.11 0.14 0.14 0.15 0.15 0.16 0.15 0.13 0.11 0.09
P (kPa) 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22 88.22
k (MJ/kg) 2.48 2.48 2.47 2.46 2.46 2.45 2.45 2.45 2.46 2.46 2.47 2.47
c (kPa/°C) 0.06 0.06 0.05 0.05 0.05 0.06 0.06 0.06 0.05 0.05 0.05 0.05
es 1.15 1.25 1.65 2.00 2.07 2.26 2.29 2.33 2.17 1.96 1.65 1.32
ed 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11
D 0.57 0.60 0.69 0.74 0.74 0.71 0.71 0.73 0.75 0.72 0.69 0.64
Dþc
c
Dþc s
0.43 0.40 0.31 0.26 0.26 0.29 0.29 0.27 0.25 0.28 0.31 0.36
−2
R so(MJ m 18.75 22.68 26.64 29.95 31.89 32.77 32.31 30.35 27.22 23.51 19.59 17.42
day−1)
Rs (MJ m−2 day−1) 8.16 12.23 17.94 20.35 16.41 17.07 12.68 13.21 12.02 13.39 14.38 10.03
Rn 4.75 6.84 9.74 11.28 9.96 10.34 8.37 8.32 7.39 7.17 6.44 4.83
λETo (MJ m−2 2.88 4.66 8.52 11.08 10.23 11.21 9.76 9.61 8.25 7.59 6.10 3.88
day−1)
ET (in -mm) 21.74 51.36 93.89 122.12 112.74 109.81 119.51 123.56 90.92 83.64 67.22 42.76
Rainfall (mm) 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
Difference −9.17 −34.69 −50.13 −27.27 190.06 618.59 875.99 660.34 496.98 99.66 −54.92 −20.86
3 Impact Assessment of Hydrologic Attributes and Slope Instability
3.3 Drainage Network Evolution and Landslide 103
Fig. 3.1 Drainage network of 1972
3.3 Drainage Network Evolution and Landslide
3.3.1 Drainage Network Analysis of 1972
The drainage lines drawn from SOI Topographical map of 1972 reveals less
drainage density and wider watershed areas of overland flow (Fig. 3.1). The pos-
sibility of head ward extension and further branching indicates potential erosion and
land failure. Only few 1st order fingertip branches could reach their maximum
length, allowing others to extend rapidly by head ward erosion facilitated by slope
failure. This channelization is also facilitated by the human intervention along the
road as the upslope drainage is concentrated on the road section and flow down
slope as a collected whole. The same is true for the settlement also as the collected
water from roof-top starts flowing down slope with increased energy and concen-
trated erosion is thus facilitated and this may initiate smaller channels and may
increase gradually by either lengthening or widening.
Order wise spatial distribution of drainage network of sub-watersheds in the
Shivkhola shows that maximum no. of 1st order streams (43) are found in sub-
watershed 1and it also covers a large part (4.0 km2) within this sub-watershed. Sub-
104
Table 3.4 Order wise spatial distribution of drainage network of sub-watersheds in Shivkhola-1972
S.no. Order no. No of stream segments Total length of stream Mean length Area (in km2) Mean total area (in km2)
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
1 4th 43 7 2 1 35.5 3.5 2.75 5.0 0.83 0.5 1.37 5.0 4.0 1.18 1.06 2.30 0.09 0.16 0.53 2.035
2 3rd 9 2 1 0 3.75 0.5 2.00 – 0.42 0.25 2.00 – 0.91 0.12 0.54 – 0.10 0.06 0.54 –
3 3rd 17 5 1 – 7.5 3.5 0.5 – 0.44 0.7 0.5 – 1.82 0.95 0.54 – 0.10 0.19 0.54 –
4 3rd 8 3 1 – 4.5 1.75 1.5 – 0.56 0.58 1.5 – 0.81 0.39 0.49 – 0.10 0.13 0.49 –
5 3rd 5 2 1 – 1.75 1.25 0.75 – 0.35 0.62 0.75 – 0.60 0.41 0.30 – 0.12 0.20 0.30 –
6 3rd 21 4 1 – 12 4.5 2 – 0.57 1.12 2.00 – 3.02 2.06 0.90 – 0.14 0.51 0.90 –
3 Impact Assessment of Hydrologic Attributes and Slope Instability
Fig. 3.2 Spatial distribution of drainage up to 4th order stream, 1972
watershed 1 ranks 2nd in terms of areal coverage. On an average most of area of

sub-watershed 2, 3, 4, 5 and 6 are dominated by first order streams, followed by 2nd
and 3rd orders (Table 3.4). Order wise stream network analysis of 1972 depicts that
the whole basin is being mostly dominated by the impact of 1st order streams and
their gradual headward extension (Fig. 3.2). The central-mid section is character-
ized by the cumulative effects of streams due to existence of 3rd and 4th order
streams with maximum discharge.
3.3.2 Drainage Network Analysis of 1987
The drainage lines depicted in 1987 SOI (Survey of India) Topographical map,
showed a further branching and extension of the channels (Fig. 3.3). The watershed
Fig. 3.3 Drainage network of 1987
areas of overland flow were reduced as the drainage was served by channelized flow
and characterized by active down cutting processes over the steep slope along both
sides of the main Shivkhola River. The analysis of drainage for a period during
1972–1987 revealed the extreme dynamism in the extension of the channels. The
formation of numerous small channels have increased the erosion of surface and
sub-surface soil as well as entraining the small soil particles and reducing the
cohesive strength of the soil which could induce slope failure in near future. Because
of the increased down cutting of the channels the slope surface were getting dis-
sected and rugged gradually making some part of the region inaccessible. The
interfluves are further sharpened indicating the stage of early maturity. This stage
was characterized by maximum extension of slope without remarkable presence of
flat surface. In such condition slope instability gets maximum attention from plan-
ners, administrators and academicians due to their frequent manifestation in slope
failure.
Fig. 3.4 Spatial distribution of drainage up to 5th order, 1987
Order wise spatial distribution of drainage network of sub-watersheds in the

Shivkhola Watershed (1987) shows sub-watershed 6 and 1 are occupied by max-
imum number of 1st order streams and 4 and 5 are with minimum number of 1st
order streams (Fig. 3.4). Fifth order stream is mainly found in sub-watershed 1.
Sub-watershed 1 ranks 1st in terms of areal coverage which is followed by sub-
watershed 6, 3, 2, 4 and 5 (Table 3.5). So, sub-watershed 1 and 6 are susceptible to
soil erosion due to drainage concentration. As the drainage network of 1987 is
associated with the development of first order stream over the steep marginal area
and their continuous branching due to headward erosion, the first order stream
Table 3.5 Order wise spatial distribution of drainage network of sub-watersheds in Shivkhola
Watershed (1987)
Watersheds
1 2 3 4 5 6
No. of stream
1st order 65 17 36 16 12 72
2nd order 18 5 7 3 2 16
3rd order 4 1 3 1 1 4
4th order 1 – 1 – – 1
5th order 1 – – – – –
Total length of stream
1st order 35 7.5 13 8 6 26
2nd order 8.5 1.5 4.5 2.8 1 7.3
3rd order 2.5 1.5 2.5 1.5 0.8 4.3
4th order 3.0 – 0.5 – – 1.5
5th order 3.75 – – – – –
Mean length of stream
1st order 0.53 0.4 0.4 0.5 0.5 0.4
2nd order 0.47 0.3 0.5 0.9 0.5 0.5
3rd order 0.63 1.5 0.8 1.5 0.8 1.1
4th order 3.00 – 0.5 – – 1.5
5th order 3.75 – – – – –
Area in sq. km.
1st order 3.56 0.9 1.5 1 0.7 2.4
2nd order 1.53 0.2 0.8 0.3 0.2 1.1
3rd order 0.4 0.3 0.5 0.4 0.2 1
4th order 0.94 – 0.2 – – 0.7
5th order 1.30 – – – – –
Mean area in sq. km.
1st order 0.06 0,05 0.04 0.06 0.06 0.03
2nd order 0.09 0.04 0.11 0.1 0.1 0.07
3rd order 0.10 0.3 0.17 0.4 0.2 0.25
4th order 0.94 – 0.2 – – 0.7
5th order 1.30 – – – – –
covers the large part of all the sub-watershed in Shivkhola. Generally mid-section
of sub-watersheds are attributed with the 2nd and 3rd order streams where the
drainage concentration and maximum number of stream confluence points pro-
motes the channel to flow with high discharge and so high rate of erosion is found
in this section.
3.3.3 Drainage Concentration and Stream Confluence Point
The number of confluence points/junction points per unit area is an important index
in a river basin analysis. Confluence point is a place of union of two or more
streams over space which depends on the others geomorphic and geo-hydrologic
attributes of the basin. It is assumed that greater number of confluence points
indicate greater amount of erosion and drainage accumulation within the basin. The
sub-watershed 1, 2, 3, 4, 5 and 6 are experienced with 67, 15, 32, 14, 11 and 69
confluence points respectively. On the basis of these distributed confluence point in
each sub-watershed, it is inferred that the topography of sub-watershed 1 and 6 are
heavily dissected followed by sub-watershed 3, 2, 4 and 5. The southern part or the
right hand part of the main Shivkhola watershed (sub-watersheds-2, 3, and 6) is
attributed with large number of confluence points in comparison to northern part of
the basin (Fig. 3.5) mainly due to topographic configuration. The southern part is
registered with dominant erosion and changing slope surface by frequent and
gradual drainage convergence. It is also assumed from the drainage confluence map
Fig. 3.5 Stream junctions points of the Shivkhola watershed

that the main stream of all the individual watersheds (sub-basins) is experiencing
large no. of stream union where the valley deepening as well as valley lengthening
is quite common. Such processes steepen the valley side slope and make the slope
material to move downward under the influence of gravity.
3.3.4 Drainage Network Orientation
Drainage orientation is the key to understanding the way in which a drainage

network fills the space that is available to it (Jarvis 1976). Drainage orientation is
related to the strike and dip slopes of the underlying rocks and is very sensitive to
fold axes and rock jointing. Basically, the regional slope surface and the underlying
rock structure play a dominant role in changing the drainage network orientation in
the Shivkhola Watershed. The interior and exterior links, length, and frequency of
streams are related with the azimuth of flow direction. The main drainage line of the
Shivkhola is oriented towards the south-east up to lower Paglajhora, and then it has
got its course in easterly direction with sinuosity within the sub-watershed-1. The
left hand two main tributaries are oriented in south-east direction through sub-
watershed-4 and 5 respectively. Sub-watershed 2, 3 and 6 consist of the tributaries
which are oriented in north-east direction. It is to be assumed that large part of the
sub-watershed-1; 2, 3 and 6; 4 and 5 are characterized by south-east and east; north-
east and south-east slope facets according to orientation of main drainage lines
where the landslide potentiality is high. The study depicts that the drainage ori-
entation plays an important role for promoting landslip as the availability of
moisture content depends on it which helps to reduce cohesion and increase pore-
water pressure in the soil and induces landslip.
3.3.5 Drainage Density and Landslide Potentiality
According to Horton (1945) ‘Drainage density in an important indicator of the

linear scale landform elements in a drainage basin’. The views of Horton (1945) and
Strahler (1964) and others on drainage density are almost identical. The latter states,
“in order to find out drainage density the length of the streams of all hierarchical
orders should be measured, totaled up and divided by the geographical area of the
drainage basin”. Dury (1969) has used ‘the expression of texture’ and says ‘in order
to express the texture of dissection, we must use not the water course but the valley
bottom (the thalweg), whether wet or dry’. The thalweg length divided by area is
equivalent to drainage density as suggested by Horton and others.
This may be taken as the index of dissection of a drainage basin. Two areas may
have identical drainage density but while one of them may have higher dissection
index because of the presence of a large number of streams, the other may have
lower dissection index because of the presence of small number of streams.
Drainage density is an effective indicator of slope failure as the drainage efficiency

from a rainfall is determined by the drainage length in an area. The length of
drainage per unit area is thus an indicator of surface flow and potential instability.
The source of the drainage channels are the slide prone areas and the valley sides
are also the zones of slope failure. So the potential slope failure zones are extended
along the drainage lines and are concentrated at the source points. The greater
drainage density is thus an indicator of potential instability. At mid-central areas,
near Paglajhora, numbers of drainage channels from the upper marginal watershed
region meet the main stream, and so drainage density is maximum indicating the
excess surface water (Fig. 3.6). This is the main cause of slope instability at
Paglajhora.
3.3.6 Upslope Contributing Area and Landslide Potentiality
Upslope Contributing Area (UCA) is an effective indicator of drainage concen-

tration over space. The surface water accumulates in a cumulative rate away from
the water divide as more Upslope Contributing Area helps in accumulation of more
water and so indicates more surplus moisture and instability of slope and soil. The
calculation of contributing area (Fig. 3.7) has to be made considering multiple flow
direction where the cumulative flow at a point (surrogated by the upslope area)
should be distributed among more than one neighboring down slope pixel (Borga
et al. 1998).The topographic irregularities at concave and convex slope are
responsible for the convergence and divergence of flow and thus necessitate the
implication of topographic index by Quinn et al. (1991). Another concept of
specific contributing area (total contributing area divided by the contour length) is
computed by distributing flow from a pixel among its entire lower elevation
neighbour pixel (Borga et al. 1998).
Quinn et al. (1991) proposed that Fraction of Flow (Fi) allocated to each lower
neighbour is to be determined by:
Si Li
Fi ¼ ð3:9Þ
R Sj L i
where the summation is for the entire lower neighbour; S is the directional slope,
and L is an effective contour length that acts as the weighting factor. The value of L
used here is 10 m of the pixel size of the cardinal neighbour and 14.14 m of the
pixel diagonal for diagonal neighbour.
The study on Landslide Potential Index (LPI) of Upslope Contributing Area
depict that the percentage of occurrence of landslides and its distribution among the
different groups of contributing area shows the increasing Landslide Potential Index
(LPI) with increasing value of upslope contributing area (Table 3.6). The regions of
contributing area of less than 5 Km2 experiences the LPI of 76.56 and that of 10
Km2 experiences the LPI of 100 (Table 3.7). Figure 3.7 shows the grid wise value
Fig. 3.6 Drainage density (km/km2) map of the Shivkhola Watershed
of Upslope Contributing Area which gradually increases away from the water
divide and the maximum of 20.98 km2 is registered at the lower most portion of
every sub-watershed. The more contributing areas are registered along the main
river which experience maximum flow. Sub-watershed-I and VI are experiencing
maximum upslope contributing area and maximum overland flow in the area of
study. It is to be assumed that the maximum amount of soil erosion is expected in
sub-watershed-I due to maximum overland flow.
Fig. 3.7 Upslope contributing area map of Shivkhola Watershed
Table 3.6 Drainage density and landslide potentiality

Classes Number of pixels % of Landslide pixels % of FR
(F1) (Npix (Ni)) Npix (Ni) (F2) (Npix(Si)) (Npix(Si))
0–1.90 5,560 16.78 90 2.67 0.16
1.90–3.80 5,453 16.46 109 3.23 0.20
3.80–5.71 3,289 9.93 158 4.68 0.47
5.71–7.61 5,049 15.24 137 4.06 0.27
7.61–9.51 3,477 10.49 159 4.71 0.45
9.51–11.41 2,728 8.23 465 13.78 1.67
11.41–13.31 1,875 5.66 544 16.13 2.85
13.31–15.21 2,191 6.61 542 16.07 2.43
15.21–17.12 1,942 5.86 697 20.66 3.52
17.12–19.02 1,567 4.73 572 16.96 3.58
Table 3.7 Frequency distribution and Landslide Potential Index (LPI) of upslope contributing
area
Upslope con- Number Number of landslide Landslide Landslide potential
tributing area of pixels occurrences pixels occurrence index (LPI) = F2/
(Km2) in % in % ratio F1 × 100
<5.00 85.33 84.48 0.76 76.56
5.00–10.00 8.00 6.90 0.66 66.67
10.00–15.00 1.33 1.72 1.00 100
15.00–20.00 4.00 5.17 1.00 100
>20.00 1.33 1.72 1.00 100
3.4 Drainage Morphometry and Slope Instability
Geo-hydrologic and geometric parameters of the drainage play a significant role in

changing the quality of soil and water of a small hilly sub-watershed and make the
hill slope more vulnerable to slope instability. The morphometric study on river
basin was first introduced by Horton (1932). Later on, this idea was developed by
Miller (1953), Schumm (1956), Coates (1958), Melton (1958), Maxell (1960),
Strahler (1957), Chorley (1969), and Mulder and Syvitsky (1996). During 1950s
and 1960s the morphometric proposals made by Horton again were developed in
the studies of drainage basin morphometry, which aimed to analyse regularities of
stream networks and subsequently led to the variation in network characteristics
and to investigation of statistical or topological properties of network. The hierar-
chical ranking of the stream in a river basin was studied by Strahler (1964). Sig-
nificant contribution in various aspects of morphometric parameters have been
studied by host of scientists from time to time e.g. Morisawa (1962), Gregory and
Wallings (1968, 1973), Singh (1978a, b, 1981), Singh et al. (1986), and Richards
(1981). Recently Chandra et al. (1984) studied watershed with the help of remote
sensing method using simulation model. Quantitative geomorphology provides a
systematic approach to the analysis of a complex landscape of any size and origin
of the drainage basin. Some important parameters i.e. Basin Shape, Stream Length,
Length of Overland Flow, Form Factor, Circularity Ratio, Elongation Ratio,
Compactness Factor, Elipticity Index, Bifurcation Ratio, Relative Relief, Drainage
Density, Stream Frequency, Constant of Channel Maintenance, Ruggedness Index,
Dissection Index, Slope Inclination, and Upslope contributing area of the sub-
watersheds have been considered to develop the priority scale.
Basin shape is referred to the shape of the outline of the drainage basin that is
determined as shape of projected surface on the horizontal plane of the basin
map. Mulder and Syvitsky (1996) have indicated that a majority of the rivers have
elongated basins. Generally, the shape of the basin affects the flow pattern and
consequently affects the surface run-off and erosional mechanism within the
catchment. The basin shape may be expressed through Form Factor, proposed by
Horton in 1932. The cumulative length of channel segments increases with the
3.4 Drainage Morphometry and Slope Instability 115
channel order, it being the lowest for the first- order channels. The law of stream
length applied in the present study revealed that the mean length of the channel
segments of each of the successive orders of the shiv-khola watershed formed
approximately a direct geometric series. A plot of logarithms of mean stream
lengths against their respective order showed the positive exponential form of
relationship between stream order and mean stream length of the different sub-basin
of the siva-khola watershed. The length of the overland flow is the length of no
channel flow from the basin perimeter to the nearest channel. It is an important
independent variable affecting the quantity of water required to exceed a certain
threshold of erosion. Horton (1945) and Schumm (1956) noted that the length of
overland flow approximately equals half the reciprocal of drainage density. When
the rainfall intensity exceeds soil infiltration capacity, the excess water flows over
the land surface as overland flow.
According to Gregory and Wallings (1973) the form factor is the governing
factor of the water courses which enter the main streams. Form factor has been
introduced by Horton (1932) that shows the shape of the basin.
Form Factor defined as the ratio of the basin area to the square of the basin
length using the following equation:
Au
Rf = ð3:10Þ
L2 b
where,
Au Basin area (km2)

Lb Basin length (km).
If the basin is wider, the form factor will be comparatively higher, and conse-
quently much narrower basins have low form factor values (Gregory and Wallings
1973). The calculated values of form factor of the six sub-basin for the shiv-khola
watershed are 0.12, 0.40, 0.45, 0.36, 0.25 and 0.52 respectively which reveals the
sub-basin-I and sub-basin-V are more elongated than other basins and that is why
the erosional activities become prominent within the two sub-basin. The circularity
index, (Rc) is expressed as the shape of the basin that was used by Miller (1953).
The circularity ratio is dimensionless, whose values vary from 0 to 1. The circu-
larity ratio obtained from the ratio of the basin area (Au) to the area of a circle (Ac)
having equal perimeter as the perimeter of the drainage basin.
Au
Rc = ð3:11Þ
Ac
where,
Au Area of the basin

Ac Area of the circle with same perimeter as the basin.
The deviation of the values of circularity ratio from 1 shows the irregularity of
the catchment. Here in case of sub-basin-I, the values of the circularity index
deviates more and showing most irregularity and elongation within the whole
drainage basin. Besides, sub-basin-II and IV are also indicating the lower values of
circularity index and consequently revealing another two irregular segments of the
Siva-khola watershed. The sub-basin III, V and VI are more regular in respect of the
values of the index (values which are closer to 1). A circular basin is more likely to
have a shorter lag time and a higher peak flow than an elongated basin. Elongation
Ratio indicates how the shape of the basin deviates from a circle (Schumm 1956). It
is an index to mark the shape of the drainage basin. The value of R varies from 0
(highly elongated shape to the unity (1.00) circular shape. So, the higher values of
R mean more circular shape of the basin and vice versa.
Elongation ratio is defined as the ratio of the diameter of a circle which has same
area as the basin to the maximum basin length.
Dc
R= ð3:12Þ
Lbm
where,
Dc Diameter of the circle having same area as the given drainage basin
Lbm Maximum length (km).
Compactness factor of the basin is used to express the basin shape, which is
indicated by the deviation of the basin area from a circle of an equal area. Com-
pactness factor is expressed as the shape of the basin that was used by Horton and
was devised by Gravelius (Gupta 1999). The compactness factor was obtained from
the ratio of the perimeter of the watershed to the circumference of a circle whose
area is equal to that of the drainage basin.
P
Cc = pffi ð3:13Þ
½2pA
where,
P Perimeter of the basin (km)

A Area of basin (km2).
3.4.1 Determination of Composite Ranking Coefficient Value

and Instability Rank
The form factor reveals that the sub-basin 1 and sub-basin 5 are more elongated
than other basins and that is why erosional activities become prominent within these
two sub-basins. The value of the circularity index for sub-watershed 1 deviates
Table 3.8 Basin geometry and composite ranking co-efficient value and instability analysis
Sub- Form Rank Elon- Rank Circularity Rank Elipticity Rank Compactness Rank Length of Rank Composite Instability
basin factor gation index (C) index (E) co-efficient overland rank rank
(F) ratio® (Cc) flow (km)
1 0.12 6 0.39 1 0.3 1 6.76 1 0.62 5 0.15 3 17 5
2 0.4 3 0.71 4 0.55 3 1.97 4 1.06 2 0.13 5.5 21.5 3
3.4 Drainage Morphometry and Slope Instability
3 0.45 2 0.76 5 0.77 6 1.72 5 0.63 4 0.13 5.5 27.5 1

4 0.36 4 0.68 3 0.54 2 2.17 3 1 3 0.15 3 18 4
5 0.25 5 0.56 2 0.68 4 3.09 2 1.09 1 0.15 3 17 6
6 0.52 1 0.81 6 0.69 5 1.5 6 0.51 6 0.16 1 25 2
117
more from 1, showing most irregularity in shape. Besides, sub-basin 2 and 4 are
also indicating the lower values of circularity index and consequently revealing
another two irregular segments of the siva-khola watershed. The sub-basin 3, 5 and
6 are more regular.
The ranks were assigned to the sub-watersheds according to their forms and
relative shapes in ascending order of magnitude favouring easy drainage. Sub-basin
6 was thus ranked V for maximum form factor. All the ranks for Form Factor,
Elongation ratio, Circularity Ratio, Elipticity Index, Compactness Coefficient and
length of overland flow were assigned accordingly and finally cumulative value
were assigned as the composite ranking for drainage efficiency. Thus sub-basin 5
and 1 are more efficient in drainage followed by sub-basin 4 (Table 3.8).
3.5 Conclusion
Rainfall is the significant landslide triggering factor in the Shivkhola Watershed as

most of the landslides are rain-induced. Not only that, all the recorded landslides
occurred during the excess moisture period. The analysis of drainage network of
1972 and 1987 depicts that there is continuous branching of channels and these are
developed through headward erosion and sharpening of interfluves between two
sub-watersheds. Sub-watershed 1 and 6 are attributed with maximum number of
stream, stream length, and mean area at each orders which are the indicators
of more surface run-off and potential erosion and slope failure. Maximum number
of stream confluence points is found in sub-watershed 1 and 6 which is followed by
3, 2, 4 and 5. The potential slope failure zones are extended along the drainage lines
and concentrated at the source points. The greater drainage density is an indicator of
potential instability. At the zones near Paglajhora, mid-central areas, numbers of
drainage channels from the upper marginal watershed region meet the main stream.
Here drainage density is maximum indicating the excess surface water. The max-
imum LPI is recorded within the range between 5.00 and 6.50 km/km2 and is
followed by 3.50–5.00 km/km2 showing the effects of intensity of drainage density
on slope instability. In the saucer shaped basin, the concentration of flow at the
central lower portion caused the maximum density but due to gentle slope, the
landslide occurrences are less. The moderate density on steeper slope caused havoc
on slope instability due to cumulative effects of other triggering factors. The surface
water accumulates in a cumulative rate away from the water divide as more Upslope
Contributing Area (UCA) helps in accumulation of more water that indicates more
surplus moisture and instability of slope and soil. Upslope Contributing Area of
20.98 km2 is registered at the lower most portion of every sub-watershed. The more
contributing areas are registered along the main river which experience maximum
length and thus maximum flow. The percentage of occurrence of landslides and its
distribution among the different groups of contributing area shows increasing
Landslide Potentiality Index (LPI) with increasing value of upslope contributing
area. The regions of upslope contributing area of less than 5 Km2 experiences the
3.5 Conclusion 119
LPI of 76.56 and that of 10 Km2 and more experiences the LPI of 100. The study on
basin geometry states that sub-basin 1 and 5 are more efficient in drainage and more
susceptible to soil erosion and slope instability.
References
Indian J Power River Val Dev 190–194
Himalayas. Chang Env Scenerio Indian Subcont (Bd) 390–399
Basu SR, De SK (2003) Causes and consequences of landslides in Darjiling-Sikkim Himalayas.
India Geograph Pol 76(2):37–52
Borga M, Dalla Fontana G, Da Ros D, Marchi L (1998) Shallow landslide hazard assessment using
a physically based model and digital elevation data. J Environ Geol 35(2–30):81–88
Chandra S, Sharma KP, Kashyap O (1984) Application of remote sensing methods to hydrology:
watershed studies using simulation model for Upper Yamuna catchments’ School of
Hydrology. University of Roorkee, Roorkee
Chorley RJ (1969) The drainage basin as a fundamental geomorphic unit. In: Chorley RJ (ed)
Water, earth and man. Methuen, London, pp 77–100
Coates DR (1958) Quantitative geomorphology of small drainage basins of Southern Indiana,
Office of Naval Research, Geography Branch, Project NR 389-042. Technical report 10
Dury GH (1969) Hydraulic geometry. In: Chorley RJ (ed) Water, earth and man. Methuen and Co
Ltd, London, pp 319–330
Dutta DE et al (1966) Landslides and erosion in Kalimpong subdivision, Darjeeling district and
their bearing on North Bengal floods. Bull Geol Surv India Ser B 15(1):61–78
Froehlich W, Gil E, Kasza I, Starkel L (1990) Thresholds in the transformation of slopes and river
channels in the Darjeeling Himalaya, India. Mt Res Dev 10(4):301–312
Gabet JE, Burbank WD, Putkohen KJ, Pratsituala AB, Ojha T (2004) Rainfall threshold for
landsliding in the Himalayas of Nepal. Geomorphology 63(1–4):131–143
Ghosh AMN (1950) Observation of the landslides of the 11th and 12th June, 1950 in the
Darjeeling Himalaya. Unpublished geological survey of India report
Ghosh S, Van Westen CJ, Carranza E, Jetten V (2009) Generation of event- based landslide
inventory maps in a data-scarce environment; case study around Kurseong, Darjiling District,
West Bengal, India. In: Malet JP, Remaitre A, Bogaard T (eds) Landslide processes: from
geomorphologic mapping to dynamic modeling: proceedings of the landslide processes.
European Centre on Geomorphological Hazards (CERG), Strasbourg, pp 37–44
Gilbert GK (1909) The convexity of hilltops. J Geol 17:344–350
Glade T (2000) Modelling landslide triggering rainfall thresholds at a range of complexities. In:
Bromhead E, Dixon N, Ibsen ML (eds) Landslide in research, theory and practice. Thomas
Telford, Cardiff, pp 633–640
Gregory KJ, Walling DE (1968) The variation of drainage density within a catchment. Bull Int
Assoc Sci Hydrol 13:16–68
Gregory KJ, Walling DE (1973) Drainage basin form and process: a geomorphological approach.
Eward Arnold, London 45 pp
Gupta SNP (1999) Morphological character of hillside slope in the Damodar baisn in Bihar. Indian
Geog Stud 13:16–21. Geography Research Centre, Patna
Horton RE (1932) Drainage basin characteristics. Trans Am Geophys Union 14:350–361
Horton RK (1945) Erosional devepolment of streams and their drainage basins, hydrological
approach to quantitative morphology. Geol Soc Am Bull 56:275–370
Howard AD (1997) Badland morphology and evolution: interpretation using a simulation model.
Earth Surf Process Landf 22:211–227
Ijjasz-Vasquez EJ et al (1992) Sensitivity of a basin evolution model to the nature of run-off
production and to initial conditions. Water Resour Res 28:2733–2741
Jarvis RS (1976) Stream orientation structures in drainage networks. J Geol 84:563–582
Jensen ME, Burman RD, Allen AG (eds) (1990) Evapotranspiration and irrigation water
requirements. ASCE, New York
Kirkby MJ (1980) The stream head as a significant geomorphic threshold. In: Coates DR, Vitek JD
(eds) Threshold in geomorphology. Allen and Unwin, Winchester, pp 53–73
Larson RA (1995) Slope failure in southern California: rainfall threshold, prediction and human
causes. Environ Eng Geosci 1:3930401
Maiti R (2007) Identification of potential slope failure zones of shiv-khola watershed; Darjiling
Himalaya, through critical analysis of slope instability—a step towards rational and scientific
management of land, soil and water. UGC sponsored minor research project [F.31-210/2005
(31.03.2007)]
Maxell JC (1960) Quantitative geomorphology of the San Dimas experiment forest, California.
Technical report 19, Office of Naval Research Project 389-042
Melton MA (1958) Geometric properties of mature drainage basin systems and their controlling
agents. J Geol 66:442–460
Miller VC (1953) A quantitative geomorphic study of drainage basin characteristics in the Clinch
mountain area, Virginia and Tenneses. Department of Geology, Columbia University, Contract
N60NR271-30. Technical report 3, pp 1–30
Resour Res 25:1907–1918
Morisawa ME (1962) Relation of quantitative geomorphology to streams flow in representative
watersheds of the Appalaihian Plateau province, Department of Geology, Columbia University,
ONR project technical report, vol 2, pp 389–442
Mulder T, Syvitsky JPM (1996) Climate and morphologic relationships of river implications on
river loads. J Geol 104:509–523
Nautiyal SP (1951) A geological report on the hill slope stability in and around Darjeeling, WB.
Unpublished report of the geological survey of India
Nautiiyal SP (1966) On the Stability of certain hill slopes in and around Darjeeling, WB. Bull Geol
Surv India Ser B 15(1):31–48
Pal C (2006) Unpublished report of the investigation on the slope instability along NH-55 between
Tindharia to Mahanadi. Geological Survey of India
Paul DK (1973) Evolutionary ecology. Harper and Row Publications Inc., New York, pp 1–65
Penmann HL (1963) Vegetation and hydrology. Technical communication No. 53. Common-
wealth Bureau of Soils, Harpenden
Polemio M, Petrucci O (2000) Rainfall as a landslide triggering factor: an overview of recent
international research. In: Bromhead E, Dixon N, Ibsen ML (eds) Landslide in research, theory
and practice, proceeding of the 8th international symposium on landslides. Thomas Telford,
Cardiff, pp 1219–1226
Polloni G et al (1996) Heavy rain triggered landslides in the Alba area during November 1994
flooding event in the Piemonte Region (Italy). In: Senneset K (ed) Landslide—Glissements de
Terrain, vol 3. A.A. Balkema, Rotterdam, pp 1955–1960
Quinn P et al (1991) The prediction of hillslope flow paths for distributed hydrological modeling
using digital terrain models. Hydrol Process 5:59–79
Richards KS (1981) General problems in morphometry. In: Goudie A (ed) Geomorphological
technique. George Allen and Unwin, London, pp 26–30
Sarkar S (2011) Evolution of the Paglajhora slump valley in the Shivkhola basin, the Darjiling
Himalaya, India. Geogr Pol 84(Special Issue, Part-2):117–126
Schumm SA (1956) Evolution of a drainage system and slope in badland at Perth Amboy, New
Jersey. Geol Soc Am Bull 66:597–646
References 121
Sengupta CK (1995) Detailed study of geofactors in selected hazard prone stretches along the
surface communication routes in parts of Darjiling and Sikkim Himalaya, Phase-I, Part-I
(Rongtong-Kurseong Road Section). In: GSI (ed) Annual progress report (F.S. 1993–1994)
Singh S (1978a) Physiographic regions, landforms and erosion surfaces of Ranchi plateau. Natl
Geogr 13(1):43–65
Singh S (1978b) A quantitative analysis of drainage texture of small drainage basins of the Ranchi
plateau, in morphology and evolution of landforms. Geology Department, Delhi University,
New Delhi
Singh S (1981) Estimation of drainage density. Natl Geogr 16(2):81–89
Singh S, Gardiner V, Ojha SS (1986) Spatial variation of drainage density in the Palamau upland.
India Natl Geogr 21(1):83–99
Smith TR, Bretherton FP (1972) Stability and conservation of mass in drainage basin evolution.
Water Resour Res 8:1506–1529
Starkel L (1972) The role of catastrophic rainfall in the shaping of the relief of the Lower Himalaya
(Darjeeling Hills). Geogr Pol 21:103–147
Strahler AN (1957) Quantitative analysis of watershed geomorphology. Trans Am Geophys Union
38:913–920
Strahler AN (1964) Quantitative geomorphology of drainage basins and channel network. In:
Chow VT (ed) Handbook of applied hydrology. McGraw Hill, New York, pp 439–476
Toll DG (2001) Rainfall induced landslides in Singapore. Proc Inst Civil Eng Geotech Eng
149:211–216
Wieczorek GK, Guzzetti F (2000) A review of rainfall thresholds for triggering landslides.
Mediterranean Storms, proceedings of the EGS plinius conference, Maratea, Italy, Oct 1999,
pp 407–414
Zezere JL (2000) Rainfall triggering of landslides in the area north of Lisbon (Portugal). In:
Bromhead E, Dixon N, Ibsen ML (eds) Landslide in research, theory and practice, proceeding
of the 8th international symposium on landslides. Thomas Telford, Cardiff, pp 1629–1634
Chapter 4
Surface Run-off, Soil Erosion and Slope
Instability
Abstract The dynamic nature of a landscape results from the interaction of surface
run-off with rocks and soil being guided by geo-hydrologic variables. The estimation
of surface run-off and its better understanding reveals a clear idea about the degree
and amount of surface erosion and slope vulnerability over the space. In this chapter
Soil Conservation Service (SCS) Run-off Curve Number (CN) model proposed by
United State Department of Agriculture (USDA 1972) is used to determine the
surface run-off from six individual sub-watersheds for predicting the periodical
spatial distribution of slope instability and soil erosion. The determined Curve
Number (CN) under antecedent moisture condition-III (AMC-III) for sub-watershed
I, II, III, IV, V and VI are 85.02, 73.52, 87.36, 87.76, 85.57 and 89.85 respectively.
Sub-watershed I contributes maximum run-off from a rainfall of 90.5 mm
(4,52,359.4 m3) which is followed by VI, III, IV, V and II. Landslide Potentiality
Index Value (LPIV) is derived for each watershed which reveals that Sub-watershed
I and VI is the significant landslide prone unit of the study area. Finally, considering
both run-off and LPIV an instability scale has been made which reveals that Sub-
watershed VI, I and III have to be paid more attention for a proper management of
land, water and soil during the months of July, August and September. All the
necessary constructions, plantation and related preparedness through raising
awareness and making task forces during pre-monsoon dry period are of utmost
importance for managing landslip and soil erosion at Shivkhola Watershed.
4.1 Introduction
Runoff, being the most important hydrologic variable, draws the attention of
hydrologists, water resource planners, local govt. etc. for water resource planning
and applications. The objective mostly sought by hydrologists is the accurate and
timely prediction of runoff of a given point in a drainage basin by either using a range
of equation and models or direct measurement at gauging station. The stream
measuring stations are more common in large river system and reservoirs where

124 4 Surface Run-off, Soil Erosion and Slope Instability
continuous monitoring generates a database of temporal variation in run-off. The

gauge station at all target points are not possible to set and so, inspite of having some
limitations models are to be used for estimating run-off. Run-off from a drainage
basin is influenced by various climatic (type of precipitation, intensity of rainfall,
duration of rainfall, areal distribution of rainfall, direction of storm movement,
antecedent precipitation, evaporation and transpiration) as well as physiographic
factors (land use, type of soil, area of the basin, shape of the basin, elevation, slope,
orientation, type of drainage network, indirect drainage and artificial drainage). The
geology or soil materials contribute to a large degree of the infiltration rate, and thus
affect run-off, vegetation and the practices incident to agriculture and forestry
(Schwab et al. 2002). But, it is difficult to quantify the amount of surface run-off from
a drainage basin on the basis of these factors. Run-off is estimated by various
methods such as (i) Empirical formula and tables (Binnie’s run-off coefficients,
Barlow’s percentage run-off coefficients-K, Strange’s Tables, Inglis and De Souza’s
formulae, Lacey’s formula, A.N. Khosla’s formula, U.P. Irrigation Department’s
formula, Indian Council of Agricultural Research’s formulae-1971, Parker’s
formulae, C.C. Vermuel’s formulae, I.G. Justin’s formulae. J. Rodier’s Coefficients-
1967 and F.V. Zaleskkii U.S.S.R’S formula-1967); (ii) Estimating Losses (A.N.
Khosla’s relation, U.S. formula and David Lloyd’s formula); (iii) Infiltration
Method; (iv) Rational Methods (R.L. Gregory and C.E. Arnold’s modification-1932,
M. Bernard’s Coefficient-1938, American Society of Civil Engineering Coefficients,
W.S. Kerby formula-1959, C.F. Izzard Formula-1944 and Ben Chie Yen, Yung
Yuan Shen and Ven Te Chow-USA); and (v) Unit hydrograph method and sys-
tematic unit graph method. U.P. Irrigation Research Institute Roorke (1960)
developed a statistical correlation between run-off (R) and precipitation (P) in cms
for Himalayan and Bundelkhand Region Rivers in Uttar Pradesh. Soil Conservation
Research Demonstration and Training Centre Dehradun of ICAR (1971) analysed
run-off and rainfall from 17 sub-watersheds in the Nilgiri hills and also a reliable
regression equations have been reported for the estimation of nine important physio-
graphic characters i.e. catchment perimeters in km-Lp, main stream length in km-Ls,
compactness co-efficient Ce, rotundity factor RF, form factor FF, shape index SI, total
watershed relief in m-RT, drainage density (km/km2)-DD, and time of concentration
in minutes TC for the catchments. A rainfall-runoff modeling is developed by Singh
(1988). Chow et al. (1988) also estimated surface runoff. A geomorphology based
artificial neural networks (GANNs) was put forward for the estimation direct runoff
over watersheds by Zhang and Govindaraju (2003). Mishra et al. (2003) presented a
modified SCS-CN Method. Saragni et al. (2007) evaluated three unit hydrograph
models to predict the surface runoff from a Canadian watershed. Recently, an
integrated approach for estimating surface run-off using Remote Sensing and GIS in
the field of hydrological research is applied by Durbude et al. (2001), Jasrotia and
Singh (2006), Tripathi et al. (2002) and Zade et al. (2005).
In the present approach Remote Sensing data is used as the basic information
input for computing runoff using the Soil Conservation Service (SCS) Run-off
Curve Number (RCN) model proposed by United State Department of Agriculture
(USDA-1972). The SCS models are mostly used in hydrology and water resource
4.1 Introduction 125
planning of agricultural areas in the United States and other countries as well. In
India an empirical model for estimation of flood discharge in Ramganga basin
using Remote Sensing data was derived by Chandra and Sharma (1978). This
model of SCS Curve Number is used for estimation of runoff by correlating
generalized land cover with hydrologic soil groups and data is derived from the
SCS table (Ragan and Jackson 1980; Tiwari et al. 1991 etc.). The SCS model is
attractive because the major input parameters are defined in terms of land use and
soil type. The advantage of a model having parameters defined in terms of land
use or land cover is that the user can experiment with alternate forms of land
development and assess the impact of that changes might have (Ragan and
Jackson 1980). These models are originally developed for predicting runoff
volumes from agricultural fields and small watersheds (Horn and Schwab 1963;
Harrold 1957). In the present study, the evaluation of surface run-off and land-
slide potentiality in the basin hierarchy of the Shivkhola watershed (Fig. 4.1) will
Fig. 4.1 Six sub-watersheds with drainage

help to predict the periodic slope susceptibility to soil saturation, soil erosion, and
soil slip and on the basis of which we can infer a rational management procedure
to reduce slope vulnerability.
4.2 Run-off Estimation
The runoff equation for small mountain watershed can be written as:
Qt ¼ f ðPt; S; IaÞ ð4:1Þ
(Hand book of hydrology 1972).

