Vous êtes sur la page 1sur 686

International Conference

On
Infrastructure Development
for
Environmental Conservation
& Sustenance
28 - 30 October, 2015

International Conference on

Infrastructure Development for


Environmental Conservation & Sustenance
28 - 30 October, 2015
Editors
DR.G.Ranganath
DR.S.Suresh Babu
Supported by

SERB, Department of Science &Technology, Government of India


Department of Space, Government of India
National Bank for Agriculture and Rural Development
Ministry of Environment & Forests, Government of India
University Grants Commission, New Delhi
Indian National Science Academy, New Delhi

Organized by
Department of Civil Engineering,
Adhiyamaan College Of Engineering,
Dr. M.G.R Nagar, Hosur,
Krishnagiri, Tamilnadu-635109, India.

INDECS 2015

Infrastructure Development for Environmental Conservation & Sustenance

Copyright 2015 Editors

Department of Civil Engineering


Adhiyamaan College of Engineering

Dr. M.G.R Nagar, Hosur,


Krishnagiri, Tamilnadu - 635 109, India.

Disclaimer

This book or any part of thereof, may not be reproduced


In any form or solid as a second hand book without the
Written permission of the editors. Authors are sole
responsible for the views, opinions and research outcome
present herein, and the editors are not liable for it.

Cover page design by: Mr.D.A.Ananth, Indigo Multimedia


Printed in India @Divya Printers Bangalore, +919686294383

International Conference on
Infrastructure Development for
Environmental Conservation & Sustenance
28 - 30 October, 2015
ORGANIZING COMMITTEE
PATRON

Dr. T.BANUMATHI,
Chair person, AERI Trust, Chennai.

CHAIRMAN

DR. G. RANGANATH,
Principal, Adhiyamaan College of Engineering, Hosur.

CONVENER
DR. S. Suresh Babu,
Head, Department of Civil Engineering & Dean (R&D)
Adhiyamaan College of Engineering, Hosur.

MEMBERS
DR. H. Karibasappa
DR.G.Nandhini Devi
Prof. N. S. Madhu
Mr.R.Srinivasan
Mrs.V.Priya
Mrs.S.S.Revathy
Mr.R.Chinnasakkan
Mr.B.M.Purushothaman
Mr.V.S.Satheesh
Mr.R.Thirumalai
Mr.M.Manikandan
Mr.R.Yuvarajasekaran
Mr.P.Satheesh Kumar
Ms.V.A Narmadha

CONFERENCE SECRETARIAT
Mr.S.Gobinath

DR. L. Yesodha
Prof. A.Krishna Mohana Reddy
DR. M. S.Vinaya
Mrs.K.J.Jegidha
Mr.G.Saravanan
Mrs.M.Mahalakshmi
Mr.S. Roopan Kumar
Mr.R.Venkob Rao
Mr.S.Sethuraman
Mr.S.Veera Ragavan
Mr. Sadat Ali Khan
Mr.S.Nirmalraj
Mr.V.Visakh

PREFACE
The conference entitled Infrastructure Development for Environmental Conservation
& Sustenance, is being held at our institution organized by research department of
civil

engineering

aims

to

bring

together

practicing

engineers,

academicians,researchers,industries, architects, and all other alike quarters of the


sustenance practices for the safe habitat management, under challenging environments.
The main theme of this conference is to focus on the diversified facets of civil
engineering practices converging to the Infrastructure Development practices and
technology as the broader domain, with more treatise on habitat sustenance practices.
Eminent authorities have been invited to deliver keynote lectures on different aspects
on

the

conference

themes.

In

all,

there

are

sixty

six

(66) contribution papers including ten (10) Keynote Lectures. The written contribution
for the keynote speakers and authors have been included in the conference proceedings.
Keeping the objective of the conference, the contributions from the authors have been
under the following domains:
a) Geospatial technology for sustainable & built-in environment,
b) Structural Health Monitoring & eco-friendly practices.
The conference organizing committee thanks all the members of advisory committee,
reviewers of the technical abstracts and host of volunteers who are responsible for this
event.

DR. S. Suresh Babu


Convener-INDECS-15

DR. G. Ranganath
Conference Chair-INDECS-15

FOREWORD
It is pleasure to write a foreword to the proceedings of the international conference
INDECS organized by the department of civil engineering of our institution. This
unique conference provides a great opportunity to practicing engineers, academicians
and research connoisseurs, architects to showcase their competence and contributions
on the subject and also provides a platform for the meaningful deliberations, besides
promoting the development of contacts with a like researches across the world who
have contributed their research findings for the INDECS-15.
I appreciate the department of civil engineering of our institution, for having
volunteered in this endeavor, to intake the task of bringing closer interactions amongst
the academia and researchers through this INDECS-15, to have concretized but humble
beginning towards supplementing the unified approach for facing the unprecedented
challenges and threats. I appreciate the experts from the Home land and abroad for
their consent to deliver keynote lectures, authors of contribution papers and reviewers
of the papers. I also thank the sponsors, conference advisory committee, conference
organizing committee and host of volunteers who are responsible in successfully
organizing this event.

Dr.Mrs.Banumathi .T
Chairperson, AERI Trust

CONTENTS
Keynote Speakers

Sustainability Outlook 2030: A Case Study of Israel,


Valerie Brachya,Shlomo Hasson, Eran Feitelson

1-15

Coal and Sustainable Development


John L. Daniels

16-20

Early warning systems for minimizing loss of life and damage to Infrastructure for
Sustainable Development,
Tad.S. Murty and Abdulraheem Masoud

21-34

Assessing Sustainability of Ground Improvement Methods: Quantitative Triple Bottom


Line Framework and Case Study
Krishna R. Reddy and Rajiv K. Giri

35-44

Transport of Heavy Metal Ions Through Soils


P V Sivapullaiah

45-51

Geosynthetics for Ground Improvement


Sushovan Dutta, Maheboobsab B. Nadaf , B. S. Asha and J. N. Mandal

52-61

Seismic Behavior of RC Frames with various Masonry In-Fill Configurations


S Suresh Babu, Sidramappa V. Itti, R Srinivasan

62-68

Compressibility and Chemical Compatibility of Clayey Soil-Bentonite Backfills for


Slurry-Trench Cutoff walls
Y. J. Du , R. D. Fan , S. Y. Liu , Krishna R. Reddy

69-78

Stability Aspects Performance of Non Perforated vs Perforated Quarter Circle


Breakwaters A Study
Arkal Vittal Hegde

79-85

Role of Structural Engineers in Sustainable Construction

86-91

Ranjith Dissanayake and Chaminda Bandara

Contribution Papers

Shortterm shoreline changes along Karnataka coast, India: A Geospatial


Approach,
Ateeth Shetty, K.S. Jayappa, Praveen G. Deshbhandari, A.S. Rajawat, Ratheesh
Ramakrishnan

92-100

Traffic congestion study for Coimbatore district


Using GIS Anitha Selvasofia.S.D , Prince Arulraj.G

101-105

SCS-CN and GIS based approach for estimation of surface runoff in Sita-Swarna
basin, Udupi district
Anil, B, Shivanna, M.S.Vinaya

106-112

Prioritization of Erosion Potential Area through Morphometric Analysis in


Venkatapura Watershed, Karnataka: A RS and GIS Approach,
Praveen G. Deshbhandari , C. Krishnaiah

113-120

Evaluation of Water Quality Index For Surface and


Area of Bangalore
Prasad C S M V , Dr. Inayathulla M

121-128

Hydrologic Analysis of Watershed Using


IDF Curves & Runoff Models
M Inayathulla , M. Raghunath, C. Rangaraj& Y A Narayanaswamy

129-141

Groundwater Prospectus Map for Kolar Taluk Subwatersheds,


M Inayathulla Y A Narayanaswamy, Shashishankar.A & Chalapathi k

142-145

Environmental Conservation Assessment Used by GIS and Remote Sensing Bharamasagar at Chithradurga Dist, Karanataka,
V. Praveena Kumara, Bgayalakshmi, Lingananda M V.Sivani Shobha udaya

146-151

Subsurface Waters In Urban

Performance Of Geopolymer Concrete In Acidic Environment,

152-155

Shankar H. Sanni , Dr. R. B. Khadiranaikar

Light Weight Flyash Aggregate Concrete - An Environmental Friendly


Approach,
Geena George , Dr. Asha.k

156-161

Experimental Investigation On Behaviour Of Steel Concrete Composite


Columns, R.Thirumalai

162-166

Effect Of Fibre Reinforcement On Strength And Stiffness Improvement Of


Cohesive Soil , Suchit Kumar Patel, Baleshwar Singh

167-173

Behavior Of Circular Footing Resting On Reinforced Sand,


Sangameshwar Patil, P.G Rakaraddi

174-184

Seismic evaluation and retrofitting of reinforced concrete buildings,


Dr.A.Vijayakumar Dr.D.L.venkatesh babu, Dr.A.R.santhakumar

185-193

Construction Accidents Indian Scenario,


D.Kamalnataraj , Dr. V. Ramasamy

194-199

Ecological Behaviour Of Washbowl From Discarded Printed Circuit Board,


A.G.Ganeshkumar, Dr.G.Ranganath, Dr Channankaiah

200-208

Sustainable transporation planning is key to emissions and energy conservation ,


P.Ponnurangam

209-218

Slope stability analysis by finite element approach,


Barnali Ghosh , S. K. Prasad

219-223

Simulation modeling of sustainable infrastructure development in PERI urban


areas - a case study of Chennai,
Abirami krishna. A, Devi priyadarisini. K, Dr.Umadevi. G

224-233

Reliability Analysis of Post Tensioned PSC Girders Designed As Per IS


1343:2012
Bharath B.M. , Dr. K. Manjunath Puja M. Singh

234-237

Effect Of Number Of Loading Cycles On Dynamic Properties Of Solani River


Sand, J. Chavda1, B. K. Maheshwari, G. R. Dodagoudar

238-242

Concrete Performance With Admixtures Of Fly Ash And Copper Slag


Concerning Mechanical and Durability Aspects For Normal Strength,
Velumani.M , Nirmalkumar.K Sathish.S

243-251

Experimental study on stress-strain behaviour of masonry prism with different


types of brick and mortar,
Dr. K.vidhya , Dr. S. Kandasamy

252-257

Experimental investigation of the combined load response of model piles driven


in sand,
Prasad Pujar , P.G Rakaraddi

258-266

Non-destructive testing of thin elements using impact echo method,


S.Christi , K.Sivasubramanian , V.Venkatesan , K.N.Lakshmikandhan ,
P.Sivakumar

267-271

A Review On Characterization And Application Of Fly Ash Zeolites,


Channabasavaraj.W

272-279

De-fluoridation using tea residue,


Priya.V,Dr.Samathkumar.M.C, Dr..Balasubramanya.N

280-284

Solubilisation Of Metals From E-Waste Using Penicillium Chrysogenum Under


Optimum Conditions,
J. Senophiyah mary , Dr. T. Meenambal

285-293

Behavior Of Single Pile In Horizontal And Sloping Ground Under Combined


Loading,
Jyoti S Patil , P G Rakaraddi

294-307

Experimental investigation on strength and fire resistance of high performance


concrete,
K.Athi Gajendran, Dr.R.Anuradha, Dr.G.S.Venkatasubramani

308-317

An experimental study on concrete using selected waste materials as partial


replacement of fine aggregate,
Mr. A.Arunraj

318-322

Influence of clay minerals on bituminous mix,


Mrs.M.Mahalakshmi, Dr.S.Suresh babu , Dr.K.E.prakash, Mr.S.Vimal

323-326

Performance Of Masonry Wall Composites With Natural Fibers,


S.Deepak , R.Anuradha and G. Prince arulraj

327-332

Flexural Behavior Of Reinforced Concrete Beam Using Coconut Shell And Fly
Ash,

333-341

V.Divya

Investigation On Conductive Concrete Compositions,


M.Amirthavarshini , R.Selvaraj and N.Chella Kavitha

342-346

Durability Studies On Raw And Calcined Clay Based Geopolymer Concrete,


R.Priyanka and, R.Selvaraj

347-354

Rapid Chloride Permeability Test On granite Powder Concrete,


A.Arivumangai and Dr. T. Felixkala

355-358

Mechanical And Durability Properties Of Concrete Using M-Sand As Fine


Aggregate, S. MuraliKrishna & Dr.T.Felix Kala

359-365

Experimental Investigation Of Hpc Using Metakaolin And Flyash With Partial


Replacement Of Cement,
M.Narmatha , Dr.T.Felixkal

366-371

Feasibility Study Of Coir Fiber With Coir Pith, As Concrete Reinforcement


Material,
Priya.V, Dr. Sampath kumar.M.C , Dr.Balasubramanya.N

372-375

A Study On Replacement Of Natural Sand By Blends Of Flyash And Bottom


Ash In Concrete, Sadat Ali Khan , S Gobinath Veeraragavan

376-384

A study on Glass Fiber Reinforced concrete deep beams with direct loading,
Vengatachalapathy.V1, Ilangovan.R2, A.R.Santhakumar3

385-392

GEMBA KAIZEN- A New Management Tool for Construction Industry


Jagadeesha Kumar B G 1, Rishi Rai 2 Souvik Chakraborty

393-395

Abstracts

396-411

SUSTAINABILITY OUTLOOK 2030: A CASE STUDY OF ISRAEL


Valerie Brachya1,Shlomo Hasson2, Eran Feitelson3
1.

Jerusalem Institute for Israel Studies and lecturer at the Hebrew University
Jerusalem and Tel Aviv University, Israel
Valerie@jiis.org.il
2. Professor Emeritus and Head of the Shasha Center for Strategic Studies,
Hebrew University, Jerusalem, Israel
mshasson@mscc.huji.ac.il
3. Professor of Geography and Head of the School for Environmental Studies,
Hebrew University, Jerusalem, Israel
msfeitel@mscc.huji.ac.il

Abstract
Although the future is highly uncertain, policy makers need forward looking long term strategic
outlooks in order to be aware of possible situations and to enhance opportunities and minimize risks.
Scenario building provides a way of thinking about plausible futures and their consequences. Outlooks
and scenarios have been used at the global level but less at the national level of policy making, which
often responds to crises rather than acts to fulfill a vision. Scenarios prepared in Israel for mapping a
strategic route to reach a vision for sustainability towards 2030 helped to identify the strategies and
policies towards the triad of environment, wellbeing and resilience. They included integrated
governance, risk management, urban vitality and sustainable patterns of consumption.
Keywords: Scenarios, Outlooks, Sustainability, Strategies, Resilience
Introduction
Sometimes, we wait too long for evidence to
justify action
Sometimes, we act too quickly on new
opportunities and oversee spill-over effects.
Often, we approach the future on the basis of
our experiences in the past: policies function in
the short, but not in the long run[ 1]
Policy in all areas of governmental
responsibility commonly focuses on providing
urgent responses to unanticipated changes or
coping with a crisis situation in one field or
another. The establishment of an investigating
committee, for example, to review why a crisis
occurred is a valid and understandable
mechanism for coping with an unanticipated
result[ 2]. However, it would be better if such a
review were undertaken in advance as a
forward looking process to identify and plan
how such risk situations could be avoided. The
task is then how to enable policy makers to be
aware of what could happen, recognize
desirable paths and avoid undesirable ones,

create and harness opportunities and minimize


risks and vulnerabilities.
Looking ahead to the long term future is full of
uncertainties which cannot necessarily be
predicted by models or anticipated accurately
by forecasts. Sudden processes such as
economic crises, social instability, political
changes or extreme events (earthquakes, fires,
floods and so on) can cause dramatic
unforeseen changes. Other processes are slow
but steady and their trends can be foreseen.
Looking at the long term future can put on the
table a range of possible futures with a high
probability and the paths that lead towards
them. It can also identify desired futures or a
vision of the future and the paths which lead
back from them to current conditions,
backcasting rather than forecasting.
Preparation of a long term Overview or
Outlook can provide a framework for strategic
directions. It leaves the immediate decisions up
to the current operating administration, but it
can provide well defined paths along which
Page 1

key intervention points and conscious choices


can be made and concrete actions taken to
promote a desired pathway towards
sustainability and avoid an unsustainable
pathway. It may also indicate how the choice
of mechanisms for action can be assessed
through evaluating the efficiency and
effectiveness of policy packages comprising of
complimentary and supporting implementation
mechanisms
and
avoiding
conflicting,
unacceptable and contradictory mechanisms.
One way of coping with unknown futures is
through scenario building. A scenario is a
story that describes a possible or plausible
future, but it is neither a vision nor a forecast.
It describes what might happen in the future
and the processes that led to it. In this way
scenarios enable policy makers to ponder the
significance of current developments and the
direction in which they are headed. Scenarios
are intended to inspire thinking about models
and processes that are often ignored. This does
not mean that one of the scenarios will
necessarily materialize in whole or in part.
What is important is the act of disengaging
from existing thought processes and acquiring
experience in thinking about the unexpected
and about what appears impossible. In this way
policy makers can cultivate thinking outside
the box, improve the environment for
thinking by challenging existing perspectives,
and examine new opportunities[ 3].
Scenario building has been widely used in a
variety of fields, such as energy, finance,
security, environment and sustainability (Rand,
Shell, World Energy Outlooks, World
Economic Forum scenarios [4]). The widely
acclaimed scientific panel on climate change
IPCC [5] and the UN Environment Program
(UNEP) Global Environment Outlook GEO
[6] used scenario building, to demonstrate the
implications on the environment of global
markets,
security
risks,
technological
developments and the importance of good
public administration. Environmental Outlooks
have been prepared by UNEP on a global
scale, Global Environmental Outlooks (GEO
1,2,3,4,5) and for a group of countries,
members of the OECD (OECD Environmental
Outlooks 2030, 2050 [7]), and for Africa
(UNEP)[8]. Few countries have adopted the
process on a national level, to contribute to
long term thinking in central or local
government.