Qt is the depth of runoff over the catchment during selective time period “t”
Pt is the depth of rainfall over the catchment during time period “t”
S is the potential maximum retention of water by the soil in equivalent depth
over the catchment
Ia is the initial abstraction during the period between the beginnings of rainfall
Pt and runoff Qt in equivalent depth over the catchment.
The relation between S and Ia is very important to understand as a part of S is

retained in the soil in the form of interception, infiltration, depression storage and
absorption initially in the form of Ia. Ia varies according to soil condition and
Antecedent Moisture Condition (AMC) which is the existing moisture condition of
soil and land and expressed as the amount of rainfall received in past 5 days.
ðP 0:2SÞ2
Q¼ ð4:2Þ
ðP þ 0:8SÞ2
(Equation applicable to small hilly watershed. For a given storm 20 % (0.2S) of

the potential maximum retention is the initial abstraction before runoff begins)
where,
25400
S¼ 254 ð4:3Þ
CN
Q Actual direct runoff (in mm)

P Total Rainfall (in mm)
S Potential maximum retention (in mm)
CN SCS Runoff Curve Number.
4.2 Run-off Estimation 127
4.2.1 Determination of Runoff Curve Number (RCN)
The Runoff Curve Number (RCN) is a quantitative descriptor of the land and soil
complex and is commonly assigned based on information acquired from field
surveys and interpretation of Aerial photograph or Satellite Imageries. The RCN of
a Soil—Vegetation—Land (SVL) complex in a specific antecedent moisture
condition (AMC) takes on values from 0 to 100. This number is derived from the
character of the soil, vegetation, including crops and the land use of that soil as well
as intensity of use. When CN equals to 100, S becomes zero (In water-logged areas
or in wet paddy field). This leads to Q (Runoff) = P (Precipitation). In other cases,
when S → ∞, CN → 0, this gives Q = 0.
The CN value is derived on the basis of Soil—Vegetation—Land (SVL) and
Antecedent Moisture Condition complex. The soils are grouped into four types
according to their hydrologic character. The land use, with respect to their hydro-
logic i.e. draining condition is to be analyzed with due importance. The determi-
nation of CN value requires the following data input
1. Land use—Land cover class.
2. Hydrologic Soil groups.
3. Hydrologic (Draining) condition.
4. Antecedent Moisture Condition (AMC).
4.2.1.1 Land Use and Land Treatment
Land use is the watershed cover and it includes every kind of vegetation, litter and
mulch and fallow as well as non agricultural uses such as water surfaces (lakes,
swamps etc.) and impervious surfaces (roads, roofs etc.). Forest area includes all
lands classed as forest under any legal enactment dealing with forest or adminis-
tered as forest whether state owned or private. Tree crops include woody perennial
plants that reach a mature height of at least 8 ft and have well defined stems and a
definite crown shape. Plantation is an important land use of the study area. Tea is
the single plantation crop of this region and occupies a considerable section of land,
and exerts a great influence on the economy and culture of the concerned society.
Open space includes the parks, golf course, cemeteries, lawns etc. where the grass
or trees cover ranges from greater than 50 % to less than 75 %. Urban district
category reflects the commercial, business and industrial sectors of the study area
where 70–85 % areas is impervious in nature. Residential district means the urban
residential houses, either reflects the compact settlement area or the scattered houses
of tea gardens, forest etc. and here 65 to 12 % area is impervious in nature. Land
use and land cover map of the Shivkhola watershed is prepared with the help of
LISS-III Satellite Image, Google Earth Image in consultation with Survey of India
(SOI Topo-sheet. Later on, with proper ground truth verification with GPS a land
use and land cover map is developed. In the present study broadly the Shivkhola
Fig. 4.2 Land use and land cover map of the Shivkhola watershed
Watershed is classified into ten individual land use pattern such as i. Bare Surface,
ii. Agricultural land, iii. Jungle, iv. Roads, v. Settlement, vi. Tea garden, vii. Open
forest, viii. Degraded forest, ix. Mixed forest and x. Dense forest. (Fig. 4.2). Again,
hydrological soil group wise land use and land cover is attributed (Table 4.1) to
extract Curve Number following Empirical Curve Number Table under different
hydrological condition for each of the six sub-watersheds which the entire study
area is divided into.
Table 4.1 Land use/land cover and their hydrologic conditions

Land use/land cover Intensity and characters Hydrologic
conditions
Bared Steep slope Without any residue cover Poor
surface Gentle slope Without any residue cover Poor
Agricultural land Contoured and terrace Poor
Jungle Upslope 50–70 % ground is covered by plant Fair
Down slope <50 % ground is covered by plant Poor
Roads Concrete Average 65 % area is impervious area Poor
Metal Average 25 % area is impervious Fair
Settlement Compact 65 % average impervious area Poor
Disperse 25 % average impervious area Fair
Tea garden Upslope More than 75 % area is covered by tea Poor
plantation
Down slope More than 75 % area is covered by tea Good
plantation
Open forest Upslope Grazed but not burned Fair
Down slope Grazed but not burned Fair
Degraded Upslope Forest and tree crops heavily grazed Poor
forest or burned
Down slope Forest and tree crops heavily grazed Poor
or burned
Mixed Upslope Wood and grass combination 50 and Fair
forest 50 %
Down slope Wood and grass combination 50 and Good
50 %
Dense forest >90 % ground is covered by plant Good
4.2.1.2 Hydrologic Soil Groups
Soils are originally assigned to hydrologic soil groups based on measured rainfall,
run-off, and infiltrometer data (Musgrave 1955). Most of the soil groupings are
based on the premise that soils found within a climatic region that are similar in
depth to a restrictive layer or water table, transmission rate of water, texture,
structure, and degree of swelling when saturated, will have the similar run-off
responses (National Engineering Handbook, USDA 1993). The given four hydro-
logical soil groups are used in determining hydrologic soil cover complexes, which
are used in the method for estimating runoff from rainfall. The soil properties play
an important role in the estimation of runoff from the rainfall. In this concern the
properties can be represented by hydrologic parameters that influence the minimum
rate of infiltration obtained from a bare soil after prolonged wetting. Hydrologic soil
group-A, having high infiltration rates and low run-off potential even when thor-
oughly wetted and consisting chiefly of deep, well to excessively drained sands or
gravels, having high rate of water transmission (Tables 4.2 and 4.3). The USDA
Table 4.2 Infiltration rates used in hydrologic soil groupings

Class Hydrologic Rate/hour Remarks
soil group Inches Millimeters
Very D Less than Less than High clayey soil
low 0.1 2.5
Low C 0.1–0.5 2.5–12.5 Shallow soils, clay soils, soils low
in organic matter
Medium B 0.5–1.0 12.5–25.0 Sandy loams, silt loams
High A Greater Greater Deep sands well aggregated soils
than 1.0 than 25.0
Table 4.3 Permeability ratings used in hydrologic soil groupings

Group Hydrologic soil group Rate/hour
Inches Millimeters
Very slow D Less than 0.05 Less than 1.30
Slow C 0.05–0.20 1.31–5.00
Moderately slow B and C 0.20–0.80 5.01–20.00
Moderate B 0.80–2.50 20.01–50.00
Moderately rapid B 2.50–5.00 50.01–130.00
Rapid A 5.00–10.00 130.01–250.00
Very rapid A Greater than 10.00 Greater than 250.0
soil texture normally included in this group is sand, loamy and loamy-sand and
sandy-loam. These soils have a transmission rater greater than 0.76 cm/h. Soils
having moderate infiltration rates and moderately low run-off potential character-
ized by Hydrologic soil group-B, when thoroughly wetted and consisting chiefly of
moderately deep to deep and moderately well to well drained soils with moderately
coarse textures (Tables 4.2 and 4.3). These soils have a moderate rate of water
transmission.
The USDA soil textures normally included in this group are silt-loam and loam.
These soils have a transmission rate between 0.38 and 0.76 cm/h. Soils having slow
infiltration rates when thoroughly wetted and consisting chiefly of moderately deep
to deep and moderately well to well drained soils with moderately fine to moder-
ately coarse textures includes Hydrologic soil group-C having moderately high run-
off potential (Tables 4.2 and 4.3). These soils have a moderate rate of water
transmission. The USDA soil textures normally included in this group is sandy clay
loam. These soils have a transmission rate between 0.13 and 0.38 cm/h. Soils
having very slow infiltration rates and high run-off potential when thoroughly
wetted and consisting chiefly of clay soils with a high swelling potential with a
permanent high water table comprising Hydrologic soil group-D (Tables 4.2 and
4.3). These soils have a clay pan or clay layer at or near the surface with shallow
soils cover of nearly impervious material.
The USDA soil textures normally included in this group are clay loam, silt clay
loam, sandy clay, silt clay and clay. These soils have a very low rate of water
transmission between 0.0 and 0.13 cm/h. Some soils are classified in group D
because of a high water table that creates a drainage problem; however, once these
soils are effectively drained, they are placed into another group. The following
parameters were used to assess run-off curve number and all these parameters are
related with hydrologic soil group.
(i) Infiltration rate: See Table 4.2.
(ii) Soil permeability: These are tentatively suggested rates through saturated
undisturbed areas under about 0.5 in. (12.7 mm) head of water.
(iii) Draining parameter of the soil: In well drained condition water is removed
from the soil readily but not rapidly. Well drained soils are commonly
intermediate in texture and moderate to rapid infiltration and permeability
rate with hydrologic soil group A and B, although soils of other textural
classes may also be well drained (Tables 4.2 and 4.3). In moderately well
drained situation water is removed from the soil somewhat slowly so that
the profile is wet for a small but significant part of the time. Moderately well
drained soils commonly have a slowly permeable layer within or immedi-
ately beneath the solum or some combination of these conditions consisting
hydrologic soil group of B and C (Table 4.3). The soil remains wet for a
large part of the time in poor drained condition. The water table is com-
monly at or near the surface during a considerable part of the year.
(iv) Effective depth of the soil: This refers to depth of soil material that is
readily penetrated by plant roots, and is important to the moisture regime
of the soils. The effective depth figures are those observed during field
profile examination. In case of those soils which are lying over disinte-
grated and weathered rock material (murrum), the total depth of such a
layer has been converted into effective soil layer by assuming percent soil
material in the layer. The total depth is then calculated by adding this
assumed depth to overlaying thickness of soil layers.
(v) Average clay content within the soil at various depths: This pertains to
the average clay content in the whole profile depth (Table 4.4). To derive
this figure clay content of each horizon is multiplied by thickness of the
horizon. Sum of all the horizons is then divided by total thickness of the
profiles. Example—Hydrologic soil group-A have less than 10 % clay and
Table 4.4 Average clay contain according to thickness

Horizon Thickness (cm) Clay (%) Avg. clay contain
1 15 20 300 1845/75 = 24.4 %
2 20 23 460
3 25 29 725
4 15 24 360
Total 75 1,845
Hand book of Hydrology (1972)
more than 90 % sand or gravel. The saturated hydraulic conductivity of all

soil layers exceeds 40.00 µm/s. The depth to water table is greater than
60 cm. Soil group B possesses between 10 and 20 % clay and 50–90 %
sand. Soil group C is characterized by 20–40 % clay and less than 50 %
sand. Soil group D consists greater than 40 % clay and less than 50 % sand.
4.2.1.3 Hydrologic Condition
Hydrologic condition indicates the effects of cover type and treatment on infiltration
and runoff and is generally estimated from density of plant and residue cover on
sample areas or refers to the state of vegetation growth. A poor hydrologic con-
dition refers to pasture heavily grazed with sparse vegetation. A fair condition is for
pasture moderately grazed with between half and 3/4th of the basin under plant
cover i.e. the basin has 50 to 25 % plant cover. A good hydrologic condition
indicates that the soil usually has a low runoff potential for the given hydrologic soil
group, land cover type and treatment. Some factors to consider in estimating the
effects of cover on infiltration and runoff are—i. Canopy or density of lawns, crops
or other vegetation areas; ii. Amount of year-round cover; iii. Amount of grass or
close seeded legumes in rotations; iv. Percent of residue cover; and v. Degree of
surface roughness (Maidment 1972).
4.2.1.4 Antecedent Moisture Condition
Antecedent Moisture Condition (AMC) refers to the water content present in the
soil at a given time. The AMC value is intended to reflect the effect of infiltration on
both the volume and rate of runoff according to the infiltration curve (Sing 1975).
The amount of rainfall in past few days reflects the condition of Antecedent
Moisture and is as essential indicator of amount of initial abstraction (Ia) and
percentage runoff (Table 4.5).
The Soil Conservation Service (SCS) developed three antecedent soil moisture
conditions and labeled them as I, II, III. These AMCs correspond to the following
soil condition:
Table 4.5 Relation between initial abstraction (Ia) and potential retention (S)
Relation between initial abstraction (Ia) Region/soil type Antecedent moisture
and potential retention (S) condition (AMC)
Ia = 0.1S Used in black soil region AMC II and AMC III
Ia = 0.2S Small hilly watershed AMC I
Ia = 0.3S All other regions –
Table 4.6 Seasonal rainfall limits for the antecedent soil moisture conditions (AMC)
Condition General description 5 days’ antecedent
rainfall (mm)
Dormant Growing
season season
I Optimum soil condition from about lower plastic 13 36
limit to wilting point
II Average value for annual flood 13–28 36–53
III Heavy rainfall to light rainfall and low temperature 28 53
within 5 days prior to the given storm
Source Soil Conservation Service (1972)
AMC I: Soils are dry but not to the wilting point; satisfactory cultivation has
taken place.
AMC II: Average condition.
AMC III: Heavy rainfall to light rainfall and low temperature have occurred
within the last 5 days; Saturated soil (Table 4.6).
Initial Abstraction (Ia) consists mainly of interception, infiltration and surface

storage, all of which occur before runoff begins. An exact determination of Ia is
very difficult. However, for practical purpose, Ia can be related to S. Based on
analysis of data from a large number of small watersheds, the Soil Conservation
Service (1969) found Ia to be roughly equal to 0.2S (Table 4.4). It can also be
estimated by relating to the antecedent soil-moisture index. Potential maximum
retention (S) includes Ia and the infiltration occurring after runoff begins. This later
infiltration is controlled by the rate of infiltration at the soil surface or by the rate of
transmission in the soil profile (Hand Book of Hydrology 1972). The magnitude of
‘S’ will be reduced during a successive period of storm (such as one day per week).
The maximum limit of ‘S’ depends on the character of soil and soil cover complex.
The ‘S’ factor is also related with antecedent moisture condition (AMC) deter-
mined by the total rainfall in the 5 days period preceding a storm.
4.2.2 Calculation for Antecedent Moisture Condition

or Cumulative Effects of Rainfall in Last Storm
of the Year
4.2.2.1 Antecedent Rainfall Conditions and Curve Numbers

(for Ia = 0.2S)
The rain water does not completely drains down slope as an immediate effect, but
has an effect for next 5 days, on an average. So for any hydrological calculation
5 days’ antecedent rainfall is considered as important to calculate its cumulative

effects (Schwab et al. 2002).
The hydrological condition for the first day of a long duration storm is different
to that of 5th day, even if the rainfall in each day remains same. The long time for
accumulation of water to the channel is also an important factor of consideration
(Handbook of Hydrology 1972). The authors studied 6 years’ catastrophic average
rainfall for the month of May, June, July, August. It revealed that last storm
occurred in the month of September and October. July is the maximum catastrophic
rainfall month for all the years. But, 5 days’ antecedent average rainfall of 90.5 mm
(Table 4.7) for the month of Sept’ 2013 is taken into account and on the basis of
which weighted curve number is converted from AMC-II to AMC-III following
Schwab et al. (2002) conversion table (Table 4.8) to calculate surface run-off from
the Shivkhola Watershed.
4.2.3 Estimation of Discharge (in m3) from Individual

Sub-watershed
The effectiveness of Remote sensing Data using sophisticated technique of SCS

Curve Number for determining runoff and discharge of watersheds arising out of a
certain rainfall is studied by Tiwari et al. (1991) and Schwab et al. (2002). In the
present study the possibilities for conservation and treatment of surface water
draining down the six sub watersheds are analyzed. The calculation of surface run-
off from the individual sub-watersheds is associated with the following steps.
Step-I The identification and marking up of the sub-watersheds draining the
water from its upslope contributing area up to the point of junction with
main stream.
Step-II The division of the main basin into 6 sub-watersheds according to
convenience of further calculation.
Step-III Analysis of SOI Topographical map (1987), LISS III Satellite Imagery-
2002, Google Earth Image and intensive field investigation for sub-
watershed wise land use classification (Tables 4.1 and 4.9).
Step-IV Determination of Curve Number under different hydrological condition
for different hydrological soil group and for individual land use in AMC-
II following empirical curve number table.
Step-V The Curve Number in AMC II condition for each land use category is
then applied in order to estimate Weighted Curve Number (Ragan et al.
1980) for each sub watershed applying the following formula.
CN1 A1 þ CN2 A2 . . .. . .. . .CNn An

Weighted CN ¼ ð4:4Þ
A1 þ A2 . . .. . .. . .. . .. . .An
Table 4.7 Significant rainfall in post-monsoon
2007 Rain 2008 Rain 2009 Rain 2011 Rain 2012 Rain 2013 Rain
4.2 Run-off Estimation
(mm) (mm) (mm) (mm) (mm) (mm)

20th June 95.5 23rd May 103.5 10th June 222.72 7th June 125 3rd June 146.5 24th May 88.9
26th June 183.5 28th July 150 28th June 93.5 9th June 100 19th June 133 28th May 101.6
31st July 134.5 29th July 160 29th June 120.5 23rd June 203 7th July 175 27th June 111.7
3rd Oct 100 30th July 112.5 20th July 120 26th June 179 12th July 200.5 14th July 103.6
4th Oct 90 19th Aug 150 21st July 124.5 29th June 196.5 9th Sep 98.7 18th July 102
31st Aug 120.5 23rd July 100 7th July 273.5 27th Sep 95.5 25th July 92.4
17th Sep 89.5 27th July 120.5 8th July 162.5 5th Aug 114.3
3rd Oct 130 18th Aug 145.5 21st July 146.5 25th Aug 89.4
23rd Sep 106.2 28th July 148.5 28th Aug 102.6
30th July 100 16th Sep 115.5
10th Aug 191.5 25th Sep 116.3
31st Aug 107.5 26th Sep 90.5
Days 5 8 9 12 6 12
Total 603.5 1,016 1153.42 1933.5 849.2 1228.8
Avg. 120.7 127 128.15 161.12 141.53 102.4
135
Table 4.8 Conversion

Curve number for condition II Factor to convert curve
factors from different AMC
number for condition II to
antecedent moisture condition
condition Condition Condition
I III
10 0.40 2.22
20 0.45 1.85
30 0.50 1.67
40 0.55 1.50
50 0.62 1.40
60 0.67 1.30
70 0.73 1.21
80 0.79 1.14
90 0.87 1.07
100 1.00 1.00
Source Schwab et al. (2002)
1, 2, …, n; Land use of different category

CN1 Curve Number of respective land use (1)
A1 % area under respective land use.
Step-VI The Weighted C.N. are then converted to AMC III condition (as the
antecedent average rainfall corresponds to the AMC III condition) fol-
lowing the conversion Table No. 4.8 proposed by Schwab et al. (2002).
Step-VII The Potential Maximum Retention (S) in mm is then calculated applying
the formula 4.2 (Table 4.11).
Step-VIII The estimation of actual direct run-off (Q-mm) for each sub-watershed
applying the formula-2 (Table 4.11).
Step-IX The runoff is then transferred to discharge in m3 by multiplying with
sub-watershed area (Schwab 2002) (Table 4.11).
Step-X Month wise surface run-off from each sub-watershed was calculated
considering 30 years’ monthly average rainfall (since 1979–2009).
4.3 Relationship Between Land Use and Land Cover,

Hydrological Soil Group, Curve Number (Under
Average-AMC), Expected Run-off (m3)
The analysis of land use/land covers of all sub-watersheds states that the large part
of the Shivkhola watershed is covered with tea garden (4.69 km2), open forest
(3.05 km2), jungle (3.09 km2) settlement (2.26 km2) and degraded forest
(2.26 km2). Dense forest, agricultural land, roads account for about 1.2, 0.36,
Table 4.9 Hydrological soil group wise land use and land cover and determined curve number (CN) under different hydrological condition for individual
watersheds
Land use/land cover Hydrological soil Hydrological Curve number Distribution of land uses in individual
group condition (CN) watersheds (area in km2)
I II III IV V VI
Agricultural land B Poor 74 0.19 0.00 0.00 0.12 0.00 0.00
Settlement Compact D Poor 92 0.07 0.02 0.05 0.00 0.00 0.75
Disperse C Fair 80 0.55 0.02 0.10 0.06 0.06 0.25
Concrete Concrete D Poor 92 0.48 0.11 0.18 0.12 0.00 0.32
road Metalled C Fair 80 0.26 0.00 0.05 0.18 0.13 0.12
Dense forest A Good 30 0.80 0.14 0.00 0.00 0.00 0.00
Bared Steep slope C Poor 91 0.15 0.00 0.02 0.03 0.03 0.23
Surface Gentle B Poor 85 0.04 0.00 0.00 0.00 0.00 0.18
slope
Tea garden Up slope B Poor 74 0.98 0.05 0.31 0.41 0.15 0.87
Down slope A Good 62 0.37 0.07 0.30 0.08 0.30 0.24
Mixed forest Up slope B Fair 76 0.62 0.01 0.06 0.03 0.05 0.06
Down slope A Good 58 0.06 0.02 0.00 0.00 0.00 0.12
Degraded Up slope C Poor 77 0.40 0.12 0.16 0.14 0.14 0.52
forest Down slope B Poor 66 0.16 0.08 0.00 0.00 0.00 0.18
Open forest Up slope C Fair 73 0.75 0.14 0.09 0.31 0.23 0.37
Down slope B Fair 60 0.28 0.09 0.10 0.06 0.06 0.14
4.3 Relationship Between Land Use and Land Cover, Hydrological Soil Group ...
Jungle Up slope C Fair 70 0.73 0.16 0.45 0.04 0.23 0.77

Down slope B Poor 67 0.20 0.21 0.26 0.16 0.13 0.24
137
1.93 km2 respectively and mixed forest covers 1.73 km2. Concrete road, compact
settlement, bare surface (upslope), degraded forest, tea garden (upslope), agricul-
tural land, and down slope jungle depict the poor hydrological condition with less
retention and high surface run-off. Fair and good hydrological condition is expe-
rienced in settlement, metal road, dense forest, down slope tea garden, mixed forest,
open forest and jungle which encompass the hydrological soil group of C, A and B.
The sub-watershed I and VI contribute maximum area of hydrological soil group C
and D with poor hydrological condition. Curve Number (CN) is a descriptor of
surface moisture retention and run-off which is also an expression of land use/land
cover and soil characteristics. The poor hydrological condition is attributed with
higher Curve Number (CN) values and maximum surface run-off. Settlement, road,
bare surface, upslope degraded forest are attributed with the Curve Number (CN) of
more than 76 which is an indicator of minimum retention. Dense forest, down slope
jungle, mixed forest, tea garden and open forest are characterized by fair to good
hydrological condition with the Curve Number (CN) value of less than 75. Dense
forest, down slope tea garden, and down slope mixed forest experience good
hydrological condition and minimum Curve Number (CN <65) value under
hydrological soil group of A where retention is very high and run-off is very low.
Curve Number (CN) value of compact settlement, concrete road, up slope bared
surface are 92, 92 and 91 respectively with poor hydrological condition and provide
maximum run-off (Table 4.9). The sub-watershed I and VI contribute maximum
area of hydrological soil group C and D with poor hydrological condition.
Settlement, roads, bare surface, and degraded forest largely distributed in sub-
watershed VI which is followed by sub-watershed III, IV, II and V. Only sub-
watershed I (0.19 km2) and IV (0.12 km2) are experiencing agricultural land with
CN value of 74. Dense forest is distributed in the sub-watershed I and II covering
the area of about 0.84 and 0.14 km2. Steep slope of the sub-watershed I and VI is
dominated by tea garden that covers the area of about 0.98 and 0.87 km2 respec-
tively with poor hydrological condition characterized by CN value of 74. Dense
forest is distributed in the Sub-watershed I and II covering the area of about 0.84
and 0.14 km2. Steep slope of the Sub-watershed 1 and VI is mainly covered by tea
gardens and that shows poor hydrological condition.
The sub-watershed I rank 1st in terms of average areal coverage of mixed
forest, open forest and jungle which is followed by sub-watershed VI, III, II, V
and IV. It is to be assumed that as the sub-watershed I and VI covers large
upslope contributing area and are attributed with the maximum CN value, these
two are the significant run-off producing units of the Shivkhola watershed
(Table 4.9). Sub-watershed VI is considered as the watershed of poorest hydro-
logical condition followed by IV, III, and V. Watershed II shows favourable
hydrological condition.
4.4 Sub-watershed Wise Analysis of Curve Number (AMC-III) ... 139
4.4 Sub-watershed Wise Analysis of Curve Number

(AMC-III), Potential Retention (S) and Run-off
The determined Curve Number (CN) under antecedent moisture condition-III

(AMC-III) for sub-watershed I, II, III, IV, V and VI are 85.02, 73.52, 87.36, 87.76,
85.57 and 89.85 respectively. The potential retentions (Table 4.10) are very low for
the sub-watershed VI (28.69 mm), IV (35.46 mm) and III (36.75 mm) which
indicates that these 3 sub-watersheds are significant run-off contributors irrespective
of areal coverage. In terms of surface run-off, sub-watershed-I contribute
4,52,349.40 m3, sub-watershed-II 50,641.50 m3, sub-watershed-III 1,90,278.00 m3,
sub-watershed-IV 1,07,109.00 m3, sub-watershed-V 66,174.00 m3 and sub-water-
shed-VI 3,48,948.30 m3. Estimated LPIV is very high for Sub-watershed I and VI
(Table 4.10). On the basis of the estimated potentialities of run-off and landslide
potentiality index value (LPIV) an instability scale is made which reveals that sub-
watershed I is very much prone to soil erosion and slope instability, followed by VI,
IV and III. Considering the greater contributing area, sub-watershed I not only
became the most contributor of the run-off of the entire Shivkhola Watershed but
also became a spatial unit of high magnitude landslide potentiality. Running water
on steep slope in the form of turbulent, concentrated flow or sheet wash dislodge
materials from slope surface and thus deserves due importance.
4.5 Periodic (Monthly) Relationship Between Surface

Run-off and Slope Susceptibility
The month wise estimated surface run-off (Table 4.11) states that the month of July
experiences the higher amount of runoff and which is followed by August, June and
September mainly during the rainy season. The potentiality to soil erosion, soil
saturation and landslip of the sub-basin I and VI are greater because these two
contribute considerable surface run-off to the Shivkhola watershed. The time span
between October and March is attributed with lower amount of surface runoff
throughout the watershed. This time span is called as ‘the dry period’ during which
the potentiality to soil erosion, soil saturation and soil slip are very low, whereas
that are more frequent during the wet period (July, August and September). On the
basis of the surface runoff from the whole Shivkhola watershed, a year is divided
into two distinct periods:—(i) less susceptible period (October–April) and (ii) more
susceptible period (June–September). Thus all the required constructions and other
preparedness are to be completed during less susceptible period.
140
Table 4.10 Calculated CN (AMC-III), retention (S) and run-off and the determination of instability rank
Sub- Weighted Converted WCN from Maximum poten- Actual direct Run-off Landslide potential- Instabili-
watershed curve number condition AMC-II to tial retention (S- run-off (Q- (m3) ity index value ty rank
(WCN) AMC-III mm) mm) (LPIV)
I 69.69 85.02 44.75 52.66 4,52,349.40 14.01 1
II 54.06 73.52 91.48 31.85 50,641.50 5.09 6
III 72.80 87.36 36.75 57.66 1,90,278.00 9.41 3
IV 74.69 87.76 35.46 58.53 1,07,109.00 7.36 4
V 71.13 85.57 42.83 53.80 66,174.00 6.52 5
VI 76.93 89.85 28.69 63.33 3,48,948.30 11.86 2
4 Surface Run-off, Soil Erosion and Slope Instability
Table 4.11 Month wise surface run-off distribution from individual watersheds
Watershed Parameters Jan Feb Mar Apl May Jun July Aug Sept Oct Nov Dec
I Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 85.02 85.02 85.02 85.02 85.02 85.02 85.02 85.02 85.02 85.02 85.02 85.02
S 44.75 44.75 44.75 44.75 44.75 44.75 44.75 44.75 44.75 44.75 44.75 44.75
Q 0.27 1.14 15.23 56.48 255.01 677.32 943.74 732.64 537.41 138.74 0.23 2.91
Run-off 2139 9793 130826 485163 2190535 5818179 8106727 6,293,378 4,616,352 1,191,777 1,978 24,997
II Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 73.52 73.52 73.52 73.52 73.52 73.52 73.52 73.52 73.52 73.52 73.52 73.52
S 91.48 91.48 91.48 91.48 91.48 91.48 91.48 91.48 91.48 91.48 91.48 91.48
Q 0.38 0.03 5.55 34.88 215.28 629.06 893.6 683.89 490.78 106.15 0.42 0.14
Run-off 604 48 8,825 55,459 342,295 1,000,205 1,420,824 1,087,401 780,340 168,779 668 223
III Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 87.36 87.36 87.36 87.36 87.36 87.36 87.36 87.36 87.36 87.36 87.36 87.36
S 36.75 36.75 36.75 36.75 36.75 36.75 36.75 36.75 36.75 36.75 36.75 36.75
Q 0.65 1.89 18.12 61.62 262.77 668.08 952.72 741.76 545.99 145.55 0.59 4.13
Run-off 2,145 6,237 59,796 203,346 867,147 2,204,664 3,143,976 2,447,808 1,801,767 480,315 1,947 13,629
IV Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 87.76 87.76 87.76 87.76 87.76 87.76 87.76 87.76 87.76 87.76 87.76 87.76
S 35.43 35.43 35.43 35.43 35.43 35.43 35.43 35.43 35.43 35.43 35.43 35.43
4.5 Periodic (Monthly) Relationship Between Surface Run-off ...
Q 0.74 2.04 18.65 62.52 264.08 687.54 954.21 742.93 547.42 146.72 0.67 4.37
Run-off 1,354 3,733 34,129 114,412 483,266 1,258,198 1,746,204 1,359,562 1,001,779 268,498 1,226 7,997
V Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 85.57 85.57 85.57 85.57 85.57 85.57 85.57 85.57 85.57 85.57 85.57 85.57
S 42.83 42.83 42.83 42.83 42.83 42.83 42.83 42.83 42.83 42.83 42.83 42.83
Q 0.34 1.29 15.87 57.66 256.84 679.41 945.88 734.74 539.45 140.33 0.30 3.16
Run-off 418 1,587 29,042 70,922 315,913 835,674 1,163,432 903,730 663,524 172,606 369 3886.8
(continued)
141
142
Watershed Parameters Jan Feb Mar Apl May Jun July Aug Sept Oct Nov Dec
VI Rainfall 12.57 16.67 43.76 94.85 302.8 728.4 995.5 783.9 587.9 183.3 12.3 21.9
CN 89.85 89.85 89.85 89.85 89.85 89.85 89.85 89.85 89.85 89.85 89.85 89.85
S 28.69 28.69 28.69 28.69 28.69 28.69 28.69 28.69 28.69 28.69 28.69 28.69
Q 1.31 3.02 21.67 67.4 270.9 695.07 961.88 750.49 554.82 152.86 1.22 5.82
Run-off 7,218 16,640 119,401 371,374 1,492,659 3,829,836 5,299,959 4,135,200 3,057,058 842,259 6,722 32,068
CN Curve Number, S Potential Maximum Retention (mm), Q Actual direct run-off (mm), (Rainfall in mm and Run-off in m3 )
4 Surface Run-off, Soil Erosion and Slope Instability
4.6 Conclusion 143
4.6 Conclusion
The Shivkhola watershed contributes significant amount of surface run-off during

the catastrophic rainfall months of a year when slope surface gets saturated due to
minimum retention. Area under the hydrological soil group of D and C (mainly sub-
watersheds VI, I and IV) provide maximum run-off where Curve Number (CN)
value is very high. The settlement areas of sub-watershed VI, concrete as well as
metal roads and bare surface of sub-watershed VI and I are to be given proper
attention for arresting run-off through constructing catch-water drains, proper
plantation and diversion of run-off. Proper drainage from settlement area through
concrete drains is to be made to protect the slope of highest priority. Upslope
degraded forest of all the sub-watersheds are to be brought under thorough plan-
tation and proper management. July and August experience pronounced run-off
from all the sub-watersheds. The people should be made aware about the possible
consequences of the excessive run-off during monsoon that results from either
deforestation or construction. All the preparation to manage landslide through
necessary constructions, plantations, preparation of task force, and detailed plan for
monitoring of slope conditions are to be made during pre-monsoon period.
Immediate attention is to be paid for sub-watershed I and VI as these two are the
great contributors of surface run-off and experience maximum landslide occur-
rences in the whole Shivkhola watershed. Some remedial measures such as plan-
tation of first growing and rapid water holding capacity grasses over the exposed
surface, construction of horizontal as well as vertical drains over large upslope
contributing area which can reduce length of overland flow and can minimize the
effects of maximum overland flow, and plantation along first order stream to reduce
the amount of discharge are to be followed to reduce surface run-off and also to
reduce soil erosion. As a whole, deforestation should be checked immediately by
providing primary needs to all the poor income family those who collect forest
wood for fulfilling domestic as well as commercial demand.
References
Chandra S, Sharma KP (1978) Application of remote sensing to hydrology. In: Proceedings of

symposium on hydrology of rivers with small and medium catchments, Roorkee, India, vol II,
pp 1–13
Chow VT, Maidment DR, Mays LW (1988) Applied hydrology. McGraw Hill, New York
Durbede DG, Purandara BK, Sharma A (2001) Estimation of surface run-off potential of a
watershed in semi-arid environment-a case study. J Indian Soc Remote Sens 29(1 and 2):48–58
Handbook of Hydrology (1972) Soil conservation department. Ministry of Agriculture, New Delhi
Harrold LL (1957) Minimum water yield from small agricultural watershed. Am Geophys Union
Trans 38:201–208
Horn DL, Schwab GO (1963) Evaluation of rational run-off co-efficients for small agricultural
watershed. ASAE Trans 6(3):195–198, 201
Jasritia AS, Singh R (2006) Modelling un-off and soil erosion in a catchment area, using the GIS,
in the Himalayan region, India. Environ Geol 51:29–37
Kerby WS (1959) Time of concentration of overland flow. Civil Eng 29:60
Maidment DR (1972) Handbook of hydrology. McGraw-Hill, New York, pp 1–75
Musgrave GW (1955) How much of the rain enters the soil? Water: U.S. Department of
Agriculture Yearbook, Washington, DC, pp 151–159
Ragan MR, Jackson JT (1980) Run-off synthesis using LANDSAT and SCS models. J Hydraul
Div 106:667–678
Sarangi A, Madramootoo CA, Enright P, Prasher SO (2007) Evaluation of three unit hydrograph
models to predict the surface run-off from a Canadian watershed. Water Resour Manage
21:1127–1143
Schwab GO, Fangmeier DD, Elliot WJ, Frevert RK (2002) Soil and water conservation
engineering, 4th edn. Wiley, New York, pp 34–35
Singh VP (1988) Hydrologic systems: rainfall-run-off modelling, vol 1. Prentice Hall, Englewood
Cliffs
Tiwari KN, Pal DN (1991) Establishing SCS run-off curve number from IRS digital data base.
J Indian Soc Remote Sens 19(4):24–251
Tripathi MP, Panda RK, Pradhan S, Sudhakar S (2002) Run-off modeling for small watershed
using satellite data and GIS. J Indian Soc Remote Sens 30(1 and 2):39–52
U.S. Department of Agriculture, Natural Resources Conseration Service (1993) National
engineering handbook, title 210-VI. Part 630, chapters 9 and 10. Washington, DC. http://
directives.sc.egov.usda.gov/
U.S. Soil Conservation Service (SCS) (1972) Hydrology. Natural engineering handbook, Sect. 4.
GPO, Washington, DC
Zade M, Ray SS, Dutta S, Panigrahy S (2005) Analysis of run-off pattern for all major basins of
India drived using mote sensing data. Curr Sci 88(8):1301–1305
Zhang B, Govindaraju R (2003) Geomorphology based artificial neural networks (GANNs) for
estimating of direct run-off over watersheds. J Hydrol 273:18–34
Chapter 5
Geomorphic Threshold and Landslide
Abstract The present study established the link between critical rainfall (cr),
critical slope angle (cs), critical height (ch) and landslide. The critical rainfall was
estimated incorporating geo-technical parameters such as angle of internal friction
(u), slope angle (⊖), upslope contributing area (UCA), transmissivity (T), wet soil
density (ps), and density of water (pw). Cohesion (c), angle of internal friction (u),
unit weight of the materials (γ), and slope angle (⊖) were taken into account to
estimate critical slope height. The thickness of total soil (h), thickness of saturated
soil (z), wet soil density (Ps), density of water (Pw), friction angle (u) and slope
steepness (⊖) were considered to derive critical slope angle. Study attempted to
calculate critical rain to slope failure and its return period. The temporal probability
of the landslide events were estimated applying Binomial and Poisson Probability
Distribution Model based on past landslide occurrences. The probability model
suggests that occurrences of major landslides with more than 90 % certainty could
be expected in every 7.5 years.
Keywords Critical slope Critical height Critical rainfall Return period

Probability model
5.1 Introduction
Geomorphic threshold is significant parameters in analyzing the stability condition

of particular spatial unit in a quantitative way. According to White et al. (1996) ‘the
minimum or maximum level of some quantity needed for a process to take place or
a state to change is generally defined as threshold’. Varnes (1978) studied the role
of minimum intensity and duration of rainfall to cause a landslide of shallow soil
slips, debris flows, debris slides or slumps. Crozier (1997) opined a maximum
threshold, beyond which there is 100 % chances of occurrences of the process at
any time when the threshold value is exceeded. The most commonly investigated
threshold parameters such as critical slope, critical slope height and critical rainfall

146 5 Geomorphic Threshold and Landslide
(cumulative rainfall, antecedent rainfall, intensity and duration of rainfall) in rela-

tion to landslide phenomena has been attempted to identify in the present study.
Starkel (1972) for the first time, observed the geomorphic effects of an extreme
rainfall event in the eastern Himalaya (mainly Darjiling), India. Froehlich et al.
(1990), investigated the same area and found that shallow slides and slumps on
steep slope segments occured when 24 h rainfall reaches 130–150 mm or contin-
uous 3 days rainfall totals 180–200 mm. Campbell (1975), Cotecchia (1978), Caine
(1980), Innes (1983), Pomeroy (1984), Cannon and Ellen (1985), Neary and Swift
(1987) Keefer et al. (1987), Cannon (1988), Kim et al. (1991), Li and Wang (1992),
Ceriani et al. (1992), Larsen and Simon (1993), Wilson and Wieczoreck (1995),
Wieczorek (1987, 1996), Wieczorek et al. (2000), Terlien (1997, 1998), Crosta
(1998), Crozier (1999), Glade et al. (2000), Crosta and Frattini (2001), Aleotti
(2004), Guzzetti et al. (2004), Gabet et al. (2004), Giannecchini (2006), Glade
(1998), Zezere et al. (2005), Cardinali et al. (2006) and Dahal et al. (2006b) tried to
establish rainfall-intensity thresholds for predicting the slope failure accurately.
Caine (1980) first established worldwide rainfall threshold values for landslides.
Montgomery and Dietrich (1994) introduced a physically based model for the
topographic control on shallow landsliding in terms of geomorphic threshold.
Recently Guzzetti et al. (2007) reviewed rainfall thresholds for the initiation of
landslides worldwide and proposed new empirical thresholds based on the statis-
tical analysis of the relationship between rainfall and landslide occurrences. They
defined intensity-duration threshold as:
I ¼ 73:90D0:79 ð5:1Þ
where, I is the hourly rainfall intensity in millimeters (mm hr-1) and D is duration in
hours.
Brunsden et al. (1981), Manandhar and Khanal (1988), Dhital et al. (1993),
Upreti and Dhital (1996), Gerrard and Gardner (2000), Dhital (2003), Dahal et al.
(2006a), and Dahal (2006) while other works, such as Caine and Mool (1982),
Dhakal et al. (1999), De Vleeschauwer and De Smedt (2002), De Smedt (2005)
focused mainly on landslide risk assessment in Himalayan terrains by analyzing
physical properties of landslides and debris flows, effects of regional and local
geological settings, and recommendations for environmental-friendly preventive
measures.
In the Shivkhola watershed of Darjiling Himalaya, the physical processes and
human actions (formation of road-cut benches and concentration of human settle-
ment) are active on the slope in a systematic interactive combination which make
the slope steeper than repose angle and thus the instability is introduced into the
system. The present study attempts to identify the critical values of rainfall, slope
height and slope angle beyond which there is a greater probability of slope insta-
bility. The formation of road-cut benches to develop communication network
lengthens the steep slope, removes the lateral and basal support, and disturbs the
soil, favours infiltration and through-flow, helping in the increase of wet soil depth.
All those changes and their combined manifold after-effects help to generate
geomorphic threshold and the shear stress to increase over shear strength. Some-
times, moderate levels of rainfall for few days in the basin become more critical
because of weak lithological composition. Moreover, the concentration of human
settlement in Tindharia, Gayabari and Shiviter generates enormous pressure on
slope materials and favours threshold values by reducing the shearing strength. The
study shows that only 105.88 and 88.928 mm daily rainfall is the critical rain for
initiation of slide at Tindharia and Lower Paglajhora respectively. So there is every
possibility for the generation of geomorphic threshold and initiation of slide due to
hydrologic factor. In this way, there is a frequent occurrence of debris slide which
reduces the slope angle on landslide scar face to that of repose angle to attain
temporary stability through internal feed back in a process of homeostatic
adjustment.
Study involves the measurement of upslope contributing area, contour length,
slope angle, transmissivity, depth of soil, depth of the saturated soil, density of water,
wet soil density, unit weight of the materials, pore-water pressure, cohesion, and
angle of repose for the determination of ‘intrinsic threshold’ (Schumm 1977a, b)
condition such as critical slope angle, rainfall and slope height beyond which slope
materials may undergo chemical decomposition and thereby lose its former strength
and the slope may collapse without an extrinsic type of threshold being crossed. The
response to threshold crossing may induce dramatic erosion and striking changes of
the concerned landforms which is shaped primarily by the disturbances rather than by
normal events. The recoveries of such disturbances are often a long and slow process
which is mainly accomplished through Self-organized Feed Back Mechanism in the
geomorphic system.
5.2 Materials and Methods
5.2.1 Angle of Internal Friction/Angle of Repose

and Cohesion
The angle of repose is the angle at which the debris attains temporary stability on
slope is very essential to estimate as the debris slope attains instability. The
geo-technical factor like angle of repose of the debris is measured after Bloom
(1991). The tangent of angle of repose of dry granular materials is slightly greater
than, but approximately equals to the co-efficient of sliding friction of the material
or its mass friction (u) (Van Burkalow 1945; Bloom 1991). The cohesion and angle
of internal friction is measured by tri-axial compression test (Fig. 5.1) following
Mohr Stress Diagram (Fig. 5.2).
The saturated conductivity of the sail varies from 10−2 m s−1 for the soil depth
less than 0.5 m–10−5 m s−1 for soil depth between 1 and 2 m (Fenti 1992). Based on
these and other data (Metteotti 1996) estimated the transmissivity (T) of saturated
soil to lie between 5 and 30 m−2 day−1, with a mean value of 15 m−2 day−1.
Fig. 5.1 Tri-axial soil testing mechanism
Fig. 5.2 Mohr stress diagram
The upslope contributing area (b) was estimated considering multiple flow direction
where the cumulative flow at a point (surrogated by the upslope area) should be
distributed among more than one neighboring down slope pixel (Borga et al. 1998).
5.2.2 Determination of Threshold Slope Angle for Initiation

of Slide
The formation of road-cut benches introduces the steep back slope and the slope on
the landslide scar is greater than the angle of repose. This situation is mainly
responsible for instability. The stability equation for a mass of loose, friable
cohesion less debris after (Melnikov and Chensokov 1969) is as following.
5.2 Materials and Methods 149
Shear Stress W:cos h tan u

Safety Factor ¼ ¼ 1 ð5:2Þ
Shear Strength W:sin h
tan u Co-efficient of friction

u Angle of repose
W Weight of soil
θ Slope on scar face.
tan u
or; 1 ð5:3Þ
tan h
or; tan u tan h ð5:4Þ
i.e. (Angle of repose/Angle of internal friction) ≥ (Slope on scar face).