Use of Scenarios
Scenarios have been developed by a range of
disciplines for a variety of purposes to
understand the processes, trajectories and
driving forces likely to have an impact on the
future, their likely outcomes and to develop
preparedness to cope with their consequences.
There is no single accepted methodology for
the preparation of scenarios [3]. Some
scenarios focus on a well defined problem;
others focus on the relevant actors and the
implications of their possible actions. A
typology of scenarios would include
predictive,
explorative
and
normative
scenarios, each with a different emphasis :
what will happen, what can happen and how to
reach desirable targets. Some scenarios are
prepared through deductive processes,
harnessing available data and expertise. Others
are prepared through inductive processes,
building synergy through dialogue between
different approaches or between opposing
stakeholders. Some scenarios have a strong
quantitative base, others are based on
qualitative assessments, sometimes followed
up with a quasi-quantitative interpretation of
their consequences.
The IPCC scenarios [5] for example focused
on a single issue, global warming, and the
implications of its consequences for climate
change and sea level rise.
Scenarios for climate change were first
presented in 1990 and then in 1992. They
followed a sequential approach as follows:

Storylines or narratives to generate


socio-economic drivers responsible for
greenhouse gas emissions

Emissions scenarios, based on the


storylines

Climate change scenarios based on


the emissions

Environmental scenarios based on the


impacts and consequences of possible climate
changes

Vulnerability scenarios, based on


risks and responses to the impacts
In 2006, IPCC decided to conduct a new
scenario process, normative or anticipatory
rather than exploratory, with a rather different
process:

Generation of 4 alternative future


radiative
forcings
to
2100
(termed
Representative Concentration Pathways
RCP)
Page 2


Socio-economic conditions likely to
generate the RCP's

Climatic conditions likely to be


generated by the RCP's

Impacts, vulnerability and mitigation


The IPCC process was a scientifically based
approach, whose purpose was to arouse global
concern over changing conditions likely to
have crucial consequences for socio-economic
systems.
The UNEP GEO scenarios from GEO3 [6]
onwards used scenarios based on the work of
the Global Scenario Group, particularly led by
the Tellus Institute in Boston and by the
Stockholm Environment Center. The GEO3
scenarios were holistic and explorative, not
problem focused. Their purpose was to show
trends and present storylines, including a
vision for 30 years ahead. The GEO4
scenarios focused on actors, including
political, institutional and cultural aspects, for
a vision to 2050. The GEO5 scenarios were
normative, to identify the gap between
conventional worlds and sustainable worlds
and to propose how to close the gap.
The Millenium Ecosystem Assessment (MA)
scenarios [9] were prepared through
deliberative dialogue with the research and end
user communities. They used existing
scenarios of global driving forces and their
likely consequences but added a stage to
identify those aspects relevant to the future
condition and likely trends of ecosystems.
The UK is an example of the use of scenarios
at a national level. The UK government
commissioned a wide range of scenario
studies, one of which was the Foresight Land
Use scenarios FLUF [10] which were
prepared in the context of a review of future
land use in the UK. They focused attention on
5 main drivers and generated 3 scenarios :
adaption to change, societal and institutional
resistance to change and the spatial
concentration of population.
The Land use project provided a base for the
UK National Ecosystem Assessment (UK
NEA) scenarios [11]. They constituted,
together with the Natural England scenarios, a
look into plausible futures of the natural and
built environment and the consequent impacts
on ecosystem services.
The major drivers were land use change,
resource consumption and climate change.

The team constructed matrices of the condition


and trends in ecosystem services for each
habitat under 6 scenarios. The cumulative
levels of increase and decrease in ecosystem
services were presented in a comparison of
outcomes of the alternative scenarios.
Israel Sustainability Outlook 2030
In 2010 the Ministry of Environmental
Protection in Israel and a think tank, the
Jerusalem Institute for Israel Studies, with the
support of a philanthropic fund, initiated the
preparation of a Sustainability Outlook for
Israel [12]. At its center was a process of
scenario building. The purpose of the 'Israel
Sustainability Outlook 2030' was to bring to
the attention of decision-makers the
implications of present and anticipated socialeconomic-environmental trends, formulate a
realistic vision of sustainability to which Israel
should aspire, identify the gaps between
existing trends and the possibility of realizing
the vision and identify road maps which could
lead towards a more sustainable future
The process of preparing the Outlook consisted
of several stages:
1. the diagnosis of current development
through a systematic mapping of the variables,
identifying the links between them
and
developing an understanding of relationships ,
such as economic growth, social development,
consumption modes , impacts on the
environment, and responses of the State and
society.
2. scenario building, through the deductive
method, which included:
Identifying driving forces by means
of a matrix of influences and
coordinate systems that identify
unknown variables of high influence;
Creating a matrix of scenarios;
drafting the narratives of the
scenarios;
assessing the significance of the
scenarios
3. creation of a vision, beginning with a
positive vision, continuing to a defensible
vision, and concluding with a satisfactory
vision. The positive vision aspires to the
maximum , while the defensible
vision
identifies thresholds that must not be crossed.
As such it is a minimal vision.
4. testing the vision
in relation to the
scenarios, to locate opportunities and
difficulties in its realization
Page 3

5. formulating strategies to move towards the


desirable vision, whatever the scenario which
actually materializes
6. policy packages for implementing all the
strategies in the most effective and acceptable
way to all relevant stakeholders
The process began with the formulation of a
research question, which locates the projects
center of gravity. The question formulated
was: What are the factors that shape the
environment and human wellbeing, as well
as their resilience?
This focus is a little different from the
sustainability triad that usually consists of
environment, society and economy. The
Outlook 2030 team put wellbeing at the center
of the discussion, while economic activity was
defined as being the most significant element
that
influences
the
achievement
of
sustainability but not its main goal. Other
driving forces affecting sustainability are
social, geopolitical, cultural and behavioral,
and issues relating to governance.

Israels population is expected to


grow to 11 million by 2030, both as a
result of natural growth and as a result
of expected rise in life expectancy.
The direct significance of these
demographic trends and the increase
in 1 person households is an increase
in the demand for natural resources,
consumer products and services.

The Israeli economy grew in the last


decade by 20% and since the 1990s
by 75% and has proven resistant to
crises. If there is no change in current
trends, the Israeli economy is
expected to develop in line with
global markets. A rise in per capita
GDP is expected to be reflected in a
rise in disposable income and an
increase in purchasing power and the
intensification of the impacts of
household consumption on the
environment.

The goal of strengthening social and


environmental resilience was added to the
sustainability triad because it constitutes a
necessary and complementary variable in a
world of high levels of risk and uncertainty.
An increase of resilience would be
characterized by boosting the capacity to
contend with crisis situations, identifying the
extent and distribution of risk, reducing the
intensity of anticipated damages, increasing the
capacity of a community to cope with risk and
crisis situations and use them as opportunities
for positive change.

The Israeli economy transformed


from local production to foreign trade,
which currently constitutes 61% of
GDP. This trend can be expected to
continue in the future. The transition
to foreign trade raises the level of
Israels dependence on the import of
food products and raw materials for
industry and therefore intensifies its
exposure to risks.

The general rise of GDP does not


reflect polarization between different
social strata. The Gini index and the
per capita disposable income level
reveal widening of gaps within Israeli
society. The disparity in per capita
disposable
income
(excluding
expenditure on food) is increasing
between the top and bottom deciles,
with a small improvement in the level
of the lowest decile.

The increase in disposable income is


expected to lead to a rise in per capita
consumption. Pressures on the
environment can be attributed to the
consumption patterns of the upper
socioeconomic strata. Emulation of a
similar lifestyle by lower deciles as
disposable incomes rise will improve
standards of living but will
significantly increase pressures on the
environment.

Stage 1 - Diagnosis of past, present and


future trends
The trends observed in Israel over the last 20
years indicate that a "business as usual"
scenario, based on the continuation of present
policies and trends, would not lead to a
sustainable environment or to wellbeing or to
their resilience. Current environmental policy
is likely to bring about improved
environmental performance and the abatement
of negative impacts of economic activity on
the environment; but trends of population
growth and rising standards of living are
generating ever-increasing pressures on the
environment and intensifying the depletion of
natural resources and the rate of environmental
degradation. The following were found to be
highly significant:

Page 4

Current development trends are


generating sprawl and the loss of open
landscape
in
contradiction
to
increasing the efficient use of land as
well as to reducing energy use by
reducing distances travelled.

An overview of past and present trends


indicate on the one hand, an expected
improvement in the management of hazards
and pollutants, but on the other hand, signals
an expected deterioration of the environment
as a result of increasing pressures of
consumption.
Interpretation of the trends
As part of the diagnosis stage, experts from a
range of disciplines were requested to respond
to a series of questions concerning the
identification of crucial issues and likely
processes which will affect the future:
1.
What are the main environmental
issues today in Israel in your field of expertise?
To what extent are they similar or different
from the main issues in the global dialogue?
2.
What are the recent past and present
processes affecting your area of expertise
which impact on the environment in Israel?
Can they be predicted with a relatively high
level of probability?
3.
What factors determine
these
processes? Can they be forecasted or are they
incapable of prediction?
4.
What global and regional geoenvironmental processes are likely to
influence the context for sustainability within
Israel (such as water stress and droughts)
5.
What global and regional geopolitical processes are likely to influence the
context for sustainability in Israel, such as the
increased influence of China and India and the
possible decrease of the United States and
Europe and possible developments in the
Middle East.
6.
What are likely to be the influences of
the above processes on the environment
(problems,
opportunities,
vulnerabilities,
coping capacity, resilience)? Can they be
controlled through governmental intervention
in Israel?
7.
How are the above processes likely
to influence wellbeing in Israel?
8.
Does current environmental policy in
Israel cope with the problems , realize the
potential opportunities to reduce risks and
vulnerabilities and to strengthen coping
capacity with change (resilience)?

9.
What
are
the
likely
future
developments in your field? Identify possible
trends and assess their probability (best and
worst possibilities) in relation to likely trends
and processes and in relation to intensive but
unpredictable events,
10.
What is needed in order to make best
use of opportunities for change and to
strengthen coping capacity and resilience to
certain and unpredictable changes?
11.
What steps should be taken to reduce
risks and vulnerability and to overcome their
adverse consequences?
Below are the key messages that emerge from
each expert opinion and are relevant to the
formulation of a business as usual scenario
and the alternative scenarios.
Key economic messages:
Privatization of the economy has resulted in
separation between the regulator, on the one
hand, and the producer and employer on the
other, thus enabling government, the state as
regulator to set environmental controls and to
promote the inclusion of social and
environmental liability in business. The
transition from production of goods to the
provision of services is likely to reduce
pressures on the local environment generated
by industrial production Given the dependence
on global market developments, the
government will need to adopt measures
designed to promote innovation, including
cleantech industries and especially those
areas in which Israel has a competitive
advantage.
Recommendations for the future:

Making Israel a leader in innovation


will require extensive investment in education,
research
and
development,
training
institutions, and promotion of technological
and non-technological innovation.