In the present study the slope angle of Tindharia and Lower Paglajhora area is
53°20′ and 48°20′ which always outweighs the angle of repose. The other indefinite
slope stability model for cohesion less material and slope parallel seepage after
(Borga et al. 1998) also supports the Eq. (5.5).

h Ps tan h
¼ 1 1 ð5:5Þ
z Pw tan u
H Thickness of total soil

Z Thickness of saturated soil
Ps Wet soil density
Pw Density of water
u Angle of repose
θ Slope on scar face.

tan h
For maintaining the stability ‘h’ is needed to be greater than ‘z’ and 1 tan u
should be positive (Table 5.1).

tan h
So 1 ð5:6Þ
tan u

tan h
or; 1 ð5:7Þ
tan u
or; tan u tan h ð5:8Þ
or; u h ð5:9Þ
(Angle of repose) ≥ (Slope on scar face)

Table 5.1 Calculation of threshold angle

Melnikov and Chenso- Borga et al. (1998) Carson (1977a, b, c)
kov (1969)

(W cos θ tan u)/W tan h
¼ PPws 1 tan tan α = (1-u/γz cos2 α) tan u)
u 1
h
z
sin θ > 1 (Introducing pore water pressure
or, u h (Dry condition) in saturated condition)
(Dry condition) h—Thickness of total soil α = Threshold angle of failure
tan φ = Co-efficient z—Thickness of saturated u = Pore water pressure on potential
of friction soil sliding surface
u = Angle of repose Ps—Wet soil density z = Depth of potential shear plane
W = Weight of soil Pw—density of water γ = Bulk unit weight of the sliding materials
θ = Slope on scar face u—Angle of repose u = γw z cos2 α) (γw = Unit weight
θ—Slope on scar face of soil water)
Cohesion is ignored in the long term due to
weathering and unloading processes
u = 133.75 g/cm2
5.2.3 Determination of Threshold Rainfall to Initiate Slide
Campbell (1975), Caine (1980), Larsen and Simon (1993) established that the
empirical threshold condition to initiate landslide refers to relational value based on
statistical analysis of the relationship between rainfall and landslide occurrences
where as the physical thresholds are usually determined with the help of hydrologic
and stability models that take into consideration of various attributes such as slope
(b), upslope contributing area (a), transmissivity (T), wet soil density (ps), density
of water (pw), slope angle (θ), angle of internal friction (u), relation between rainfall
and pore-water pressure etc. In the absolutely unstable condition the role of rain-
water to initiate the threshold for sliding could be determined. If the hydrological
factors like rain fall and seepage flow are considered the threshold condition for
absolute instability can be predicted. The critical rainfall was derived (cr) after
Borga et al. (1998) (Eq. 5.10).

b:ps tan h
rcr ¼ T:sin h 1 ð5:10Þ
apw tan u
5.2.3.1 Estimation of Average Catastrophic Rainfall to Obtain Return

Period and Probabilistic Recurrence Interval of the Critical
Rain
The Selim Hill Tea Estate situated 250 m North West of Tindharia registered 52 days
having more than the critical rain fall to initiate threshold condition during the years
2005–2010 (Table E.1, Appendix E). Average day wise rainfall for catastrophic days
in 2005, 2006, 2007, 2008, 2009, and 2010, are 120.7, 127, 128.5, 161.12, 141.53
and 102.4 mm respectively which are greater than the estimated threshold rainfall for
initiating slide at Tindharia and Lower Paglajhora. This indicates high possibility of
frequent slide in those places.
5.2.4 Threshold Slope Height to Initiate Slide
(Skempton and Hutchinson 1969) in their experience in the development of steep

slope and its evolution through slides in the glacial till of County Durham found a
critical slope height of 45 m at 30–35° steepness. Terzaghi (1962) calculated the
critical slope height of the cliff at which failure occurs. In the present study critical
height for slope failure is determined after Culmann (1866) and Carson (1977a, b, c).
4c sin h:cos u
Critical slope height ðhc Þ ¼ ð5:11Þ
c 1 cosðh uÞ
(Culmann 1866; Carson 1977a, b, c)
hc Critical (threshold) slope height

c/ Cohesion
u Angle of internal friction
γ Unit weight of materials
θ Slope angle.
Sc
Critical slope height ðhc Þ ¼ ð5:12Þ
W
(Terzaghi 1962)
Hc slope height of the cliff at which failure occurs

Sc Compressive strength of the rocks
W Unit weight of the rocks.
5.2.5 Application of the ‘Poisson’ and ‘Binomial’ Probability

Distribution Models to Estimate the Temporal
Probability of Landslide Events
The frequent occurrence of landslides in the unstable terrain of Skivkhola watershed

and their continuous monitoring through intensive field investigation and in con-
sultation with others research works made by Starkel (1972), Basu and Sarkar
(1985, 1988), Basu and Maiti (2001), Maiti (2007a, b), Ghosh et al. (2009), Sarkar
(2011) and author himself have provided most reliable earlier landslide frequency
data since 1968 (Table D.1, See Appendix D). To determine the temporal proba-
bility of rain-induced landslide events the exceedance probability of one or more
landslides were attempted by considering the landslide as random point events.
Two major discrete probability distribution models such as the ‘Poisson distribu-
tion’ and the ‘Binomial distribution’ were mostly applied to calculate the exceed-
ance probability of landslide (Coe et al. 2004; Crovelli 2000). According to
‘Poisson distribution model’ the occurrences of landslide events that is experi-
encing ‘n’ landslides during the time ‘t’ could be expressed by:
ðktÞn
P½NL ðtÞ [ ¼ n ¼ eðktÞ ð5:13Þ
n!
where, λ = average rate of landslide occurrence.

n = 0, 1, 2, 3, … n.
So, the exceedance probability or the probability of experiencing landslide
events during the time ‘t’ could be expressed as
P½NL ðtÞ [ ¼ 1 ¼ 1 P½N L ðtÞ ¼ 0 ¼ 1 ekt ¼ 1 el

t
ð5:14Þ
where, l ¼ 1k and µ = mean recurrence interval between successive landslide events.

In the same way, the exceedance probability could be assessed by using the
binomial probability distribution model with the help of following expression.

t 1 t
P½NL ðtÞ [ ¼ 1 ¼ 1 P½NLðtÞ ¼ 0 ¼ 1 ð1 PÞ ¼ 1 1 ð5:15Þ
l
To estimate the temporal probability of the landslide events in the Shivkhola

watershed, the mean recurrence interval of known landslide event years was
deduced that is 2.75 (16 known events year in 44 years). In the same way, mean
recurrence interval of known major landslide events was also deduced that is 3.66
(12 major events year in 44 years). Then, both Poisson and Binomial distribution
models were being applied to determine the exceedance probability.
5.3 Result and Discussion
5.3.1 Geo-technical Properties and Critical Slope Angle
The investigation with the temporal change in the slide scar reveals that the slope
evolution is subjected to a complex interaction between physical and anthropogenic
processes. The human actions in various developmental activities lead to the
development of geomorphic threshold in the form of slope steepness, slope height
and threshold rainfall.
5.3 Result and Discussion 153
Table 5.2 Calculated geotechnical parameters of two major landslide locations

Sl.No. Parameters Lower paglajhora Tindharia
1 Major principal stress (kg/cm2) 1.83 2.553
2 Minor principal stress (kg/cm2) 0.76 0.92
3 Normal stress (kg/cm2) 1.10 1.4044
4 Angle of rupture (Degree) 35°30′ 33°
5 Angle of internal friction (Degree) 21° 24°
6 Cohesion (kg/cm2) 0.06 0.06
7 Shear strength (kg/cm2) 0.5 0.73
The derived geo-technical parameters of two major landslide locations show

more or less identical results (Table 5.2). The thickness of the soil and that of the
saturated soil during monsoon are measured to be 4.5 m (Tindharia T.E.) and
7.25 m (Lower Paglajhora). The wet soil buck density is measured to be 1.96 g/cc
and density of water is 1.07 g/cc. The angle of internal friction varies from 21° to
26° with an average of 24°.
In the present study the angle of repose in sun dry condition for the concerned
material varies from 21–26°. The basic requirement for the short term stability of
the slope at marginal escarpment of Tindharia and Lower Paglajhora are to maintain
the slope angle to be nearer or less than 21° (Table 5.3). A steep slope will decline
by slope failure to an angle of repose slope to attain short term stability. This
concept leads to the concept of limiting or Threshold slope angle. It is clearly
observed from the Fig. 5.3 that middle section, extreme lower most part, Sep-
oydhura and some northern marginal parts are registering the slope angle of less
than 24°. Large parts of Tindharia, Gayabari, 14 Miles Bustee, Lower Paglajhora
and Shiviter are facing the slope angle of more than 24° and also are considered as
the most vulnerable part of the Shivkhola Watershed. Hill Cart Road is passing
through the escarpment slope which is greater than threshold slope (55–67°) and
also exhibiting as very high landslide prone section of the watershed. Analysis also
reveals that more than 60 % area of the Shivkhola watershed experiences above
threshold slope angle of 24° (Figs. 5.3 and 5.4). In the Shivkhola Watershed high to
very high landslide hazard risk is being found at the place with slope angle between
24° and 40°. Low to moderate level of risk is observed in the area pertaining to the
slope angle of less than 24°. It could be demonstrated that 50 % area of the
Shivkhola basin is dominated by the high intensity of landslide with a slope more
than 24°.
Table 5.3 Threshold slope angle

Methods Melnikov and Borga et al. Carson (1977a, b, c)
Chensokov (1969) (1998)
Threshold θ = 21–26° θ = 21–26° α = 9°51′ (Considering pore
slope angle water pressure)
Fig. 5.3 Areal coverage of below and above of threshold slope angle (24°)
Fig. 5.4 Pixel wise distribution of slope angle

5.3.2 Calculated Critical Rainfall to Initiate Debris Slide
The calculated critical rainfall (Table 5.4) of two major landslide prone parts of the
Shivkhola watershed is 105.88 mm/day (Tindharia T.E.) and 88.93 mm/day (Lower
Paglajhora).
5.3.2.1 Return Period of Rainfall
The return period of the total of the catastrophic rainfall and average daily rain of
catastrophic days was calculated on the basis of the rainfall data during 2001–2006
following Gumbel (1954) (Table 5.5).
T ¼ ðN þ 1Þ=m ð5:16Þ
T Return period
N No of years
m Rank in ascending order
(Gumbel 1954)
5.3.2.2 Probabilistic Recurrence Interval of Rainfall
The calculation of recurrence interval of the total of the catastrophic rainfall and
average daily catastrophic rain of days recording more than the calculated threshold
rain is done by log probability law following Chow (1951), (1954) and Schwab
et al. (2002).
Xc ¼ x ð1 þ CvKÞ ð5:17Þ
Xc Calculated rainfall
X mean value
Cv Coefficient of variation
K Log probability Frequency Factor (calculated from the table of Chow 1954).
The daily average catastrophic rain (more than the calculated threshold) that can
be experienced at a recurrence interval of 20 years (with 5 % probability) is
164.97 mm and that at a recurrence interval of 5 years (with 20 % probability) is
131.793 mm (Table 5.6).
The calculation shows that 105.88 and 88.928 mm daily rainfall is the threshold
rain for Tindharia and Paglajhora respectively and the analysis of return period
shows that 120.7 mm daily rainfall can occur at a recurrence interval of 1.4 years
156
Table 5.4 Critical rainfall (mm/day) for setting instability (Borga et al. 1998)
Location Transmissivity (T) Slope Contour Run-off Wet soil density Density of Friction Critical
(m−2 day−1) (θ) length (m) area (g/cc) water (g/cc) angle (°) rainfall
Lower 15 48°20′ 22.00 968 1.96 1.96 21 88.928
Paglajhora
Tindharia T.E. 15 53°20′ 27.00 1404 1.96 1.96 21 105.88
5 Geomorphic Threshold and Landslide
Table 5.5 Return period of catastrophic rainfalls after Gumbel (1954)
Year No of days of Total of the Arranged in Rank T = (N + 1)/m Average Arranged in Rank T = (N + 1)/m
catastrophic catastrophic descending (Gumbel daily rain descending (Gumbel

rainfall rainfall order (1954)) order (1954))
2001 5 603.5 1933.5 1 7 120.7 161.12 1 7
2002 8 1016 1228.8 2 3.5 127 141.53 2 3.5
2003 9 1153.42 1153.42 3 2.33 128.15 128.15 3 2.33
2004 12 1933.5 1016 4 1.75 161.12 127 4 1.75
2005 6 849.2 849.2 5 1.4 141.53 120.7 5 1.4
2006 12 1228.8 603.5 6 1.16 102.4 102.4 6 1.16
Mean 1130.7367 130.1500
a
S.D. 452.7617 19.8087
b
C. 0.4000415 0.152199
V.
a
S.D. Standard Deviation
b
C.V. Coefficient of Variation
157
Table 5.6 Amount of rain

P% T (Years) K Xc (mm)
fall at certain probability and
with specific return period 99 1.01 −2.001 90.539
(after Chow 1951, 1954) 50 2 −0.083 128.507
20 5 0.083 131.793
5 20 1.759 164.971
1 100 2.669 182.985
following Gumbel (1954) and 128.507 mm daily rain has a recurrence interval of
2 years with 50 % probability following Chow (1951), (1954). That means there is
every possibility for the generation of geomorphic threshold for initiation of slide
due to hydrologic factor. At Paglajhora the critical rainfall is 88.93 mm which is
less than the estimated rainfall of 90.54 mm at the recurrence interval of 1.01 year
with 99 % probability. So it can be inferred that Paglajhora is a place of higher
probability of rainfall triggering landslide phenomena in every rainy season (Maiti
2007a, b).
5.3.2.3 Relationship Between Major Landslide Events and Triggering

Antecedent Rainfall in the Shivkhola Watershed (Based on July
Rainfall)
A relationship between antecedent cumulative rainfall and landslide events of 1993,

1998, 2003, 2007 and 2010 was established on the basis of the data recorded from
earlier research works done by Ghosh et al. (2009b) and the collection of rainfall
data from nearby Selim Hill Tea Estate by author himself. Only two days ante-
cedent cumulative rainfall of 211.3 mm invited the slope failure at several places of
Tindharia and Gayabari and Mahanadi (Table 5.7) in 1993. The 1998 landslide
event took place due to 300–600 mm cumulative rainfall in the past 2/3 days. The
2 days’ antecedent cumulative rainfall of 390 mm was responsible for 1998 land-
slide events. The major event of 2003 happened due to incessant rainfall of 500 mm
in 2 days. 17 and 18th July, 2007 received rainfall of 124.5 and 100 mm respec-
tively. These 2 days’ antecedent cumulative rainfall of 224.5 mm caused havoc
slope failure at Tindharia and Upper and Lower Paglajhora. Again 2007 faced
landslide events on 8th September when 6, 7 and 8th September’s antecedent
cumulative rainfall amount was 275 mm.
In 2010, major and prominent landslide events happened as a result of 5 days’
rainfall of 345 mm at 14 Mile near lower Paglajhora, Nurbong, Gitingia, and
Shiviter. Antecedent Cumulative rainfall induced landslide analysis shows that the
continuous and uniform rate of minimum amount of rainfall (approx. less than
80 mm/day) for few consecutive days can cross the geomorphic threshold and can
introduce slope instability condition.
Table 5.7 Major landslide events and antecedent cumulative rainfall

Landslide events 1 day 2 day 3 day 4 day 5 day
location
Tindharia and gaya- 110 mm 211.3 mm 265 mm 305 mm 340 mm
bari and mahanadi 2nd July, 3rd July,
1993 1993
(Landslide)
Chunabhati, tidharia, 150 390 450 485 520
paglajhora, mahanadi 6th July, 7th July 8th July
(along NH-55), 1998 (Landslide) (Landslide)
jogmaya
Gayabari T.E., Along 197 mm 500 mm 527 565 590
NH-55, tindharia and 8th July, 9th July,
shiviter. paglajhora 2003 2003
(Landslide)
Tindharia and upper 124.5 224.5 255.5 300 315
and lower paglajhora 17th July, 18th July,
2007 2007
(Landslide)
Tindharia, lower pag- 91 mm 187 mm 275 mm
lajhora and shiviter 6th Sep- 7th Sep- 8th Sep-
tember tember tember
2007 2007 2007
(Landslide)
14 Mile near lower 111.7 mm 345 mm 245 295 365
paglajhora, nurbong, 15th June, 16th June,
gitingia, shiviter 2010 2010
(Landslide)
5.3.3 Calculated Critical Height to Initiate Slide
The determined critical slope height after Culmann (1866) at Tindharia and Lower
Paglajhora are 5.89 and 7.80 m respectively. The height of the vertical back wall
along the road should be restricted to almost 5.89 m at both the places (Table 5.8) in
Tindharia. The landslide affected area with more than 5.89 m height corresponding
to the average threshold slope angle of 24° must be identified and shaped to that of
safe height (below 5.89 m.). At Paglajhora, the critical height for slope failure is
7.80 m.
5.3.4 Temporal Probability of Landslide Events
The occurrences of major landslide events with more than 90 % certainty could be
expected in every 7.5 years in case of Poisson distribution model. If we consider
landslide event, then we can say that it can be expected in every 13 years with
Table 5.8 Critical height for initiation of slide

Sl. No. Slope parameters Tindharia Lower paglajhora
1 Upslope contributing area (a) 1404 m2 968 m2
2 Contour length (b) 27 m 22 m
3 Slope angle (⊖) 53°20′ 48°20′
4 Angle of internal friction (u) 21° 21°
5 Transmissivity (T) 15 m−2 day−1 15 m−2 day−1
(Borga et al. 1998) (Borga et al. 1998)
6 Wet soil buck density (Ps) 1.96 g/cc 1.96 g/cc
7 Density of water (PW) 1.07 g/cc 1.07 g/cc
8 Cohesion (kg/cm2) 0.06 0.06
9 Critical height for initiation of slide 5.89 m 7.80 m
Culmann (1866)
c0 sin h cos /
hc ¼ 4c1cos ðh/Þ
10 Critical height for initiation of slide 9.30 m 9.30 m
Terzaghi (1962)
Hc ¼ ScW
0.9
0.8
P
R
O 0.7
B
A
B
0.6
I
L 1
I
0.5
T
Y 2
0.4
3
0.3
4
0.2
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25
Return period in years
Fig. 5.5 Landslide probability and its return period

100 % certainty. In case of Binomial distribution the 100 % certainty of the major
landslide events are to be expected at the return period of 19 years (Fig. 5.5). In the
mountain region physical and anthropogenic processes are active on slope in an
interactive combination. Construction of settlement, road and associated defores-
tation destabilize soil and slope. Slope is steepened, soil becomes loose and friable,
lateral support is removed, soil becomes saturated by hydrological intervention. All
these together lead to instability and threshold condition are achieved. Ultimately
slope failure occurs and that helps to achieve temporary stability (Fig. 5.6).
ANTHROPOGENIC
PROCESS
Concentration of
Settlement
Road-cut Bench
Reduction in
Plying of vehicles soil strength I
N
S
Roof-top & Waste Steepening
T
water disposal of Slope A
B
I
Deforestation L
I
Increase in
T
slope length
Y
S
L Reduction
OR
I Of Slope
D Through EQUILI
PHYSICAL Reduction in T
I Internal BRIUM
PROCESSES lateral and H
N Feedback
basal E
I
support S
N
H
G
SLOPE O
L
Increase in D
depth of
Saturated of G
soil E
N
SOIL/DEBRIS E
Increase the
depth of soil R
A
Infiltrated T
Through I
Flow O
N
WATER Soil becomes

Reduced
loose &
Surface
friable
flow
Fig. 5.6 Homeostatic adjustment through self-organized internal feedback in a complex system of
interaction among physical and anthropogenic process
5.4 Conclusion
Threshold slope angle for two major landslide locations i.e. Tindharia and Pag-
lajhora varies from 21° to 26°. Large parts of Tindharia, Gayabari, 14 Miles Bustee,
Lower Paglajhora and Shiviter are facing the slope angle of more than average
threshold slope angle and shows absolute instability. It is also examined that more
than 60 % area of the Shivkhola Waterhsed is beyond the threshold slope angle.
The calculated critical rainfall at Tindharia and Lower Paglajhora are 105.88 and
88.93 mm/day respectively. 120.7 mm daily rainfall can occur at a recurrence
interval of 1.4 years following Gumbel (1954) and 128.507 mm daily rain has a
recurrence interval of 2 years with 50 % probability following Chow (1951, 1954).
That means there is every possibility for the generation of geomorphic threshold for
initiation of slide due to hydrologic factor in Shivkhola watershed. Antecedent
Cumulative rainfall induced landslide analysis shows that the continuous and
uniform rate of minimum amount of rainfall (approx. less than 80 mm/day) for few
consecutive days can cross the geomorphic threshold and can introduce slope
instability condition. On the other hand, one or two day’s heavy showers that are
more than 200 mm/day rain may cause devastating slope failure in the Shivkhola
Watershed. Critical slope height with 48–53° steepness may range from 5.89 to
7.80 m. Special care should be taken to reduce the height of back wall along the
main road to that of 6.00 m of less. The Interaction between physical and
anthropogenic processes leads to the initiation of geomorphic threshold that leads to
slope failure reducing the slope through internal feedback mechanism and may
proceed towards the new state of equilibrium (Fig. 5.6).
References
Aleotti P (2004) A warning system of rainfall-induced shallow failure. Eng Geol 73:247–265
Indian J Power River Valley Dev 190–194
Basu SR, Sarkar S (1988) Ecosystem vis–a–vis Landslides, a case study in Darjeeling Himalayas.
Impact of Development on Environment (Ed). The Geog Soc India, Cal II:45–53
Bloom AL (1991) Geomorphology, a systematic analysis of the Cenozoic Landforms. Prentice
Hall of India Pvt. Ltd., New Delhi, pp 76–177
Borga et al (1998) Shallow Landslide hazard assessment using a physically based model and
digital elevation data. J Environ Geol 35(2–30):81–88
Brunsden D, Jones DKC, Martin RP, Doornkamp JC (1981) The geomorphological character of
part of the Low Himalaya of Eastern Nepal. Z Geomorph NF Suppl-Bd 37:25–72
Caine N (1980) The rainfall intensity–duration control of shallow landslides and debris flows.
Geogr Ann 62A:23–27
Caine N, Mool PK (1982) Landslides in the kolpu khola drainage, middle mountain nepal. Mt Res
Dev 2:157–173
References 163
Campbell RH (1975) Soil slips, debris flows, and rainstorms in the santa monica mountains and
vicinity, Southern California. US Geol Surv Prof Paper 851:1–20
Cannon SH, Ellen SD (1985) Rainfall conditions for abundant debris avalanches, San Francisco
Bay region. Calif Geol 38(12):267–272
Cannon SH (1988) Regional rainfall-threshold conditions for abundant debris-flow activity. In:
Ellen SD, Wieczorek GF (eds) Landslides, floods, and marine effects of the storm of January
3–5, 1982, in the San Francisco Bay Region, California. US Geol Surv Prof Paper 1434:27–34
Cardinali M, Galli M, Guzzetti F, Ardizzone F, Reichenbach P, Bartoccini P (2006) Rainfall
induced landslides in December 2004 in Southwestern Umbria, Central Italy. Nat Hazards
Earth Syst Sci 6:237–260
Carson MA (1977a) Application of the concepts of Threshold slopes to the laramie mountains,
Wyoming. Inst Br Geogr Spec Publ 3:31–48
Carson MA (1977b) Threshold and characteristic angles of straight slopes. In: Proceedings of the
4th Guelph symposium on geomorphology, Norwich Geo Books, pp 19–34
Carson MA (1977c) Angle of repose, angles of shearing resistance at angle of talus slopes. Earth
Ceriani M, Lauzi S, Padovan N (1992) Rainfall and landslides in the Alpine area of Lombardia
Region, central Alps, Italy. In: Interpraevent international symposium, Bern, vol. 2, pp. 9–20
Chow VT (1951) General formula for hydrologic frequency analysis. Am Geophys Union Trans
32:231–237
Chow VT (1954) The long-probability law and its engineering applications. ASCE 80 (Separate
No. 536)
Cotecchia V (1978) Systematic reconnaissance mapping and registration of slope movements. Bull
Int Ass Eng Geol 17:5–37
Coe JA, Michael JA, Crovelli RA, Savage WZ, Laprade WT, Nashem WD (2004) Probabilistic
assessment of precipitation-triggered landslide using historical records of landslide occurrence,
Seattle Washington. Environ Eng Geosci X 2:103–122
Crosta G (1998) Regionalization of rainfall threshold: an aid to landslide hazard evaluation.
Environ Geol 35(2–3):131–145
Crosta G, Frattini P (2001) Rainfall thresholds for triggering soil slips and debris flow. In:
Proceeding of the EGS 2nd plinius conference 2000, mediterranean storms, Siena, pp 463–488
Crovelli RA (2000) Probability models for estimation of number of coasts of landslides. USGS,
Denver, Colorado
Crozier MJ (1997) The climate-landslide couple: a southern hemisphere perspective. In: Matthews
JA, Brunsden D, Frenzel B, Glaeser B, Weiss MM (eds) Rapid mass movement as a source of
climatic evidence for the Holocene. Gustav Fischer Verlag, Stuttgart, pp 333–354
Crozier MJ (1999) Prediction of rainfall-triggered landslides: a test of the antecedent water status
model. Earth Surf Proc Land 24:825–833
Culmann C (1866) Graphische Statik, Zurich
Dahal RK, Hasegawa S, Masuda T, Yamanaka M (2006a) Roadside slope failures in Nepal during
torrential rainfall and their mitigation. In: Marui H, Marutani T, Watanabe N, Kawabe H,
Gonda Y, Kimura M, Ochiai H, Ogawa K, Fiebiger G, Heumader J, Rudolf-Miklau F,
Kienholz H, Mikos M (eds) Proceeding of the Interpraevent international symposium, niigata
2006, Disaster mitigation of debris flow, slope failures and landslides, vol 2. Universal
Academy Press, Tokyo, pp 503–514
Dahal RK, Hasegawa S, Yamanaka M, Nishino K (2006b) Rainfall triggered flow-like landslides:
understanding from southern hills of Kathmandu, Nepal and northern Shikoku, Japan. In:
Proceedings of the 10th international congress of IAEG, The geological society of London,
IAEG2006 Paper number, vol 819 pp 1–14 (CD-ROM)
Dahal RK (2006) Geology for technical students—a textbook for bachelor level students. Brikuti
Academic Publication, Exhibition Road, Kathmandu, Nepal 756 pp
De Vleeschauwer C, De Smedt F (2002) Modeling slope stability using GIS on a regional scale.
In: Proceedings of the first geological belgica international meeting, vol 12. Aardkundige
Mededelingen, leuven, 11–15 Sept 2002, pp 253–256
De Smedt F (2005) Slope Instability analysis using GIS on a regional scale: a case study of
Narayanghat-Mungling highway section, Nepal, a dissertation report presented at Universiteit
Gent. Vrije Universiteit Brusssel, Belgium
Dhakal AS, Amada TK, Aniya M (1999) Landslide hazard mapping and application of GIS in the
Kulekhani watershed. Nepal, Mt Res Dev 19(1):3–16
Dhital MR, Khanal N, Thapa KB (1993) The role of extreme weather events, mass movements,
and land use changes in increasing natural hazards, A Report of the preliminary field
assessment and workshop on causes of recent damage incurred in southcentral Nepal,
ICIMOD, Kathmandu, 19–20 July 1993, 123 pp
Dhital MR (2003) Causes and consequences of the 1993 debris flows and landslides in the
Kulekhani watershed, central Nepal. In: Rickenmann D, Chen C-L (eds) Proceedings of the 3rd
International conference debris-flow hazards mitigation: mechanics, prediction and assessment,
vol 2. Millpress, Rotterdam, Netherlands, pp 931–942
Fenti V (1992) Indagini geologic-techniche sull’ area del dispositivo di misura (in Italian). In:
Marchi L (ed) II Basino attrezzato del Rio Cordon, Quadevni di Ricerca, n. 13, Ragione
Veneto. Dipartimento Foreste, Venezia Mestre: (pp 109–122)
Froehlich W, Gil E, Kasza I, Starkel L (1990) Thresholds in the transformation of slopes and river
channels in the Darjeeling Himalaya. India Mt Res Dev 10(4):301–312
Gabet EJ, Burbank DW, Putkonen JK, Pratt-Sitaula BA, Oiha T (2004) Rainfall thresholds for
landsliding in the Himalayas of Nepal. Geomorphology 63:131–143
Gerrard J, Gardner RAM (2000) Relationships between rainfall and landsliding in the Middle
Hills, Nepal. Norsk Geogr Tidsskr 54:74–81
inventory maps in a data-scarce environment; case study around Kurseong, Darjiling District,
West Bengal, India. In: Malet P, Remaitre A, Bogaard T (eds) Landslide processes: from
European centre on geomorphological hazards (CERG), Strasbourg, France, pp 37–44
Giannecchini R (2006) Relationship between rainfall and shallow landslides in the southern Apuan
Alps (Italy). Nat Hazards Earth Syst Sci 6:357–364
Glade T, Crozier M, Smith P (2000) Applying probability determination to refine landslidetrig-
gering rainfall thresholds using an empirical Antecedent Daily Rainfall Model. Pure Appl
Geophys 157:1059–1079
Gumbel EJ (1954) Statistical theory of extreme values and some practical applications. Applied
mathematics series 33. US bureau of standards, Washington, DC
Guzzetti F, Cardinali M, Reichenbach P, Cipolla F, Sebastiani C, Galli M, Salvati P (2004)
Landslides triggered by the 23 November 2000 rainfall event in the Imperia Province, Western
Liguria, Italy. Eng Geol 73:229–245
Guzzetti F, Peruccacci S, Rossi M, Stark CP (2007) Rainfall thresholds for the initiation of
landslides in central and southern Europe. Meteorol Atmos Phys. doi10.1007/s00703-007-
0262-7. Accessed 7 Aug 2007
Innes JL (1983) Debris flows. Prog Phys Geog 7:469–501
Keefer DK, Wilson RC, Mark RK, Brabb EE, Brown WM, Ellen SD, Harp EL, Wieczorek GF,
Alger CS, Zatkin RS (1987) Real-time warning during heavy rainfall. Science 238
(4829):921–925
Kim SK, Hong WP, Kim YM (1991) Prediction of rainfall triggered landslides in Korea. In: Bell
DH (ed.) Landslides, vol 2. A.A. Balkema, Rotterdam, pp 989–994
Larsen MC, Simon A (1993) A rainfall intensity-duration threshold for landslides in a
humidtropical environment. P R Geogr Ann 75(1–2):13–23
Li T, Wang S (1992) Landslide hazards and their mitigation in China. Science Press, Beijing 84 pp
References 165
Maiti RK (2007a) Irrational resource extraction introducing instability in slope and hydrodynam-
ics- a case study at lish-chunkhola basin, Darjiling Indian. J Geograph Environ 8, 9:41–51
Vidyasagar University
Maiti R (2007b) Identification of potential slope failure zones of shiv-khola watershed; Darjiling
Himalaya, through critical analysis of slope instability- a step towards rational and scientific
management of land, soil and water, UGC Sponsored Minor Research Project [F.31-210/2005
(31.03.2007)]
Manandhar IN, Khanal NR (1988) Study on landscape process with special reference to landslides
in Lele watershed, central Nepal. Report submitted to Research Division, Tribhuvan
University, unpublished, 53 pp
Metteotti G (1996) Valutazione del rischio di franosita per unbacino di tipo alpino (in Italian), Ph.
D. Dissertation, University of Padova, Italy
Melnikov M, Chensokov M (1969) Safety in open cast mining. Mir Publications, Moscow
shallow landsliding. Water Resour Res 30(4):1153–1171
Neary DG, Swift Jr LW (1987) Rainfall thresholds for triggering a debris avalanching event in the
southern Appalachian Mountains. In: Costa JE, Wieczorek GF (eds) Debris flows/avalanches:
process, recognition and mitigation. Geol Soc Am Rev Eng Geol 7:81–92
Pomeroy JS (1984) Storm-induced slope movements at East Brady, northwestern Pennsylvania.
US Geol Surv Bull 1618:16 pp
Sarkar S (2011) Evolution of Paglajhora slump valley in the shivkhola basin, The Darjiling
Himalaya, India. Geogr Pol, 84(Special Issue Part-2):117–126
Schumm SA (1977a) The fluvial system. Willy, New York
Schumm SA (1977b) Drainage basin morphology, bechmark papers in geology, 4. Hutchinsons
and Ross, Pennsylvannia, Dowden
Schwab et al. (2002) Soil and water conservation Engineering. Wiley, New York, pp 18–47
Skempton AW, Hutchinson JN (1969) Stability of natural slope and embankment section. In:
Proceedings of the 7th international congress soil mechanics engineering mexicom,
pp 291–340
Starkel L (1972) The role of catastrophic rainfall in the shaping of the relief of the lower Himalaya
(Darjeeling Hills). Geogr Pol 21:103–147
Terlien MTJ (1997) Hydrological landslide triggering in ash covered slopes of Manizales
(Columbia). Geomorphology 20:165–175
Terlien MTJ (1998) The determination of statistical and deterministic hydrological landslide
triggering thresholds. Environ Geol 35(2–3):124–130
Terzaghi K (1962) Stability of steep slopes on hard unweathered rock, Geothnique 12:251–270
Upreti BN, Dhital MR (1996) Landslide studies and management in Nepal. ICIMOD, Nepal 87 pp
Van Burkalow A (1945) Angle of repose and angle of sliding friction; an experimental study. Geol
Varnes DJ (1978) Slope movement types and process. In Schuster RL, Krizek RJ (eds) Landslides
analysis and control. Special Report 176, Transportation Research Board, National Academy
of Sciences, Washington DC, pp 12–33
White ID, Mottershead DN, Harrison JJ (1996) Environmental systems, 2nd edn. Chapman and
Hall, London 616 pp
Wieczorek GF (1987) Effect of rainfall intensity and duration on debris flows in central Santa Cruz
Mountains, California. In: Crosta G, Wieczorek GF (eds) Debris flows/avalanches: processes,
recognition and mitigation. Reviews in Engineering Geology, vol 7. Geological Society of
America, pp 23–104
Wieczorek GF (1996) Landslide triggering mechanisms. In: Turner AK, Schuster RL (eds)
Landslides: investigation and mitigation, Transportation Research Board, Special Report 247.
National Research Council, Washington, pp 76–79
Wieczorek GF, Morgan BA, Campbell RH (2000) Debris flow hazards in the blue ridge of central
Virginia. Environ Eng Geosci 6(1):3–23
Wilson RC, Wieczorek GF (1995) Rainfall threshold for the initiation of debris flow at La Honda,
California. Environ Eng Geosci 1(1):11–27
Zezere JL, Trigo RM, Trigo IF (2005) Shallow and deep landslides induced by rainfall in the
Lisbon region (Portugal): assessment of relationships with the North Atlantic Oscillation. Nat
Hazards Earth Sys Sci 5:331–344
Chapter 6
Slope Stability Model and Landslide
Susceptibility Using Geo-technical
Properties of Soil
Abstract The present study deals with the assessment of geo-technical parameters
i.e. surface inclination (⊝), soil depth (z), cohesion (c), angle of internal friction (φ),
soil saturation index (m), soil density (γs) and density of water (γw) and to construct
1D (one dimensional) Slope stability model for preparing the slope instability map
under dry, semi-saturated and saturated condition of the landslide prone small hilly
Shivkhola Watershed of Darjeeling Himalaya. To determine the spatial distribution
of slope instability in the watershed, safety factor value for 50 different locations
were being estimated and with the help of GIS tools. The probability or the chances
of landslide phenomena in each class of slope instability maps were extracted by
means of frequency ratio (FR) which shows that the probability/chances of land-
slide events could be expected as very high in the high to very high landslide
susceptibility area and vice versa in all three conditions. The analysis of slope
instability under three conditions also suggested that there was an aerial expansion
of very high landslide susceptibility in saturated condition in comparison to dry and
semi-saturated condition. This aerial expansion was the outcome of complete sat-
uration and reduction of shearing strength of the slope materials above the failure
plane surface. Finally, an accuracy assessment was made by ground truth verifi-
cation of the existing landslide locations where the classification accuracy for dry,
semi-saturated and saturated conditions was 93.86, 94.58 and 85.44 % respectively.

Keywords Slope stability model RS and GIS Landslide susceptibility Safety

factor (FS) Frequency ratio (FR) Accuracy assessment
6.1 Introduction
Slope instability resulted from complex geological setting combined with various
geomorphological, hydrological and geo-technical factors such as slope, relief,
aspect, rainfall, drainage, upslope contributing area, cohesion, angle of internal
friction, wet soil density, depth of the soil, shear stress, shear strength etc. But, the

168 6 Slope Stability Model and Landslide Susceptibility …
factors conducive to slope instability, can be recognized at various levels of

abstraction from the slope itself. The cohesion and pore-water pressure both directly
control the magnitude of stress of the slope materials. These direct factors can be
influenced by other factors recognized at successively more remote levels of
abstraction. For example, pore-water pressure may be related to the rate of infil-
tration through the ground surface, which in turn, may be related to the density of
vegetation cover which is again subject to change as a result of climatic conditions or
land use activity. These chains of relationships may be critical in reducing the slope
stability condition over time to a point where the triggering of movement may occur.
Landslide susceptibility is thus a function of the degree of the inherent stability of the
slope together with the intensity of causative factors capable of reducing the excess
strength. So, the identification of the causative factors is the basis of many methods
of landslide susceptibility assessment. In most of the case the landslide is the critical
mechanism of erosional processes and in such condition landslide is inevitable and
necessary part of the natural landscape process system. Although the occurrences of
landslide hazards and its impact on human society cannot be prevented fully by
analyzing the slope stability condition, but the better understanding of geo-technical
attributes of the soil can contribute to greater knowledge and understanding about
the spatial distribution of slope instability which are very much essential for land use
planning. Many approaches to assess slope stability and landslide hazards were put
forward by Montgomery and Dietrich (1989, 1994), Carrara et al. (1991), Hammond
et al. (1992) combined a contour based steady state hydrologic model with the
infinite slope stability model (simplified for cohesion less soils) to define slope
stability classes based upon slope and specific catchment area. Numerous models in
connection to the slope stability, shallow and deep seated landslides were introduced
and verified by Varnes (1958), Young (1963), Vanmarcke (1977), Burton and Ba-
thrust (1998), Bradinoni and Church (2004), Smedt (2005) and Bhattarai and
Aoyama (2001). The geotectonic factors of slope instability were studied in details
by Brudsen (1979), Windisch (1991), Carson (1975, 1977) and Borga et al. (1998).
A comprehensive list of stability factors commonly employed in the factors mapping
approach was given by Crozier (1986), Guzzetti et al. (1999a, b, 2003) and Tiwari
and Marui (2001, 2002, 2003, 2004).
A more sophisticated approach represents the terrain in terms of differences of
inherent stability based on the Safety Factor (FS). Simply, the value FS is assumed
to be 1.00 at the moment of failure and the values successively greater than 1.00
represents the increasing stability and hence low susceptibility to slope failure.
Determination of FS permits limiting equilibrium analysis of a slope and is par-
ticularly helpful in designing the type and magnitude of remedial measures required
to achieve an acceptable FS. A considerable amount of information such as the
geometry of the slope, pore-water of the slope materials, angle of internal friction
and cohesion are required to assess the stress parameters of the slope materials
(Glade 1998) and Safety Factor value.
The present study encompasses the assessment of geo-technical parameters of
the collected soil samples from 50 landslide locations selected through stratified
random sampling with representatives of different landuse and slope classes. The
geo-technical attributes include surface inclination (β), soil depth (z), cohesion (c),
angle of internal friction (φ), soil saturation index (m), soil density (γs) and density
of water (γw). Before planning any land use, the better understanding and inves-
tigation of geo-tectonic parameters and the preparation of a stability distribution
map by applying GIS tools are very much popular and accepted approach in the
present time. The major objective of the present study is to study geo-technical
parameters and to identify the potential stability sites in the Shivkhola Watershed
through estimating the strength parameters and the application of 1D slope Stability
model to prepare landslide susceptibility maps. The validity of the prepared land-
slide susceptibility maps under dry, semi-saturated and saturated condition were
evaluated by means of a frequency ratio (FR). Finally, an accuracy assessment was
made by ground truth verification of the existing landslide locations where the
classification accuracy for dry, semi-saturated and saturated condition was 93.86,
94.58 and 85.44 % respectively.
6.2 Materials and Method
6.2.1 Slope Stability Model Concept
Two forces are responsible to determine the stability condition i.e. driving force
(shear stress) and resisting force (shear strength). Shear stress is given as,
τ = γ D sin ⊝ cos ⊝ (where γ, soil density; D, depth of the soil; ⊝, slope angle) and
shear strength of Mohr and Columb defined as, S = c + τ tan φ (where c: cohesion;
φ: friction angle; τ: shear stress). Saturated slope material increases instability with
increasing pore water pressure. The pore water pressure depends on unit weight of
water (γw) and the height of water (Dw) above the failure plane surface. The height
of the water shows the ground water condition in the soil. In this case the shear
resistance of the soil is given by the following:
S = c + ðccw mÞ D cos2 € tan u ð6:1Þ
where, m is saturation index which shows the saturation condition of the soil.
If the value of m equals to 1, the soil is completely saturated and the value of 0
indicates complete dry condition.
In many investigations of natural slope stability, infinite slope analysis had
frequently been used because of its relative simplicity where the thickness of the
soil is smaller than the length of the slope.
For realistic modeling 3D failure mechanism should be considered which
includes different depth of sliding surface throughout the slope failure mass. Soeters
and Westen (1996) recommended using infinite slope stability analysis in order to
conduct deterministic analysis of the large area and due to complication in estab-
lishing vertical depth of failure plane in 3D mechanism. Monte Carlo Method, a
simplified approach was considered by them reducing 3D depth to 2D equivalent
depth based on equal factor of safety. However, it is not simple to analyze 2D

rotational slide due to variation in depth of sliding surface. Hence, 2D depth of
rotational slide (Eq. 6.2) was then converted to equivalent translational depth
without the impact of ground water (Eq. 6.3) keeping the same factor of safety.
P
c þ c tan [ X 2
i1 zi cos b
FS ðRotationalÞ ¼ PX
c i1 zi sin b cos b ð6:2Þ
c þ c z cos2 b tan u
FS ðTranslationalÞ ¼ ð6:3Þ
c z sin b cos b
where, γ = unit weight of the soil; z = depth of the failure surface below the terrain
surface; β = the terrain surface inclination; φ = angle of internal friction;
c = cohesion.
The safety factor (FS) under the influence of ground water (semi-saturated and
saturated condition) of cohesive soil could also be calculated by applying the
revised 1D slope stability model with the help Eq. 6.4.
c þ ðc m cw Þ cos2 b tan u
FS ¼ ð6:4Þ
c z sin b cos b
where, γ = unit weight of soil; m = soil saturation index; Zw = height of water table
above failure surface; Z = depth of failure surface below the terrain surface;
γw = unit weight of water; β = the terrain surface inclination; φ = angle of internal
friction and c = cohesion.
In completely dry condition, cohesion (c) and wetness index value (m) become
zero (0) and in case of cohesion less soil, safety factor could be determined with the
help of following equation.
c D cos2 h tan u
Fs ¼ ð6:5Þ
c D sin h cos h
tan u
Or; FS ¼ ð6:6Þ
tan €
The safety factor for cohesion less soil with the influence of ground water can be
estimated by:
ðc mcw Þ D cos2 € tan u

FS ¼ ð6:7Þ
c D sin € cos €
6.2 Materials and Method 171
When the value of wetness index (m) becomes 0, the safety factor is determined
by:
ðc mcw Þ D cos2 € tan u

FS ¼ ð6:8Þ
c tan €
6.2.2 Cohesion (c) and Friction Angle (φ)
The shear strength of the soil was described as the function of normal stress on the
slip surface, cohesion, and angle of internal friction. The angle of internal friction
(u) and cohesion are the two important physical properties of the soil which
determines angle of rupture, shearing strength, safety factor as well as stability
condition of the slope materials by developing ‘stress circle’. The relationship
within all these properties to other characteristics of the soil was introduced by
Terzaghi (1950) and Wu and Siddle (1995). The geo-technical factors like angle of
repose of the debris were measured after Bloom (1991). The tangent of angle of
repose of dry granular materials is slightly greater than, but approximately equals to
the co-efficient of sliding friction of the material or its mass friction (u) (Van
Burkalow 1945; Bloom 1991). The range of cohesion and friction angle of different
soil was also adopted from Foundation of Engineering Geology (Tony Waltham
2002) for the analysis of slope stability. All the tests were carried out under drained
condition using 100 mm diameter and 25 mm thick specimen in Geotechnical
Laboratory, GSI, Kolkata.
The major stress (σ1), minor stress (σ3) and cohesion (c) were estimated through
tri-axial soil testing mechanism (Fig. 5.1) from Geo-technical Laboratory of GSI,
Kolkata (22/com/soil/GTL/ER/O6-07). On the basis of these three major attributes a
Mohr stress circle was developed to obtain angle of internal friction and angle of
rupture. At first, a circle was drawn through σ3 and σ1 with the centre on the
horizontal axis; the centre of the circle was obviously (σ1 + σ3)/2 and the radius was
(σ1 − σ3)/2 (Fig. 5.2).
The values of confining pressure, σ3, and compressive stress, σ1 were plotted on
horizontal axis where stress difference was σ1 − σ3. On a plane parallel to the
greatest principal stress axis (2α = 0) the normal stress across the plane was σ3 and
the shearing stress was 0. If the plane makes an angle of 45° with the greatest
principal stress axis (2α = 90), the shearing stress is at a maximum and the normal
stress is (σ1 + σ3)/2. If the plane makes an angle of 90° with the greatest principal
stress axis (2σ = 180°), the shearing stress is 0 and the normal stress is σ1 (Billings
1987).
r1 þ r3 r1 r3
rn ðnormal stressÞ ¼ cos 2a ð6:9Þ
2 2
r1 r3 tan2 ð45 þ u2 Þ
C ðcohesionÞ ¼ ð6:10Þ
2 þ tan ð45 þ u2 Þ
For the determination of the cohesive strength and the angle of internal friction a
series of experiments were done with different values of confining pressure (σ3).
The Mohr Circle shows that as the confining pressure is increased, the stress as well
as the stress difference must be increased to produce rapture. A line which is the
tangent of the ‘Mohr Circle’ is called as the ‘Mohr Envelope’. The angle that this
line makes with the horizontal axis of the diagram is the angle of internal friction, /
(Fig. 5.2). Along any potential plane of rupture within a rock:
l ¼ s=n ¼ tan/ ð6:11Þ
where µ is the coefficient of internal friction,

τ= shearing stress along plane,
n= normal stress along the lane,
/= the angle of internal friction.
The τ is at a maximum when α = 45°; whereas n is at a minimum when α = 0 and

at a maximum when α = 90°. Shear fracture develops when n and τ combine to
make the shear stress most effective. Actual shear fracture makes an angle of less
than 45° with the greatest principal stress axis (Billings 1987).
The intercept on the vertical axis, τo, is the cohesive strength of the rock. The
curve for the Mohr envelope is:
s ¼ so þ rn tan / ð6:12Þ
where, the angle that fractures should theoretically make with the greatest principal
axis is
a ¼ 45 /=2 ð6:13Þ
So, if the angle of internal friction is 30°, the fractures would make the angle of
30° with the greatest principal stress axis. On the other hand the shear fracture
should theoretically form at 30° (Billings 1987).
The frictional angle of a particular soil depends on various factors (Cernnica
1995) like:
• State of compaction and void ratio increasing the density. The relation is not
necessarily the linear one.
• Coarseness, the shape of the particle and the angularity of the particle of the soil:
Angular grains interlock more effectively than rounded ones, thereby creating a
larger friction angle.
• Mineralogical content of the soil: Hard gravel particles result in higher friction
angles. Soft grains which may crush more easily, thereby reducing the
interlocking or bridging effects. For sand, however, the mineralogical content

seems to make little difference except if the sand contains mica. In that case the
void ratio is usually larger, thereby resulting in loose interlocking and lower
friction angle.
• Particle size distribution of the soil: The soil having the well-graded size will
have higher frictional angle than the soil having poorly graded soil.
Cohesion (C) is the attraction of particles to each other which is not directly
governed by a friction law but does provide a measure of strength of a material.
Thus sands do not exhibit cohesion, while soil which contains clay show cohesion.
It can be measured, as in soil mechanics, by the Mohr-Coulomb Equation
(Eq. 6.10).
6.2.3 Surface Inclination/Slope (β)
Slope gradients are sometimes considered as an index of slope instability, and

because of the availability of a digital elevation model (DEM), slope can be
numerically evaluated and depicted spatially (O’ Neill and Mark 1987; Gao 1993).
Firstly, the contour map at 20 m interval was prepared and digitized from the
topographical map 73B/8 (1987) at the scale of 1:50,000 and subsequently used for
generating Digital Elevation Model on ARC GIS platform. Then slope gradient
map (Fig. 2.6) was extracted from DEM and it was the classified after Anbalagan
(1992) and Dhakal et al. (2000) into ten equal classes.
6.2.4 Soil Saturation Index/Wetness Index (m)
Simple models have been developed for estimating the soil saturation of the
mountainous region as the wetness index is defined in TOPMODEL by Beven and
Kirkby.
a
m ¼ 1n ð6:14Þ
tan h
where a, is the contributing area per unit contour length and ⊝ is the slope of the
pixel.
More acceptable soil saturation model was applied by Montgomery and Dietrich
(1994), Borga et al. (1998) and Pack et al. (1998). The model envisages that the soil
saturation index can be determined with the help of topography, soil type, and
rainfall intensity of the area under study. But in practical sense, the soil is not
completely dry or fully saturated in the area, therefore it can be imagined that the
soil is half saturated. The soil saturation index is either fixed for stationary scenarios
i.e. dry, semi-saturated and full saturated soils, given by m = 0, 0.5 and 1.00 or can
be calculated on the basis of available rainfall data (De Smedt 2005). On the basis
of this assumption, wetness index equation can easily be derived and it is possible
to see the effect of few days’ consecutive rainfall in 1 day, if the soil is half
saturated. In the present study wetness index (m) value of 0.00, 0.50 and 1.00 under
dry, half-saturated and full-saturated conditions were taken into account.
6.2.5 Depth of Failure Surface/Depth (z) of Soil Below

the Terrain Surface
Repeated and continued field studies for long duration were made for the proper
cognition of the processes and their interaction. The depth of the failure surface was
measured by holding a measuring tape at both the margins of scar and the other tape
was allowed to hang, the reading was then taken from the base of the hanging tape.
The margin of the scars was surveyed by prismatic compass. The intensive survey
of the sliding scar for 40 different landslide locations was carried on by Abney’s
level at 0.5 m interval along radial lines originating from lower most part of the
scar. The altitude of the points at 0.5 m interval along the radial lines was then
estimated using Sine rule in reference to the central base point of known altitude
determined by GPS (Basu and Maiti 2001). The total thickness of soil and that of
saturated soil for 10 sites during monsoon were measured from slope cutting. After
estimating the approximate depth of all known points, a soil depth map (z/D) was
prepared using Arc GIS tool (see Chap. 2, Fig. 2.13).
6.2.6 Soil Density (γs) and Density of Water (γw)
Specific unit weight of water and unit weight of the soil were estimated by
examining the soil samples collected from 50 landslide locations during field
investigation. The density of soil and water varies from place to place due to in situ
geo-hydrologic condition (Fig. 2.16). The saturated soil density was also consulted
and adopted from the field experiences done by Deoja (1991).
6.2.7 Identification of Major Landslide Location/Landslide

Inventory Map
To determine the frequency ratio (FR) and to assess the overall classification accuracy,
the major landslidelocationofthe Shivkhola Watershed (Chap. 2,Fig. 2.1) wasdetected
by intensive field investigation with GPS, clinometers, and Abney’s level. Besides,
LISS-III Satellite Image (2010), SRTM data (2008) and Google earth image (2010) had
been incorporated with the surveyed landslide locations by thorough rectification and it
was then modified and mapped accordingly (Fig. 2.1).
6.3 Application of 1 Dimension Slope Stability Model

and Stability Analysis
With the help of derived geo-technical parameters i.e. cohesion, friction angle,
slope angle, unit weight of the soil, unit weight of water, soil depth, and saturation
index value from 50 landslide location points of the Shivkhola watershed the safety
factor values (FS) for dry, semi-saturated and saturated condition were being
estimated by applying the 1D slope Stability model (Eq. 6.4). The safety factor
values were transformed into raster value domain on ARC GIS Platform. Finally,
the landslide susceptibility maps/safety factor distribution maps were prepared by
‘slicing’ operation and then stability classes for each condition (dry, semi and
saturated condition) had been performed by studying the cumulative frequency and
their abrupt change points of the safety factor values (the instability threshold
boundaries). A 3 × 3 ‘majority filter’ technique was also applied to all the prepared
safety factor distribution maps as a post-classification filter to reduce the high
frequency variation. Higher the value of ‘FS’, greater is the propensity of slope
stability and vice versa. To assess the chances/probability of landslide phenomena
in each class to all the prepared maps under various conditions frequency ratio (FR)
was extracted by means of a ratio between landslide frequency/landslide events (%)
and landslide susceptibility area (%). FR value is approaching to 1 indicates equal
chances of landslide events, 0 indicates lesser chances and more than 1 shows
greater probability. Finally, an accuracy assessment was made after Congalton
(1991).
6.4 Shear Stress, Shear Strength and Safety Factor

and Stability Analysis
Safety factor (FS) refers to the ratio between the shearing forces (simply known as
stress, τ) and resistance of the materials to shearing forces (shearing strength, λ).
Increase in shearing stresses produces rapture within the underlying rock beds and
decrease the cohesive strength as well as the shearing strength due to which the slope
materials start to move downward and make the slope most vulnerable to land slip.
ðs ¼ r1 r 3
sin 2aÞ
FS ¼ 2
ð6:15Þ
ðkÞ ¼ ½rn tan u þ c
If the computed factor of safety is less than unity, the slope is clearly unstable
and likely to fall. If the factor of safety is greater than unity we cannot assume that
the slope is stable as we may not have chosen the most critical mode of failure
(i.e., failure surface) (Figs. 6.1, 6.2 and 6.3).
Data Used
Cohesion =Lab.  = lab. Z= [topography m

(C)-lab. Field ]= DEM (assumpti
Test test
Test. Stat. on)
FS=Factor of Safety
Susceptibility Map Susceptibility Map Susceptibility Map

Dry Condition. Semi-saturated Saturated
Condition. Condition.
Verification
GPS Survey
Accuracy Study
[RS& GIS Based]
Instability Model
Fig. 6.1 Flow chart for landslide susceptibility maps

6.4 Shear Stress, Shear Strength and Safety Factor and Stability Analysis 177
Soil Sample Analysis Sites

Shiviter T.E.
S.F.-0.980
Cohesion-0.07
S.F.-0.951
Cohesion-0.162
Nurbong T.E.
S.F.-0.978 S.F.-0.999
Cohesion-0.105 Cohesion-0.035
Gayabari Lower Sepoydhura T.E.

Tindharia T.E.
S.F.-0.99
S.F.-1.07 S.F.-0.99
S.F.-0.96 Cohesion-o.123
Cohesion-0.05
Cohesion-0.15
Cohesion-0.149
Fig. 6.2 Major landslide locations with ‘FS’ value in the watershed
6.4.1 Shear Stress
Shear stress is a force which deforms a mass of material by one part sliding over
another along one or more failure plains. This is a force per unit area which is
parallel to the surface of a body. This kind of stress is perpendicular to the normal
stress. The rupture angle is obtained by developing Stress circle (Fig. 7.2). The
shear stress is obtained by the following formula.
r1 r3
s ðshesr stressÞ ¼ sin 2a ð6:16Þ
2
where, σ1 = major stress; σ3 = minor stress; and α = rupture angle.
6.4.2 Shear Strength
This is a measure of the ability of a material to resist shear stress. This is an

important parameter in determining the engineering and geomorphic parameters of
the materials. Shear strength of the soil depends on the normal stress, angle of
internal friction and cohesion which is determined applying the following formula:
Fig. 6.3 Landslide susceptibility under saturated condition
Shear Strength ðkÞ ¼ ½rn tan/ þ C ð6:17Þ
where, rn = normal stress; / = friction angle; and C = cohesion of the soil.
6.4.3 Safety Factor Based Stability Classes
See Table 6.1.
Table 6.1 The Stability Classes and Remarks

Safety factor (FS) Slope condition Remarks
FS < 1.5 Stable Only major destabilizing factors lead to instability
1.25 < FS < 1.5 Moderately stable Moderate destabilizing factor lead to instability
1 < FS < 1.25 Quasi stable Minor destabilizing factor lead to instability
FS < 1 Unstable Stabilizing factors are needed for stability
6.5 The Result and Discussion 179
6.5 The Result and Discussion
6.5.1 Analysis of Strength Parameters (Shear Stress

and Strength) for 10 Major Landslide Locations
The Mohr envelope is the graphical representation of geo-technical parameters

which indicate the stability condition of different locations. Cohesion is low
everywhere because of the entrainment of finer particles from the sub-surface layer
by leaching processes, disintegration and decomposition by physical as well as
chemical processes and obviously by soil saturation. In Tindharia (Fig. 6.2) and
Lower Paglajhora (Fig. 6.4) the internal friction is around 19° and angle of rupture
is 35°16′ and 35°30′ respectively. Mahanadi (Fig. 6.5) and Gayabari Lower Slope
(Fig. 6.6) is experiencing the safety factor more than one which indicates slightly
higher shear strength than shear stress. Cohesion of the soil is very low that ranges
between 0.03 and 0.16. Water acts in a number of ways for removal of lateral
support by toe erosion, decrease in shearing resistance, entrainment of finer matrix
from the slope material and setting the coarser fraction to move, and increase in
Fig. 6.4 Landslide susceptibility under semi-saturated condition

Fig. 6.5 Landslide susceptibility under dry condition
Fig. 6.6 Graphical presentation of landslide susceptibility in all conditions

pore water pressure to reduce the cohesion within the soil particles. All these as a
whole, affect the slope stability. At Tindharia Proper, 14 miles Bustee, Nurbong T.
E., Central part of the Basin, Shiviter T.E. the angle of rupture is 33°, 34°, 34°30°
32° and 34°15′ respectively (Table 6.2, Figs. F.1, F.2, F.3, F.4, F.5, F.6, F.7, F.8,
F.9, and F.10, Appendix-G). The geotechnical parameters of the soil from different
locations of the Shivkhola Watershed are represented in the following developed
Mohr Stress Circles which depict the visible interpretation about the strength of
slope materials (Appendix-G, Figs. F.1, F.2, F.3, F.4, F.5, F.6, F.7, F.8, F.9, and
F.10).
6.5.2 Stability Analysis Based on Safety Factor for 10 Major

Landslide Locations
A stability analysis has been made on the basis of the calculated Safety Factor’s
value (Table 6.3) from various sites of the Shivkhola watershed which reveals that
most of the sites such as Tindharia, Lower Paglajhora (14 miles bustee), Lower
Paglajhora (main slide), Tindharia Railway Station, Shiviter, Sepoydhura, Nurbong
T.E. and also the mid-section of the Watershed are registered with the Safety
Factors value of less than 1 and it can be assumed that the slope materials of all
these places are under stress and instable in nature. Remaining soil sample analysis
sites (Mahanadi and Gayabari) are attributed as Quasi Stable part.
6.5.3 Comparative Stability Analysis Between Dry,

Semi-saturated and Saturated Soil Condition Based
on 1D Slope Stability Model
The shivkhola watershed exhibits a wide range of elevation (300–2,400 m). Slope
ranges between 7° and 65°. The central middle section and lower section of the
watershed is attributed with very gentle slope gradient of less than 20° whereas
marginal part and extreme north, south and west are characterized by very steep
slope of more than 50°. Angle of internal friction varies between 18° and 33°
(Table C.1, see Appendix C). Generally, slope materials having coarse grains over
the steep slope shows friction friction angle than the materials with finer particles
deposited along the foothills zone. At Lower and Upper Paglajhora, Tindharia
Upslope, Shiviter and Nurbong the friction angle and cohesion of the soil is very
low. Cohesion of the soil is high in the mid and lower part where more than 50 %
particles are composed with finer particles.
The intensive field work at 50 sites in the basin reveals that depth of soil varies
from 0.45 to 3.75 m in the Shivkhola Watershed. The central mid-section and lower
part of the watershed are registered with maximum soil depth whereas marginal part
182
Table 6.2 Result of geo-technical parameters and safety factor for 10 major landslide sites
Location r3 rn r1 s Φ C λ α F.S.
1. Tindharia tea garden 0.90 1.31818 2.15440 0.59036 19°28′ 0.123 0.588929 35°16′ 0.9975760
2. Lower Paglajhora (14 Miles Bustee) 0.82 1.28942 2.3212 0.69594 22° 0.149 0.6699594 34° 0.962668
3. Lower Paglajhora 0.96 1.4583 2.43794 0.6987 19° 0.162 0.6641329 35°30′ 0.950526
4. Mahanadi tea garden 0.98 1.5227 2.86 0.85192 25° 0.145 0.85504667 32°30′ 1.003670
5. Gayabari lower slope (780 M) 0.89 1.27784 2.13032 0.5286 22° 0.05 0.56628 34° 1.0712826
6. Tindharia Rly. Stn. 0.92 1.4044 2.553 0.74598 24° 0.105 0.730279 33° 0.9789525
7. Shiviter T.E 0.97 1.402407 2.3351579 0.63508 21°30’ 0.07 0.6224228 34°15′ 0.9800699
8. Sepoydhura T.E 0.88 1.3098 2.3897 0.681316 25°30’ 0.05 0.67474 32°15′ 0.990349
9. Nurbong T.E. (sinking zone) 0.93 1.2965 2.0725 0.5333 21° 0.0355 0.5331 34°30′ 0.99977
10. Central part of the watershed 0.87 1.403035 2.76818 0.853036 26° 0.15 0.834305 32° 0.978043
r1 = major principal stress; r3 = minor principal stress; rn = normal stress; s = shear stress; Φ = angle of internal friction; C = cohesion; λ = shear strength;
α = rupture angle; and F.S. = safety factor
6 Slope Stability Model and Landslide Susceptibility …
Table 6.3 Stability analysis for 10 critical locations
Location sites Under saturated Remarks
condition
S.F Stability
1. Tindharia tea garden 0.997576 Unstable Further concentration of settlement should be restricted, reduction of slope steepness, draining
of seepage water by boreholes to reduce destabilizing factors
2. Lower Paglajhora 0.962668 Unstable Surface and sub-surface drainage diversion and monitoring and draining of seepage water by
6.5 The Result and Discussion
(14 miles Bustee) boreholes, restriction on heavy loaded vehicles to reduce destabilizing factors
3. Lower Paglajhora (main 0.950526 Unstable Sub-surface drainage monitoring and draining of seepage water by boreholes, jhora training
slide) through concrete structure
4. Mahanadi tea garden 1.003670 Quasi Drainage network management, plantation and afforestation, construction of retaining walls
stable and concrete drains along the road side are to be taken into consideration in quasi stable area
5. Gayabari lower slope 1.071282 Quasi Plantation and afforestation, construction of retaining walls and concrete drains along the road
(780 M) stable side are to be taken into consideration in quasi stable area
6. Tindharia proper slide 0.978952 Unstable Human intervention should be stopped completely, application of geotextile method can bring
(near rly. station) stability over steep slope
7. Shiviter T.E 0.980069 Unstable Establishment of horizontal drains to reduce upslope contributing area above the slide and
application bio-engineering method to stabilize the slope
8. Sepoydhura T.E 0.990349 Unstable Deforestation and haphazard concentration of human settlement and their activities should be
restricted to get rid of slope failure
9. Nurbong T.E. (sinking 0.99977 Unstable Draining of seepage water by boreholes and cementing of the cracks over steep slope to reduce
zone) destabilizing factors
10. Central part of the 0.978043 Unstable Afforestation, plantation and other soil conservation technique must be employed. Concrete
watershed from Gayabari drains to monitor surface drainage and construction of retaining walls TO to duce the
destabilizing factors
183
(north, south and western part) is attributed with minimum soil depth. Due to steep
slope and active soil erosion process, the marginal area is associated with close slip
plane below the slope surface due to which soil layers get saturated very easily and
causes landslide. Tindharia, Gayabari Upper, Sepoydhura Upslope, Upper Pag-
lajhora, Mahanadi, Shiviter, Gitingia are falling in the marginal minimum soil depth
area of the Shivkhola watershed. On the other hand, Shivkhola R.F., and both sides
of the main river is characterized by maximum soil depth with low intensity of
landslide phenomena. It can be inferred that shallow seated slope instability is
intimately related with minimum soil depth with steep slope.
In dry condition, around 15 sq km of the watershed is attributed by moderate to
very low landslide susceptibility and only 5 sq km is characterized with high to very
high landslide susceptibility with high frequency ratio. Under semi-saturated con-
dition 0.88 sq km area is attributed with high landslide susceptibility. The area of
high and very high landslide susceptibility has been increased in saturated soil
condition and around 2.5 sq km area is registering high landslide susceptibility with
high frequency ratio. For dry and semi-saturated condition, the area under high and
very high landslide susceptibility is 5.36 and 8.5 sq km respectively and the area
under low to very low landslide susceptibility is around 9 and 5 sq km Under
saturated soil condition small area (3.5 %) is experienced with low landslide
susceptibility having frequency ratio of 0.00. Around 14 sq km out of total area
(21 sq km) of the Watershed is being characterized by moderate to very high
landslide susceptibility. Under saturated condition, around 6 sq km areas show the
moderate landslide susceptibility with frequency ratio value of 0.39. Upper and
Lower Paglajhora, Tindharia, Shiviter and Nurbong are the places where landslide
susceptibility ranges from high to very high. Lower middle section and few places
of extreme south are characterized by moderate to very low landslide susceptibility
(Table 6.4) under all soil saturated condition.
6.5.3.1 Stability Analysis Under Dry Condition
The slope stability map (Fig. 6.5) under dry condition reveals that only 1.58 % area
of the watershed is experienced with very high landslide susceptibility where the
chance of landslide phenomena is also very high according to the derived frequency
ratio value (21.09). High landslide susceptibility is being found in 24.88 % area in
the watershed. Maximum area of the Shivkhola watershed (30.25 %) is characterized
by moderate landslide susceptibility where the frequency of landslide is 20.835.
More than 40 % area of the Shivkhola watershed falls under the low to very low
landslide susceptibility with frequency ratio value of 0.27 and 0.00 respectively.
6.5.3.2 Stability Analysis Under Semi-saturated Condition
Under semi-saturated condition the value of safety factor varies from 0.158 to 2.58
and 4.38 % area is under very high landslide susceptibility that is around 2.50 %
Table 6.4 Frequency ratio analysis for dry, semi-saturated and saturated soil condition
Safety Landslide Area in Percentage Landslide Frequency
factor susceptibility sq km of area frequency ratio
Frequency ratio analysis under dry condition
0.158–0.40 Very high 0.320933 1.58 8 (33.33 %) 21.09
0.40–0.70 High 5.045979 24.88 9 (37.5 %) 1.50
0.70–1.00 Moderate 6.136645 30.25 5 (20.83 %) 0.67
1.00–1.50 Low 6.602401 30.51 2 (8.33 %) 0.27
1.50–2.83 Very low 2.188175 10.79 0 (0.00 %) 0.00
Frequency ratio analysis under semi-saturated condition
0.158–0.40 Very high 0.88796 4.38 9 (37.5 %) 8.56
0.40–0.70 High 7.691486 37.88 10 (41.67 %) 1.10
0.70–1.00 Moderate 6.393539 31.47 4 (16.67 %) 0.53
1.00–1.50 Low 4.15209 20.44 1 (4.16 %) 0.20
1.50–2.83 Very low 1.169058 5.76 0 (0.00 %) 0.00
Frequency ratio analysis under saturated condition
0.158–0.40 Very high 2.430299 11.97 12 (50 %) 4.17
0.40–0.70 High 7.985823 39.36 8 (33.33 %) 0.85
0.70–1.00 Moderate 6.406416 31.58 3 (12.5 %) 0.39
1.00–1.50 Low 2.763299 13.60 1 (4.17 %) 0.31
1.50–2.83 Very low 0.708297 3.50 0 (0.00 %) 0.00
greater than dry condition (Fig. 6.4). 37.88 % area of the basin is dominated by high
landslide susceptibility and 31.47 % area is registered with moderate landslide
susceptibility and equal chances of landslide occurrence phenomena. About 25 %
of the watershed is characterised with low to very low landslide susceptibility
condition. Frequency ratio value under semi-saturated condition revealed that the
probability of landslide occurrence was very high in the area of very high landslide
susceptibility which was followed by high, moderate and low. Under dry condition
moderate, low and very low landslide susceptibility area did not have experienced
major landslide events and so derived frequency ratio values tent to ‘0’.
6.5.3.3 Stability Analysis Under Saturated Condition
Under the complete saturated soil condition pore water pressure becomes very high
and reduces the cohesive strength of the soil and mountain slope become most
unstable. The values of safety factor under saturated condition ranges between 0.13
and 2.30. The area of very high landslide susceptibility had been increased in
comparison to dry and semi-saturated condition. More than 50 % area of the
Shivkhola watershed is attributed by high to very high landslide susceptibility.
31.585 % areas are under moderate landslide susceptibility where the probability of
landslide activities is low. Under the saturated condition 11.97 % area with very
Table 6.5 Accuracy analysis

Class Classified Number Producers Users Accuracy
name total correct correct accuracy total
Accuracy study under dry condition
Very low 0 6 0 0.00 0.00
Low 4 3 0 75 0.00
Moderate 10 9 9 90 100.00
High 17 15 14 88.23 93.33
Very 18 17 15 94.44 88.24
high
Total 50 50 38
Overall classification accuracy = 93.86 %
Overall Kappa statistics = 0.8919
Accuracy study under semi-saturated condition
Very low 0 6 0 0.00 0.00
Low 0 7 0 0.00 0.00
Moderate 13 10 9 76.92 90.00
High 20 16 16 80.00 100.00
Very 17 16 15 94.12 93.75
high
Total 50 50 40
Accuracy study under saturated condition
Very low 0 4 0 0.00 0.00
Low 0 7 5 0.00 71.43
Moderate 12 7 6 58.33 85.71
High 16 13 11 81.25 84.62
Very 22 16 16 72.73 100
high
Total 50 50 38
high landslide susceptibility shows the greater chances of landslide phenomena

(Fig. 6.3). Extension of Very high landslide susceptibility under saturated condition
is due to saturation of slope materials and heavy pore water pressure.
6.5.4 Accuracy Result
The comparison between assumed true data and randomly selected data from the
classified image shows that the overall classification accuracy for dry, semi-satu-
rated and saturated conditions are 93.86, 94.58 and 84.44 % respectively.
The accuracy results in different landslide susceptibility classes under dry, semi-
saturated and saturated conditions are stated in Table 6.5.
6.6 Conclusion
Stability analysis on the basis of calculated safety factor values and stability classes
from major landslide location shows that the places of Paglajhora, Tindharia,
Shiviter T.E., Sepoydhura, and Nurbong are unstable and only Gayabari and
Mahandi are quasi-stable in nature. Analysis of the geo-technical parameters reveals
that the cohesion of soil ranges between 0.03 and 0.16 and angle of internal friction
varies between 18° and 34°. But the major landslide location sites in the study area
are registered with the friction angle from 18° to 24°. There is a tendency of an
increase in areal extension of high to very high landslide susceptibility area in
saturated soil condition rather than dry and semi-saturated soil. This situation
indicates that soil saturation and reduction of soil cohesion is the important land-
slide contributing factor in the Shivkhola Watershed. Under saturated condition
more that 50 % area are characterized by high to very high landslide susceptibility.
More than 35 % area is dominated by high landslide susceptibility under
semi-saturated condition and only 1.58 % area is experiencing very high landslide
susceptibility under dry condition (Fig. 6.6). The prepared landslide susceptibility
maps for all three conditions (dry, semi-saturated and saturated) express a
reasonable accuracy i.e. 93.86, 94.58 and 85.44 % respectively. Moderately steep
slope with low friction angle and less cohesion is very much prone to slope
instability in the Shivkhola Watershed.
The prepared maps may offer useful tool for the installation and the continuous
monitoring of the geotechnical attributes measuring apparatus/instruments such as
field shear box for measuring the shear strength properties of the soil, permeameter,
double ring infiltrometer, and piezometer to measure soil pore water pressure. The
monitoring result must be served to the local Govt. Local people are to be made
aware of the triggering geotechnical factors through and they are to be brought into
the active monitoring and management system.
There is no any unique and generalized model available for the preparation and
identification of slope instability sites as the instability condition depends upon
various factors which vary from one place to others. So, one simple model could
not be accepted for all landslide locations. Considering all the landslide triggering
factors of the study area, one dimensional slope stability model has been adopted to
conceive the spatial distribution slope instability. In the present work, our approach
is to determine the potential instability location in connection to spatial distribution
of geotechnical parameters. Based on the analysis priority was fixed and manage-
ment options could be followed up for the Shivkhola watershed. It is observed from
the study that the areal extent of potential slope instability and the chances or
probability of slip under saturated condition is very high as a result of soil saturation
and increased pore water pressure, less cohesion, and low friction angle. Steep slope
sites i.e. Paglajhora (lower and upper), Tindharia, Nurbong, Shiviter and Mahanadi
are very much subjected to slope instability. The mid-central steep slope of the
watershed must be brought under immediate attention as the propensity of areal
increase in slope failure is very high as a result of drainage concentration and the
percolation of water through weak lithological composition.
References
Eng Geol 32:269–277
Basu, S. R. and Maiti, R. K. (2001) Unscientific mining and degradation of slopes in the
Darjeeling Himalayas, Changing Env. Scenerio of the Indian Subcontinent (Bd) pp 390 – 399
Bhattarai P, Aoyama K (2001) Mass movement problems along Prithwi highway Nepal. Annual
Report of Research Institute for Hazards in Snowy Areas, vol 23. Niigata University, Niigata,
pp 85–92
Billings MP (1987) Structural geology, 3rd edn. Prentice Hall, New Delhi
Bloom AL. (1991) Geomorphology, a systematic analysis of the Cenozoic Landforms. Prentice
Hall, New Delhi, pp 76–177
Borga M et al (1998) Shallow Landslide hazard assessment using a physically based model and
Brardinoni F, Church M (2004) Representing the landslide magnitude frequency relation. In:
Kirkby JM, Darby ES (eds) Earth surface processes and landforms, vol 29, no 1. Capilano river
basin, British Colombia, pp 115–124
Brudsen D (1979) Mass movement. In: Embelton C, Thornes J (eds.) Process in geomorphology.
Wiley, Hoboken, pp 130–186
Burton A, Bathurst JC (1998) Physically based modeling of shallow landslide erosion and
sediment yield at a catchment scale. Environ Geol 35(2–3):89–99
Carrara A et al (1991) GIS technique and statistical models in evaluating landslide hazard. Earth
Surf Process Land 16(5):427–445
Carson MA (1975) Threshold and characteristic angles of straight slopes. Proceedings of the 4th
Guelph Symposium on Geomorphology, Norweich Geo Books, 19–34
Cernnica JN (1995) Geo-technical engineering: soil mechanics. Willy, Hoboken
Congalton R (1991) A review of assessing the accuracy of the classification of remotely sensed
data. Remote Sens Environ 37:35–46
Crozier MJ (1986) Landslides: causes, consequences and environment. Croom Helm Australia Pty
Ltd, London, p 252
Deoja BB et al (1991) Mountain risk engineering handbook. International centre for integrated
mountain development (ICIMOD). Kathmandu, p 875
Dhakal AS et al (2000) Landslide hazard mapping and its evaluation using GIS: an investigation of
sampling schemes for a grid-cell based quantitative method. Photogrametric Eng Remote Sens
66(8):981–989
Gao J (1993) Identification of topographic settings conducive to landsliding from DEM in Nelson
County. Earth Surf Process Land 18:579–591
Guzzeti F, Cardinali M, Reichenbach P, Carrara A (1999a) Comparing landslide maps: a case
study in the upper Tiber River Basin, central Italy. Environ Manage 18:623–633
References 189
Guzzeti F, Cardinali M, Reichenbach P, Carrara A (1999b) Landslide hazard evaluation: an aid to

Guzzeti F, Cardinali M, Reichenbach P, Ardizzone F, Galli M (2003) Impact of landslides in the
Umbria region, central Italy. Nat Hazards Earth Syst Sci 5:1–17
Hammond C et al (1992) Level I stability analysis (LISA) documentation for Version 2. General
Technical Report INT-285, USDA forest Service, Intermountain Research Station, p 121
Resour Res 25(8):1907–1918
shallow land sliding. Water Resour Res 30(4):1153–1171
Neill and Mark (1987) On the frequencu distribution of land slope, Earth Surface Processes and
Landforms, vol.12, ISSUE-2, pp127–136
Pack RT, Tarboton DG, Goodwin CN (1998) Terrain stability mapping with SINMAP, Technical
description and users guide for version 1.00, Report and software. http://www.engineering.usu.
edu/dtarb/. Accessed 15 June 2007
Smedt F (2005) Slope instability analysis using GIS on a regional scale: a case study of
Narayanghat-Mungling highway section, Nepal, a dissertation report presented at Universiteit
Gent. Vrije Universiteit Brusssel, Belgium
Soeters R, Westen CJ (1996) Slope instability recognition, analysis and zonation. In Turner AK,
Schuster RL (eds) Landslides: investigation and mitigation. Transportation Research Board
Special Report 247, 129–177
Terzaghi K (1950) Mechanism of landslides, in application of geology to engineering practice.
Barkley Volume, Geological Society of America, pp 83–123
Tiwari B, Marui H (2001) Shearing behaviour of landslide sliding and mining scarp soil during
drained ring shear test. In: Proceedings of 15th international conference on soil mechanics and
geotechnical engineering, vol 1, Istambul, pp 295–298
Tiwari B, Marui H (2002) Mechanism of shear zone formation and its effect in residual shear
strength. In: Proceedings of 3rd international conference on landslides, slope stability and
safety of infrastructure, pp 4–133
Tiwari B, Marui H (2003) Estimation of residualshear strength for bentonite-kaolin-Toyoura sand
mixture. J Jpn Landslide Soc 40(2):124–133
Tiwari B, Marui H (2004) Objective oriented multi-stage ring shear test for the shear strength of
the landslide soil. J Geotech Geoenv Eng ASCE 130(2):217–222
Soc America Bull 56:669–707
Vanmarcke EH (1977) Reliability of earth slopes. J Geotech Eng Div ASCE 103:GT11
practice: highway research board, Special Report 29, vol 544. NAS-NRC Publication,
pp 20–47
Waltham, T. (2002) Foundations of Engineering Geology
Windisch EJ (1991) The hydraulics problem in slope stability analysis. Can Geotech J 28
(6):903–909
Wu W, Siddle RC (1995) A distributed slope stability model for step forested basins. Water
Resour Res 31:2097–2110
Chapter 7
Application of Analytical Hierarchy
Process (AHP) and Frequency Ratio (FR)
Model in Assessing Landslide
Susceptibility and Risk
Abstract To prepare landslide susceptibility map of the Shivkhola watershed, one of

the landslide prone part of Darjiling Himalaya, RS and GIS tools were being used to
integrate 10 landslide triggering parameters like lithology, slope angle, slope aspect,
slope curvature, drainage density, lineament, upslope contributing area (UCA), road
contributing area (RCA) settlement density, and land use and land cover (LULC).
Analytical Hierarchy Process (AHP) was applied to quantify all the factors by
estimating factors weight on MATLAB Software with reasonable consistency ratio
(CR). Frequency ratio model (FR) was used to derive class frequency ratio or class
weight incorporating both pixels with and without landslides and to determine the
relative importance of individual classes. All the required data layers were prepared in
consultation with SOI Topo-sheet (78B/5), LIIS-III Satellite Image (2010) by using
Erdas Imagine 8.5, PCI Geomatica, and ARC GIS Software. The weighted linear
combination (WLC) method was followed to combine factors weight and class weight
and to determine the landslide susceptibility coefficient value (LSCV or ‘M’) on GIS
platform. Greater the value of ‘M’, higher is the susceptibility of landslide. The
Shivkhola watershed was classified into five landslide susceptibility zones by aver-
aging window lengths of 3, 5, 7, and 9 and taking into account the landslide threshold
boundaries value of 7.05, 9.29, 11.5, and 13.8. The overall classification accuracy rate
is 92.22 % and overall Kappa statistics is 0.894. The elements like weighted LULC
map, RCA (road contributing area) map and settlement density map were developed
and their weighted linear combination was performed to prepare landslide risk
exposure map. Then by integrating landslide susceptibility map and landslide risk
exposure map landslide hazard risk co-efficient values were derived and a classifi-
cation was incorporated on ARC GIS Platform to prepare landslide hazard risk map of
the Shivkhola watershed. To evaluate the validity of the landslide hazard risk map,
probability/chance of landslide hazard risk event has been estimated by means of
frequency ratio (FR) between landslide hazard risk area (%) and number of risk events
(%) for each landslide hazard risk class. Finally, an accuracy assessment was made
through a comparative study between true GPS derived data and a set of randomly
selected pixels points from the classified image corresponding to the true data from 50
locations on ERDAS Imagine (8.5) which depicts that the classification accuracy of
the landslide hazard risk map was 92.89 with overall Kappa statistics of 0.8929.