The fields of green growth, cleantech,


efficiency, and reduction (of raw materials,
energy, water, land-use, and waste) are
potential avenues for market leadership.

Israel could find itself in competition


with global markets and exposed to risks
related to imports and exports.

If Israels foreign debt increases, it


will be inherited by future generations.

Under a scenario in which Israel does


not take measures to encourage innovation, it
could lose its place in global markets and
experience a decline in GDP per capita.
Page 5

Key energy messages:


Israel was dependent on fuel imports and
suffered the consequences of being an
electricity island, disconnected from
international networks. Coal was selected as a
strategic source for energy production, and
measures were taken to reduce environmental
consequences in accordance with accepted
practices at the time; the production,
transmission, and distribution of electricity are
carried out by a government company (the
electricity company), whose goal is ensuring
the reliable supply of electricity.
The recent inclusion of natural gas in the
energy
market
significantly
reduces
dependence on fuel imports and paves the way
for private electricity producers, both
constituting substantive structural changes for
Israel. To date, the scope of renewable energy
development has been very limited; a
significant reduction in air pollution from the
burning of fuel for energy production has been
achieved, but it is still unclear how Israel will
meet its obligations to reduce greenhouse gas
emissions.
Recommendations for the future:

Maintenance of a long-term reserve of


natural gas for Israel by limiting its export;

Import of liquid natural gas (LNG),


its storage, and construction of a regasification facility at sea (stationary or mobile
container);

Increase the scope of electricity


production from renewable energy

Construction of a smart bi-directional


electricity network to accommodate heavy
demands, including electric motor vehicles;

Use of natural gas for transportation


(CNG, methanol);

Key messages for land and the use of space


Land is perceived more as a means of
production and less as a resource for the
achievement of national objectives. However,
its price does not include external costs such as
payment for infrastructures or environmental
damage. The lack of full pricing of land
facilitates a trend towards the expansion of
built-up areas at the expense of open space and
does not promote the renewal of existing urban
areas. Urban renewal will only occur when it is
considered economic. In the absence of
urban renewal, residential, commercial and

industrial development will spread to space


outside the cities.
Recommendations for the future:

Strengthen cities and urban life,


implement urban renewal project
and implement measures to
restrict urban sprawl;
Use underground space
Require infrastructures to pay for
their use of land,
Integrate infrastructures into shared
spaces or corridors, including
underground tunnels within cities.

Key messages for water


The provision of water was historically
perceived as a strategic foundation for
achievement of national objectives; the water
supply network was established and managed
at the national level.
The recent master plan for the water system
approved in July 2011 recognizes that the
water system has to operate in conditions of
uncertainty, such as a decrease in natural water
sources and requires water supply for urban
and industrial consumers to be determined by
market allocation Plans to double the potential
water supply will be achieved through
desalination and more wastewater for
agriculture.

Key messages for ecosystems and ecosystem


services
Human development has resulted in the
transformation of natural ecosystems into
pasture, afforestation, agricultural production,
artificial bodies of water, and urban areas.
Whilst these activities have benefitted the
population (for example, by providing food)
they have weakened other functions (such as
natural protection against flooding). Biological
diversity in Israel has enjoyed relatively good
protection.
Local ecosystems today are unable to meet the
needs of the population. Israel is increasingly
dependent on imported biological goods
(grains, beef, timber lumber) from dry
ecosystems. Development in Israel has
decreased, fragmented, and contaminated
natural ecosystems, particularly
forests,
scrubland, freshwater systems, and coastal
areas. The loss of these systems as control
mechanisms exposes Israel to possible future
Page 6

extreme events. Cultural services (landscape


and heritage) have generated great value in
terms of tourism and recreation.
Global warming poses a threat to Israels
ecosystems, including invasive species,
frequency and intensity of floods, water loss,
soil erosion and fires.
Recommendations for the future:

Cancel water subsidies for agriculture


and return abandoned lands to their natural
ecosystems
Promote agriculture which is not
dependent on ground resources
(such as greenhouse cultivation)
able to absorb higher water costs

Give priority in land use to


ecosystems that are important in providing
control mechanisms;

Strengthen resistance to climate


change by preserving areas of sharp climatic
transition.
Linkages and relationships between the
variables
The wide range of driving forces identified by
the indicators and the experts provided the
basis for creating a concept map (see fig. 1).

Assessment of the variables and the


relationships between and among
them
provided the basis for the Influence Diagram
(see fig 2) . The team identified active factors
that mainly influence and, to a lesser degree,
are subject to influence, and critical factors
that both influence and are influenced.

Stage 2 Building scenarios


There is no context in Israel for scenario
building as was created by the Foresight Unit
in the UK. The Israel 2030 scenarios are a
stand alone project created by the project team
and invited experts. The process did not
include public participation or consultation.
Before developing scenarios for the future, a
scenario of the past was constructed (19482011) described in terms of the system
variables. Continuation of present trends of the
variables was the basis for the business as
usual scenario and variations in the variables
generated a further 6 scenarios.

The scenarios were classified into four groups


by the degree of dominance of various
mechanisms. One group is dominated by the
market mechanism; in another group the social
mechanism is dominant; the third group
includes scenarios where the mechanism of
state intervention is dominant; and in the
fourth group it is the geopolitical mechanism
that is dominant.
The result was 7 scenarios , which were
expanded by descriptive storylines/narratives.
Three scenarios presume a neoliberal ideology
and market superiority
The "business as usual" scenario
market dominance: An emphasis on
markets and growth; continued
population growth and current
consumption patterns; reaction to
short-term problems; development of
a city-state; absence of an integrated
vision; focus on local conflicts; the
Israeli-Arab conflict continues to play
a central role on the agenda;
degradation and fragmentation of
ecological systems (in fact, marginal
attention is given to the environment).
Unregulated market scenario A
strong, powerful market along with
deepening
degradation
of the
environment and increasing social
discontent neoliberalism is the
reigning ideology; absence of
environmental and social regulation
and the development of a city-state
dominated by Tel Aviv; growing
social disparities adversely affecting
young people and the middle-class;
privatization of state land.
Regulated market scenario A strong,
powerful market; growing attention to
the environment but with a splintered
and polarized society; promoting
growth; creation of a bureaucraticinstitutional mechanism; integration
between economy and environment;
absence of social inclusion; restraint
on the concentration of wealth.
Grassroots
initiative
scenario,
social
mechanism dominant
Community mosaic Strengthening
of social resilience and reduction of
environmental degradation but a
significant drop in economic growth
emphasis placed on inclusion
(integration of all elements of
Page 7

society); promotion of grassroots


processes; promotion of social
businesses, new entrepreneurship
networks and organizational and
community structures; emphasis on
local economy.
State as developer scenario, state mechanism
dominant
Intervening/developing
state

Adoption of a post-neoliberal
approach
with
social
and
environmental emphases, government
intervention in market processes as
a result of social unrest and outburst
of waves of protest throughout the
country which led to the rehabilitation
and intervention of the bureaucraticinstitutional system; drop in economic
growth.
Two scenarios were developed in which the
geopolitical mechanism is dominant
Unregulated
fortress
state

Environmental, economic and social


deterioration caused by political
seclusion worsening of geopolitical
status along with boycott and
sanctions; cessation of foreign
investments; ongoing environmental
neglect; increase of environmental
risks; growing reliance on local
natural resources.
Regulated fortress state Strong
regulation
following
political
seclusion, drop in economic growth
while protecting the environment and
building social resilience worsening
of geopolitical status along with
boycott and sanctions; cessation of
foreign investments; government
intervention;
development
of
innovation in defense industries;
establishment of inclusion and
regulation mechanisms in the social
and environmental areas; protection
of reserves and effective use of
resources;
protection of
open
landscape areas for security reasons.

aspects but regulation could mitigate the


severity of the damage. An analysis indicates
that the level of sustainability in each scenario
is far from the desirable situation and that we
should be able to reach a higher level of
sustainability whatever the scenario.

The different scenarios have different levels of


sustainability (environment, wellbeing and
resilience). A scenario from the "market
superiority" group can reach a relatively high
level of sustainability only in a "regulated
market." The social scenarios would reinforce
wellbeing and the environment but their
economic price would be high. In the
geopolitical situation of a fortress state,
sustainability would be compromised from all

The understanding that the "business as usual"


scenario will not lead to the vision for 2030
and that incremental improvements of existing
systems, beneficial as they may be, will only
perpetuate the existing situation without
leading to the desired long-term outcome, led
to the conclusion that deep strategic
intervention is needed to bridge the gaps that
exist between each of the scenarios and the
vision.

Stage 3 a vision of sustainability for Israel


2030
Each member of the project team prepared his
own vision; they were combined into a
commonly acceptable vision for sustainability
in Israel . The vision consisted of a summary
statement followed by clauses detailing
important themes.
"Israel in 2030 will be a country
whose citizens live in an environment
that provides economic wellbeing,
social resilience and personal security
while enabling a diverse range of
community lifestyles. It will be a
country that promotes innovation and
enterprise, thriving urban life,
inclusion and access for all of the
population
to
employment
opportunities and services. It will be a
country where there is absolute
decoupling of economic growth from
deterioration of the environment and
the continual rise in material
consumption. In 2030 the quality of
life in Israel of the current generation
will be high but will include
responsibility for protecting natural
resources for the present and future
generations."
The vision was elaborated to address diverse
aspects including: economy, society, urbanism,
infrastructure,
water,
energy,
ecology,
governance and international relations.