192 7 Application of Analytical Hierarchy Process (AHP) …

Keywords Landslide susceptibility Landslide risk Analytical hierarchy process

(AHP) Frequency ratio (FR) RS and GIS Accuracy assessment
7.1 Introduction
The identification of the causative factors is the basis of many methods of landslide
susceptibility assessment. In most of the cases, the landslide is the critical mech-
anism of erosional processes and in such condition, landslide is inevitable and
necessary part of the natural landscape process system. Although the occurrences of
landslide hazards and its impact on human society cannot be prevented fully by
analyzing the slope stability condition, but the better understanding of geo-technical
attributes of the soil can contribute to greater knowledge and understanding about
the spatial distribution of slope instability which are very much essential for land
use planning. Landslides are the results of two interacting sets of forces; the pre-
condition factors, naturally induced which govern the stability conditions of slopes,
and the preparatory and triggering factors, induced either by natural factors or by
human intervention. Landslide analysis is mainly done by assessing Susceptibility,
Hazard and Risk (Einstein 1988). RS and GIS based landslide hazard zonation
approach had been studied by Anabalangan (1992), Muthu and Petrou (2007) and
Caiyan and Jianping (2009). Rowbothan and Dudycha (1998), Donati and Turrini
(2002), Lee and Choi (2003), Lee et al. (2004a, b), Lee and Pradhan (2006, 2007),
Pradhan and Lee (2010a, b, c), Sarkar and Kanungo (2004), Sharifikia (2007),
Pande et al. (2008) and Nithya and Prasanna (2010) studied and applied the
probabilistic model for landslide susceptibility and risk evaluation. Guzzetti et al.
(1999a) summarized many landslide hazard evaluation studies. Jibson et al. (2000)
and Zhou et al. (2002) applied the probabilistic models for landslide risk and hazard
analysis. Atkinson and Massari (1998) and Vijith and Madhu (2008) introduced the
logistic regression model for landslide hazard mapping. Landslide hazards were
evaluated by using fuzzy logic, and artificial neural network models were used in
the works of Gokceoglu et al. (2000) and Pistocchi et al. (2002). Landslide
Susceptibility mapping using either multivariate or bivariate statistical approach
considered the historical link between landslide controlling factors and the distri-
bution of landslides (Guzzetti et al. 1999b, c).
The models in connection to the slope stability, shallow and deep seated land-
slides were introduced and verified by Varnes (1958), Young (1963), Vanmarcke
(1977), Burton and Bathrust (1998), Bradinoni and Church (2004). The geotectonic
factors of slope instability were studied in details by Brudsen (1979), Windisch
(1991), Carson (1975, 1977) and Borga et al. (1998). Comprehensive list of sta-
bility factors commonly employed in the factors mapping approach was prepared
by Crozier (1986) and Tiwari and Marui (2001, 2002, 2003, 2004). Analytical
Hierarchy Process (AHP), a semi-quantitative method based on decomposition,
comparative judgement, and synthesis of priorities are often very much useful for
regional susceptibility studies as suggested by Saaty (1980), Soeters and Van

Westen (1996), Mwasi (2001), Nie et al. (2001), Yagi (2003), Komac (2006),
Yalcin and Bulut (2007), Kamp et al. (2008) and Yalcin (2008). The frequency ratio
(FR) model has become very popular as realistic quantitative approach in the
landslide susceptibility mapping. This approach is related with the historical
landslide events and their areal coverage. Lee and Pradhan (2007) argued that
frequency ratio model provides a correlation between the historical slide locations
and various influencing factors under consideration. Intarawichian and Dasananda
(2011) applied frequency ratio model to analyze slope instability and ascribed the
model as a popular quantitative method.
The present study deals with the estimation of factor’s weight and class
frequency ratios using ‘AHP’ and ‘FR’ model respectively. Integration between
factor’s weight (FW) and class frequency ratio (FR) was performed with the help of
a liner combination model. This is done to derive pixel wise landslide susceptibility
index values (LSIV) on GIS platform and to prepare landslide susceptibility map.
Landslide hazard risk analysis is the assessment of the probability of the damage
of land and associated resources of different magnitudes that may occur in a region
due to landslides. Earlier attempts to reduce landslide risk is largely a history of
management of landslide terrain by construction of protective structures or moni-
toring and warning systems, or the ever-increasing sophisticated methods for
mapping and delineating areas prone to landslide (Dai and Lee 2002). Risk of
landslide is normally defined as the expected number of lives lost, persons injured,
property damages and disrupted economic activities due to particular landslide
hazard for a given area and reference period (Varnes 1984). To reduce the risk from
the landslide events, the knowledge about potentiality to slope instability is
crucially needed. Information of landslide events are described in the form of
landslide susceptibility map of the concerned region and the preparation of this map
depends largely on the complex sets of knowledge of controlling slope movement
factors. Landslide analysis is mainly done by assessing susceptibility, hazard and
risk (Einstein 1988). The process of creating the maps involves several qualitative
or quantitative approaches (Soeters and Van Westen 1996; Guzzetti et al. 1999a;
Van Westen et al. 2008). Jibson et al. (2000), Praise and Jibson (2000) and Zhou
et al. (2002) have applied the probabilistic models for landslide risk and hazard
analysis. Landslide hazard risk map was made integrating landslide susceptibility
map and landslide risk exposure map on ARC GIS platform to identify the spatial
distribution of potential risk prone area in a representative drainage basin, over
which the attributes of land, soil and water exhibit a spatial order away from the
water divide in an interacting combination with human actions.
Tectono-statigraphically, the study area, Shivkhola Watershed is located in the
southern escarpment slope of Darjiling Himalaya, where high grade metamorphic
rocks of the Darjiling and Chungthang groups are thrusted over low grade meta-
morphic rocks of the Daling Group along the MCT (Main Central Thrust, Mallet
1875; Sinha-Roy 1982). Main Central Thrust (MCT) and Main Boundary Thrust
(MBT) are passing through the study area (Fig. 7.1). The MCT (a major ductile
shear zone) has divided two major litho-tectonic units, the Higher Himalayan
Fig. 7.1 Tectono-stratigraphy and past landslide in the Shivkhola watershed (Mandal and Maiti
2013)
Crystalline Sequence (HHCS) and the Lesser Himalayan Sequence (LHS) in

Darjiling Himalaya. The HHCS comprises of quartzo-feldspathic gneisses of both
igneous and sedimentary origin which suffers high grade of metamorphism (Catlos
et al. 2001). The LHS is dominated by garnet-biotite-mica schist and chlorite schists
in the upper part and slates and phyllites in the lower. The picturesque landslide
affected areas are Paglajhora, Tindharia, Mahanadi, Jogmaya and Shiviter. During
rainy season water percolates through the exposed rock joints and entrains the finer
particles and reduces the cohesive strength of the soil.
The rapid urbanization and expansion of tourism in Darjiling Himalaya are
putting unprecedented pressure on land and soil with the gradual elimination of
vergin forest land after independence. Lack of land use planning coupled with
vulnerable geological structure and heavy frequent rainfall have led to the formation
of vicious cycle of soil erosion and landslide during and after monsoon seasons,
causing devastating damage to human lives and properties. Significant studies in
Darjiling Himalaya identified the causes and consequences of major landslide
occurrences phenomena (Dutta 1966). Since 1968, the Shivkhola watershed of

Darjiling Himalaya faced 128 approachable landslide events, of which 76 events
were identified as reactivated (not 70 m away from old slided area) and 52 as fresh
events (70 m away from the old slided area) (Appendix D, Table D.1). The
considered landslide events took place in 16 years and out of which 12 years were
recognized as the major landslide years. All the landslide events occurred during the
monsoon period being triggered by continuous and heavy showers. Rainfall on all
the major landslide events date was more than the critical rainfall calculated after
Borga et al. (1998). Most of the landslide events occurred in the lithological unit of
Darjeeling Gneiss, Daling, Damuda and Siwalik.
In the present study of landslide hazard risk mapping in Shivkhola Watershed of
Darjiling Himalaya prioritized class ranking value (PCRV) and prioritized factors
rating value (PFRV) for each thematic data layers and their consistency checking
was accomplished through pair-wise comparison matrix as described by Saaty
(1980, 1990, 1994), and Saaty and Vargas (2001). Landslide hazard risk map was
made integrating landslide susceptibility map and landslide risk exposure map on
ARC GIS platform to identify the spatial distribution of potential risk prone area.
In the Shivkhola Watershed, Lower Paglajhora, Tindharia, Shiviter and Maha-
nadi are the major and prominent landslide location sites where settlement,
communication lines, and tea garden area are being affected severely. Since 1968,
Paglajhora alone has had 10 landslide events, all in the above mentioned landslide
event years. The majority of these landslides was dangerous as in most of the events
Hill Cart Road (NH-55) was affected and the communication line between Siliguri
and Darjiling was completely interrupted, from days to month. Paglajhora sinking
zone faced massive slope failures in 1998, 2002, 2005 and 2011 which indicates
that the occurrence of landslides in the region is ongoing. This poses a tremendous
threat to upslope settlement and Hill Cart Road (life line between Siliguri and
Darjiling Town). The landslide events at Tindharia also used to cut-off the Hill Cart
Road and brought tremendous threat to tourists, upslope settlements and tea
gardens. In Shiviter, around 8 acres of land were destroyed by the destructive slope
failure in the past 10 years. The physiographic configuration (arcuate) that provide a
favourable condition for producing hydrostatic pressure, proximity to Main Central
Thrust (MCT) and Main Boundary thrust (MBT), intensely fractured and sheared
bed rock, toe cutting and headward erosion of debris covered slope by fast flowing
tributaries, immense pressure over the fragile slope materials by man-made concrete
structure, moderate to steep slope gradient, improper drainage and accumulation of
highly anisotropic materials with a great thickness and low shearing resistance have
made these landslide locations in the Shivkhola watershed most unstable in char-
acter. The main purpose of the present study is to prepare landslide susceptibility
map and landslide hazard risk map applying RS and GIS semi-quantitative
approach and to compare risk dominated part of Paglajhora, Tindharia, and Shiviter
with the prepared risk map by incorporating landslide inventory statistics and
frequency ratio (FR) analysis.
The thematic data layers of all the landslide inducing factors were integrated to
prepare landslide susceptibility map using a linear combination model in GIS. The
Analytical Hierarchy Process (AHP) was used to derive prioritized factor rating
value (PFRV) and a FR Model was applied to obtain prioritized class rating value
(PCRV) for all the landslide triggering factors considered in the study. The inte-
gration between PFRV and PCRV were made in a linear combination model on GIS
platform to estimate landslide susceptibility index value (LSIV) for each pixel and a
suitable classification technique was incorporated to prepare the landslide suscep-
tibility map of the Shivkhola watershed. The data used in the present study are
Satellite image (IIRS P6/Sensor-LISS- III, Path-107, Row-052, date-18/03/2010),
modified SRTM data with scene size 1° lat. and 1° long. (date-5th April, 2008),
Google Earth Image (1st September, 2010), Geological Map (Geological Survey of
India, East Kolkata) and Topographical Map (78B/5, Survey of India). Data layers
for landslide inducing factors were generated using ERDAS Imagine 8.5, Arc View
and ARC GIS Software.
7.2.1 Landslide Susceptibility Assessment
The following section presents the methods and results of the landslide suscepti-
bility analyses in this study.
7.2.1.1 Determination of Landslide Triggering Factors
The landslide triggering factors were identified by interviewing the local people and
an investigation of the landslide sites in the watershed through intensive fieldwork.
During 10 days field work in July 2011, landslide locations were identified and
marked with GPS. Lithological structure, land use and land cover type around the
landslide scar, slope angle, construction of human structure and their role to
promote landslide, drainage network, altitude and slope aspects were investigated to
determine landslide triggering factors. The landslide triggering factors including
lithology, slope angle, drainage, slope aspect, slope curvature, lineament, upslope
contributing area (U.C.A.), land use/land cover, road contributing area (RCA) and
settlement density were taken into account to prepare landslide susceptibility map
of the Shivkhola watershed and their hierarchical arrangement was made on priority
basis. Shivkhola watershed is a small mountain basin where rainfall is uniformly
distributed over the entire area, so rainfall intensity was not considered in the
landslide susceptibility calculation (Mandal and Maiti 2011, 2012, 2013).
7.2.1.2 Generation of Landslide Inducing Factor Maps
First, the contour map at 20 m interval was prepared and digitized from the SOI
Topo-sheet (1987, 78B/5) at the scale of 1:50,000 and subsequently employed for
generating the Digital Elevation Model (DEM) using ARC GIS Software. Then
slope gradient, slope curvature and slope aspect map were derived from DEM and
classification was made to derive all these parameters in raster value domain fol-
lowing the earlier works of Dhakal et al. (2000). Surface curvature is a topographic
attribute that describes the convexity/concavity of a terrain surface. Curvature
depicts the slope gradient or slope direction (aspect), usually in a particular
direction. A positive curvature indicates that the surface is upwardly convex at a
grid cell and a negative curvature indicates that the surface is upwardly concave at
that grid cell. A value of zero indicates that the surface is flat. The expected values
of all three output raster images for a hilly area can vary from −0.5 to 0.5; for steep,
rugged mountains the value can vary between −4 and 4.
The lithological map of the study area was collected from Geological Survey
India (GSI), Kolkata (Eastern Region) and necessary modifications were incorpo-
rated after intensive field investigation. Final lithological map was prepared and
transformed into raster value domain on ARC GIS Platform. Class weight value for
each lithological class was assigned according to rock mass strength, described by
GSI. A Drainage density map (length of drainage/m2) was made from the topo-
graphical map (78B/5, 1987) and classified into ten equal intervals.
The lineament exhibits the zone of weakness surface of some linear to curvi-
linear features such as fracture, joint, fault etc. in the geological structure. There are
no basic differences between these three features. All these linear to curvilinear
features were identified as the same deformed surface where the propensity of slope
instability is very high. To generate lineament map (distance from lineament in
meters) of the Shiv-khola watershed, PCI-GEOMATICA was used and in the
extraction process 3 SRTM bands of wavelengths were taken into account: Near
Infrared (Band-I; 0.7–1.3 µm), Red (Band-II; 0.6–0.7 µm) and Green (Band-III;
0.5–0.6 µm). The ‘Lineament extraction’ algorithm was used to prepare lineament
map. The study area was classified into ten classes on the basis of the distance (m)
from lineaments.
Upslope Contributing Area is an effective indicator of drainage concentration
over space. The place with more contributing area encompasses more soil saturation
that reduces soil cohesion. The specific contributing area (total contributing area
divided by the contour length) is computed by distributing flow from a pixel among
its entire lower elevation neighbour pixel (Borga et al. 1998). Quinn et al. (1991)
proposed that the Fraction of Flow (Fi) allocated to each lower neighbour (i) is
determined by using Eq. 7.1. An upslope contributing area map was prepared based
on calculated contributing area value for each (0.25 km2) grid and it was divided
into 6 equal classes (Fig. 3.7, Chap. 3).
SiLi
Fi ¼ P ð7:1Þ
SiLi
where, the summation (Σ) is for the entire lower neighbour; S is the directional
slope, and L is an effective contour length that acts as the weighting factor. The
value of L used here is 10 m of the pixel size of the cardinal neighbour and 14.14 m
of the pixel diagonal for diagonal neighbour.
The road contributing area (RCA) map (Fig. 7.6) was made by multiplying road
contributing length (RCL) with road contributing width (RCW) with eight equal
classes from the concerned topographical sheet and it was converted into raster
value domain on ARC GIS Platform. The Settlement Density Map (Fig. 7.4) was
prepared by applying 3 × 3 karnel in ARC GIS platform and the whole basin was
classified into seven equal density classes. Land use and land cover (LULC,
Fig. 7.5) map of the Shivkhola watershed is prepared with the help of LISS-III
Satellite Image (2010) and Google Earth Image in consultation with SOI Topo-sheet
(78B/5). After verifying the ground truth with GPS a land use and land cover map is
developed in GIS. The Shivkhola Watershed was classified into 10 individual land
use type: (i) bare surface, (ii) agricultural land, (iii) jungle, (iv) roads, (v) settlement,
(vi) tea garden, (vii) open forest, (viii) degraded forest, (ix) mixed forest and (x)
dense forest (Fig. 4.2, Chap. 4).
7.2.1.3 Landslide Inventory Map
A Landslide Distribution Map/Inventory Map (Fig. 2.1, Chap. 2) was created to

determine landslide affected area (%) and frequency of landslide for each class of
the landslide inducing factors/factors. Mandal and Maiti (2011) identified major and
minor landslide locations during field investigation and mapped them by evaluating
the SOI topographic map (78B/5), satellite image (IRS LISS-III, 2000), SRTM data
(2008), and Google Earth Image (2010). Several field investigations were con-
ducted to identify the landslide locations and to cross-check the prepared landslide
map. Then, the map was digitized and converted into raster value domain in ARC
GIS Software. All the landslide triggering factor maps were linked with the
prepared landslide inventory map to understand the degree of importance of each
factor in landsliding.
7.2.1.4 AHP and Quantification to Each Factor Map/Prioritized Factor

Rating Value
AHP is a decision making and semi-quantitative value judgement approach which

serve the objectives of the decision makers. This process is employed to support the
decision on the instability rank of the factors by estimating prioritized factor rating
value (PFRV). In the AHP, different factor preference and their conversion into
Table 7.1 Scale of preference between two parameters

Scale Degree of Explanation
preference
1 Equally Two activities contribute equally to the objective
3 Moderately Experience and judgement slightly to moderately favour
one activity over another
5 Strongly Experience and judgement strongly or essentially favour
one activity over another
7 Very strongly An activity is strongly favoured over another and its
dominance is showed in practice
9 Extremely The evidence of favouring one activity over another is
of the highest degree possible of an affirmation
2, 4, 6 and 8 Intermediate Used to represent compromises between the references
values in weight 1, 3, 5, 7 and 9
Reciprocals Opposites Used for inverse comparison
numerical value was accomplished with the help of comparative oral judgment and
synthesis of priorities. A couple comparing matrix was constructed on the basis of
the preference of a factor as compared with the other factor and arithmetic mean
method was applied to arrange landslide triggering factors hierarchically and to
determine prioritized factor rating value/eigenvector (PFRV) with reasonable
consistency ratio (CR) on MATLAB software after Saaty (1980) (Table 7.2). To
develop pair-wise comparison matrix, each factor/class was rated against every
other factor by assigning a relative dominant value ranging between 1 and 9 on the
basis of the relative importance of the factors in terms of landslide frequency. The
value also varies between the reciprocals 1/2 and 1/9 for inverse comparison
(Table 7.1).
Another appealing feature of the AHP is the ability to evaluate pair-wise rating
inconsistency. The eigenvalues enable to quantify a consistency measure which is
an indicator of the inconsistencies or intransivities in a set of pair-wise ratings.
Saaty presented that for a consistent reciprocal matrix, the largest eigenvalue λMax is
equal to the number of comparisons n (Table 7.2).
In AHP, an index of consistency, known as the CR (Consistency Ratio), is used
to indicate the probability that the matrix judgements were randomly generated
(Saaty 1994).
CR ¼ CI=RI ð7:2Þ
where RI is the average of the resulting consistency index depending on the order of
the matrix given by Saaty and CI is the consistency index that is expressed in the
following equation. If the value of CR is smaller or equal to 10 %, the inconsistency
is acceptable, but if the CR is greater than 10 %, the subjective valued judgement
needs to be revised.
A measure of consistency, called consistency index CI, is defined as follows:
Table 7.2 Landslide triggering factors and determined prioritized factor weights
Factors 1 2 3 4 5 6 7 8 9 10 Prioritized
rating
Slope 1 2 3 4 5 6 7 8 9 9 0.2944
Lithology 1/2 1 2 3 4 5 6 7 8 9 0.2150
Drainage 1/3 1/2 1 2 3 4 5 6 7 8 0.1537
Lineament 1/4 1/3 1/2 1 2 3 4 5 6 7 0.1087
Curvature 1/5 1/4 1/3 1/2 1 2 3 4 5 6 0.0764
UCA 1/6 1/5 1/4 1/3 1/2 1 2 3 4 5 0.0535
RCA 1/7 1/6 1/5 1/4 1/3 1/2 1 2 3 4 0.0375
LULC 1/8 1/7 1/6 1/5 1/4 1/3 1/2 1 2 3 0.0266
Settlement 1/9 1/8 1/7 1/6 1/5 1/4 1/3 1/2 1 2 0.0193
density
Slope aspect 1/9 1/9 1/8 1/7 1/6 1/5 1/4 1/3 1/ 1 0.0149
2
CI = 0.0615; R.I (random index) = 1.49 and CR = 0.0413 (consistent). RCA road contributing area; UCA
upslope contributing area; LULC land use and land cover
Source Saaty (1980)
CI ¼ kMax n/n 1 ð7:3Þ
Saaty and Vargas (2001) randomly produced reciprocal matrices using scales 1/9, 1/8,
1/7, …, 1, …, 8, 9 to evaluate a so called random consistency index (RI). The average
RI of 500 matrices is given in Table 7.3.
7.2.1.5 Frequency Ratio Model and Prioritized Class Rating Value
Frequency ratio (FR) model is also a well accepted and popular quantitative
approach for the preparation of landslide susceptibility map. Lee and Talib (2005),
Pourghasemi (2007), Lee and Pradhan (2007), Jadda (2009), Avinash and
Ashamanjari (2010), Intarawichian and Dasananda (2011) successfully applied
‘FR’ model to generate landslide susceptibility zoning map. To obtain frequency
ratio (FR) for each class of all the data layers a combination has been established
between landslide inventory map and criterion maps using following equation.

NpixðSi Þ NpixðNi Þ
Fri ¼ P P ð7:4Þ
NpixðSi Þ NpixðNi Þ
NpixðSi Þ The number of pixels containing slide in each class (i),

NpixðNi Þ Total number of pixels having class (i) in the whole watershed,
P
N Total number of pixels containing landslide,
P pixðSi Þ
NpixðNi Þ Total number of pixels in the whole area of the watershed.
Table 7.3 Random index (RI)

N 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49 1.51 1.53 1.56 1.57 1.59
201
The derived frequency ratio (FR) value of more than ‘1’ indicates, strong and
positive relationship between landslide occurrences in each class of the data layers
and high landslide susceptibility where ‘FR’ value of less than ‘1’ depicts the
negative and low landslide susceptibility. In this study, ‘FR’ values for each class
were accepted as prioritized class rating value (PCRV) or prioritized class weight
(PCW).
7.2.1.6 Linear Combination Model and Landslide Susceptibility Map
Avinash and Ashamanjari (2010) and Intarawichian and Dasananda (2011) used a
landslide susceptibility index value (LSIV) which is the summation of class-and
factor-weighted values.
‘FR’ values for each class (PCRV) or prioritized class rating value, (Table G.1,
Appendix G) as well as prioritized factor’s weighted values (PFRV) for each factor
map was taken into account in calculating the landslide susceptibility index value
(LSIV) with the following linear combination model:
X
n
LSIV ¼ ðWi FRi Þ FV ð7:5Þ
i¼1
where, n: total number of factors included in the study (n = 10); Wi: Factor’s weight
(PFRV), FV IS factor value, and FRi: Class Frequency Ratio/class weight.
The ‘LSIV’ varied from ‘4.81’ to ‘16.00’. Higher the value of ‘LSIV’, greater was
the propensity of landslide phenomena and vice versa. The LPIV based frequency
curve showed many oscillations. To classify the watershed into 5 susceptibility
zones moving averages with averaging window lengths of 3, 5, 7, and 9 were
considered for smoothing the frequency distribution curve (Fig. 7.2). After analyzing
four new curves, the Shivkhola Watershed was classified into 5 landslide suscep-
tibility zones i.e. very low, low, moderate, high, and very high with class boundaries
were demarcated at the significant changes of gradient of the curves. The abrupt
change points on frequency curve (landslide threshold boundaries) were 7.05, 9.29,
11.5, and 13.8 which were recognized as class boundaries to classify the
map. A 3 × 3 ‘majority filter’ technique was applied to the map as a post-classifi-
cation filter to reduce the high frequency variation.
To verify the landslide susceptibility map, landslide density under each
susceptibility class was computed. The landslide inventory map was crossed with
prepared landslide susceptibility map to derive landslide affected pixels for each
susceptibility classes (zones). Research by Sarkar and Kanungo (2004) indicates
that the higher the landslide density, greater is the probability and larger the area is
affected by landslide in each landslide susceptibility class.
Fig. 7.2 Frequency distribution of landslide susceptibility index value of the Shiv-khola
watershed in West Bengal, India
7.2.1.7 Accuracy Assessment of the Landslide Susceptibility Map

with Field Data (GPS Survey)
Accuracy assessment is a general term for comparing the classification with

geographical data that are assumed to be true, in order to determine the classifi-
cation process that was accomplished by using Erdas Imagine (8.5). True data were
derived by ground truth verification with the help of GPS from the existing 50
landslide locations. Simultaneously, a set of randomly selected 50 reference pixels
points from the classified image corresponding to the true data (GPS record) were
used for evaluating the validity of landslide susceptibility map after Congalton
(1991).
7.2.2 Landslide Risk Assessment
In this study, landslide risk map was made with the help of following principle
(Mandal and Maiti 2012).
Landslide Hazard Risk = Landslide susceptibility × Landslide Risk Exposure/
Intensity of Risk Elements.
The entire methodology to prepare the landslide hazard risk map of the Shiv-
khola Watershed could be summarized under following heads (Fig. 7.3).
Fig. 7.3 Methodology of Geo-referencing and Digitization of the DATA

landslide hazard risk map (SOI Topo-sheet, Satellite Image LISS III)
Settlement Density Road Contributing Weighted Land use

Map Area Map & Land Cover Map
RASTER
INTEGRATION
RISK EXPOSURE MAP
Multiplication of risk exposure LANDSLIDE HAZARD

map with Landslide RISK MAP
Susceptibility Map
7.2.2.1 Preparation of Weighted Risk Factor/Element Maps (Weighted

Land Use/Land Cover Map, Road Contributing Area Map
and Settlement Density Map)
In the present study, land use and land cover, road network, and settlement were
considered as important risk factors/elements because these three are worst affected
by landslide events in the study area. To derive the weighted risk factor maps the
developed numerical scale of 1–10 was applied to assign the scores for each class of
the risk factor maps. Weighted land use/land cover map (Fig. 7.5) is the expression
of the intensity of risk induced land use pattern in the Shivkhola watershed.
A weighted land use/land cover map was developed assigning more weightage
values considering the landslide contributing units to the significant landslide
triggering land use pattern i.e. tea garden area, degraded forest, bare surface and
agriculture. Weighted value to each class/range of the road contributing area (RCA)
map was assigned considering the intensity and impact of road network on land-
slide and thus a weighted road contributing area map (Fig. 7.6) was made on GIS
platform. Settlement Density Map was prepared by assigning more weighted values
to high density class and low for low density class for raster integration and a
weighted settlement density map (Fig. 7.4) was made accordingly. Three risk factor
maps were classified into low, moderate and high intensity zones.
7.2.2.2 Integration Between Weighted Land Use and Land Cover, Road
Contributing Area and Settlement Density and the Development
of Risk Exposure Map
To integrate risk factor maps prioritized class rating values (PCRV) and prioritized
factor rating value (PFRV) were obtained for each class and each risk factors maps
developing couple-comparing matrix (Table 7.4) according to Saaty (1980). Then,
Fig. 7.4 Settlement density map
a linear combination model was performed on ARC GIS platform to derive land-
slide risk exposure co-efficient value. The derived landslide risk exposure co-effi-
cient value ranges between 0.19 and 7.24 and here the values of 0.75, 1.45, 2.55,
3.85, 4.90, and 6.15 were taken into account as threshold points to classify the
watershed into 7 landslide risk intensity zones. Risk exposure map (Fig. 7.7) shows
the intensity of risk elements over the space which were delineated here as very low
to very high ranges. The intensity of risk elements are high to very high at Tindharia
Upper as well as Lower, very few parts at Gayabari upper slope, Giddapahar,
Shiviter, and upper Paglajhora.
Table 7.4 Determined prioritized class rating value (PCRV) and prioritized factor rating value
(PFRV) of three landslide hazard risk element maps
Weighted LULC Low Moderate High Very high PFRV-
map 0.691
PCRV 0.06 0.12 0.25 0.55
Weighted RCA map Very Low ML M MH H VH PFRV-
low 0.218
PCRV 0.02 0.04 0.05 0.09 0.13 0.24 0.43
Weighted settlement Very Low ML M MH H VH PFRV-
density map low 0.091
PCRV 0.03 0.04 0.05 0.11 0.14 0.27 0.37
PCRV prioritized class rating value, PFRV prioritized factor rating value, LULC land use and land
cover, RCA road contributing area, ML moderately low, M moderate‚ MH moderately high,
H high, VH very high
7.2.2.3 Preparation of Landslide Hazard Risk Map
The prioritized class rating value (PCRV) against each class of landslide suscep-
tibility map and risk exposure map were estimated by developing couple-comparing
matrices on MATLAB Software (Table 7.5) after Saaty (1980). Then, raster inte-
gration between landslide susceptibility and landslide risk exposure was performed
by overlay analysis on ARC GIS platform and finally landslide hazard risk
co-efficient (R) values were derived for each pixel. To classify ‘R’ value the same
method was followed like the landslide susceptibility map and the Shivkhola
Watershed was divided into four landslide hazard risk zones i.e. low, moderate,
high and very high.
7.2.2.4 FR Study to Establish the Validity of Landslide Hazard Risk

Zones
To evaluate the validity of landslide hazard risk map frequency ratio (FR) value was
estimated for each landslide hazard risk class by means of a ratio between landslide
hazard risk area (%) and landslide hazard risk events (%). The ‘FR’ value
approaching towards ‘0’ indicates lower landslide probability and the value
approaching away from ‘0’ or toward ‘1’ or more than ‘1’ denotes greater chances of
landslide risk event in future. The records of the landslide hazard risk events in the
unstable terrain of the Skivkhola watershed were collected from the study accom-
plished by Basu and Ghatowar (1988), Basu and Sarkar (1985, 1988), Basu and
Ghosh (1993), Basu and Maiti (2001), Maiti (2007a, b), Ghosh et al. (2009), and
author himself that were taken into account to estimate frequency ration (FR). The
records depict that since 1968–2011, the Shivkhola watershed faced 128
Table 7.5 Prioritized class rating value (PCRV) of landslide susceptibility and landslide risk exposure with consistency ratio (CR)
Landslide susceptibility Very low (VL) Low (L) Moderately low Moderate (M) MH H VH Consistency
map (ML) ratio
PCRV 0.02 0.04 0.12 0.13 0.17 0.19 0.33 (CR) = 0.02
Landslide risk exposure Very low (VL) Low (L) Moderately low Moderate (M) MH H VH Consistency
map (ML) ratio
PCRV 0.027 0.036 0.053 0.103 0.143 0.266 0.376 (CR) = 0.049
207
approachable landslide events and amongst them 76 events were treated as reacti-
vated (not 70 m away from old slided area) and 52 as fresh events (70 m away from
the old slided area) (Appendix D, Table D.1). Out of 16 prominent landslide events
year during the period 1968–2011, 12 years had been recognized as the major
landslide hazard risk events years because in these years destructive landslide events
completely cut-off communication lines, destroyed human settlements, reduced tea
garden area and threatened human lives and properties severely. Considering 36
landslide hazard risk events which occurred within 12 landslide risk event years at
different parts of the Shivkhola ‘FR’ values were derived and probable chances of
future hazard risk events were estimated for each landslide hazard risk zone.
7.2.2.5 Accuracy Assessment of the Landslide Hazard Risk Map

with Field Data (GPS Survey)
The accuracy assessment of the landslide hazard risk map was made by using Erdas
Imagine (8.5). Accuracy assessment is a general term for comparing the classifi-
cation with geographical data that are assumed to be true, in order to determine the
classification process. Basically, the true data were derived for ground truth veri-
fication with the help of GPS from the existing/active 50 landslide location with risk
elements (settlement, road and tea and agriculture). Simultaneously, a set of ran-
domly selected 50 reference pixels points from the classified image corresponding
to the true data (GPS record) were used for evaluating the validity of landslide
hazard risk map (Congalton 1991).
7.3.1 Relationship Between Landslide Susceptibility

and Landslide Triggering Factors
Landslide susceptibility map of the Shivkhola Watershed was the product of an

interaction between factors and existing landslide. Slope gradient (Fig. 2.6, Chap. 2)
of the watershed varies from very gentle gradient (around 10°) in the mid central and
mid-lower part to that of high (more than 60°), towards the marginal part/water
divide. Most of the landslide phenomena were found in the area of above 35° slope
gradient where class weight value ranges between 5.35 and 77.56 (Appendix G,
Table G.1). South, south east, north, east and north easterly facing slope (Fig. 2.7,
Chap. 2) were registered with highest class weight values of 29.23, 23.63, 69.63, 53.
51, and 14.86 respectively. All these slope facets were associated with moderate to
high landslide susceptibility and large number landslide events. The derived ‘FR’ and
class weight values revealed that high to very high landslide susceptibility zones are
characterized by high positive and negative curvature. Lower Paglajhora, Gayabari
(Lower), Shiviter (Lower), Tindharia T.E. were characterized by upwardly concave

or negative curvature and highest class weight value ranging from 52.09 to 165.71.
The marginal part of the watershed mainly Upper Paglajhora, 14 miles (upslope)
bustee, Gayabari (Upper), and Tindharia (Upper) registered high positive curvature
(Fig. 2.8, Chap. 2) with maximum landslide frequency and class weight of more than
40. Lithologically, Darjiling Gneiss, Gorubathan, Lingste Granite and Reyang
Formation (Fig. 2.4, Chap. 2) showed the maximum number of landslide phenomena.
Probability of Landslide phenomena was very high for the lithological composition
of gneiss, mica-schist and granulite. Class weight values of Lingtse Granite,
Gorubathan Formation and Chungtung Formation were 48.98, 50.31 and 23.10
respectively (Table G.1, Appendix G). All these lothological groups were accom-
panied with large number of landslide activities and greater chances of landslip
probability in future. Drainage density (Fig. 3.6, Chap. 3) was very high at Lower
Paglajhora, Gayabari and Shiviter T.E. which were attributed by high landslide
susceptibility and high frequency ratio (>2.5). The value of the drainage density
increased away from the marginal part to the central part. The area having more that
11 km length of drainage/km2 were attribute with highest class weight (>140) and
greater probability of landslide phenomena. The values of Upslope Contributing
Area (UCA) increases away from the water divide and the maximum of 20.98 km2
experienced the lower most portion of the watershed (Fig. 3.7, Chap. 3). The upslope
contributing area having less than 5 km2 experienced less saturation excess run-off
and less intensity of landslide. The more contributing areas were registered along the
main rivers that had maximum length and thus maximum flow. The study envisaged
that the place with the UCA of 5.00–10.00 km2 and 10.00–15.00 km2 were attributed
as high ‘FR’ value of 1.21 and 1.48 and class weight values of 20.93 and 48.39
respectively. These places were very much prone to landslide. In Shivkhola
Watershed tea garden, jungle, roads and settlement (Fig. 7.5) were characterized by
high ‘FR’ of 1.23, 1.15, 1.98 and 1.14 and highest class weight of 23.73, 15.62, 99.31
and 14.09 (Table G.1, Appendix F). The analysis showed that tea garden, road and
settlement were dominated by high intensity of landslide and could be treated as
maximum probable areas of landslide occurrences. Road contributing area (RCA)
was high at the places of Tindharia, Paglajahora, Mahanadi and Shiviter where the
landslide frequency was also very high (Fig. 7.6). At all those places the RCA ranged
from 0.008 to 0.014 km2 and the class weight value ranged between 29.84 and 169.
12. In the study area, construction of roads and slope modification caused by human
intervention were responsible for slope instability. The moderate to high intensity of
human settlement at Tindharia, Gayabari, Shiviter, Mahanadi and Lower Paglajhora
(Fig. 7.4) depicted high ‘FR’ and maximum class weight as well as greater proba-
bility of landslip (Fig. 7.7).
In Shivkhola Watershed, Lower Paglajhora, Shiviter and Tindharia were very
highly susceptible to landslide; Upper Paglajhora, Gayabari, 14 Miles Bustee and
Nurbong T.E. were characterized by high landslide susceptibility; Mahanadi and
Giddapahar were of moderate landslide potentiality; and marginal waxing slope of
water divide and low-central wanning slope were registered with low landslide
susceptibility (Fig. 7.8).
Fig. 7.5 Weighted land use and land cover map
The study revealed that around 50 % area of the Shivkhola watershed was
classified as being in the moderate to very high landslide susceptibility with 71 %
landslide phenomena. Low to very low susceptibility zones together accommodate
27 % of the landslide phenomena (Table 7.6). Landslide density in each suscep-
tibility class was derived to evaluate the intensity of landslide activities. The
landslide density value ranges from 0.031 to 0.25. The calculated density value of
Fig. 7.6 Weighted road contributing area map
0.25 and 0.15 for very high and high landslide susceptibility zones of the watershed
depicts the higher intensity of landslide activities compared to other landslide
susceptibility zones. Here, frequency study shows that more than 19.58 % area is
attributed with high to very high landslide probability, around 48.8 % with mod-
erate landslide probability and remaining area with low landslide probability
(Table 7.6). In landslide susceptibility classes of high and very high, the ‘FR’
values are 2.41 and 3.39 that indicate greater chances of landslide probability. The
determined landslide density and frequency ratio reveals that the areas with high
Fig. 7.7 Risk exposure/intensity map
and very landslides susceptibility are expected to have fresh landslide phenomena
and here lies the validity of the present landslide susceptibility mapping approach.
A relationship was established between landslide potential index and landslide
affected pixels which show that 27.22, 45, 50.03, 76.03 and 95.62 % landslide
affected areas are distributed in 8.75, 28.66, 45, 78 and 92 % landslide susceptible
areas. Around 35 % landslide affected pixels are distributed in 27 % of high to very
high landslide potentiality zones that indicate the higher probability of landslide
activities (Fig. 7.9). On the other hand 73 % landslide susceptible areas are
attributed with 65 % landslide affected pixels.
Fig. 7.8 Landslide susceptibility map
Table 7.6 Relationship between landslide susceptibility (%), landslide frequency (%) and
frequency ratio (FR) and landslide density
Landslide Pixel % Landslide pixel % Frequency Landslide
susceptibility (15 × 15 m) (B) (15 × 15 m)(a) (A) ratio (FR) density (a/b)
(b) (A/B)
Very low 7,707 9.03 245 4.47 0.50 0.031789282
Low 35,386 41.46 1,247 22.74 0.54 0.035239925
Moderate 34,364 40.26 2,676 48.8 1.21 0.077872192
High 6,932 8.12 1,074 19.58 2.41 0.154933641
Very high 964 1.30 242 4.41 3.39 0.251037344
7.3.2 Accuracy Result of Landslide Susceptibility Map
The comparison between true data and randomly selected data from the classified
image was made on GIS Platform that showed the overall classification accuracy of
92.22 % and overall Kappa Statistics was 0.894. The class wise accuracy result is
shown in Table 7.7 that indicates acceptable results.
Fig. 7.9 Pixel wise distribution of landslide potentiality index
Table 7.7 Accuracy assessment/comparison of landslide susceptibility with field data