Stages 4 and 5 - long-term strategic


directions

Page 8

The team developed nine robust strategies


relevant to a large number of the scenarios that
will enable the realization of the vision by
bridging the gaps between the scenarios and
the vision. Outlook 2030 recommends
implementing
all
of
the
strategies
simultaneously.
a) Promoting innovation and enterprise:
Stimulating innovation leading to
more efficient resource use, product
development, the development of a
circular economy, and a transition
from the consumption of products to
the consumption of services.
b) Integrated risk management and
resilience: Risks must first be
identified, defined and evaluated.
Risk management requires not only
measures to reduce risks as far as
possible but also measures to increase
the resilience of society and the
environment to coping with the
residual risks which cannot be
mitigated
c) Developing a long-term integrated
vision: Creating long-term systemic
thinking, periodically updated, by
reinforcing the connection and
coordination
between
proactive
planning and regulation. According to
this strategy a joint lexicon of
terminology and language should be
created for the public administration
to delineate a long-term development
policy incorporating the economy,
environment and society.
d) Cultivating values beyond material
consumption: Generating a deep
psychological
transformation
in
consumer behavior in order to reduce
the consumption of material resources
and replace it with other, less
material, patterns of consumption,
such as sharing and collaborative
consumption (this is a term now well
known by those familiar with
behavioral patterns of consumption)
e) Cultivating
community
life:
Cultivating a new value system based
on solidarity, mutual support,
community activity, intercultural
encounters and the creation of new
economic opportunities.
f) Urbanity and the revival of city
centers: Concentrating development
in existing urban centers while
increasing their density and social

diversity to create an urban


renaissance. Developing urbanism
and creating thriving public spaces
allowing
random,
spontaneous
encounters
that
lead
to
interdisciplinary,
cross-cultural
connections and networks that create
synergy between diverse ideas and
outlooks and provide fertile ground
for innovation.
g) Expanding the concept of security to
include the environment, society and
economy: Preparation for civilian
crises that could harm national
security and harm the society and
economy may be just as significant as
preparation for terrorist attacks or
conflicts with hostile countries.
h) Development of leadership with a
sense of responsibility for future
generations:
Development
and
training of a leadership that will lead
society to the vision of sustainability.
i) Adoption of evaluation methods and
systems: Establishing evaluation
systems to measure sustainability, to
facilitate the identification of critical
points and benchmarks that can help
policymakers plan courses of action
and evaluate their results.
Based on the above, and with an understanding
of the current and future trends and processes
in Israel and around the world, a further five
strategies were proposed:
1.

Sustainable
consumption.
Consumption patterns drive economic
growth and are currently taxing
environmental resources beyond
reasonable ecological boundaries. In
order to change the lifestyle of the
public as a whole and adopt less
material consumption patterns that
would still provide a high level of
wellbeing, this strategy recommends
promoting consumption models that
provide the consumer with access to
desirable benefits or values without
requiring individual purchase of
material products (service provision).
Instead, they would facilitate effective
use of assets, products, services and
skills that are in a state of
"underutilization" during most hours
of the day or year (sharing
consumption). To promote models of
"sustainable consumption" it is
Page 9

2.

3.

necessary to remove barriers on the


one hand, and develop missing tools
on the other, while maximizing the
use of social networking platforms to
make information accessible and
create interpersonal connectivity.
Models of sustainable consumption
already exist in the areas of travel,
clothing,
household
equipment,
children's
equipment
and
entertainment.
This
strategy
recommends expanding them to the
areas of time, space and skills in the
public and private spheres. The
development
of
"sustainable
consumption" largely depends on
business sector initiatives and civil
societys awareness level. But the
government can support the transition
to "sustainable consumption" by
educating for social values, providing
access
to
information
and
underutilized public spaces, and
building trust between suppliers and
users.
Vibrant urbanism. Urban living
generates opportunities to reduce
pressures on environmental resources,
supports a less material lifestyle,
creates platforms for promoting
innovation and creates diverse and
resilient communities. Urbanism as a
lifestyle in public spaces and
especially along city streets can be
achieved by mixed land use, reducing
distances for pedestrians and reducing
dependence on and use of private
vehicles. This strategy recommends
making urbanism a budgetary
priority, developing evaluation tools
to assess projects by their contribution
to urbanism and integrating functions
which gather people together, such as
higher education campuses and open
air markets. It recommends that the
business
sector
choose
urban
locations (as opposed to selecting out
of
town
locations),
promote
alternative means of worker access
other than by private vehicles and
develop attractive urban real estate
models, especially for the young and
elderly populations [13].
Strengthening
social-community
resilience. In a world of uncertainty it
is important to strengthen the ability
to adjust and contend with changes
and risks, turning them into new

4.

5.

opportunities rather than seeing them


only as disturbances. The purpose of
this strategy is to promote the creation
of communities that enjoy a high
quality of life and wellness while
realizing the individual and the
community's potential. It recommends
strengthening resilience at the local
level, transferring powers to local
authorities and to the community,
identifying vulnerable populations
and finding ways to reduce their
vulnerability,
promoting
active
community initiatives and developing
tools
to
evaluate
community
resilience.
Integrated management of marketable
natural resources (natural capital).
Without a policy for the management
of imported natural resources as well
as integrated management of its local
natural capital, Israel might find itself
unprepared for the competition over
export and import of raw materials
and manufactured goods in global
markets. This strategy recommends
taking
measures
to
manage
marketable natural capital for the
economy, similar to the management
of financial capital and human capital
(labor). They include an integrated
management policy for all marketable
natural resources and the development
of a national plan to combine
physical,
economic
and
environmental aspects of their
development. It also recommends a
review of the criteria for setting
royalties to reflect public values and
externalities generated by resource
extraction.
Innovation
for
sustainability.
Economic trends indicate that Israel
has to find ways that enable it to stand
at the forefront of environmental,
technological
and
businessorganizational innovation in order to
compete in global markets. This
strategy recommends recognizing the
fact that innovation carries risks and
that opportunities for innovation
should not be blocked due to claims
that their effectiveness has not yet
been proven. It recommends that the
business sector in Israel take an active
role in international forums that
discuss innovation for sustainability,
create research and development
Page 10

partnerships between companies from


different countries and be willing to
try to implement innovative ideas in
Israel. It also proposes that the
government initiate an evaluation of
the drivers of innovation science,
skilled human resources, financial
drivers and regulation and recruit
civil society as a partner to support
and review the testing of innovation.
Stage 6 Implementing the strategies
A wide variety of policy tools will be needed
for the implementation of the nine robust
strategies and the five further strategies.
Outlook 2030 proposed that the tools and
means for implementation are bundled into
"packages," in order to create synergy between
them, identify the preconditions necessary to
enable their implementation and add
complementary tools which could overcome
reservations about them, and make the policy
tools effective and efficient, socially and
politically acceptable and award them a high
likelihood of realization [14]. The following
three policy tool packages were proposed:
1. The whole of government
package. The purpose of this package
is primarily to ensure coordination
between all the bodies responsible for
a comprehensive long-term overview
of the economy and society. The
package proposes the establishment of
a government integration body that
would include representatives of the
following bodies: the Finance
Ministrys budget department, the
economic council in the Prime
Minister's office, the planning
administrations in the Interior
Ministry and the Ministry of
Environmental
Protection.
The
purpose of establishing the integrating
body is to initiate and promote
coordinated plans for action at the
national level.
2. The "risk management" package.
The purpose of this package is to
expand
activities
relating
to
identifying and evaluating risks and
building resilience to them. The
activity will take place on three
levels: 1. Establishing a unit for the
coordination and management of
environmental risks which will focus
on risks that result from incremental
processes as well as risks that result
from sudden environmental events; 2.

3.

Requiring the financial sector to


include environmental considerations
when making decisions on financial
investments; 3. Expanding the focus
on risks and resilience-building within
local
government.
Additional
measures here include leadership
development and strengthening local
government
with
a
particular
emphasis on the needs of weak local
authorities in the periphery.
The
"community-business"
package. The common denominator
between the measures that constitute
the third package is that they are
meant to advance non-governmental
activities: local-community socialenvironmental activity and business
activity. In the community component
the critical measures are providing
access to credit and a budget for
community activity and lowering
barriers to enable the use of public
assets. Complementary measures to
encourage community activity would
be decentralizing powers to the
community level, developing strong
local leadership and stimulating localregional democracy. The second
component consists of encouraging
business models that emphasize the
transition from the consumption of
products to the provision of services
and the transition from material
consumption
to
sustainable
consumption.

Stakeholders were identified who would


benefit from and be interested in promoting
each policy package and would be capable of
adapting the required measures during the
process of their implementation. Adjustments
are usually needed as unanticipated difficulties
may arise, the parties involved may act
differently than had been expected or new
parties not identified in advance may express
interest .
Conclusions
The Sustainability Outlook for Israel 2030 was
presented to top level policy makers in many
government ministries, including the Ministry
of Finance, the Ministry of Defense, the
Ministry of Environmental Protection and the
Council for Economic Affairs in the Prime
Ministers Office. Although no formal decision
was taken in government to adopt the
Page 11

recommended strategies and policy packages


for their implementation, in fact many different
actions are in process which promote many of
them. Unfortunately, some government
decisions and actions are not in line with the
strategies and may even worsen current trends.
It is therefore essential to establish a follow up
mechanism and process, which does not
currently exist in the Israeli case example.
An Outlook is a process, not a product. It helps
policy makers break out of the trap of viewing
the future in terms of continuing from the past
with some incremental improvement. Ideally it
should be an ongoing process, whereby reports
are produced at regular intervals updating
baseline information and the scenarios as new
information and directions come to light. By
focusing on the long term agenda, it can enable
fruitful discourses and partnerships to be
created between government, the business
sector and civil society.
The example of Israel could provide a model
for such a process at national level or could be
adapted for creating a similar process at a
Municipal level [15].

5.

6.

7.

8.

9.

10.

11.

12.
References
1. European
Environment
Agency
Bridging Long term Scenario and
Strategy
Analysis
(Blossom)
Presentation and Main Report 2011
2. Parliamentary Committee on the
Water Crisis in Israel, Knesset
2001(in Hebrew)
3. Hasson S. Scenario Building,
background report prepared for
Sustainability Outlook for Israel
2030, Jerusalem Institute for Israel
Studies (in Hebrew)
4. Davies P.K. et al Enhancing Strategic
Planning with Massive Scenario
Generation, Rand Corporation 2007 ,
New Lens Scenarios, Shell 2013,

13.

14.

15.

World Energy Outlook, International


Energy Agency 2014, Strategic
Foresight practice, World Economic
Forum
Intergovernmental Panel on Climate
Change Working Group 3 Special
Report on Emissions Scenarios 2000
United Nations Environment Program
Global Environmental Outlook (GEO
6 is due in 2017)
Organization
for
Economic
Cooperation
and
Development
Environmental Outlook to 2030 2008
and to 2050 The Consequences of
Inaction. 2012
United
Nations
Environmental
Program Africa Environment Outlook
2013
United Nations Millenium Ecological
Assessment Ecosystems and Human
Well-Being 2005
Land Use Futures: making the most
of land in the 21st century UK
Government Office for Science 2010
National Ecological Assessment
Chapter 25 The UK NEA Scenarios:
Development of Storylines and
Analysis of Outcomes
Jerusalem Institute for Israel Studies
Sustainability Outlook for Israel 2030
2012 (in Hebrew with English
Summary
on
project
website
kayamut2030.org)
Jerusalem Institute for Israel Studies
Urban Sustainability (project in
progress, English summary on project
website ukayamut.com)
Feitelson E Background paper on
Policy Packages prepared for
Sustainability Outlook for Israel 2030
(in Hebrew)
For further information on the use of
scenarios
see
Alcamo
J.
Environmental Futures Elsevier 2008

Page 12

Figures:
Fig. 1 Concept Map of the Projects Research Question (Source: Hasson S. for Sustainability in Israel
2030)

Fig. 2 Influence Diagram (Source: Hasson S. for Sustainability in Israel 2030)

Page 13

Fig. 3 Building Scenarios (Source: Hasson S. for Sustainability in Israel 2030)

Page 14

Fig. 4 Risk Management Package(Source: Feitelson E.. for Sustainability in Israel 2030)

Page 15

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

COAL AND SUSTAINABLE DEVELOPMENT


John L. Daniels1
1

Professor and Chair, Department of Civil and Environmental Engineering, University of North
Carolina at Charlotte, Charlotte, North Carolina, 28223, USA, jodaniel@uncc.edu

Abstract
Sustainable development depends upon electrification and infrastructure, and these in turn depend on coal. In
recorded world history, more electricity has been derived from coal than any other fuel source. It remains so and by
2040, coal is forecasted to account for 40% of generated electricity worldwide, ahead of natural gas, nuclear, petroleum,
and all renewable sources. A byproduct of coal combustion is coal ash, the properties of which are useful to civil,
commercial, and residential construction. This manuscript challenges the notion that coal should be discouraged as fuel
source. Instead, it should be leveraged to encourage the continued rise in sustainable development indicators, while also
facilitating the capacity to invest in mitigation strategies, alternative fuels, and research that would yield an engineering
understanding of climate change.
Keywords: Coal, Sustainable Development, Climate Change, Carbon

1. Introduction
Worldwide sustainable development depends on access
to energy. Much of the worlds energy policy is guided
by the premise that the use of coal to generate electricity
is contributing to global climate change and as such
should be reduced as soon as possible. Government
regulation, litigation, academic research, and popular
media all align with this narrative. Advocacy groups
discourage investments in companies which buy, sell, or
otherwise use coal, and seek to block coal-fired power
plant construction in the U.S. and in U.S.-financed
projects in the developing world. The objective of this
paper is to critically evaluate this anti-coal premise in the
context of sustainable development, using considerations
of benefits, costs, and uncertainty to suggest an
alternative perspective.
2. Background
In government, achieving policy objectives typically
requires influencing public opinion, financial incentives,
laws/regulations or combinations thereof. Environmental
policy in particular has been driven primarily by
regulation, the benefits of which are widely appreciated
in terms of clean air, water, and soil that collectively
support human and ecological health. In developed
countries, virtually all measurable contaminant
indicators in air, water, or soil have either been reduced
to levels for which direct exposure yields no measurable
risk or have been contained to specific locations to which
exposure routes are controlled (e.g., risk-based corrective

action at legacy industrial waste sites). For example, the


U.S. Environmental Protection Agency (EPA) uses 30
parts per billion (ppb) of sulfur dioxide (average annual
value) as a federal standard for healthy air. As of 2013,
the EPA estimates that more than 90% of U.S. locations
have levels that are less than 5 ppb, a value that
continues to decrease [1].
The air has never been cleaner. So the next frontier
in environmental policy is focused on regulating carbon
dioxide (CO2) as part of an effort to reduce its presence
in the atmosphere from its current level 0.040% to
0.035% [2]. In the U.S., for example, more regulations
aimed specifically at reducing coal use have been
promulgated in the past six years than in the preceding
century. These regulations cover all aspects of coal
exploration, mining, processing, transport, combustion,
and disposal. While cumulatively significant, many of
these regulations are individually incremental in nature,
with the exception of the EPAs proposed rules entitled
Carbon Pollution Emission Guidelines for Existing
Stationary Sources: Electric Utility Generating Units,
[3] and Standards of Performance for Greenhouse Gas
Emissions From New Stationary Sources: Electric Utility
Generating Units [4]. These two regulations address
current and future coal-fired electric generating units in
the U.S. In terms of current generation, proposed rules
would require nationwide idling of current units to result
in an approximately 30% reduction in CO2 emission. For
future generation, the regulation eliminates coal as a
viable fuel source unless carbon capture technology is
utilized. Specifically, the rule for future generation limits
CO2 emission to a standard of 1,100 lb CO2/MWh (500

* Corresponding

author. Tel.: +17046871219


Fax: +17046871739; E-mail: jodaniel@uncc.edu

Page 16

Author et. al. / INDECS-15

kg CO2/MWh), while current coal fired plants generate


2249 lb CO2/MWh (1021 kg CO2/MWh) on average.
That difference can only be achieved by carbon capture
and storage (CCS) technology, which has not been
viably demonstrated on a commercial scale.

in turn depend on coal. In recorded world history, more


electricity has been derived from coal than any other fuel
source. Figure 1 presents data for the India, the U.S. and
the world for the past forty years.

The foregoing discussion on benefits and costs illustrates


an inherent challenge to sustainable development, what
some authors have referred to as the energy-economyenvironment dilemma [7]. Simply put, we need lowcost energy to meet basic human needs, however the
provision and use of such energy is thought to prevent
these needs from being met. The Intergovernmental
Panel on Climate Change (IPCC) notes that the largest
known human contribution to climate change comes
from the burning of fossil fuels, accounting for 78% of
the total increase in greenhouse gas emissions from
1970-2010 [8]. And yet, sustainable development
depends upon electrification and infrastructure, and these

70
60
50
40
30

India
United States
World

20
10
0
1975

1985

1995

2005

2015

Figure 1. Percent of electricity derived from coal from


1975-2015; data from the International Energy Agency
[9]
Worldwide, coal use continues to rise, accounting
for more than 40% of electricity generation [9]. In India,
approximately 70% of electricity is generated through
coal combustion. In the U.S., the percent of electricity
generated from coal has decreased significantly in the
past five years because of regulation, anti-coal advocacy,
and the widespread availability of natural gas. Hydraulic
fracturing technology in particular has enabled natural
gas supplies to increase twenty-fold in the past decade
[10], as shown in Figure 2.
1.2
Dry Shale Gas Production
(billion cubic meters/day)

2. Coals Role in Sustainable Development

Percent of Electricity Derived From Coal

80

By statute, the EPA is required to consider the


benefits and costs of any proposed regulation. The extent
and basis of benefit/cost consideration was recently
debated for a rule entitled National Emission Standards
for Hazardous Air Pollutants from Coal- and Oil-fired
Electric Utility Steam Generating Units and Standards of
Performance for Fossil-Fuel-Fired Electric Utility,
Industrial-Commercial-Institutional,
and
Small
Industrial-Commercial- Institutional Steam Generating
Units [5], known in its abbreviated form as the Mercury
and Air Toxics Standard (MATS). As the shorter title
implies, the rules were intended to reduce emission of
mercury and other constituents from coal-fired power
plants. For this particular rule, EPA asserted that it did
not have to consider compliance costs. It did however
provide an estimate in the amount of US$9.6
billion/year. The estimated monetized benefits varied
depending on how indirect effects were considered.
Direct benefits of reducing the emission of mercury and
other elements was estimated to be three orders of
magnitude less than compliance costs, at US$4-6
million/year. However a byproduct of meeting these
reduced emissions would also have been to lower
exceedingly fine particulate matter, e.g., those particles
with diameters less than 2.5 m (PM2.5). The reduction
in PM 2.5 particulate matter was not the intent of the rule
however its removal was estimated to yield health
benefits, primarily in the form of reduced premature
mortality, in the amount of US$37-90 billion/year.
While EPA asserted that costs should not be considered,
it included these co-benefits to conclude that overall
benefits outweighed the costs in any case. The state of
Michigan sued the EPA for this rule and on June 29,
2015, the U.S. Supreme Court ruled against the EPA [6].
In writing for the majority, Justice Antonin Scalia wrote
The Agency must consider costincluding, most
importantly, cost of compliancebefore deciding
whether regulation is appropriate and necessary.