Class Classified Number Producers Users Accuracy
name total correct correct accuracy total
Very low 0 5 0 0.00 0.00
Low 4 3 0 75.00 0.00
Moderate 11 10 9 90.91 90.00
High 16 15 13 93.75 86.67
Very 19 17 17 89.47 100.00
high
Total 50 50 39
7.3.3 Analysis of Landslide Hazard Risk
Study represents that the watershed is dominated by high landslide hazard risk
followed by very high, moderate and low landslide hazard risk. In terms of areal
coverage, 17.92, 25.37, 29.35 and 27.34 % area of the watershed is recognized by
low, moderate, high and very high landslide hazard risk respectively (Fig. 7.10).
Figure 7.11 shows the spatial distribution of landslide hazard risk where Lower
Paglajhora, Tindharia and Shiviter are registered with high risk exposure due to
Fig. 7.10 Distribution of pixel/area (%) within landslide hazard risk classes
high intensity of risk elements (road, settlement and tea garden) that leads to high to
very high landside hazard risk. The developed landslide hazard risk map of the
Shivkhola Watershed exhibits a clear picture about the spatial location of vulner-
ability and risk of landslide hazard. The marginal parts and lower most segment of
the basin experiences low intensity of risk elements and are also least affected by
slope instability. Landslide hazard risk is very high at lower paglajhora, Shiviter
Tea Garden Area, Tindharia etc. Lower Gyabari and Sepoydhura are the places of
moderate to low probability of landslide hazard risk. Moderate level of risk is found
at middle as well as lower section of the Shivkhola Watershed. It can be concluded
that the watershed is dominated by the moderate to high level of landslide hazard
risk. Landslide hazard risk is too high at Mahanadi, Tindharia and Shiviter because
of higher intensity of the risk elements.
After thorough analysis of the intensity and magnitude of damage, 36 landslide
events are marked as risky out of a total of 128 since 1968. Again the distribution of
landslide among various landslide risk zone shows that very high, high and mod-
erate risk areas experienced 62, 43 and 22 landslide events respectively and out of
Fig. 7.11 Landslide hazard risk map with few significant risk events (field photo)
those 16, 13 and 7 are identified as riskful events. Frequency ratio for each low,
moderate, high and very high landslide hazard risk zone are 0.00, 0.66, 1.42 and
2.48 respectively which also shows very low chance, tendency towards equal
chance, high and very high chance of landslide hazard risk events in the respective
hazard risk zones (Table 7.8). Analysis of frequency ratio conclude that there is a
greater probability of future occurrence of risk events at the places of Lower
Paglajhora, Tindharia and Shiviter in future and these three locations are presently
existing in high to very high landslide hazard risk zone where frequency ratio is
more than ‘1’.
The determined threshold slope angle of Paglajhora, Tindharia and Shiviter
ranges from 13° to 37° and cohesion of the soil from 0.01 to 0.70 with greater
percentage of sand particle. Very fragile and fragmented lithological composition
helps easy percolation of rain water that generates adequate pore-water pressure for
promoting downward movement of slope materials at Paglajhora and Tindharia.
The existence of moderate to high intensity of risk elements and human intervention
associated with all favourable geomorphic and geo-hydrologic landslide triggering
Table 7.8 Frequency ratio (FR) analysis and landslide probability for each risk zone
Landslide risk No. of Area Frequency land- Frequency of Frequency of risk Frequency Probability/chance of
class pixels (%) slide events risk event events (%) ratio (FR) the risk event
Low (L) 8,982 27.34 1 0 0.00 0.00 Very low
Moderate (M) 9,641 29.35 22 7 19.44 0.66 Tendency towards equal
High (H) 8,333 25.37 43 13 36.11 1.42 High
Very high (VH) 5,888 17.92 62 16 44.44 2.48 Very high
Total 32,844 100 128 36 100
217
factors have recognized Lower Paglajhora, Tindharia and Shiviter as high to very
high landslide hazard risk zone in the Shivkhola Watershed.
The thickness of the soil and that of the saturated soil during monsoon are
measured to be 4.5 m (Shiviter T.E.) and 7.25 m (Lower Paglajhora) and 1.28 m
(Shiviter T.E) and 1.30 m (Lower Paglajhora) respectively at the back wall of the
landslide scar though it is a bit lower further upslope on steeper section. The wet soil
buck density is measured to be 1.96 g/cc and density of water is 1.07 g/cc. The angle
of internal friction varies from 19° to 23° with an average of 21°. The calculated
critical rainfalls of two major landslide prone parts of the Shivkhola watershed are
105.88 mm/day (Shiviter T.E.) and 88.93 mm/day (Lower Paglajhora) after Borga
et al. (1998). Following Chow (1951, 1954, 1964) the calculated rainfall of
90.539 mm which is less than the critical rainfall of those two places at the
recurrence interval of 1.01 year with 99 % probability.
The determined safety factor (FS) of all three locations are less than ‘1’ which is
measured considering the stress parameters such as major principal stress (r1 ),
minor principal stress (r3 ), normal stress (rn ), shear stress (s), angle of internal
friction (Φ), cohesion (C), shear strength (λ), and rupture angle (α) applying Direct
Shear Test Mechanism and developing Mohr’s Stress Circle (Table 7.9). All the
above mentioned parameters have recognized Paglajhora, Tindharia and Shiviter as
significant unstable sections in the Shivkhola watershed. Not only that the slope
steepening developed by road-cut benches and toe erosion, plying of heavy loaded
vehicles and its enormous pressure on more fragile slope materials, depletion of
forest cover in a rapid pace, continuous and regular orographic rainfall in rainy
season, easy percolation of water through fragmented rock-soil composition and
increase pore water pressure have caused destructive slope failure and damaged
human structure and disrupted normal life by cutting-off the communication lines at
all these three locations and have treated them as most significant landslide hazard
risk prone sectors of the Shivkhola Watershed.
A comparative study has been made here to establish the interrelationship
between all the landslide triggering factors (slope angle, slope curvature, slope
aspect, lithology, drainage density, upslope contributing area, settlement density,
road contributing area and land use and land cover.) considered in the present study
as well as to figure out the different levels of abstraction of all the parameters in
landslide hazard risk zones (Table 7.8). Study revealed that high and very high
landslide hazard risk zones covering the places of Pahlajhora sinking zone,
Tindharia, Shiviter, and 14 Miles Bustee are closely associated with 35–45° slope
angle; north-east, south-east, north, east and south slope aspect; high posivite and
high negative curvature; drainage density of 3.5–6.5 km/km2; upslope area of more
than 5 km2; high settlement density; moderate to high road contributing area; and
land use/land cover of settlement, road, degraded forest and open forest
(Table 7.10).
Table 7.9 Measured geo-technical parameters for Tindharia, Shiviter and Paglajhora
Location r3 rn r1 s Φ C λ α F.S.
1. Tindharia Tea Garden 0.90 1.31818 2.15440 0.59036 19°28′ 0.123 0.588929 35°16′ 0.9975760
2. Lower Paglajhora (14 Miles Bustee) 0.82 1.28942 2.3212 0.69594 22° 0.149 0.6699594 34° 0.962668
3. Lower Paglajhora 0.96 1.4583 2.43794 0.6987 19° 0.162 0.6641329 35°30′ 0.950526
4. Tindharia Rly. Stn. 0.92 1.4044 2.553 0.74598 24° 0.10 5 0.730279 33° 0.9789525
5. Shiviter T.E. 0.97 1.402407 2.3351579 0.63508 21°30′ 0.07 0.6224228 34°15′ 0.9800699
r1 = major principal stress; r3 = minor principal stress; rn = normal stress; s = shear stress; Φ = angle of internal friction; C = cohesion; λ = shear strength;
α = rupture angle; and F.S. = safety factor
219
220
Table 7.10 Comparative study between landslide triggering factors and landslide hazard risk zones
Hazard Lithology Slope Aspect Curvature Drainage U.C.A Settlement R.C.A. Land use/ Location
risk class density (km2) density land cover
(km/km2)
Very high Gneiss/mica- 45°–70° North/ High Positive/ 5–6.5 >10 >1.25 High Degraded Lower pag-
schist east/ high negative forest and lajhora /shi-
south settlement viter /
(20.11 and Tindharia
12.55)
High Quatzite 30°–45° NE/SE Moderate 3.5–5.5 5–10 0.5–1.25 Moderate Road/mixed /Gayabari/
schist, phyllite positive/ and more to high and open 14 miles bu-
negative than 8 forest stee/Nur-
bong T.E
Mahanadi/
Moderate Soft sand- 20°–30° Flat Low 2.00–3.50 <5 0.25–0.50 Low Jungle Giddapahar/
stone, shale upper
paglajhora
Low Conglomerate/ <20° NW Around (0) <2.00 <5 <0.25 Low Dense forest Marginal/
shale/ mid-central
sandstone part
UCA Upslope contributing area; RCA Road contributing area
7 Application of Analytical Hierarchy Process (AHP) …
Table 7.11 Accuracy assessment/comparison of landslide susceptibility with field data

Landslide hazard Classified Number Producers Users accu- Accuracy
risk class total correct correct racy (%) total (%)
Very low 0.00 1 0 – –
Low 0.00 1 0 – –
Moderately low 0.00 3 0 – –
Moderate 19 17 15 78.95 88.24
Moderately high 8 6 5 62.50 83.33
High 17 16 16 94.12 100.00
Very high 6 6 6 100.00 100.00
Total 50 50 42
7.3.4 Accuracy Result of Landslide Hazard Risk Map
The comparison between assumed true data and randomly selected data from the
classified image has been made on GIS Platform that shows the overall classifi-
cation accuracy is 92.89 % and overall Kappa Statistics is 0.8929 %. The class wise
accuracy result is shown in Table 7.11 that indicates acceptable results.
7.4 Conclusion
Very fragile and fragmented lithological composition helped easy percolation of

rain water that generated adequate pore-water pressure for promoting downward
movement of slope materials at Paglajhora and Tindharia. The combinations of
moderate to high intensity of risk elements and human intervention associated with
all favourable geomorphic and geo-hydrologic landslide triggering factors have
recognized Lower Paglajhora, Tindharia and Shiviter as high to very high landslide
hazard risk zone in the Shivkhola Watershed. The calculated critical rainfalls of two
major landslide prone parts of the Shivkhola watershed are 105.88 mm/day
(Shiviter T.E.) and 88.93 mm/day (Lower Paglajhora) after Borga et al. (1998). Log
probability analysis after Chow (1951, 1954, 1964) shows that the rainfall of
90.539 mm is expected at the recurrence interval of 1.01 year with 99 % proba-
bility. This revealed high-potentiality of slide at these locations. Not only that, the
slope steepening caused by road-cut benches and toe erosion, plying of heavy
loaded vehicles, depletion of forest cover in a rapid pace, continuous and regular
orographic rainfall in rainy season, easy percolation of water through fragmented
rock-soil composition and increase pore water pressure have caused destructive
slope failure at all these three locations.
The derived prioritized factor rating values (PFRV) were high for Slope steep-
ness (0.2944), lithology (0.2150), drainage (0.1537), and lineaments (0.1087)
indicating as the significant contributing factors for landsliding in the Shivkhola
watershed. The slope aspect, settlement density and LULC were registered with
minimum prioritized factor rating values of 0.0149, 0.0193 and 0.0266 that proved
these factors as less significant to promote landslide activities. Road contributing
area (RCA), slope curvature and upslope contributing area were proved to be
moderately important in the present context. Analytical hierarchy process is proved
to be important to efficiently identify the landslide triggering factors of most
importance. It may again be helpful as a support in decision making process for
efficient management. The study revealed that lithological composition with steep
slope and drainage network orientation are to be given more priority in the decision
of structural construction, specially the construction of roads.
References
Eng Geol 32:269–277
Atkinson PM, Massari R (1998) Generalized linear modeling of susceptibility to landsliding in the
central Apennines, Italy. Comput Geosci 24:373–385
Avinash KG, Ashamanjari KG (2010) A GIS and frequency ratio based landslide susceptibility
mapping: Aghnashini river catchment, Uttara Kannada, India. Int J Geomat Geosci 1
(3):343–354
Basu SR, Sarkar S (1985) Some consideration on recent landslides at Tindharia and their control,
Indian. J Power River Valley Dev 1985:190–194
Basu SR, Sarkar S (ed) (1988) Ecosystem visavis Landslides, a case study in Darjeeling
Himalayas. Impact of development on environment. Geog Soc India Cal II:45–53
Basu SR, Ghatowar L (1988) Landslides and soil erosion in the gish drainage basin of the
Darjeeling Himalaya and their bearing on North Bengal floods. Studia Geomorph Carpatho
Bale 22:105–22
Basu SR, Ghosh L (1993) A comprehensive study of landslides and floods in the lish basin of the
Darjeeling Himalaya, Indian J Power River Valley Dev 43:196–203
Borga M et al (1998) Shallow landslide hazard assessment using a physically based model and
Brardinoni F, Church M (2004) Representing the landslide Magnitude Frequency relation;
Capilano river basin, British Colombia. Kirkby JM, Darby ES (eds) Earth surface processes
and landforms, vol 29, issue 1, pp 115–124
Brudsen D (1979) Mass movement. In: Embelton C, Thornes J (eds) Process in geomorphology.
Wiley, New York, pp 130–186
Burton A, Bathurst JC (1998) Physically based modeling of shallow landslide erosion and
sediment yield at a catchment scale. Environ Geol 35(2–3):89–99
Caiyan WU, Jianping Q (2009) Relationship between landslides and lithology in the three Gorges
reservoir area based on GIS and information value model, vol 42, issue 2. Higher Education
Press and Springer, pp 165–170
Carson MA (1975) Threshold and characteristic angles of straight slopes. In: Proceedings of the
4th Guelph symposium on geomorphology, Norweich Geo Books, pp 19–34
References 223
Surf Processes 2:363–380
Catlos EJ, Harrison TM, Kohn MJ, Grove M, Ryerson FJ, Manning CE, Upreti BN (2001)
Geochronologic and thermobarometric constraints on the evolution on the main central thrust,
central Nepal Himalaya. J Geophys Res 106:16177–16204
Chow VT (1951) General formula for hydrologic frequency analysis. Am Geophys Union Trans
32:231–237
Chow VT (1954) The long-probability law and its engineering applications. ASCE 80:1–25
(Separate No. 536)
Chow VT (ed) (1964) Handbook of applied hydrology. Mc Grow-Hill Book Company, New York
Congalton R (1991) A review of assessing the accuracy of classification of remotely sensed data.
Remote Sens Environ 37:35–46
Crozier MJ (1986) Landslides: causes, consequences and environment. Croom Helm Australia Pty
Ltd., London, 252p
Dai FC, Lee CF (2002) Landslide characteristics and slope instability modeling using GIS; Lantau
Island, Hong Kong. Geomorphology 42:213–228
Dhakal AS, Amada T, Aniya M (2000) Landslide hazard mapping and its evaluation using GIS: an
investigations of sampling schemes for a grid-cell based quantitative method. Photogram Eng
Remote Sens 66(8):981–989
Donati L, Turrini MC (2002) An objective and method to rank the importance of the factors
predisposing to landslides with the GIS methodology, application to an area of the Apennines
(Valnerina; Perugia, Italy). Eng Geol 63:277–289
Dutta KK (1966) Landslips in Darjeeling and neighbouring hills slopes in June 1950. Bulletin of
the geological survey of India. Ser B 15(1):7–30
Einstein HH (1988) Landslide risk assessment procedure. In: Proceedings of the fifth international
symposium on landslides, pp 1075–1090
inventory maps in a data-scarce environment; case study around Kurseong, Darjiling district,
West Bengal, India. In: Malet JP, Remaitre A, Bogaard T (eds) Landslide processes: from
European centre on geomorphological hazards (CERG), Strasbourg, pp 37–44
Gokceoglu C, Sonmez H, Ercanoglu M (2000) Discontinuity controlled probabilistic slope failure
risk map of the Altindag (settlement) region in Turkey. Eng Geol 55:277–296
Guzzetti F, Carrara A, Cardinali M, Reichenbach P (1999a) Landslide hazard evaluation: a review
of current techniques and their application in a multi-scale study, Central Italy. J Geomorphol
31:181–216 (Elsevier, London)
Guzzetti F, Cardinali M, Reichenbach P, Carrara A (1999b) Comparing landslide maps; a case
study in the upper Tiber River basin, central Italy. Environ Manage 18:623–633
Guzzetti F, Cardinali M, Reichenbach P, Carrara A (1999c) Landslide hazard evaluation: an aid to
Intarawichian N, Dasananda S (2011) Frequency Ratio model based landslide susceptibility
mapping in lower Mae Chaem watershed, Northern Thailand. Environ Earth Sci 64:2271–2285
Jadda M (2009) Landslide susceptibility evaluation and factor analysis. Eur J Sci Res 1450-
216X33(4):654–668
Jibson WR, Edwin LH, John AM (2000) A method for producing digital probabilistic seismic
landslide hazard maps. Eng Geol 58:271–289
Kamp U, Growley BJ, Khattak GA, Owen LA (2008) GIS based landslide susceptibility mapping
for the 2005 Kashmir earthquake region. Geomorphology 101:631–642
Komac M (2006) A landslide susceptibility model using the analytical hierarchy process method
and multivariate statistics in perialpine Slovenia. Geomorphology 74:17–28
Lee S, Choi U (2003) Development of GIS based geological hazard information system and its
application for landslide analysis in Korea. Geosci J 7:243–252
Lee S, Ryu JH, Won JS, Park HJ (2004a) Determination and publication of the weights for
landslide susceptibility mapping using an artificial neural network. Eng Geol 71:289–302
Lee S, Choi J, Min K (2004b) Probabilistic landslide hazard mapping using GIS and remote
sensing data at Boun, Korea. Int J Remote Sens 25:2037–2052
Lee S, Pradhan B (2006) Landslide hazard assessment at Cameron highland Malaysia using
frequency ratio and logistic regression models. Geophys Res Abs 8. SRef-ID: 1607-7962/gra/
EGU06-A-03241
Lee S, Pradhan B (2007) Landslide hazard mapping at Selangor, Malaysia using frequency ratio
and logistic regression models. Landslides 4(1):33–41
Lee S, Talib JA (2005) Probabilistic landslide susceptibility and factor effect analysis. Environ
Geol 47:982–990
Maiti RK (2007a) Irrational resource extraction introducing instability in slope and hydrodynam-
ics- a case study at Lish-Chunkhola basin, Darjiling, Indian. J Geogr Environ 8&9:41–51
Vidyasagar University
Maiti RK (2007b) Critical analysis of slope instability on mining scars at Tindharia Cricket
Colony, Darjiling, West Bengal. In: Proceedings of eighteenth convention and national seminar
on “quarternary” climatic changes and landforms, organized at Manonmaniam Sundaranar
University, Tirunelveli, Tamilnadu, pp 189–205
Mallet FR (1875) On the geology and mineral resources of the Darjeeling district and Western
Duars. Mem Geol Surv India 2:1–72
Mandal S, Maiti R (2011) Landslide susceptibility analysis of Shivkhola watershed, Darjiling: a
remote sensing and GIS based analytical hierarchy process (AHP). J Indian Soc Remote Sens.
doi:10.10007/s12524-011-0160-9
Mandal S, Maiti R (2012) Application of RS and GIS based semi-quantitative approach in
landslide hazard risk assessment of the Shivkhola watershed, Darjiling Himalaya. Geo Risk
Assess Manag Risk Eng Syst Geohazards 6(4):203–220
Mandal S, Maiti R (2013) Integrating the analytical hierarchy process (AHP) and the frequency
ratio (FR) model in landslide susceptibility mapping of Shiv-khola watershed, Darjeeling
Himalaya. Int J Disaster Risk Sci 4(4):200–212
Muthu K, Petrou M (2007) Landslide hazard mapping using an expert system and a GIS. IEEE
Trans Geosci Remote Sens 45(2):522–531
Mwasi B (2001) Land use conflicts resolution in a fragile ecosystem using multi criteria evaluation
(MCE) and a GIS based Decision Support System (DSS)
Nie et al (2001) The application of remote sensing technique and AHP-fuzzy method in
comprehensive analysis and assessment for regional stability of Chongqing City, China. In:
Proceedings of the 22nd international Asian conference on remote sensing, vol 1. University of
Singapore, Singapore, pp 660–665, 5–9 Nov 2001
Nithya ES, Prasanna RP (2010) An integrated approach with GIS and remote sensing technique for
landslide zonation. Int J Geomatics Geosci 1(1):66–75
Pandey A, Dabral PP, Chowdhary VM, Yadav NK (2008) Landslide hazard zonation using remote
sensing and GIS: a case study of Dikrong river basin, Arunachal Pradesh, India. Environ Geol
54:1517–1529
Pistocchi A, Luzi L, Napolitano P (2002) The use of predictive modeling techniques for optimal
exploitation of spatial databases: a case study in landslide hazard mapping with expert system-
like methods. Environ Geol 41:765–775
Porghasem H (2007) Landslide hazard zoning statistical frequency ratio method in the basin
Safarood. M.Sc thesis, Tarbiat Modarres University, Noor, pp 1386
Pradhan B, Lee S (2010a) Delineation of landslide hazard areas on Penang Island, Malaysia, by
using frequency ratio, logistic regression, and artificial neural network models. Environ Earth
Sci 60:1037–1054
References 225
Pradhan B, Lee S (2010b) Landslide susceptibility assessment and factor effect analysis:
backpropagation artificial neural networks and their comparison with frequency ratio and
bivariate logistic regression modeling. Environ Models Softw 25(6):747–759
Pradhan B, Lee S (2010c) Regional landslide susceptibility analysis using back-propagation neural
network model at Cameron highland, Malaysis. Landslides 7(1):13–30
Parise M, Jibson WR (2000) A seismic landslide susceptibility rating of geologic units based on
analysis of characteristics of landslides triggered by the 17 January, 1994 Northridge,
California earthquake. Eng Geol 58:251–270
Quinn et al (1991) The prediction of hillslope flow paths for distributed hydrological modeling
using digital terrain models. Hydro Processes 5:59–79
Rowbotham D, Dudycha DN (1998) GIS modelling of slope stability in Phewa Tal watershed,
Nepal. Geomorphology 26:151–170
Saaty TL (1980) The analytical hierarchy process. McGraw Hill, New York, 350p
Saaty TL (1990) The analytical hierarchy process: planning, priority setting, resource allocation,
1st edn. RWS Publication, Pittsburgh, 502p
Saaty TL (1994) Fundamentals of decision making and priority theory with analytic hierarchy
process, 1st edn. RWS Publication, Pittsburgh, 527p
Saaty TL, Vargas LG (2001) Models, methods, concepts and applications of the analytic hierarchy
process, 1st edn. Kluwer Academic, Boston, 333p
Sarkar S, Kanungo DP (2004) An integrated approach for landslide susceptibility mapping using
remote sensing and GIS. Photogram Eng Remote Sens 70(5):617–625
Sharifikia M (2007) RS and GIS application in Geo-hazard- A case study part of central Alborz-
Iran. Ph.D. thesis submitted in Geology Department, University of Delhi, India
Sinha-Roy S (1982) Himalayan main central thrust and its implication for Himalayan inverted
metamorphism. Tectonophysics 84:197–224
Soeters R, Westen CJ (1996) Slope instability recognition, analysis and zonation. In: Turner AK
and Schuster RL (eds) Landslides: investigation and mitigation. transportation research board
special report 247. National Academy Press, Washington, DC, pp 129–177
Tiwari B, Marui H (2001) Shearing behaviour of landslide sliding and mining scarp soil during
drained ring shear test. In: Proceedings of XVth international conference on soil mechanics and
geotechnical engineering, vol 1, Istambul, pp 295–298
Tiwari B, Marui H (2002) Mechanism of shear zone formation and its effect in residual shear
strength. In: Proceedings of 3rd international conference on landslides, slope stability and
safety of infrastructure, vol 1, pp 4–133
Tiwari B, Marui H (2003) Estimation of residualshear strength for bentonite-kaolin-Toyoura sand
mixture. J Japn Landslide Soc 40(2):124–133
Tiwari B, Marui H (2004) Objective oriented multi-stage ring shear test for the shear strength of
the landslide soil. J Geotech Geoenviron Eng ASCE 130(2):217–222
Van Westen CJ, Castellanos Abella E, Sekhar LK (2008) Spatial data for landslide susceptibility,
hazards and vulnerability assessment: an overview. Eng Geol 102(3–4):112–131
practice: highway research board, special report 29, vol 544. NAS-NRC Publication,
Washington, DC, pp 20–47
Vanmarcke EH (1977) Reliability of earth slopes. J Geotechl Eng Div ASCE 103
(GT11):1247–126
Varnes DJ (1984) Landslide hazard zonation review of principle and practice. Natural hazards,
UNESCO, Paris
Vijith H, Madhu G (2008) Estimating potential landslide sites of an upland sub-watershed in
Western Ghat’s of Kerala (India) through frequency ratio and GIS. Environ Geol
55:1397–1405
Windisch EJ (1991) The hydraulics problem in slope stability analysis. Can Geotech J 28
(6):903–909
Yagi H (2003) Development of assessment method for landslide hazardness by analytical

hierarchy process (AHP). Abstract volume of the 42nd annual meeting of the Japan Landslide
Society, pp 209–212
Yalcin A, Bulut F (2007) Landslide susceptibility mapping using GIS and digital photogrammetric
techniques: a case study from Ardesen (NE Turkey). Nat Hazard 41(1):201–226
Yalcin A (2008) GIS based landslide susceptibility mapping using analytical hierarchy process and
bivariate statistics in Ardesen (Turkey): comparisons of results and confirmations. Catena
72:1–12
Zhou CH, Lee CF, Li J, Xu ZW (2002) On the spatial relationship between landslide and causative
factors on Lantau Island, Hong Kong. Geomorphology 43:197–207
Chapter 8
Landslide Mitigation
Abstract The fundamental impetus of any kind of natural hazard and risk man-
agement is an awareness of threat, a notion of responsibility and a brief that human
action might reduce the risk. Various components such as susceptibility analysis,
hazard and risk identification, consequence analysis, hazard analysis, and risk
evaluation are included in the landslide management framework. In the present
study of the Shivkhola Watershed for slope stabilization, some mitigation measures
have been proposed on the basis of community wisdom assessed through percep-
tion study on people living in four landslide prone villages such as Paglajhora,
Tindharia, Gayabari and Giddapahar. The value of experience depicts that Catch-
water drain along the junction between road and the hill slope, jhora training,
retaining wall, catchment water drainage, introduction of vegetation etc. may bring
stability of slope. Besides, author’s intensive field investigation suggested that the
construction and maintenance of buildings, introduction of landslide warning sys-
tem, and improvement of soil strength would be taken into account as landslide
mitigation measures. Here, some control works and restraint works have been taken
into account to reduce landslide hazard and risk.
Keywords Shivkhola watershed Control work Restraint work Value expe-

rience Slope stability
8.1 Introduction
Natural hazard and risk management includes of the threat, a notion of responsi-
bility. Various components such as scope definition, hazard and risk identification,
consequence analysis, hazard analysis, risk calculation, and evaluation are included
in the landslide management framework. No hazard modification method should be
treated as absolute safe. Most methods developed to address the physical hazard are
best described as control measure, reduction measures or mitigation measures-not
preventive measures. ‘Prevention’ is the ultimate form of event modification.

228 8 Landslide Mitigation
Whereas, mitigation is the desired result of risk reduction measures. Various

methods that can be selected as the risk reduction measures in the landslide prone
area are summarized in Table 11.1. Howell et al. (2006) contend that the geological
community has a crucial role to play in educating local planners and engineers
about the types of hazards facing their communities, the extent, place and economic
consequences of these hazards and how to reduce exposure to them.
A successful landslide mitigation measures in the Shivkhola watershed needs
some fundamental requirements and on the basis of these the basic goals and
objectives of management options can be achieved. The following capabilities,
resources and philosophies were suggested to provide better and effective landslide
hazard and risk management after Crozier (2004).
• A technical and scientific information base.
• An informed and capable local and regional Government.
• An appropriate statutory and legal infrastructure.
• An informed and capable professional and technical community to manage and
execute a risk reduction programme.
• A philosophical basis for determining the acceptability of risk.
• A risk reduction programme with the methods, policies, goals and objectives.
• An effective practice and experience.
• An effective system of communication and education.
Schuster and Kockelman (1996) identifies four approaches to reducing landslide
risk.
• Restricting development in the landslide prone-areas;
• Developing and implementing excavation, grading, landscaping and construc-
tion codes;
• Implementing physical measures to prevent or control landslides, such as
drainage, slope geometry modifications and structures; and
• Developing monitoring and warning system (Table 8.1).
Landslide of different types occurs frequently in geotechnically active domains
in Himalaya, North East India and in stable domains in Westerns Ghats and Nilgiri
Hills in South India. Landslides have had disastrous consequences and in 2005,
over 500 lives were lost due to landslide. As a part of management of this sig-
nificant natural hazard, National Core Group for Landslide Hazard Mitigation was
formed. This core group comprising key national ministers and institutions for
drawing a strategy for monitoring the impact of landslides, devising landslide
hazard mitigation, monitoring the activities related to landslide hazard mitigation
including hazard zonation, evolving an early warning system and protocols for
landslide hazard/risk reduction. The Geological Survey of India (GSI) was declared
as the Nodal Agency for Landside Hazard mitigation studies. The responsibility of
the GSI in accordance with landslide mitigation includes:
Table 8.1 Landslide hazard reduction measures options (Crozier 2004)

Methods for landslide Applied activities relating to the methods
mitigation
1. Physical methods • Toe buttressing
• Slope reinforcement: bolts, anchors, pins, piles
• Grouting fissures and joints
• Chemical reinforcement of the soils
• Chunam plastering
• Bioengineering
2. Hydrological • Surface water diversion
methods • Geotextile covers
• Development of artificial drains
• Draining of water from slope debris by horizontal drains
3. Site grooming • Removal of heavily weathered debris from the susceptible slope
surface
• Lowering the steepness of upslope in the landslide prone area
• Contouring the land surface to divert the flow of water
• Bioengineering
4. Regulations • Imposition of laws against waste water disposal, construction of
concrete and multi-storied building and slope clearance in and
around landslide affected area
• Building codes
• Earth work/foundation and drainage standards
5. Land use planning • Restriction on the landslide induced activities
schemes • Rules on the changing the slope coverage
6. Education • Awareness programme and peoples participation
• Radio and T.V. broadcasting about cause and effect of landslide of
the region
• Development of the communication system
7. Warning systems • Periodic survey of the landslide area and continuous monitoring
• Warning system based on the triggering agent i.e. accumulated
rainfall
8. Loss-sharing scheme • Insurance
• Coordinating and undertaking the geological studies for landslide hazard

mitigation.
• Carrying out landslide hazard zonation for different scales as per prescribed
parameters.
• Studying the factors responsible for landsliding.
• Monitoring landslide.
• Suggesting precautionary and preventive measures.
• Developing a quick response mechanism to investigate the incidences on
emergency basis.
• Evolving an early warning system.
• Developing an inventory/database on landslide hazard.

• Development of awareness strategy and awareness programmes in the hazard
prone areas.
With the establishment of National Disaster Management Act (NDMA) in 2005
some more responsibilities were entrusted to GSI as Nodal Agency which includes
assisting NDMA in drawing up the policies, plans, and guidelines in case of
landslide hazards and advise NDMA in technical matters.
Landslide mitigation works are to be conducted in order to stop or reduce the
landslide movement so that the resulting damages can be minimized. With a clear
understanding of the causes and mechanics of the landslide, the landslide control
works can be implemented in the Shivkhola Watershed. Basically, landslide
mitigation works are broadly classified into two categories such as control works
and restraint works. The control works involve modifications of the natural
conditions of landslides such as topography, geology, ground water, and other
conditions that indirectly control portions of the entire landslide movement. The
restraint works rely directly on the construction of structural elements. Landslide
control works involves surface drainage control works, sub-surface drainage control
works, soil removal works, buttress fill works, and the development of river
structure. The surface drainage control works can be accomplished with the
introduction of drainage collection work and drainage channel works. The drainage
collection works are to be designed to collect upslope water by introducing cor-
rugated half pipes horizontally at regular interval to reduce the length of over length
flow and to minimize the seepage. The drainage channel works are to be designed
to remove the collected water out of the landslide affected areas.
The sub-surface drainage control works are to be incorporated to remove the
ground water within the landslide mass and to prevent the inflow of ground water
into the landslide mass from the surrounding areas. This type of works are to be
achieved with the development of intercept under drains and trench drains,
horizontal gravity drains, drainage wells, and drainage tunnels investigating the
landslide prone area properly. Soil removal works are to be applied for small to
medium size landslides. The buttress fill is placed at the lower portions of the
landslide in order to counterweight the landslide mass. It is most effective if the
soils generated by the soil removal works are used.
Retaining walls are also to be constructed to prevent smaller sized and secondary
landslides that often occur along the toe portion of the larger landslides. Because of
the large-scale earth-movement and numerous springs that are expected in landslide
terrain, crib walls are common instead of conventional reinforced concrete retaining
walls. The anchor works utilize the tensile force of anchor bodies embedded
through the slide mass and into stable earth, and are connected to thrust blocks
located on the ground surface. The thrust blocks are anchored with a tendon that
counteracts the driving forces of the landslide to restrain the slide movement.
The protection measures which could be adopted to reduce the risk of landslide
may broadly be grouped into following heads (Selby 2005):
• Avoidance, removal or diversion of the problem.

• Reduction of the forces tending to cause materials movement.
• Increasing the forces which can resist the downward movement of the materials.
• Provision of the landslide warning system.
Various landslide mitigation practices are present in the work of Zaruba and
Mencl (1976), Schuster and Krizek (1978), and Veder (1981). Keefer et al. (1987)
introduced a landslide warning system based on the duration, amount and intensity
of rainfall in urban area. Gray and Leiser (1982) propounded biotechnical method
for landslide mitigation.
8.2 Problem Perception Over the Major Landslides Prone

Areas of the Shivkhola Watershed, Darjiling Himalaya
Before going to take a suitable management option for the landslide prone area,
Shivkhola Watershed, a perception study has been made about the problems facing
by the people living in the concerned study area. Not only that, a priority was given
on the suggestions put forwarded by them and accordingly a rational management
proposal has been established. The slope instability is the most important problem
almost everywhere in the surveyed settlements except at Gayabari. The main
settlement area is situated on relatively stable part but the settlement at tea gardens
are still prone to landslips and thus affected badly. Considering the 1st rank
problems only, almost all of the respondents of Pagla jhora, Tindharia and
Giddapahar are equivocal of this problem. Both of these two settlements are
extremely affected by slope failure. The rational management of this problem is of
utmost importance for the safety of lives, property and landmass. The survey on the
migration into these areas shows that since last 70 or more years, fresh population
are migrated to add concentrated pressure on slope in spite of being marked as
highly landslide prone (Maiti 2007).
Considering the value of experiences of interacting with such a problem, an
opinion survey is made to realize the best possible suggestions for ameliorating the
problem. The opinion survey shows that maximum of the respondents are in favour
of plantation on bare slope and restriction of slope clearing and thus the slope
should be brought under vegetation coverage. Some of them think about proper
drainage system which should take care of the waste water as well as the roof top
water to be guided to the nearest jhora with utmost care so that no water can
percolate down during the journey to the jhora. Out of a total of 163 1st rank
opinions, 32.5 % are equivocal of plantation on bare slope and 13.5 % are of
restriction on slope clearing. Nearly 14 % of 1st rank suggestion is for proper
development of drainage system.
8.3 Recommended Landslide Mitigation Practices
After discussing the problems perceived by the people living in the landslide
affected areas in the Shivkhola watershed, the applicability of various landslide
mitigation measures that could be incorporated in this connection to reduce the
degree and frequency of landslide phenomena for the concerned study area were
analyzed in detail. It is true that the suggestions for landslide mitigation measures
can only minimize/reduce the rate of harmful effect of slope failure to the com-
munities but cannot check the landslide activities in the area absolutely. The
following recommendations must be introduced immediately in the suitable places
of very high landslide susceptible and risk prone areas in the Shivkhola, such as
Paglajhola, Tindharia, Shiviter Gayabari, Nurbong, Giddapahar and Mahanadi.
Others area should also be brought under the scheme of landslide mitigation
measures for reducing or checking the future landslide phenomena.
8.3.1 Armouring the Catch-Water Drain Along the Junction

Between Road and the Hill Slope
The landslides along the Hill Cart Road especially at Tindharia, Paglajhora, and
Mahanadi and Giddapahar section are due to the concentration of the upland
drainage, along the drain at the base of back slope along the Hill Cart Road and the
Narrow Gauge Rail. A huge run-off from a large upslope area pass through the drain
along the junction between road and the hill slope (Fig. 8.1). This junction should
be completely cemented so that no significant water can pass through it causing
harm to the road. Maximum of the damage from outer margin of the road is made
Hill Slope
Runoff Contribution
Catch Water Drain
Road Possible Slide
Down Ward Pressure

Percolation Damage Starts at
Emergence Point
Through Flow
Fig. 8.1 Schematic diagram showing the damage by Percolation along the drain at the junction of
Hill side and Road
8.3 Recommended Landslide Mitigation Practices 233
due to this factor. The fore most step for managing the slope, soil and water is the
plastering of the drain for not to allow easy percolation along the hill side. The
runoff from the hill sides is easily concentrated along the drain at the junction with
road and makes its journey as through flow under the road saturating the underlying
materials. The water when emerges out at the outer margin may develop a channel
and thus in due course, by head ward extension, starts damaging the road. Once
created such landslide scars keep on increasing at a faster rate until the road is
completely wiped out.
8.3.2 Jhora Training
Considering huge discharge, laden with high amount of abrasive tools and conse-
quent basal scour and erosion during monsoon the Jhora Training seems to be
important to save the slope from failure. The technique of jhora training includes
the construction of Guide Walls along the side banks in descending steps, gabion
drops and guide structures. Slope along the jhora bed is to be broken into gabion
drops like small steps (Fig. 8.2).
The torrential water which flows down slope with huge velocity is interrupted
with breaks of slope and water is allowed to fall from restricted height (over fall)
and thus the erosive power of flowing water can be restricted (Figs. 8.3, 8.4 and
8.5). The cemented course will not allow water to infiltrate inside in the sinking
areas mainly. All the jhoras along the Hill Cart Road has to be managed at the
junction point with hill Cart Road and Narrow Gauge Rail Line. The length of the
riser and height of the heads are to be fixed by analyzing the gradient along the long
profile of the concerned jhora and the stability of the materials in response to the
Fig. 8.2 A design of jhora

training Channel Bed Head
Drop Structure
Guide Wall
Over Fall
Earth or Boulder Fill

Fig. 8.3 Jhora training at lower Paglajhora
Fig. 8.4 Jhora training at 14

Miles Bustee
velocity and discharge of the flow during storm. Though in some cases jhora
training is practiced and proved to be successful, the training of all the jhoras is yet
to be achieved. After the massive destruction in May, 2011 at Lower Paglajhora,
along the main nala which was passing through the slided area was trained with
concrete small steps to arrest the percolation or seepage of channel water (Fig. 8.3).
Fig. 8.5 Jhora training at

Paglajhora
The main problem in the Jhora Training is the clearance of seepage water from
the interior mass. The conduits or the holes made for the clearance of seepage water
is mainly clogged with clay and thus water cannot be allowed to drain due to lack of
proper maintenance. The subsidence and sinking of the materials are facilitated due
to storage of seepage water at the back of the concrete structure being facilitated by
the weight of the overlying materials.
8.3.3 Retaining Wall
Retaining walls are made to resist the pressure of dislodged materials or to resist the
pressure of earth filling deposited behind it after its formation. The retaining wall
should conform to the repose angle of the constituent slope materials. The repose
angle in dry condition can be easily measured by pouring the sun-dry materials
from vertical position on a near horizontal surface. At the time of the construction
of the retaining walls, the possible pressure from the backfill, the nature of base-
ment, angle of repose and hydraulic character of the slope forming materials are to
be studied for the fixation of depth of foundation, width and height of the berms
(steps) of the wall.
The further refined concept of angle of internal friction may be introduced which
is approximately equal to angle of repose (Van Burkalow 1945) but not exactly
same as it includes the inter-granular friction as well as interlocking. The steep, high
wall should be avoided and may be inserted with few berms. The foundation base
of the walls should be 1/10th of the height +30 cm and the wider base should have a
back slope towards the backfill. Such type of retaining wall should be a well
accepted management technique along the Hill Cart Road.
Sometimes unnecessary huge overburden is imposed on the unstable slope due
to concentration of materials by constructing huge retaining wall with large boulder
and thus fresh slide occurs. To avoid the unnecessary overburden of the concrete
retaining wall at the places of weak lithology of Lower Paglajhora boulder made
retaining wall (Figs. 8.6, 8.7 and 8.8) with wire net could have been the best
procedure. Such method not only reduce the pressure but also helps for easy coming
Fig. 8.6 Retaining wall at lower Paglajhora
Fig. 8.7 Boulder made

retaining wall at Shiviter
Fig. 8.8 Retaining wall at

lower Paglajhora
out of the seepage water from the slope materials which can reduce the pore-water
pressure and can check the landslide.
Interruption of drainage from the back fill is the main cause of failure of
retaining wall and which is generally observed at Shiviter (Fig. 8.7), Paglajhora and
Tindharia. The loose clay and silt are generally drained with water and helps in
clogging the weep holes. Only the retaining wall could not be able to manage such
slope failure. Here, Shotcrete technique, a soil-cement mixture that can be sprayed
above the retaining wall (open surface) to prevent erosion from weathering and
percolation of rain water. The steep slope of the massive slope failure section of
14 Miles Bustee would be shaped and reduced and then retaining wall may be
constructed with full care.
For the shallow debris slide (Fig. 8.7) on the way to Shiviter Tea Garden Gabion
Mattresses, a rectangular basket made up of wire mesh that are filled with small
rocks built by hands, can be laid end to end and side to side on the slided scar with a
fairly solid foundation. Several gabion mattress can be joined together to prevent
soil erosion and slope materials failure. Before the placement of the gabion mattress
the slope has to be well prepared and a filter layer is placed beneath it to resist the
movement of finer particles entrainment from the sub-surface soil that can improve
the soil strength. Sometimes the weight of the retaining wall is so high so that
additional stress cannot be resisted by the slope and thus entire slope moves down
slope. The bamboos net are fixed at regular interval to restrict the down slope
movement of dislodged materials. The thrust from the backfill is so high so that
these bamboo structures are damaged (Fig. 8.9). The introduction of vegetation on
the slide scar at the back fill of each bamboo structure in a process of developing
total green cover may be an effective management on suitable slope.
Fig. 8.9 Damaged retaining

wall, 14 Miles Bustee
8.3.4 Breast Wall
Breast walls are the construction close to the slope base built to protect freshly cut
slope from slope failure. Along the roads, the breast wall is constructed along the
inward side of the slope. It is one of the popular soil and slope conservation
measures and is frequently observed along the road cuts. Slopes need to be trimmed
and flattened before construction. More stable concave slopes are not often formed
due to the construction of breast wall at the base by further cutting and steepening.
As already mentioned the back fills are prone to more slide due to lack of easy
drainage of the seepage water. The weep holes are often clogged with clay and silt
and need to be monitored regularly and other options of immediate and easy
drainage should be open. The new road cuts are very much prone to down slope
movement of soil and slope materials and the necessity of breast wall and retaining
wall is urgently felt to restore the stability of the vulnerable slopes mainly at
Tindharia (Figs. 8.10 and 8.11), Sepoydhura and Shiviter. At Paglajhora breast wall
with jhora training has been established to check seepage pressure as well as
landslip (Fig. 8.10).
8.3.5 Catchment Water Drainage
Drainage may be used to prevent surface or sub-surface water reaching the slide
area or to remove it from the slide area (Bianco and Bruce 1991). Water in the slide
area may be the result of steady seepage flows, or it may be released from the soil or
rock mass by stress changes or chemical activity, for example loading or the
Fig. 8.10 Breast wall at

Paglajhora
Fig. 8.11 Breast wall at

Tindharia
decomposition of refuse (Bromhead et al. 1996) respectively. The diversion of

water from the potential slope failure zone is mainly done through Catch Water
Drain. The diversion is made from more damage prone areas to that of less. The
natural drainage after being interfered with, they attempt to redesign themselves
which may be detrimental to the slope. Sometimes natural processes of weathering
due to lithological characters may develop brittle and fragile slope forming mate-
rials which through interaction with surface and sub surface water may become
susceptible to down slope movement. Thus the slope has to be completely excluded
from both surface and sub surface water and for this purpose, Catch Water Drain is
the best option. The gully (Channels or streams) heads are prone to landslide where
the runoff enters into the channel head helping it to progress further uphill. The
slide scars extend uphill and the scarps along the sides also extend upslope
increasing the area of the slide scar and bringing more and more area under the
destruction with the help of accumulated surface and sub surface water. The
through flows are highly active in these regions and a high rate of 5–12 m/hr of the
subsurface flow even during spring has been observed in eastern Himalayan slope.
Water entering into the soil few meters upslope is released on the scar faces making
the slope instable. Generally a series of near parallel drains are to be constructed
above the gully head to collect the water to restrict it to take part in destruction.
These should have a non-eroding graded gradient with a cemented mortar and
should be drained into a non-erosive channel. In a sinking zone and on highly
permeable rocks, closure spacing of the drains are necessary. At Pagla Jhora region
such Catch Water Drains are to be constructed for arresting the surface water from
the large upper catchment and may be diverted to either 2nd or 4th sub-watershed.
8.3.6 Continuous Monitoring of Sub-surface Water
For efficient and effective planning for drainage arrangement in this distress area,
knowledge on the pore water pressure is needed. Thus, installation of Piezometer on
slopes at different levels in Paglajhora sinking zone is necessary. To assess the slope
instability condition, a network of monitoring instruments could be installed in
order to provide, together with others climatic data, the piezometric response deep
inside the landslide and to measure the rate of deformation at several points in the
interests of public safety (Angeli et al. 1994a, b). The installation of high-strength
steel anchors and sub-horizontal drains are to be drilled into the slope to assist in
removal of ground water in accordance with an engineering design just in the
sinking area of Lower Paglajhora, Shiviter tea garden and Nurbong Tea Estate.
Basically, the sub-surface drainage network can be managed by the following
good engineering practice.
• Introduction of sub-surface drain in the sinking zone and provision of filter
around subsurface drain.
• Provision of drain behind retaining walls.
• Use of flexible pipelines with access for maintenance.
• Prevention of inflow of surface water.
8.3.7 Landslide Mitigation by Improving Soil Strength
Usually, a simple method for increasing the shear strength of soil is to reduce the
pore water pressure and that can be accomplished by reducing the sub-surface
horizontal and downward flow of water through applying the following methods.
• Grouting or void filling by cementing materials or chemical and development of

concrete step like structure along first order drainage in Paglajora sinking zone
and Nurbong sinking area that may reduce the seepage and permeability and
also can reduce the pore water pressure within the soil.
• Development of artificial drainage upslope may act to prevent the ingress of
water into the slided area and that can be introduced in Lower Paglajhora,
Gayabari, Tindharia and Shiviter.
• A certain type of first growing vegetation can minimize pore space within the
soil and moderates infiltration characteristics, controls run-off and reduce soil
moisture condition through transpiration. This type of process is to be ensured in
Shiviter, 14 Miles Bustee, Mahanadi and Sepoydhura landslide area.
• Unstable Slope coverage by plastic/synthetic cloths in small landside scar can be
more effective mechanism for reducing rain water percolation and pore water
pressure at Paglajhora, Shiviter Landslide (on the way to Shiviter T.E.),
Tindharia Railway. Station. Landslide, Sepoydhura T.E. landslide etc.
• Chunam Plaster mechanism could be applied at some places i.e., Chunabhati,
Mahanadi, Tindharia and Paglajhora where slope surface is exposed to surface
run-off and rainfall. This technique can prevent slope from erosion and water
infiltration and restrict pore water pressure and also enhance slope stability
condition by increasing the soil strength. It covers the slope surface with a hard
water proof shell (Fig. 8.12).
• In Paglajhora and Tindharia care should be taken in the construction of drainage
measures to ensure that pore water pressure is not increased.
Fig. 8.12 Mechanism of Chunam Plaster to protect a soil slope

8.3.8 Geo-textile Method for Landslide Mitigation

(A Bio-engineering Approach)
Among many other methods applied for soil erosion control and slope stabilization,
bio-engineering approach is one of the best choices. Gray and Leiser (1982) and
Howell et al. (2006) established bio-engineering methods to protect soil erosion and
slope instability. Collison (1993) assessed the role of vegetation cover on slope
stability. Bio-engineering is the utilization of vegetation may be alone or in com-
bination with geotechnical structure, for the protection of slopes by reducing and
controlling factors that cause instability. In the Shivkhola watershed, massive
landslide area with steep slope at Lower Paglajhora (not approachable, below the
Hill Cart Road, Fig. 8.13) should be brought immediately under the geo-textiles
method (mainly the jute nets) with an integrated scheme of introduction and
maintenance of the grass and fern seedlings at the opening of the net at regular
interval so that at maturity the entire slope becomes covered with vegetation thatch.
Another section of Paglajhora (Fig. 8.14) that faced massive landslide phenomena
on May, 2011 should also be brought under this geo-textile mechanism. The
introduced vegetation will get the nutrient from the decomposed jute net and grow
to its maximum soon to offer optimum protection of slope. Jute geotextiles are the
most effective among all types of geotextiles-both natural and man-made (Thomson
and Ingold 1986). Additionally, hygroscopic nature of jute yarns in the Jute
geotextile cause them to swell by around 20 % when wet. This situation will
promote to reduce the velocity of sub-surface water as well as entrapment of soil
particles. Before netting the slope, a fertilizer is to be spread on the slope in order to
Fig. 8.13 Landslip at

Paglajhora
Fig. 8.14 Massive landslip at Paglajhora
promote faster growth of vegetation and soil on the surface of slope is first graded to
remove the unevenness present, where possible. The area may be seeded with very
limited quantity of Pennisetum species/Vetiveria species grasses known for quick
and easy growth and have root depth nearly equaling the existing soil cover, i.e.
around 40–50 cm. The netting of slopes generally will stop the downward move-
ment of slope materials as it reduces the impact of direct rain drops and acts as a
barrier to surface flow and even holds together slope material. Such method could
be very much applicable in Shivkhola Watershed mainly at the places of Lower
Paglajhora (Fig. 8.14), marginal North East part of Shiviter Tea Garden, and Tin-
dharaia where structural measures are not possible enough due to very steep slope
gradient and very fragile rock-soil composition. Proper care for recurring growth of
the vegetations on the specified and managed surface has to be taken for year long
protection of the slope from landslip.
8.3.9 Introduction of Vegetation and Slope Stability
The role of vegetation in improving slope stability is well recognized and com-
prehensive reviews may be found in several publications of Greenway (1987),
Phillips and Watson (1994), Gray and Sotir (1996). In general, vegetation influence
slope stability through both hydrological and mechanical mechanisms (Table 8.2).
Hydrological mechanism that lead to lower pore water pressure and soil moisture
Table 8.2 Effects of vegetation on slope stability (After Greenway 1987)

Hydrological mechanisms Mechanical mechanisms
1. Foliage intercepts the rainfall, causing 1. Roots reinforce the soil, increasing soil
absorptive and evaporative losses that reduce shearing strength
rainfall available for infiltration
2. Roots and stems increase the roughness of 2. Tree roots may anchor into film strata,
the ground surface and the permeability of the providing support to the upslope soil mantle
soil, leading to increased infiltration capacity through buttressing and arching
3. Root extracts soil moisture from the soil, 3. Vegetation exposed to the wind transmits
which is lost to the atmosphere via transpi- dynamic forces into the slope
ration, leading to lower pore water pressures
4. Depletion of soil moisture may accentuate 4. Roots bind soil particles at the ground
cracking in the soil, resulting in higher surface, reducing their susceptibility to soil
infiltration capacity erosion
are beneficial, while those that yield higher soil water are adverse. Mechanical
mechanisms that increase shear resistance in the slope are beneficial, while those
that increase shear stress are adverse. Simply, vegetation covered surface generally
reduce the ability of rainfall to cause slope failure through the process of inter-
ception and evaporation and consequently, increases the shear strength in under-
ground soil by the network of roots. For the interception by trees, the structure of
the canopy is to be analysed by the free throughfall co-efficient, the stemflow
partitioning coefficient, the canopy storage capacity and the trunk storage capacity.
In the present study of Shivkhola watershed, suitable vegetation cover is of great
importance at Paglajhora up and down slope, Tindharia T.E., 14 Miles Bustee and
Nurbong Tea Garden area as it affects the slope hydrology and slope stability and
determines throughfall, stem flow, root water uptake, permeability, root rein-
forcement, vegetation surcharge and interception loss from input rainfall that is
described here in an integrated vegetation-slope model (Greenway 1987; Fig. 8.15).
The connection between forest cover and deep-seated landslide initiation is less
clear than with shallow landslide; however, any land use or management practice
that alters hydrological pathways and timing could potentially influence deep-seated
mass movement activity (Phillips et al. 1990).
Dense vegetation cover and the network of tree roots are considered to be a
major contributor to soil strength and slope stability in the area where roots are
present (Phillips and Watson 1994). Various studies have also used the tensile
strength of tree roots as a measure of the contribution of vegetation to increasing a
slope’s stability and thereby reducing the incidence of landsliding (Gray and Sotir
1996; Montgomery et al. 2000). As a result of increase rate of deforestation, the
Shivkhola Watershed is facing the problems of soil erosion due to direct falling of
rain drops and slope failure due to sub-surface soil saturation very easily. Con-
sidering the importance of vegetation cover and to avoid such problems of erosion
and slope failure, afforestation programme is to be introduced in the landslide
affeceted area such as Paglajhora (Fig. 8.16), Tindharia, Mahanadi, Sepoydhura,
Shiviter and Nurbong immediately where most of the slope surface area is exposed
Fig. 8.15 Integrated slope-vegetation model, Greenway (1987)
Fig. 8.16 Afforestation at Paglajhora

Fig. 8.17 Plantation along

first order channel
to sub-aerial processes. The on-site benefit of afforestation include a marked

reduction in shallow landsliding (Pearce et al. 1987; Phillips et al. 1990),
substantially reduced rates of earth flow movement (Phillips et al. 1990), and
cessation of gullying processes (Bayfield and Meister 1998; Phillips et al. 2000).
Development of first order stream and its flow through sub-surface soil over steep
mountain slope is the major cause of soil erosion and slope instability at several
places in the Shivkhola Watershed. For freshly-exposed slope surface created by
road cutting which is basically found on the way to Shiviter T.E. and Tindhariea
T.E. and of course aside the Hill Cart Road, the vegetative turfing (introduction of
fast growing grasses, bushes, trees and bamboos) can be most effective mitigation
measures in the vulnerable areas of Shivkhola Watershed. Sometimes sprays has to
be provided to the exposed slope surface to improve the moisture condition of the
soil and to facilitate vegetal turfing.
A preventive measure of soil erosion has been taken by saplings plantation along
first order channel at Gayabari Lower slope (Fig. 8.17). In the slope failure prone
areas of the Shivkhola watershed, wild fern are to be grown profusely and can be
used for slope protection due to some unique taxonomic characters. The compound
leaves are 0.5–1 m long and convex upward and are arranged in successive vertical
layers. The rain drops falling on the convex leaves having a down slope trend are
conveyed down slope through the tips of the leaves and thus seldom get opportunity
to reach slope surface (Fig. 8.18). Thus the thick fern cover is an essential slope
protector which restricts soil from being saturated and thus pore water pressure can
be controlled to restore slope stability (Maiti 2007). The fern may be introduced to
the upper catchments of Paglajhora, 14 Mile landslide area etc.
Fig. 8.18 Introduction of

Fern Vegetation
8.3.10 Maintenance and Continuous Monitoring of NH-55,

Hill Cart Road
Landslide Monitoring is an important and integral part of landslide investigation as

well as landslide management. The major objectives of landslide monitoring system
are:
• To provide information that can assist landslide investigation and analysis.
• To determine the rate of change of the areal extent of the major landslide
location in the watershed.
• To identify the link between ground movement, rainfall and ground water that
can promote slope materials to move downward.
• To provide early warning in the landslide affected areas mainly at Paglajhora,
Shiviter and Tindharia where movements could affect life and property.
• To monitor the effectiveness of landslide management strategies taken along the
Hill Cart Road.
In a number of unstable locations, where monitoring has taken place over several
years, a monitoring strategy has been established which includes the gathering of
information on both meteorology and ground movements. Increasingly taking into
account the cost of manual data gathering, electronic systems are being used,
allowing easier data acquisition, interpretation and storage (Fort et al. 2000). It is
generally observed from the previous experience that after the destruction of slope
by landslide, emergency attention is given to the affected areas and huge amount is
spent for the restoration and rehabilitation. It can be suggested strongly for regular
monitoring of the potential failure area and a little precautionary measure may save
the slope as well as huge property from destruction.
The records of subsidence (Table 8.3) of Hill Cart Road at Lower Paglajhora are
a significant landmark of slope instability in the Shivkhola Watershed. Near about
Table 8.3 Records of the rate of road subsidence from lower Paglajhora sinking zone (By author
himself)
Years Rate of subsidence of Hill Result of the subsidence
Cart Road from Rail Bench
May’ 2006 No prominent subsidence Unaffected the Hill Cart Road
took place
July’ 2007 0.75 m Affected the Hill Cart Road
September’ 2008 1.03 m Affected the Hill Cart Road during
continuous rainfall
August’ 2009 1.50 m Road blockage due to lowering of road
bench by heavy monsoon rainfall
June’ 2010 5.00 m Failure washed away the marginal
subsided part and damaged the
transport network completely
2 km length of the Hill Cart Road which is stretching along the steep north, east and
south facing slope with fragile lithological composition is heavily affected by the
down slope movement of slope materials and the subsidence of road and rail-cut
benches (Fig. 8.19).
Fig. 8.19 Road subsidence at Paglajhora

8.3.11 Road Diversion to Avoid the Paglajhora-Sinking Zone
The problem of the road subsidence and sinking in the Paglajhora region (Gayabari-
Mahanadi-Giddapahar Sector) has been a severe bottleneck since long along the
N.H.-55. The N.H.-55 (Hill Cart Road) has been constructed in the British regime
with a very low gradient up to the fabulous hill city Darjeeling. To take advantage
of this low gradient (around 1:50) road bench, the Himalayan toy train tract was
also planned and aligned by them along the hill cart road. With time it has become
the busiest route to transport heavy traffic to the remote areas of Darjeeling district,
W.B. Till date, the journey path of the Himalayan queen (Darjeeling Himalayan toy
train) has somehow been maintained to attract the tourist from all over the world.
Recently, this toy train has become christened as World Heritage Site by UNESCO,
and the maintenance procedure has become more active as well. As most of the land
slide occurs along Hill Cart Road (NH-55), and a huge amount is spent for the Post-
Slide Management, an attempt for pre-slide management of the susceptible areas
has to be introduced with immediate effects with less efforts and investment.
A sector wise job assignment has to be made for regular supervision of the slope
stability and a serious drive for pre-slide management of potential slope failure
zones are to be introduced and it should be considered as emergency as that in post
slide condition. For all monitoring programme, it is essential that accurate records
are kept about various landslide inducing parameters and rate of change of the
landslide affected areas by thorough inspections. Not only will monitoring systems
allow the implementation of an emergency response if required; they can cite
baseline of information and can increase the knowledge about the landslide prone
area, the Shivkhola Watershed.
As new alignment of the rail bench from other route at such a low gradient to
avoid the 2 km long problematic stretch of Paglajhora Sinking Zone (Fig. 8.20) is
perhaps not a pragmatic proposition. Though, an alternative route of heavy traffic
can be possible in the area. Already a by-pass route has been developed which is
joining Giddapahar with Gayabari avoiding Paglajhora. The said road is aligned
along a ridge line, baring steep gradient with numerous zigs and is suitable for only
small/light vehicle. An alternative road from Rohini T.E. to Kurseong has been
implemented and the upslope vehicles are passing along this road and thus the
stress has been reduced a bit on NH-55 as well as on affected sites of Paglajhora,
Gayabari, and Tindharia etc. But the imposition and strict implementation of
restriction on heavy vehicle can only assure relief on these instable areas along the
Hill Cart Road (NH-55).
8.3.12 Construction and Maintenance of Buildings
Landslide in the Shivkhola Watershed is a common problem as it damages roads,

buildings and other infrastructure but much can be done to mitigate such damages
Fig. 8.20 Reconstruction of the Hill Cart Road near Paglajhora
but the lack of maintenance generally makes all the constructions more susceptible
to slope failure and that is why the regular maintenance is more important to
mitigate landslide. In the field of landslide mitigation, the local authority, the
developer, the architect and the contractor and in due course the building owner
have to play an important role in terms of ensuring landslide management activities
in the most suitable manner by taking into account the steepness of slope, geo-
technical properties of the slope materials and the relative weight of the concrete
structure for further construction in the landslide prone area of Shivkhola
Watershed.
In terms of construction techniques, there are a number of techniques for min-
imizing the impacts of ground movement through appropriate design. In this cases,
the foundation of the building and location sites are more important. Traditional
strip foundation can easily fracture, causing significant structural damage at
Tindharia, Mahanadi and Gayabari Tea Garden. Rafts can accommodate slight
movements and span minor fissures and voids that may form beneath the raft over
the lifetime of the building. Beside light-weight framed buildings, which may be of
timber with brick or concrete infill or sheet construction materials, are likely to be
least problematic in future. At Giddapahar, Tindharia Upslope and Shiviter Tea
Garden area, construction of the houses should have raft foundation with jacking
points. Light weight, low rise buildings composed of materials that will not be
prone to visible cracking and damage may be effective. Human activities such as
slope cut and fill operation over the steep slope in the surrounding of dense
settlement area (Tindharia, Mahanadi and Sepoydhura) and lack of maintenance of
waste water and roof-top water disposal adversely affect the ground stability con-
dition. Residents both in groups or individual, can work together to ensure the
management of roof-top water and household used water through pipes and its
channelization to concrete drains. Last of all, specific rules should be imposed for
the establishment of concrete houses in the landslide prone area of the Shivkhola
Watershed and that can ensure the slope stability.
8.3.13 Landslide Warning System
USGS introduced an inventory of landslides (debris flow) from the storm of 27th
June, 1995 in Madison County by using aerial photography, field investigations,
rainfall measurements from rain gauges and National Weather Service Doppler
Radar Observations (Morgan et al. 1999). The inventory data are being used to
ascertain the conditions that caused the debris flows and to develop methods of
warning system of such events in the future (Morrissey et al. 2001). In Shivkhola
Watershed, at risk prone areas of Paglajhora Sinking Zone and Tindharia slope
failure is mainly triggered by heavy and continuous rainfall. So, continuous mon-
itoring of rainfall and pore-water pressure by installing weather station and
piezometer respectively is essential. Recorded parameters as well as their possible
impact should be simulated and forecasted to the people living in the region through
local electronic media. To execute the process, local Government in association
with others non-govt organization and local people should come forward for
installing weather station and piezometer in the landslide hazard risk area of
Shivkhola Watershed mainly along the Hill Cart Road. At present Satellite systems
are used to monitor the movement of the landslide (Hutchinson et al. 2002). In UK
National Survey Authority has installed a base station from where landslide
movement is obtained at a very small time interval. This type of satellite based
system can be installed in the destructive landslide prone areas of Darjiling
Himalaya to detect landslide movement. The rate of movement then be transmitted
to the people living in the landslide risk prone area which can help to rehabilitate
the people. Besides, GPS and Total Station could be utilized to assess the areal
extension and direction of landslide masses at regular time interval. Such assess-
ment of areal extension of landslide may also impart knowledge on the possible
impact of landslide phenomena.
For combating the problem of slope failure the following remedial measures
should be adopted immediately in the Shivkhola Watershed, Darjiling Himalaya.
• Lined catch water drains at every 20 m intervals above the nala on the down-
slope and upslope should be constructed for the collection of surface run-off.
• Horizontal Perforated pipes should be placed at the slided area during the
monsoon period for draining out of the sub-surface water.
• Vegetal turfing over the steep tea garden slope is to be introduced for mini-
mizing the recharge of the ground water.
• Continuous monitoring of ground water condition with the help of necessary
accessories (Peizometer).
• Replacement of human settlement from vulnerable areas.

• Afforestation of the hill slope with local shrubs to absorb the moisture content
from the sub-surface soil.
• The Concentration of human settlement in and around landslide prone area
should not be permitted by the local Government and a strict rule is to be
regulated.
• The plying of heavy loaded vehicles through the landslide prone area should not
be allowed.
• The local community is to be brought under the landslide mitigation programme
and they are to be educated on landslide risk management.
8.4 Conclusion
The success of landslide hazard and risk mitigation measures depends on the
continual iterative processes of information input and managerial response. There
are many aspects of landslide management that are still poorly understood. It is
clear from the intensive field investigation and people’s perception that the com-
munities have to be taken collective responsibility by enacting legislation to guide
the landslide hazard management process. Continuous advancement in science and
technology in landslide mitigation measures can be efficient tools to reduce the
consequences of landslide phenomena for the unstable Shivkhola Watershed.
The Shivkhola watershed exhibits a wide range of elevation between 300 and
2,040 m with varying steepness of slope. The development and extension of the
drainage network over steep mountain slope makes the topography rugged in
nature. The interaction between slope, relief and drainage network associated with
weak lithological composition such as structural discontinuities have made the
places of Lower Paglajhora, Shiviter, Gayabari, and Tindharia most susceptible to
slope failure. It is seen that most of the slided and subsidence zones are linked with
more stream junction/confluence of the drainage lines, structural discontinuities and
lineaments. The spatial distribution of landslide locations in different lithological
unit of individual sub-watershed depicts that sub-watershed-I is lithologically
unstable followed by watershed II and V. Spatial distribution of amplitude of relief
reveals that sub-watershed-I, IV, II and III are dominated by landslide as there is a
positive relation between relief and the landslide frequencies.
The drainage concentration along the weakness planes could be checked or
reduced by reducing the upslope contributing area with the establishment of hori-
zontal as well as vertical concrete drains just above and within the sinking zone and
slide scar. The identification of structural discontinuities/weakness planes and their
cementation can reduce the seepage and pore-water pressure in the soil and promote
slope stability to some extent. The sub-watershed-I, II, III and western part of VI are
dominated by large number of lineaments/discontinuities/weaknesses planes.
Basically, the rocks of Darjiling gneiss, Daling formation, Lingtse granite,
8.4 Conclusion 253
Garubathan formation and Reyang formation are very much attributed with
weaknesses planes and associated land slide phenomena which are common in the
right hand side of the main Shivkhola River and are to be brought under such
management technique to minimize the slope failure. The potential dissection and
roughness of the topography is very high in the north, north-east, east, south and
south-east facing escarpment slope of Lower Paglajhora, Mahanadi, Shiviter,
Gayabari Up-slope and Tindharia through which the Hill Cart Road and Railway
Line is passing. At all those places slope is attributed with high positive curvature
that invites immediate drainage concentration and makes the places detrimental to
slope failure. The extreme lower segment of the slope is registered with high
negative curvature caused by active toe erosion of the main river that also invites
drainage concentration. The places of more dissection and the roughness should be
avoided for the establishment of new settlement and further construction to reduce
the landslide hazard risk. Besides, the diversion of drainage lines just above the
potentially active slided area and cementation by developing concrete steps along
the drainage lines can reduce the concentration of both surface and sub-surface
drainage water and reduce the pore-water pressure. The more convexity of the slope
could also be reduced modifying the topography up to the threshold slope angle of
24° that can control more drainage concentration and make the slope more stable.
The extreme middle and lower most part of the basin is characterized by cut and fill
terrace where slope stability is very high due to less slope steepness. Sub-water-
shed-II is more susceptible for slope instability as most of it lies on steeper slope
and is followed by sub-watershed-III and I.
The study envisages that the middle most part of a particular slope facet is more
prone to failure due to easy and maximum drainage concentration. The extreme
marginal part of the basin is registered with low landslide potentiality because of the
less drainage concentration. The areas lying in the mid-central part of the slope
segment in Lower Paglajhora, Tindharia, Gayabari and Shiviter are dominated by
moderate level of drainage density indicating the more saturation excess surface
water which is an indicator of potential erosion and instability. All the first order
tributaries are characterized by head-ward extension and branching which produces
sharpened interfluves and make the slope steeper beyond threshold slope angle. The
central part of the watershed shows the maximum concentration of drainage but due
to gentle slope gradient the propensity to slope failure is less. Watershed-III is more
prone to slope instability as it possesses higher drainage and deserves more
attention for management. In watershed-III nearly 92 % of the basin area posses the
lower value of Constant of Channel Maintenance and it means higher drainage
density and is thus detrimental to the slope and soil stability.
The upslope contributing area increases toward the main tributaries where the
concentration of channels and accumulation of water promotes more excess
moisture and instability of slope and soil. The large numbers of drainage concen-
tration at Lower Paglajhora, Shiviter, 14 Miles Bustee and Gayabari have made the
places more susceptible to soil erosion and slope failure. The places of landslip
induced by drainage concentration should be brought under the measures of catch
water concrete drains and installations of horizontal pipe with filters to extract
sub-surface water that can resist seepage and reduce pore-water pressure and bring
stability to the slope and soil.
Land use and land cover plays a significant role in the Shivkhola Watershed as it
influences surface run-off, slope material saturation, potential retention of rainfall
and pore-water pressure. Since 1950, rapid rate of deforestation and clearing of
forest cover in connection to the expansion of settlement, establishment of
communication lines and fulfillment of the socio-economic demand of the local
people have aggravated the problems of soil loss and slope failure. The establish-
ment of settlement and communication lines and transformation to flat surface
invites the concentration of drainage water and make steep back wall very much
critical to slope failure. In Shivkhola Watershed, degraded forest rank first in terms
of triggering land use factor of slope instability and that is followed by settlement,
road, open forest, tea garden, mixed forest, jungle, dense forest, bare surface and
agricultural land. The major landslide sites Lower Paglajhora is associated with
degraded forest and road contributing area (RCA); Tindharia with communication
lines and settlement concentration; 14 Miles Bustee with RCA; and Shiviter with
haphazard concentration of settlement, degraded forest and open forest. Spatially,
the sub-watershed-I, III, IV and VI are dominated by degraded forest, settlement
and road and tea garden. To keep the slope safe from slope failure, the human
intervention over steep escarpment slope should be stopped at any cost. Not only
that the places characterized by degraded forest, open forest and bare surface will
have to be brought under effective afforestation and plantation programme and that
would be monitored by the local inhabitants. A strict Government rules against the
illegal modification of slope and clearance of forest cover must be imposed for the
healthy growth of vegetation that can maintain the stability of soil and slope.
The dense forest, down slope jungle, mixed forest, tea garden, and open forest
area are characterized by fair to good hydrological condition with the curve number
(CN) value of 75 that shows maximum retention and minimum run-off as well as
lower rate of soil loss. On the other hand settlement, road, bare surface, upslope
degraded forest are attributed with the Curve Number (CN) of more than 76 which
is an indicator of minimum retention and maximum run-off and soil erosion. The
derived instability rank based on CN, retention, run-off, and landslide potentiality
suggests that Sub-watershed VI, I and III have to be paid more attention for a proper
management of land, water and soil as these three are characterized by maximum
run-off and high susceptibility to slope failure. To manage the land, water and soil
in all these three sub-watersheds a well accepted land use planning should be
launched immediately by introducing several afforestations programme with active
community participation. The people of landslide prone area should also be
restricted from constructional works during the moisture period i.e. June-
September.
The most landslide prone area, the Shivkhola watershed of Kurseong Sub-div-
ition of Darjiling Himalaya registers the continuous growth of household and
population. This continuously adds huge population pressure on the existing land
and soil resources. The growth rate of household during the period of 1981–1991
and 1991–2001 are 75 and 169 % respectively where as the growth rate of
8.4 Conclusion 255
population during the same periods are 28.83 and 253 %. Such unprecedented
growth of household and population as a result of migration from surrounding
depressed rural areas is recorded at the places of Tindharia, Giddapahar, Gayabari,
Paglajhora, South Shivkhola Tea Garden, Jogmaya Tea Garden, Selim Hill Tea
Garden etc. Everyday functioning of such huge population for livelihood collection
put tremendous pressure on the land resources damaging slope and vegetation cover
leads to soil loss and slope failure. Most of the houses are made up of concrete
floors which exert immense pressure on fragile slope materials. On the other hand
tin/asbestos roofed houses release maximum concentrated flow of water on land and
erode the slope materials easily. Except Paglajhora, most of the houses are one
storied and a considerable percentage is two storied which puts excess weight on
weak lithology. Nearly 50 % houses of all the villages do not have roof top and
waste water disposal system and this situation favours the water to enter into the
soil and thus destabilize the slope materials and ultimately threatens the entire
settlement area. To resist the problems of slope failure caused due to human
intervention on unstable slope in the Shivkhola Watershed, the landslide prone
areas should not be allowed for further concentration of settlement. The steep slopes
must be avoided for the construction of roads. The restriction on the plying of heavy
loaded vehicles and on the construction of two or three storied buildings and the
imposition of wooden structured houses, the restriction on the illegal deforestation
are to be made. Introduction of afforestation and plantation programme for the
unstable sections, and the provision of appropriate roof top water and waste water
disposal system to release the water through pipes and concrete drains are to be
regularized effectively. Finally, the people should be made aware about the
unscientific interaction between man and land or soil and that can minimize the
harmful impact of population pressure on land and bring the stability to the slope
material.
The analysis of critical slope angle, rainfall and height suggests that the Shiv-
khola Watershed is a most unstable section of Darjiling Himalaya. In the existing
weak lithological composition threshold slope of the unstable sections range from
21° to 26° with an average value of 24°. The large parts of Tindharia, Paglajhora,
Gayabari, and Shiviter possess the slope angle of more than 24° that indicates
absolute instability. Around 100 mm/day rain may become critical to the landslide
prone sections in the watershed. The slope height ranging from 5.89 to 7.80 m. is
very much critical and beyond this there is a great chance of slope failure at the
places of Tindharia and Paglajhora. The special care must be taken to reduce the
height of back wall along the main road as well as the marginal part of the slided
scar to that of 5.00 m of less with respect to the below threshold slope angle of 24°.
As the rain water is the main triggering factor for reducing cohesion, angle of
internal friction and increasing the pore-water pressure which altogether increase
the driving force to the slope materials, the most suitable way would be the
application of synthetic plastic for covering the extreme unstable section mainly
along the Hill Cart Road not to percolate the rain water through soil and slope in the
rainy season. The installations of rain gauge stations at all the major landslide
locations for the continuous monitoring of rainfall and the relevance of cumulative
rain in landsliding should be transmitted to the people through electronic media

which can aware the people to avoid the negative impact of slope failure.
In the Shivkhola Watershed, Tindharia Tea Garden, Lower Paglajhora (14 Miles
Bustee), Proper Paglajhora, Tindharia Proper (Rly. Stn.), Shiviter Tea Garden,
Sepoydhura T.E., Nurbong T.E., and Lower Gayabari are unstable under saturated
condition where safety factor values are lying below ‘1’. At all those places
cohesion and angle of internal friction is very low. Friction angle ranges between
19° and 26° and cohesion varies from 0.0355 to 0.162. One dimensional (1D) slope
stability model based landslide susceptibility maps for dry, semi-saturated and
saturated condition indicates that here is a tendency of an increase in areal extension
of high to very high landslide susceptibility area in saturated soil condition. Under
saturated condition more than 50 % area are characterized by high to very high
landslide susceptibility whereas under semi-saturated condition more than 35 %
area is dominated by high landslide susceptibility. This situation demonstrates that
soil saturation and reduction of soil cohesion is the important landslide contributing
factor in the Shivkhola Watershed. To monitor the soil saturation and cohesion,
sub-surface drainage monitoring and draining of seepage water by boreholes could
be the best stabilizing factors in the existing weak lithological composition. The
application of geotextile method and introduction of first growing grasses over
fragile lithology can bring stability over steep unstable slope. Besides the con-
struction retaining walls and breast wall, jhora training through concrete structure
and cementation of cracks and weaknesses planes can bring the stability to the soil
and land.
The spatial distribution of slope instability of six individual sub-watersheds in
the Shivkhola Basin could be well explained by means of the interaction between
morphometric, geometric and hydrologic factors. The synthesis of basin geometry
such as Form Factor, Elongation ratio, Circularity Ratio, Elipticity Index, Com-
pactness Coefficient and Length of Overland Flow reveals the composite ranking in
terms drainage efficiency. On the basis of total composite index values (CIV), first
priority should be given for rational management of land, water and soil in
sub-watershed-I and II, which is followed by IV, III, VI and V. It is also seen that
the sub-watershed-I and II are most vulnerable due to steep slope, high drainage
efficiency, friable lithological composition and more anthropogenic pressure.
Landslide susceptibility analysis based on the landslide hazard evaluation factor
rating approach (LHEF) shows that PaglaJhora, Gayabari, Tindharia, Northern and
central part of Shivitar T.E., Tindharia T.E., and Gitingia T.E. are the places of very
much susceptible to catastrophic slope failure with positive relationship between
landslide triggering factors and landslide susceptibility values. On the other hand
Analytical Hierarchy Process (AHP) based prepared landslide susceptibility map
reveals that Paglajhora, 14 Miles Bustee, Tindharia, Nurbong and Shiviter are
experiencing high to very high landslide susceptibility. But no danger of landslide
exists at the places of Gitingia, Sepoydhura, lower Gayabari, extreme mid, lower
and marginal section of the watershed and lower slope of Tindharia. The traffic,
property, and life are under threat at the places of Lower Paglajhora, Tindharia (near
railway station), Shiviter tea estate and Tindharia Tea Garden where the slope
8.4 Conclusion 257
failure causes great damages to all those properties. Study shows that the places
with moderate level of landslide susceptibility and high intensity of risk elements
are registered as high to very high landslide hazard risk areas. AHP based devel-
oped couple-comparing matrix and derived prioritized factor rating value (PFRV)
reveals that lithology, concentration of human settlement, drainage and slope are
the most significant landslide triggering factors in the Shivkhola Watershed. In the
Shivkhola Watershed landslide susceptibility and hazard risk is very much asso-
ciated with moderate to steep slope (>40°); north, north-east; east, south-east and
south facing slope; degraded forest, high settlement and road contributing area, tea
garden, and bare surface; high positive and high negative slope curvature; moderate
level of drainage density over steep escarpment slope; and weak lithology-foliated
gneiss, phyllite, mica-chlorite-schist and quartz-schist. In Lower Paglajhora and 14
Miles Bustee, except along the Hill Cart Road the landslide hazard risk is moderate,
yet there exists very high level of landslide susceptibility. On the other hand,
because of the existence of high intensity of risk elements, the moderate level of
landslide susceptibility areas of Mahanadi, Gayabari and Jogmaya are experienced
with high to very high landslide hazard risk.
Before going to introduce any management actions for the landslide prone
mountain basin, the Shivkhola, the identification and analysis of landslide triggering
factors and prepared landslide susceptibility and landslide hazard risk map should be
used as an effective tool to the local administration, planners and policy makers. The
priority must be given to high to very high risk prone areas as these areas are
dominated by risk elements such as road, settlement, tea garden and other resources
and experience frequent landslide phenomena.
The utilitarian and environmental concept about the resource utilization advocate
the harnessing of optimum utility for long in a sustainable manner and thus rational
utilization of resource becomes inevitable. The land and associated soil and water
are the basic resources available on the earth. In such a highly instable region, the
protection of slope and soil is a great challenge through rational use of these
resources for harnessing greater utility over long time. Management of already
destroyed slope is a great concern today where the danger of instability is crudely
exposed. The present work reveals that rational management of potential slope
failure zones, where the danger is not exposed yet, is of most important and to be
considered as emergency as that in case of immediate response to a fresh landslide.
Pre-slide management of slope requires the identification of susceptible zones. The
present work identifies such vulnerable zones of varied priority applying functional,
systematic and metastable approach of slope evolution where the stability is
expressed as a function of a numbers of factors. The site specific management of
slope is necessary along with the general treatment recommended above and timely
response to this instability problem only can save the region from potential
destruction and the proper execution of the suggestion made may save the resources
and ultimately the society and thus the present work will find social relevance.
References
Angeli et al (1994a) A system of monitoring and warning in a complex landslide in Northeastern