1.0
0.8
0.6

Antrim (MI, IN, & OH)

Bakken (ND)

Woodford (OK)

Barnett (TX)

Fayetteville (AR)

Eagle Ford (TX)

Haynesville (LA & TX)

Marcellus (PA & WV)

Utica (OH, PA & WV)

Rest of US 'shale'

0.4
0.2
0.0

Figure 2. Increase in shale gas production in the U.S.


since 2000; data from Energy Information
Administration [10]
Despite these factors and as with the rest of the world,
coal is forecasted to be the dominant fuel source in the
U.S. through 2040, accounting for 36% of generated
electricity, ahead of natural gas, nuclear, petroleum, and
all renewable sources, as shown in Figure 3 [11].

Page 17

Author et. al. / INDECS-15


reduces the GDP and decreases the ability for
investments in any other activity, including research and
development (R&D). It is reinforcing cycle in the sense
that investments in R&D can, in turn, yield advances
which improve GDP [16,17].

2000

Billion Kilowatt Hours

1800
1600
1400
1200

4. Benefits, Costs, and Uncertainty

1000
800
600

Coal
Natural Gas
Nuclear Power
Renewable Sources

400
200
0
2015

2020

2025

2030

2035

2040

Year
Figure 3. Forecasted sources of electricity in the U.S.
through 2040, data from the Energy Information
Administration [11]
Increased electrification contributes to multiple
indicators of sustainable development, including greater
gross domestic product, increased life expectancy, and
reduced
infant/toddler
mortality.
Specifically,
electrification is a prerequisite for providing supplies of
clean water, functional medical facilities, and
refrigerated vaccines [12]. The correlation between lowcost coal-fired electricity and global health measures has
been presented by the author [13] and others [14]. For
example, Figure 4 provides data on the under-age five
mortality in India, the U.S., and the world [15].

Under-age Five Mortality Rate,


per 1000 births

250

India
United States
World

200

150

So in returning to the energy-economy-environmental


dilemma, we note that we need low cost sources of
electricity despite their environmental concerns. While
seemingly utilitarian, a cost-benefit approach provides
useful context in which to consider uncertainty and
social considerations. Such an approach can be observed
in the evolution of air emission control policies, from
gross particulate matter control, to nitrogen/sulfur
oxides, to the currently proposed regulation of CO2. The
first coal fired power plant in North Carolina was
constructed in 1926. There was no provision to remove
the particulate matter and for decades it simply emanated
from tall flue gas chimneys in the form of coal fly ash.
By the 1970s, particulate control technology was
relatively common, achieved through electrostatic
precipitators or baghouse filters. Similarly, since 1990, a
significant fraction of coal-fired power plants have been
constructed or retrofitted to reduce emissions from
nitrogen oxides through selective catalytic reduction and
from sulfur oxides through flue gas desulfurization. In
comparison to carbon capture and storage, the costs of
these changes were relatively small in comparison to
either the U.S. gross domestic product or the revenue of
a given utility. For its part, the U.S. EPA estimated that
between 1970 and 1990, the Clean Air Act (CAA)
legislation of 1970 resulted in a variety of health-related
benefits valued in the range of US$ 6-50 trillion dollars.
Compliance with this legislation was estimated at US$
520 billion [18] and these costs were absorbed during a
time in which the GDP grew by a factor of five.
In the case of CO2 regulation, there is a significant
difference in the benefits, costs, and associated
uncertainty of the calculations. This is illustrated
conceptually in Figure 5.

100

50

$100
0
1975

Benefits
1985

1995

2005

Costs

2015

Figure 4. Under age five mortality rate for India, the


U.S., and the World; data from [15]
In addition to facilitating a globally rising standard of
living, low-cost sources of electricity such as coal
provide the capacity to make investments in the research
required to develop an engineering-level understanding
of climate change, mitigation strategies, and costeffective renewable sources of energy [13]. This is
because energy costs have a direct effect on a countrys
gross domestic product (GDP). Increasing energy costs

$10

$1

$0
Particulate

NOx/Sox

CO2 capture

Figure 5. Conceptual comparison benefits and costs for


the removal of particulate matter (e.g., 1970s),

Page 18

Author et. al. / INDECS-15


nitrogen/sulfur oxides (e.g., 1990s) and carbon dioxide
capture (e.g., currently proposed).
As shown, the benefits of cleaner air from removing
large particulate matter (e.g., coal fly ash) as well as
nitrogen and sulfur oxides are demonstrably larger than
the costs. That said, we can estimate compliance costs
with far greater precision than we can benefits, given
that the former can be readily estimated with actual
equipment costs and the latter derive from macro-level
correlations with subjective assumptions. The cost to
build or retrofit a power plant is far more predictable, as
is its effect on the economy. In the case of CO2 capture,
the benefits are even more nebulous. Efforts are made to
assign a monetary value to the presumed ability of coal
fired power plants to contribute to climate change. This
is referred to as the social cost of carbon (SCC) and
includes
predicted
reductions
in
agricultural
productivity, reductions in human health, property
damage from increased floods, and reduced functioning
of ecosystems.
To enlarge the combined uncertainty, these
correlations and assumptions are linked in series to
another suite of correlations and assumptions that
connect carbon dioxide with weather patterns and
temperature. For example, although the IPCC and others
have widely reported that there has been a hiatus in
warming for the past 17 years (e.g., no correlation
between carbon dioxide and global temperatures), a
recent publication [19] has demonstrated that perhaps
this hiatus is actually a function of data biases. The
authors employ a series of correlations, corrections, and
assumptions to ultimately conclude that warming has
indeed been occurring after all. In reality, there is no
validated equation that can relate a given or proposed
coal fired power plant to its effect on global sea level rise
nor with the frequency, severity, or location of extreme
weather, e.g., tornados, hurricanes, droughts, and floods.
Moreover, many such forecasts, when monetized,
presume credit for an entire storms worth of damage,
not merely the increase beyond that which is naturally
occurring.
The result is that there is less precision when using
CO2 emission data to project a cost on society in
comparison to the direct benefits we observe from
electrification (e.g., Figure 4). Similarly, a comparison of
GDP data with energy costs reveals a strong, predictable
correlation that has not been decoupled in developed
economies in which the service sector has
disproportionately increased [20]. Nevertheless, using a
modeling approach with a varying suite of assumptions
[21] the EPA predicts a range of SCC values from
US$13 to US$138 per ton of carbon dioxide. The bulk of
these costs are forecasted to occur at some point in the
future.
Carbon capture is not a mature technology and as such a
requirement to use it would make electricity more

expensive. In its proposed rules to regulate existing and


proposed coal-fired power plants, the EPA asserts that,
actually, there are no costs. This claim is made because
their modeling [4,21] presumes that from a pragmatic
perspective, no utility will build a coal-fired power plant
regardless of the proposed rules, largely because natural
gas has become so readily available. This logic aside,
there are other estimates of the incremental cost of
requiring carbon capture for coal fired power plants.
David and Herzog [22] note that requiring carbon
capture would add more than US$0.03/kilowatt-hour.
According to the U.S. Energy Information
Administration, North Carolinas price is approximately
US$0.09/kilowatt-hour, thus this would increase prices
by at least 33%. Similarly, the Electric Power Institute
has generated the following comparison of prices [23]:
Table 1. Comparison of cost for carbon capture
technology with coal vs. natural gas combined cycle [23]
Type
Project Name Total
Cost/kW
Plant
(US$)
Output
(MW)
Natural Gas Panda
758
$1,100
Combined
Sherman
Cycle
Plant,
Sherman,
Texas, USA
Coal with SaskPower
110
$11,800
Carbon
Boundary
Capture
Dam
Estevan,
Saskatchewan,
Canada
Coal with Southern
524
$8,800
Carbon
Company
Capture
Kemper
County,
Mississippi,
USA
As shown in Table 1, implementing carbon capture, as
would be necessary to comply with proposed regulation
of CO2 at 1,100 lb CO2/ MWh (500 kg CO2/ MWh),
increases the cost by an order of magnitude. In either
case, these costs are known with far greater accuracy
(smaller standard error bars) than any benefit (or cost) of
climate change. Moreover, such increases impact the
poorest countries in the world and poorest citizens of a
given country.
5. Conclusion
Coal fired power has been the primary source of fuel for
much of the worlds history and it is forecasted to remain
so for the foreseeable future. Its use correlates with a
growing economy in both developing and developed
countries as well as with multiple indicators of
sustainable development. The benefits and costs of coal

Page 19

Author et. al. / INDECS-15

fired power can be computed with greater precision than


the putative benefits of eliminating coal as a fuel source.
References
[1] U.S. EPA, Sulfur dioxide data
http://www.epa.gov/airquality/sulfurdioxide/
[2] Hansen, J., M. Sato, P. Kharecha, D. Beerling, R. Berner,
V. Masson-Delmotte, M. Pagani, M. Raymo, D. L. Royer,
J. C. Zachos, Target atmospheric CO2: Where should
humanity aim? Open Atmos. Sci. J. (2008), Vol. 2, pp.
217-231.