Italy. Landslide News 8:12–14
Angeli et al (1994b) Longterm monitoring and remedial measures in a coastal landslide (Italy). In:
Proceedings of the VIIth ISL, Trondheim, Norway, Balkema, Rotterdam
Bayfield MA, Meister AD (1998) East coast forestry project review. Report of Ministry of
Agriculture and Forestry
Bianco B, Bruce DA (1991) Large landslide stabilization by deep drainage wells. In: Slope
stability engineering-application and development. Thomas Telford, London, pp 319–326
Bromhead EN et al (1996) Stabilisation of an urban refuse dump and its planned extension near
Ancona, Marche, Italy, engineering geology of waste disposal. Geol Soc Eng Geol 11:87–92
Collison AJC (1993) Assessing the influence of vegetation on slope stability in the tropics. Ph.D.
thesis, University of Bristol
Crozier MJ (2004) Management frameworks for landslide hazard and risk:issues and options. In:
Glade T (ed) Landslide hazard and risk. Wiley, Chichester, pp 331–350
Fort DS et al (2000) Instrumentation and monitoring of the coastal landslides at Lyme Regis,
Dorset, UK. In: Bromhead E et al (eds) Landslides in research, theory and practice. Thomas
Telford, London
Gray DH, Sotir RB (1996) Biotechnical and soil bioengineering slope stabilisation. Wiley, New
York
Gray DH, Leiser AJ (1982) Biotechnical slope protection and erosion control. Van Nostrand
Reinhold, New York
Greenway DR (1987) Vegetation and slope stability. In Anderson MG, Richards KS (eds) Slope
stability. Wiley, Chichester, pp 187–230
Howell JH, Sandhu SC, Vyas N, Sheikh R, Rana SS (2006) Introducing bio-engineering to the
road network of Himachal Pradesh. J Indian Road Congr 67–3
Hutchinson JN et al (2002) Landslide movement affecting the lighthouse at St Catherine’s point,
Isle of Wight. In: McInnes RG, Jakeways J (eds) Instability planning and management.
Thomas Telford, London, pp 291–298
Keefer DK et al (1987) Real-time landslide warning during heavy rainfall. Science 238:921–925
Maiti R (2007) Identification of potential slope failure zones of Shiv-Khola watershed; Darjiling
Himalaya, through critical analysis of slope instability- a step towards rational and scientific
management of land, soil and water. UGC Sponsored Minor Research Project [F.31-210/2005
(31.03.2007)]
Montgomery DR, Schmidt KM, Greenberg HM, Dietrich WE (2000) Forest clearing and regional
landsliding. Geology 28:311–314
Morgan B et al (1999) Historical and potential debris-flow hazard map of area affected by the June
27, 1995, storm in Madison County, Virginia. US Geological Survey Geologic Investigations
Series Map I-2623B, scale 1:24000
Morrissey M et al (2001) Regional application of a transient hazard model for predicting initiation
of debris flows in Madison County, Virginia. US Geological Survey Open File Report 01-481
Pearce AJ, O’Loughlin CL, Jackson RJ, Zhang XB (1987) Reforestation: on-site effects on
hydrology and erosion, Eastern Raukumara range, New Zealand. Forest hydrology Interna-
tional Association of Hydrological Sciences, Wallingford, pp 489–497
Phillips CJ, Watson AJ (1994) Structural tree root research in New Zealand: a review. Landcare
research science series 7. Manaaki Whenua Press, Lincoln
Phillips CJ, Pearce AJ, Marden M (1990) Effectiveness of reforestation in prevention and control
of landsliding during large cyclonic storms. In: Proceedings, 19th IUFRO conference,
Montreal, pp 358–361
Phillips CJ, Marden M, Miller D (2000) Review of plant performance for erosion control in the
East Coast region. Landcare research contract report LC9900/111 prepared for Ministry of
Agriculture and Forestry (Unpublished)
References 259
Schuster RL, Krizek J (eds) (1978) Landslides: analysis and control. Transportation Research
Board, Washington
Schuster RL, Kockelman WJ (1996) Principles of landslide hazard reduction. In: Turner AK,
Schuster RL (eds) Landslides: investigation and mitigation, transportation research board,
National Research Council, Special Report 247. National Academy Press, Washington, DC,
pp 91–105
Selby MJ (2005) Hillslope materials and processes. Oxford University Press, Oxford (Reprinted
Volume)
Veder C (1981) Landslides and their stabilization. Springer, New York
Zaruba Q, Mencl V (1976) Engineering geology. Elsevier, Amsterdam
Appendix A
To assess size class distribution of the soil particles Keen-Box Method was applied.
The soil samples were collected from 50 different locations of the Shivkhola
Watershed (Appendix A). On the basis of analyzed soil samples, spatial zonation of
granule, sand, silt and clay were made and finally all the maps were incorporated
with landslide inventory map to determine the potential stability of the soil
particles. The depth of soil for each location was determined studying the landslide
scars, slope cutting and boring the slope surface (Appendix A) (Table A.1).

and Prediction, Springer Natural Hazards, DOI 10.1007/978-981-287-146-6
Table A.1 Soil texture and depth analysis of the watershed
262
Sample no. Location of the collected samples Latitude and longitude Saturated soil Granule (%) Sand (%) Silt (%) Clay (%)
depth (in m)
1. Mahanadi T.E. 26°53′30″N/88°24′02″E 1.55 41.59 55.84 1.47 1.05
2. Paglajhora R.F. 26°52′43″N/88°22′30″E 1.45 37.17 56.38 3.03 0.26
3. Upper Paglajhora 26°22′20″N/88°18′00″E 1.40 8.25 88.11 1.98 1.65
4. Giddapahar T.E. 26°02′20″N/88°17′05″E 1.25 6.36 87.47 2.98 3.18
5. Sepoydhura T.E. 26°52′03″N/88°23′03″E 2.00 44.29 44.40 2.98 2.49
6. Nurbong T.E. 26°49′57″N/88°23′30″E 2.55 25.68 72.51 0.75 1.05
7. Nurbong T.E. near sinking zone 26°51′37″N/88°23′28″E 2.75 46.04 50.05 2.19 1.71
8. Gitingia T.E. 26°52′03″N/88°22′45″E 1.15 22.26 73.47 2.31 1.95
9. Shiviter T.E. (Upper) 26°52′38″N/88°20′40″E 0.85 30.54 69.21 0.13 0.11
10. Lezipur T.E. 26°52′26″N/88°20′08″E 0.75 28.32 68.86 1.59 1.24
11. Shiviter T.G. (M.B) 26°52′51″N/88°19′52″E 1.85 33.60 61.92 2.03 2.43
12. Shiviter lower 26°50′20″N/88°19′20″E 2.95 17.52 79.34 1.42 1.71
13. Lower Paglajhora 26°52′55″N/88°18′24″E 0.55 27.75 69.16 1.47 1.62
14. 14 miles bustee 26°52′02″N/88°18′17″E 0.65 47.10 51.49 0.64 0.75
15. Gayabari proper 26°52′41″N/88°18′46″E 1.20 19.42 76.70 1.67 2.22
16. Shivkhola R.F. 26°52′50″N/88°18′40″E 3.75 39.63 58.34 0.93 1.09
17. Shivkhola Central 26°52′47″N/88°18′38″E 3.50 34.22 62.36 1.90 1.52
18. Jogmaya T.E.(1234M) 26°52′35″N/88°18′45″E 3.25 29.08 68.30 1.14 1.47
19. Shiviter labor line 26°52′53″N/88°18′52″E 0.95 28.24 69.22 1.05 1.40
20. 14 Mile Bustee proper 26°52′51″N/88°18′25″E 0.75 37.47 61.11 0.70 0.70
21. Tindharia proper 26°51′28″N/88°20′17″E 1.25 38.82 60.81 0.16 0.24
22. Tindharia Rly. Stn. 26°51′23″N/88°20′08″E 1.20 30.42 67.46 0.61 0.91
23. Gayabari upslope 26°52′40″N/88°18′43″E 0.45 23.97 73.71 1.03 1.29
24. Gayabari lower slope 26°52′40″N/88°18′43″E 1.55 33.50 64.81 0.81 1.12
25. Tindharia tea garden 26°51′38″N/88°20′20″E 3.15 28.73 54.32 2.95 1.87
(continued)
Appendix A
Table A.1 (continued)
Sample no. Location of the collected samples Latitude and longitude Saturated soil Granule (%) Sand (%) Silt (%) Clay (%)
depth (in m)
Appendix A
26. Tindharia managers B 26°51′25″N/88°19′19″E 2.10 36.27 49.53 2.25 1.47

27. Chunabhati 26°51′01″N/88°20′24″E 0.65 32.58 65.98 0.94 0.35
28. Mid-slope at the right sight of Tindharia 26°51′23″N/88°20′25″E 0.90 34.35 71.15 1.14 .64
29. Near chunabhati railway station 26°51′00″N/88°20′227″E 1.05 34.25 62.75 1.18 0.65
30. Extreme lower segment from Tindharia T.E. 26°51′30″N/88°19′22″E 1.76 31.72 59.35 1.99 1.48
31. Upslope from Gitangea T.E. 26°52′20″N/88°22′35″E 1.35 29.20 63.75 2.05 1.69
32. Right slope from Tindharia 26°51′10″N/88°19′17″E 1.19 35.50 64.20 0.92 0.75
33. Central part of the watershed 26°51′45″N/88°19′00″E 2.05 33.25 61.50 1.65 1.53
34. Shiviter T.E. 26°52′51″N/88°19′52″E 1.81 28.90 69.55 1.12 1.35
35. Upslope of Shiviter T.E. 26°52′55″N/88°20′45″E 1.49 35.95 66.74 0.94 0.62
36. Gayabari upslope 26°52′42″N/88°18′49″E 2.15 32.29 64.46 1.32 1.55
37. Lower segment of Shivkhola R.F. 26°51′45″N/88°18′30″E 1.63 31.85 65.21 1.34 1.18
38. Between lower Paglajhora and Mahanadi 26°52′58″N/88°18′34″E 1.32 30.68 65.48 1.41 1.23
39. Lower slope from lower Paglajhora 26°52′55″N/88°18′28″E 0.11 30.45 65.50 1.40 1.20
40. Few kms before 14 miles basti 26°52′49″N/88°18′27″E 1.05 31.85 67.45 1.39 1.27
41. Extreme upslope of paglajhora 26°57′58″N/88°17′34″E 1.45 25.30 72.92 1.89 1.26
44. Middle section btwn. Nurbong and Tindharia T.E. 26°51′10″N/88°19′42″E 1.73 33.84 63.39 1.46 1.22
45. Lower slope from Tindharia 26°51′00″N/88°19′05″E 1.67 30.95 65.21 1.18 1.24
46. Chunabhati (along road side) 26°49′00″N/88°20′26″E 1.07 34.45 64.30 1.16 0.73
47. 14 miles basti 26°52′50″N/88°18′24″E 0.75 35.90 61.25 0.74 0.59
48. Shiviter manager’s bunglaw 26°52′53″N/88°18′47″E 1.53 30.55 61.39 2.10 1.80
49. Mahanadi (along the Hill Cart Road) 26°53′20″N/88°24′12″E 1.55 32.95 63.45 1.78 0.87
50. Central part of the watershed 26°51′45″N/88°18′40″E 2.59 29.7 67.95 1.08 1.40
Source Laboratory analysis by author himself
263
Appendix B
Appendix B shows the rainfall statistics since 1979–2009. This rainfall statistics
helped to assess the return period and the recurrence interval of critical rainfall for
landsliding. The month of June, July and August are the consistent rainfall months
of a year when the propensity of slope failure is of great. Not only that the
occurrence of rainfall were incorporated with the landslide events which showed a
positive relationship between consistent rainfall and landslide in the Shivkhola
Watershed (Table B.1).

Table B.1 Rainfall statistics since 1979–2009 (rainfall in mm)
266
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1979 12.5 62.5 00 74.00 265.5 731.75 588 989.5 386.25 122.5 00 00
1980 00 00 56.25 150 168.25 249.5 739.25 757.75 447.5 268.5 75.00 35.00
1981 2.5 16.5 97.5 35 261 883.5 870.5 620.5 725.5 725.25 12.5 12.5
1982 3.25 53 00 83.5 150 462 350 800 600 275 4.75 13.00
1983 00 00 00 12.8 203.21 830.20 701.5 1087.7 765.59 293.8 00 00
1984 31 5 63 199 490.5 611 1287 884.5 570.6 4.80 00 00
1985 00 10 80.30 187.1 161.1 1123 594.6 473.25 462.1 89.5 10 00
1986 22.5 25 5 10 288.8 685.15 1276.97 646.93 591.62 223.12 29.05 10.5
1987 44 76 00 203.25 592.59 949.5 1432.2 850.2 982.8 362.05 00 8.5
1988 00 00 27 46.12 462.34 1042.6 1641 905 1042 484.18 00 00
1989 10 12 33 120 208.9 947.4 752.5 632.5 577.5 179 00 00
1990 59.05 24.4 6.7 34.25 69.72 203 416 718.13 445.2 250.5 00 00
1991 2 00 125 51.3 35.1 113 379.7 271 299 00 00 00
1992 1.00 25.00 57.5 125 568.8 889.5 1200 982.5 520 171.5 5.6 1.5
1993 2.85 37 65.25 204.5 796.5 1025.5 999.5 785.25 457.8 85 35 2.25
1994 3.5 45 51.8 79.5 420.5 675.2 795.5 1195.5 653.5 154.5 17.25 1.5
1995 4.00 28 49.5 131 214 493.9 1161.8 752.2 484.5 110.4 00 00
1996 00 00 24.6 84.2 266.1 838 1130.8 706.6 613.8 175.4 00 32
1997 60.0 27.4 30.6 19.2 204.6 29.4 677.8 608.6 551 19 00 00
1998 10 17.6 50 45.00 405.5 675.5 978.4 795.2 735 75.6 4.95 15
1999 00 4.5 22.5 57.2 178.6 748.6 1007.3 1006.3 610.7 157.6 00 00
2000 7.2 13.5 89.5 19.6 121.3 369.7 750 890.4 748 8.2 00 251.4
2001 00 00 246 87.5 145.8 641.6 1367 1035.3 395.8 26.8 27.2 10
2002 00 00 30.6 60.2 427.1 938.6 1013.1 732.9 569.8 274.6 00 00
(continued)
Appendix B
Table B.1 (continued)
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2003 5.0 00 15.00 127.6 498.4 995.8 1040.2 877.3 542.6 16.2 00 00
Appendix B
2004 00 00 00 62.8 194.4 709.8 655.8 235.4 372.6 9.2 12.4 14

2005 8.6 11.2 2.4 65.5 175.4 1000.25 1520.31 325.56 523.05 250.5 20 25
2006 6 7.25 12.35 204 180.4 1340.7 1811.8 602 648.8 279.8 00 251.8
2007 00 00 28.8 123.6 288.2 570.2 1268 828.4 363.4 129 24 6
2008 105 25.2 45.6 145.6 375.5 657.8 1155.4 985.7 578.5 211.3 54 4.5
2009 2.5 7.25 35.75 111.5 475.8 987.25 1245 1125 724.5 157.35 34.8 3.95
267
Appendix C
Appendix C consisting depth of soil, slope angle, angle of internal friction,

cohesion, wet soil density, specific yield of soil, and safety factor (for dry, semi-
saturated and saturated conditions). With the help of all these geo-technical
parameters of soil, a slope stability model (1-D Slope stability model) was applied
to determine the value of safety factor and on the basis of safety factor value, spatial
distribution of landslide susceptibility was assessed under dry, semi-saturated and
saturated condition. To establish the model the saturation index (m) of value 0.00,
0.5 and 1.00 were taken for dry, semi-saturated and saturated conditions
respectively. The cohesion (c) and friction angle (φ) of the soil were estimated
using ‘tri-axial soil testing mechanism’. Slope angle was assessed studying the
prepared slope map from DEM (Digital Elevation Model) (Table C.1).

Table C.1 Result of field measurement and laboratory test of soil samples observed and collected from different location of the watershed
270
Sample no. (Z in m) (Ѳ) Φ (c) kg/cm3 (γ in KN/m3) (γw-KN/m3) FS (dry condition) FS (semi-saturated) FS (Saturated)
° °
1. 1.55 42 28 0.42 1.86 0.92 0.883 0.737 0.5916
2. 1.45 54° 23° 0.21 2.13 1.01 0.452 0.378 0.3052
° °
3. 1.40 37 27 30′ 0.11 1.89 0.89 0.7773 0.6146 0.4519
4. 1.25 38° 31° 0.65 1.97 0.82 1.3687 1.2019 1.0351
5. 2.00 39° 26° 0.45 2.01 0.90 0.8312 0.6963 0.5615
6. 1.35 39° 19° 0.02 2.01 1.12 0.4402 0.3218 0.2033
7. 2.75 49° 19°30′ 0.03 2.25 1.09 0.3176 0.2430 0.1685
8. 1.15 35° 25°15′ 0.60 2.05 0.85 1.215 1.0756 0.9360
9. 0.85 66° 24° 0.04 1.95 0.76 0.2631 0.2245 0.1859
10. 0.75 64° 29° 0.71 2.04 0.79 1.448 1.3958 1.3435
11. 1.85 67° 24° 0.02 2.36 1.08 0.2017 0.1584 0.1152
12. 2.95 22° 28° 0.65 2.21 0.99 1.6033 1.3085 1.0137
13. 0.55 65° 21°15′ 0.01 2.00 0.84 0.2050 0.1670 0.1289
14. 0.65 46° 22° 0.32 2.05 0.82 0.8708 0.7927 0.7147
15. 1.20 51° 23° 0.79 1.88 0.77 1.0597 0.9893 0.9189
16. 3.75 59° 28° 0.67 2.06 0.78 0.5159 0.4554 0.3949
17. 3.50 57° 29° 0.66 2.10 0.76 0.5565 0.4675 0.2569
18. 3.25 15° 22° 0.33 2.22 0.81 1.6907 1.4157 1.1406
19. 0.95 63° 25° 0.25 1.98 0.77 0.5661 0.5199 0.4737
20. 0.75 64° 22° 0.04 2.35 1.01 0.2546 0.2123 0.1699
21. 1.25 48° 21° 0.05 2.23 0.98 0.3817 0.3058 0.2298
22. 1.20 52° 20° 0.07 2.22 0.88 0.3385 0.2822 0.2352
23. 0.45 46° 26° 0.91 1.99 0.85 2.5050 2.4044 2.3038
24. 1.55 20° 29° 0.52 2.03 0.73 2.0372 1.7634 1.4895
25. 3.15 24° 25° 0.06 2.11 0.75 1.0716 0.88551 0.69937
26. 2.10 13° 24° 0.58 2.13 0.74 2.52008 2.18508 1.85008
(continued)
Appendix C
Table C.1 (continued)
Sample no. (Z in m) (Ѳ) Φ (c) kg/cm3 (γ in KN/m3) (γw-KN/m3) FS (dry condition) FS (semi-saturated) FS (Saturated)
27. 0.65 36° 27° 0.86 1.96 0.69 2.12086 1.99742 1.87398
Appendix C
28. 0.90 26° 25° 0.10 1.90 0.68 1.104495 0.784422 0.7623225
°
29. 1.20 37 25° 0.25 2.30 0.91 0.7041 0.6021 0.5001
30. 2.20 28° 31° 0.81 2.22 0.87 0.9137 0.8131 0.7124
31. 1.70 38° 24° 0.41 1.99 0.54 0.8197 0.7423 0.6650
32. 1.30 36° 29° 0.75 1.97 0.53 1.3788 1.2178 1.1735
33. 2.65 27° 23° 0.06 2.11 0.77 3.546 2.9152 2.2843
34. 1.75 35° 32° 0.48 2.05 0.81 1.1772 1.000 0.8246
35. 1.15 39° 27° 0.51 2.40 1.05 0.8783 0.7582 0.6382
36. 2.18 44° 21° 0.09 2.44 1.11 0.3323 0.2627 0.2874
37. 1.72 64° 22° 0.25 2.15 0.98 0.3686 0.3237 0.2788
38. 1.65 49° 33° 0.24 1.89 0.66 0.7199 0.6214 0.5228
39. 1.40 61° 24° 0.08 1.97 0.53 0.3152 0.2820 0.2488
40. 1.35 25° 19° 0.06 2.02 0.72 0.7958 0.6642 0.5326
41. 1.49 30° 21° 0.09 2.55 1.09 0.7196 0.5774 0.4353
42. 1.40 42° 31° 0.17 2.48 1.09 0.7658 0.6191 0.4725
43. 0.85 51° 20° 0.35 1.85 0.61 0.7499 0.7013 0.6527
44. 1.45 23° 30° 0.40 1.93 0.65 1.7576 1.5285 1.2995
45. 1.50 55° 20° 0.22 2.04 0.96 0.4078 0.3479 0.2879
46. 0.95 66° 27° 0.15 2.58 1.29 0.3915 0.3348 0.2781
47. 0.65 49° 25° 0.08 2.47 1.22 0.5060 0.4059 0.3057
48. 1.15 38° 18° 0.29 2.00 0.72 0.6757 0.6009 0.5260
49. 1.23 47° 27° 0.32 1.79 0.50 0.7665 0.7002 0.6338
50. 3.05 18° 26° 0.78 1.99 0.60 1.9384 1.7121 1.4858
Z = soil depth; Ѳ = slope angle; φ = angle of internal friction; c = cohesion; γ = specific yield of soil; γw = unit weight of water; and m* = wetness index with 20 year return period
of rainfall intensity
271
Appendix D
To determine the probability of future landslide events using the probability model,
major and minor landslide event in the Shivkhola Watershed since 1968–2011
(Appendix D) were taken into account from the work done by researchers such as
Basu and Sarkar (1985, 1988) Basu Basu and Maiti (2001), Maiti (2007, 2011)
Ghosh (2009b), Sarkar (2011), and Mandal and Maiti (2011, 2012 and 2013). In
this study, the occurrences of reactivated and fresh landslides were considered
(Table D.1).

Table D.1 Occurrences of major and minor landslide event in the Shivkhola watershed since 1968–2011
274
Event date and Important location Type of the land- No. of No of reacti- No of new Triggering factors
Year slide events landslides vated slides slides
3rd, 4th, and 5th Giddapahar, Gayabari. Major and most 13 4 9 Unprecedented continu-
Oct., 1968 Paglajhora. deadful ous rainfall of 221.4 mm
in 4 days
1980 Paglajhora, 14 miles Bu- Major event 6 2 4 Due to incessant rainfall
stee, Tindharia. of 299.2 mm in 2 days
17th Sept., 1984 Shivite T.E., Nurbong T. Minor and local 5 3 2 Continuous rainfall
E., Mahanadi and
Giddapahar.
17 and 18th Oct., Maanadi, Sepoydhura, Major events 6 4 2 Continuous rainfall
1985 Gayabari
7th Sept., 1986 Mahanadi, Tindharia, Minor and local 5 2 3 Continuous rainfall
Chunabhati triggering
10 and 11th Paglajhora and Chunabh- Major event 5 3 2 Due to incessant rainfall
Sept., 1991 ati Area of 462.5 mm in 2 days
1st, 2nd, and 3rd Tindharia and Gayabari Prominent major 4 2 2 Due to heavy and con-
July, 1993 and Mahanadi event centrated rainfall of
211.3 mm in 2 days
1995 Paglajhora, Giddapahar, Minor and local 7 5 2 Cumulative rainfall and
Tindharia and Gayabari event formation of road-cut
benches with heavy
vehicles
6th, 7th, and 8th Chunabhati, Tidharia, Prominent major 10 8 2 300–600 mm cumulative
July, 1998 Paglajhora, Mahanadi event rainfall in 2/3 days
(along NH-55)
2002 Giddapahar, Tindharia Prominent major 13 6 7 Cumulative rainfall and
and Gayabari, Nurbong, event construction of road-cut
Gitingia benches and plying of
heavy loaded vehicles
(continued)
Appendix D
Table D.1 (continued)
Event date and Important location Type of the land- No. of No of reacti- No of new Triggering factors
Year slide events landslides vated slides slides
Appendix D
8 and 9th July, Gayabari T.E., Along Prominent major 12 11 1 Due to incessant rainfall
2003 NH-55, Tindharia and event of 500 mm in 2 days
Shiviter. Paglajhora,
2005 Shiviter T.E. and lower Major event 9 3 6 Heavy and continuous
Paglajhora, Jogmaya T.E. rainfall and subsidence of
road at Paglajhora
2006 Tindharia, 14 Miles Bu- Major event 10 6 4 Heavy rainfall
stee, Gitingia, Nurbong
17, 18th July and Tindharia and upper and Prominent major 6 4 2 Heavy rainfall
6th, 7th, and lower Paglajhora. event
78th, Sept., 2007
16th June, 2010 14 Mile near lower Pag- Prominent major 9 6 3 Heavy rainfall of 345 mm
morning lajhora, Nurbong, Gitin- event in 2 days
gia, Tindharia
10th May, 2011 Lower Paglajhora Prominent major 8 7 1 Heavy showers and the
event vibration caused by Si-
kkim earthquake
Total no. of landslide events year-16
No. of major landslide events year-12
Total no. of landslide events occurrences in the watershed since 1968–2011 = 128 (identified)
Total no. of reactivated landslide in the watershed-76
Total no. of fresh landslide (not associated with old slide/70 m. away from old slide) events year-52
275
Appendix E
Appendix E dealt with catastrophic rainfall events since 2005–2010 (Table E.1).

Table E.1 Analysis of catastrophic rainfall event during 2005–2010
278
2005 Rain in 2006 Rain in 2007 Rain in 2008 Rain in 2009 Rain in 2010 Rain in
mm mm mm mm mm mm
20th June 95.5 23rd 103.5 10th 222.72 7th June 125 3rd June 146.5 24th 88.9
May June May
26th June 183.5 28th July 150 28th 93.5 9th June 100 19th 133 28th 101.6
June June May
31st July 134.5 29th July 160 29th 120.5 23rd 203 7th July 175 16th 111.7
June June June
3rd Oct 100 30th July 112.5 10th July 120 26th 179 12th 200.5 14th July 103.6
June July
4th Oct 90 19th 150 17th July 124.5 29th 196.5 9th Sep 98.7 18th July 102
Aug June
31st Aug 120.5 18th July 100 7th July 273.5 27th Sep 95.5 25th July 92.4
17th 89.5 27th July 120.5 8th July 162.5 5th Aug 114.3
Sept.
3rd Oct 130 18th Aug 145.5 21st July 146.5 25th 89.4
Aug
23rd 106.2 28th 148.5 28th 102.6
Sept. July Aug
30th 100 16th 115.5
July Sept.
10th 191.5 25th 116.3
Aug Sept.
31st 107.5 26th 90.5
Aug Sept.
No of 5 8 9 12 6 12
days
Total 603.5 1016 1153.42 1933.5 849.2 1228.8
Average 120.7 127 128.15 161.12 141.53 102.4
Source Selim Hill Tea Estate 2007 (1/2 km. Crow fly distance from Tindharia)
Appendix E
Appendix F
See Figs. F.1, F.2, F.3, F.4, F.5, F.6, F.7, F.8, F.9 and F.10.
Fig. F.1 Plotting of the geotechnical attributes from Tindharia tea estate

280 Appendix F
Fig. F.2 Plotting of the geotechnical attributes from 14 Mile Bustee
Fig. F.3 Plotting of the geotechnical attributes from Paglajhora (lower)

Appendix F 281
Fig. F.4 Plotting of the geotechnical attributes from Mahanadi
Fig. F.5 Plotting of the geotechnical attributes from Gayabari lower

282 Appendix F
Fig. F.6 Plotting of the geotechnical attributes from Tindharia proper
Fig. F.7 Plotting of the geotechnical attributes from Shiviter tea estate
Appendix F 283
Fig. F.8 Plotting of the geotechnical attributes from Sepoydhura tea estate
Fig. F.9 Plotting of the geotechnical attributes from Nurbong tea estate
284 Appendix F
Fig. F.10 Plotting of the geotechnical attributes from mid-central part of the basin
Appendix G
This appendix describes the determination of frequency ratio and class weight for
each class of the landslide triggering factors such as slope angle, slope aspect, slope
curvature, lithological composition, lineaments, drainage density, land use and land
cover, upslope contributing area, road contributing area, road contributing area and
settlement density. The table also reveals the relative significance of each class of
the landslide triggering factors in terms of landslide. The determined class weight
values were used to perform the linear integration model and to prepare the
landslide susceptibility map of the Shivkhola Watershed (Table G.1).

Table G.1 Class frequency ratio/prioritized class rating value and class weight
286
Classes Number of % of Npix landslide % of (Npix(Si)) FR Class weight (W)

pixels (F1) (Ni) pixels
(Npix(Ni)) (F2)
(Npix(Si))
Slope angle (in degree) −1
0–7.17 3353 10.12 190 5.63 0.53 −45.14
7.17–14.34 3238 9.77 202 5.99 0.61 −39.43
14.34–19.92 3587 10.83 211 6.26 0.58 −42.99
19.92–24.97 2445 7.38 201 5.96 0.81 −19.60
24.97–29.75 3555 10.73 311 9.22 0.86 −14.33
29.75–34.53 2776 8.38 329 9.75 1.16 16.71
34.53–39.57 3854 11.63 413 12.24 1.05 5.35
39.57–45.95 3276 9.89 427 12.65 1.28 28.53
45.95–54.71 3557 10.74 543 16.09 1.49 50.85
54.71–67.73 3490 10.53 646 19.15 1.82 83.29
Slope Aspects (direction of slope) −2
Flat 784 2.37 24 0.711 0.3 −71.2
North 3879 11.71 665 19.72 1.68 69.63
North East 3797 11.46 443 13.13 1.15 14.86
East 4346 13.12 675 20.01 1.53 53.51
South East 6290 18.99 789 23.39 1.23 23.63
South 4556 13.75 597 17.70 1.29 29.23
South West 3332 10.06 35 0.74 0.07 −19.31
West 2870 8.66 69 2.05 0.24 −77.77
North West 3277 9.89 76 2.25 0.23 −78.62
Slope Curvature (positive, negative and zero) −3
−25.87 to −11.41 995 3.00 221 6.55 2.18 120.3
(continued)
Appendix G
Table G.1 (continued)
(Npix(Ni)) (F2)
Appendix G
(Npix(Si))
−11.41 to −5.73 785 2.37 210 6.23 2.63 165.71
−5.73 to −2.33 2111 6.37 486 14.40 2.26 128.41
−2.33 to −0.63 2431 7.34 374 11.09 1.51 52.09
−0.63–0.50 10045 30.32 388 11.50 0.38 −63.18
0.50–2.49 6302 19.02 268 7.95 0.42 −59.23
2.49–7.31 5438 16.41 464 13.76 0.84 −16.48
7.31–14.69 3343 10.09 475 14.08 1.40 40.28
14.69–24.33 895 2.70 222 6.58 2.44 146.23
24.33–46.45 786 2.37 265 7.86 3.32 325.34
Lineaments (Distance from lineament in m) −4
0.00–57.42 3381 10.20 624 18.50 1.81 82.75
57.42–126.32 3786 11.43 668 19.80 1.73 74.63
126.32–229.68 3695 11.15 451 13.37 1.20 20.25
229.68–356.00 3252 9.82 522 15.48 1.58 58.71
356.00–528.26 4799 14.48 444 13.16 0.91 −9.29
528.26–723.45 4141 12.50 286 8.48 0.68 −32.74
723.45–964.65 3887 11.73 221 6.55 0.56 −44.95
964.65–1251.75 3921 11.83 120 3.56 0.30 −71.21
1251.75–1642.20 1419 4.29 37 1.10 0.26 −75.74
1642.20–2925.40 850 2.57 0 0.00 0.00 0.00
Drainage density (length of drainage/sq. m) −5
0–1.90 5560 16.78 90 2.67 0.16 −85.62
1.90–3.80 5453 16.46 109 3.23 0.20 −81.82
(continued)
287
288

(Npix(Ni)) (F2)
(Npix(Si))
3.80–5.71 3289 9.93 158 4.68 0.47 −53.77
5.71–7.61 5049 15.24 137 4.06 0.27 −74.68
7.61–9.51 3477 10.49 159 4.71 0.45 −56.08
9.51–11.41 2728 8.23 465 13.78 1.67 68.64
11.41–13.31 1875 5.66 544 16.13 2.85 188.32
13.31–15.21 2191 6.61 542 16.07 2.43 145.57
15.21–17.12 1942 5.86 697 20.66 3.52 257.10
17.12–19.02 1567 4.73 572 16.96 3.58 263.22
Geology (Lithological Composition) −6
Darjiling gneiss 6695 20.21 692 20.52 1.02 1.55
Chungtung formation 4203 12.69 525 15.56 1.23 23.10
Lingtse granite 3150 9.51 475 14.08 1.48 48.98
Gorubathan formation 2945 8.89 448 13.28 1.49 50.31
Reyang formation 5925 17.89 621 18.41 1.03 3.00
Damuda formation (Gondwana) 3203 9.67 490 14.53 1.50 51.17
Siwalik groups 7010 21.56 122 3.62 0.17 −84.41
Land use and land cover (LULC) −7
Tea 2310 6.97 290 8.60 1.23 23.73
Jungle 2657 8.02 312 9.25 1.15 15.62
Open forest 531 1.60 19 0.56 0.35 −66.03
Degraded forest 1522 4.59 112 3.32 0.72 −28.22
Dense forest 2114 6.38 194 5.75 0.90 −10.04
Bared surface 4758 14.36 379 11.24 0.78 −22.15
(continued)
Appendix G
(Npix(Ni)) (F2)
Appendix G
(Npix(Si))
Road 1074 3.24 216 6.40 1.98 99.31
Settlement 3037 9.17 352 10.44 1.14 14.09
Agricultural land 6880 20.77 566 16.78 0.81 −19.54
Mixed Forest 9281 28.01 933 27.66 0.99 −1.28
Upslope contributing area (UCA in km/sq.km) −8
<5.00 11421 34.47 1089 32.29 0.94 −6.46
5.00–10.00 7520 22.70 923 27.36 1.21 20.93
10.00–15.00 6611 19.95 993 29.44 1.48 48.39
15.00–20.00 5215 15.74 253 7.50 0.48 −53.30
2364 7.14 113 3.35 0.47 −54.01
>20.00
Road contributing area (RCA-km/sq.km) −9
<0.002 5720 17.26 0 0.00 0.00 0.00
0.002–0.004 5307 16.02 79 2.34 0.15 −86.92
0.004–0.006 4220 12.74 440 13.04 1.02 2.46
0.006–0.008 4370 13.19 461 13.67 1.04 3.68
0.008–0.010 4003 12.08 527 15.62 1.29 29.84
0.010–0.012 3522 10.63 572 16.96 1.60 60.60
0.012–0.014 2957 8.93 608 18.03 2.02 103.80
2532 7.64 686 30.34 3.97 169.12
>0.014
Settlement density (no of settlement/sq. m) −10
Very low 6445 19.45 235 6.97 0.36 −65.34
Low 5780 17.45 329 9.75 0.56 −44.89
(continued)
289
290

(Npix(Ni)) (F2)
(Npix(Si))
Moderately low 4858 14.66 374 11.09 0.76 −24.82
Moderate 4397 13.27 499 14.79 1.11 11.68
Moderately high 4265 12.87 591 17.52 1.36 37.05
High 3774 11.39 658 19.5 1.71 72.54
Very high 3612 10.90 687 20.37 1.87 88.39
Appendix G
Index
A G
Analytical Hierarchy Process (AHP), 192, 198, Geo-textile Method, 242
199 Gravitational force, 24
Anchor Works, 230 Grouting or void filling, 241
Angle of plane sliding friction, 25
Antecedent Moisture Condition, 132, 133
Asequent landslides, 31 H
Heave, 31
Higher Himalayan Crystalline Sequence, 194
B Hydrologic Soil groups, 129, 131
Bio-engineering Methods, 242 Hypsometric curve, 63
Biological force, 25
Breast Wall, 238
Bulk density, 83 I
Bureau of Indian Standards (BIS), 46 Initial Abstraction, 133
Buttress Fill, 230 Insequent Landslide, 31
Internal friction or shearing resistance, 25
C
Catch Water Drain, 239, 240 J
Chunam Plaster, 241 Jhor Training, 233, 235
Cohesion, 25, 85–89, 147, 171–173 Jute Textile Method, 242
Consequent landslides, 31
Consistency ratio, 199
Control works, 230 K
Keen-Raczowski Test, 83
D
Digital Elevation Model (DEM), 61, 62 L
Landslide density, 202
Landslide hazard risk, 193, 195, 203, 206, 214,
E 216
Expansion forces, 24 Landslide Inventory Map, 174, 198
Landslide Potentiality Index Value, 58, 59
Landslide risk exposure, 191, 205
F Landslide susceptibility, 200, 209, 210, 214
Frequency Ratio Model, 200, 202 Landslide Warning System, 251, 252
Friction angle, 86, 87, 89, 147, 171, 173 Lesser Himalayan Sequence, 194

292 Index
Lineament, 67, 68, 197 Shotcrete technique, 237

Liner combination model, 202 Slope aspect, 71, 72
Slope curvature, 71–73
Slope stability model, 169, 171
M Slurry, 2
Main Boundary Thrust, 193, 195 Soil Conservation Service (SCS), 125
Main Central Thrust, 193 Soil Removal Works, 230
Major Stress, 86, 171 Solifluction, 31, 32
Minor Stress, 171 Spread, 2
Mohr-Coulomb equation, 25, 38 Stress circle, 148, 171
Monte-Carlo Method, 43 Sturzstroms, 31
‘Mull’ and ‘Mor’, 74 Sub-surface drainage control works, 230
Surface curvature, 71, 72, 74
Surface drainage control works, 230
N
National Disaster Management Act, 230
T
Threshold angle of stability, 25
P Threshold Rainfall, 146, 150, 151
Particle density, 83, 84 Threshold Slope Angle, 151, 152
‘Poisson’ and ‘Binomial’ probability distribu- Threshold Slope Height, 151
tion models, 152 Topple, 2
Pore-water pressure, 34 Translational slides, 2, 31
Porosity, 83, 84 Triggering factors, 97
Potential maximum retention, 139
Precondition factor, 192
U
Upslope contributing area, 37, 111
R
Restraint works, 230
Retaining Wall, 230, 235, 237 V
Rotational slides, 2, 31 Vegetation-Slope Model, 244, 245
Run-off Curve Number, 124, 134, 136 Vegetative turfing, 248
Run-off equation, 126 Volumetric expansion, 82
Run-out distance, 1
W
S Water-pressure force, 24
Safety factor, 175–178 Weighted Curve Number, 134, 136
Shear strength, 175, 177, 178 Wetness Index, 173
Shear stress, 175–178

Semi-Quantitative Approaches For Landslide Assessment & Prediction PDF

Uploaded by

Copyright:

Available Formats

Semi-Quantitative Approaches For Landslide Assessment & Prediction PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Semi-Quantitative Approaches For Landslide Assessment & Prediction PDF

Uploaded by

Copyright:

Available Formats

Springer Natural Hazards

ISBN 978-981-287-145-9 ISBN 978-981-287-146-6 (eBook)

Library of Congress Control Number: 2014951664

Springer Singapore Heidelberg New York Dordrecht London

© Springer Science+Business Media Singapore 2015

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Midnapore, India, August 2014 Sujit Mandal

It is a moment of jubilance and euphoria to me because of the candid endeavor so

Midnapore, India, August 2014 Sujit Mandal

2 Geo-spatial Variability of Physiographic Parameters

2.3 The Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 74

3 Impact Assessment of Hydrologic Attributes

4 Surface Run-off, Soil Erosion and Slope Instability . . . . . . . . . . . . 123

4.3 Relationship Between Land Use and Land Cover,

5 Geomorphic Threshold and Landslide. . . . . . . . . . . . . . . . . . . . .. 145

6 Slope Stability Model and Landslide Susceptibility

6.3 Application of 1 Dimension Slope Stability Model

7 Application of Analytical Hierarchy Process (AHP)

8 Landslide Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 227

8.3.4 Breast Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

AHP Analytical hierarchy process

MBF Main boundary fault

Sc Compressive strength of the rocks

1.1 Key Concepts on Landslide

© Springer Science+Business Media Singapore 2015 1

slope movement types. The continuous downward movement of weathered rock

1.2 A Brief Account of Some Destructive

1.3 Relevance of the Landslide Study

Table 1.1 Some destructive landslide events

Table 1.1 (continued)

bustee Slide from Siliguri 1,118 m Foliated gneiss unconsolidated

Table 1.2 (continued)

Fig. 1.2 Location of major landslide prone area in India

higher Himalayan range in the higher reaches of Sikkim-Darjeeling Himalayas

landslide-related damages further indicated that lithology and geomorphology also

Fig. 1.7 Slope failure and

Fig. 1.8 Slope failure and destruction of settlement at Kedarnath (www.dailymail.co.uk)

Fig. 1.9 Elimination of road by landslide, Kedarnath (Source www.srinistuff.com)

Landslide Society, www.hils.org.np/publication.hils.org.np). Similarly, due to a

geoscientists, environmentalists, engineers, and stakeholders needs to be estab-

1.4 Landsliding in Darjiling Himalaya

extensive. The widespread landslides occurred throughout the Darjiling District in

Fig. 1.15 Road damage at

Fig. 1.16 Elimination of Hill

Fig. 1.17 Destruction of toy

Fig. 1.18 Massive damage of

Fig. 1.19 a–d Destruction in Jogmaya Tea Garden Area

1.5 Mechanism and Causes of Landslides

as conditionally stable slope (FS = 1.00–1.30). Basically the triggering factors

ðcz cos2 b uÞ tan u

The causes of landslides are usually related to instabilities in slopes. It is

Table 1.3 Basic causes of landslide

1.6 Types and Classiﬁcation of Landslide

A. Slope movements of superﬁcial deposits (slope detritus, weathering material),

morphology of the moving mass with consideration of mechanism, material and

(c) Gelifluction of clays

1.7 Objectives of the Present Work

1.8 Applied Methodology

Risk ¼ Hazard Exposure: ð1:9Þ

% of pore space ¼ 100 % % of solid space ð2:5Þ