[16] Neal, H.A., Smith, T.L. and McCormick, J.B. Beyond


Sputnik, U.S. Science Policy in the 21st Century,
University of Michigan Press, Ann Arbor, MI (2008)
[17] Solow, R.M. Technical Change and the Aggregate
Production Function The Review of Economics and
Statistics, 39(3) (1957) 312-320.
[18] U.S. Environmental Protection Agency. The Benefits and
Costs of the Clean Air Act, 1970 to 1990; Report
Prepared for the U.S. Congress. Washington, DC, (1997)

[3] Carbon Pollution Emission Guidelines for Existing


Stationary Sources: Electric Utility Generating Units;
Proposed Rule, U.S. EPA, Washington DC, 2014.

[19] Karl, T. R., Arguez, A., Huang, B., Lawrimore, J.H.,


McMahon, J.R., Menne, M.J., Peterson, T.C., Vose, R.S.,
and Zhang, H-M. Possible artifacts of data biases in the
recent global surface warming hiatus Science vol. 348,
Issue 6242, pp. 1469-1472

[4] Standards of Performance for Greenhouse Gas Emissions


From New Stationary Sources: Electric Utility Generating
Units. U.S. EPA, Washington DC, 2014.

[20] Stern, D.I. Economic Growth and Energy Encyclopedia


of Energy, Volume 2. 2004 Elsevier Inc.

[5] National Emission Standards for Hazardous Air Pollutants


from Coal- and Oil-fired Electric Utility Steam Generating
Units and Standards of Performance for Fossil-Fuel-Fired
Electric Utility, Industrial-Commercial-Institutional, and
Small
Industrial-CommercialInstitutional
Steam
Generating Units. U.S. EPA, Washington DC, 2012.
[6] Supreme Court of the United States, Syllabus, Michigan et
al. V. Environmental Protecion Agency. No. 1446.
Argued March 25, 2015Decided June 29, 2015.
[7] Holdren, J.P. Science and Technology for Sustainable
Well Being Science (319) (2008) 425-434.

[21] Regulatory Impact Analysis for the Proposed Standards


of Performance for Greenhouse Gas Emissions for New
Stationary Sources: Electric Utility Generating Units.
U.S. EPA, Washington DC, 2013.
[22] David, J., and Herzog, H. The cost of carbon captur
MIT, Cambridge, MA
[23] Phillips, J.N. The Need for Continued Research &
Development of Coal Power, Electric Power Research
Institute, April 13, 2015.

[8] IPCC, 2014: Climate Change 2014: Synthesis Report.


Contribution of Working Groups I, II and III to the Fifth
Assessment Report of the Intergovernmental Panel on
Climate Change [Core Writing Team, R.K. Pachauri and
L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, 151 pp.
[9] International Energy Agency Statistics, Paris, France, as
compiled by the World Bank, World Development
Indicators (2015)
[10]Energy Information Administration, Natural Gas Weekly
Update for week ending July 15, 2015, accessed
7/25/15:http://www.eia.gov/naturalgas/weekly/archive/201
5/07_16/index.cfm
[11]International Energy Outlook 2013 with Projections to
2040. U.S. Energy Information Administration. U.S.
Department of Energy, available at www.eia.gov/ieo/
[12] The Welfare Impact of Rural Electrification: A
Reassessment of the Costs and Benefits, The World Bank,
2008, Washington, DC.
[13] Daniels, J.L. Coal and Sustainable Science Policy Sixth
International Conference of ICSAELS, October 11-15
(2011), Chongqing, China
[14] Epstein, A. The Moral Case for Fossil Fuels, Penguin
Group, LLC, New York, 2014, 248 p.
[15] United Nations Inter-agency Group for Child Mortality
Estimation (UNICEF, WHO, World Bank, UN DESA
Population, as compiled by the World Bank, World
Development Indicators (2015)

Page 20

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

EARLY WARNING SYSTEMS FOR MINIMIZING LOSS OF LIFE AND DAMAGE TO


INFRASTRUCTURE FOR SUSTAINABLE DEVELOPMENT
Tad.S. Murty1 and Abdulraheem Masoud2
irst Author
in

Adjunct Professor, Department of Civil Engineering, University of Ottawa, K1L0A9, Ottawa, Canada.
2

Master student, Department of Civil Engineering, University of Ottawa, K1L0A9, Ottawa, Canada.

JPEG Format

Abstract
For sustainable development, two major factors need to be considered: Climate Change and Natural Hazards. While recognizing the
fact that climate change is extremely relevant, in this article, we cannot do justice to such a complex phenomenon; it would need at
least a dozen papers, if not more. Hence, I will deal mainly with natural disasters. Countries like China and India, with populations
of over one billion, are particularly vulnerable to natural disasters. Here, we will discuss some of the fundamentals and the problems
with early warning systems, with particular attention to coastal infrastructure. There are three types of natural hazards: permanent,
evanescent and episodic. Permanent hazards like tides, wind waves and coastal erosion do not require early warning systems.
However, they are routinely monitored. Evanescent hazards like droughts, climate change, and sea level rise have no clear
beginning or ending. Hence, it is very difficult to develop early warning systems for those hazards. Only episodic hazards have clear
beginnings and endings, and early warning systems are in use in many countries, with varying success. Some examples of such
hazards are: earthquakes, cyclones (hurricanes, typhoons), storm surges, tsunamis, river floods, landslides, etc. The only major
natural hazard for which there is no early warning system is (are?) earthquakes. At the moment, NATECH and multi-tier
hazards are also on the rise. Some of the problems in the mathematical modeling of these hazards are the following: aliasing,
sampling bias, clustering, precision versus accuracy, multi-pli connected systems in a hydrodynamic sense, and modeling of coastal
inundation from storm surges and tsunamis. In an Eulerian system, this becomes a mathematically ill-posed problem, because of
multiple paths. However, this can be better studied in a Lagrangian modeling approach.
Keywords: Sustainable, infrastructure, hazard

1. Introduction
Sustained development not only means protecting the
already existing infrastructure, but also erecting new
infrastructure in an environmentally friendly manner. Since
climate change is an ongoing process, it would be unwise and
even dangerous not to factor this in all developmental
projects. The main threat to infrastructure is from natural
hazards, which may vary not only from country to country,
but even in different parts of any country. Whereas Early
Warning Systems (EWS) are mainly developed to protect
life, they are at the same time also useful for attempts to
reduce damage to infrastructure. It is unrealistic to expect
that EWS can somehow help to entirely eliminate damage.

What is more probable is that these can assist in minimizing


the damage to infrastructure.

2. Types of natural hazards


Natural hazards can be broadly classified into three
categories. Permanent hazards like astronomical tides, wind
waves and coastal erosion, do not need EWS, since they are
routinely monitored. Evanescent hazards are those that have
no clear beginning or ending, hence it is very difficult to
develop EWS for them. Examples of such hazards are
climate change, sea level rise and droughts. On the other
hand, episodic hazards have a well-defined beginning and
ending. Examples include, but not limited to: earthquakes,
cyclones, river floods, landslides, storm surges and tsunamis,
tornadoes, thunderstorms, and meso-scale weather systems

Page 21

Murty et. al. / INDECS-15

such as squall lines. EWS are in existence for such hazards,


except for earthquakes.
In recent years, another type of hazard, referred to as
NATECH hazard is also occurring. This is the situation in
which a natural hazard triggers a technological hazard. Some
example are weather systems causing air plane accidents,
earthquakes causing landslides, deforming railway tracks
leading to train accidents disturbing radio-active dumps,
thereby releasing radio activity into the atmosphere.

Another example of vertically transverse waves is light


waves. Tides, storm surges and tsunamis are long gravity
waves which can cause coastal inundation, thereby causing
loss of life and damage to coastal infrastructure. On the other
hand, wind waves are short period (and short wavelength)
gravity waves. They cannot climb land as they break at the
coastline.
The third type of waves are horizontally transverse waves
(also called Rossby waves, named after their inventor, Carl
Rossby, a famous Swedish meteorologist) (Figure 3).

Multi-tier hazards are those in which one natural hazard leads


to other natural hazards. In 2013 such a hazard occurred in
Uttarkhand in northwest India. A cloud burst caused river
floods which then triggered landslides. In this article, we will
mostly concentrate on coastal hazards.

3. Ocean waves
There are essentially three types of waves in the oceans
(Thompson, 1961). The first type are longitudinal waves,
because as the wave propagates in a horizontal direction, the
particles move back and forth in the same directions due to
compression and rarefaction. These are also called sound
waves or acoustic waves (figure 1).

Fig. 3 horizontally transverse waves, (from Thompson, 1983


)

As the wave moves from west to east, the particles move


towards north and south. These waves exist due to the socalled Beta effect, ie. variation of the Coriolis parameter with
latitude. These waves are very important for ocean
circulation studies and weather forecasting in midlatitudes. Figures 4 and 5 respectively show the short and
long
wave
ocean
spectra.

Fig. 1 sound waves, (from Thompson, 1983)

The second type is vertically transverse waves. As the wave


propagates in a horizontal direction, the particles move up
and down in the direction of gravity. Those are called gravity
waves (figure 2).

Fig. 4 short wave ocean spectra, (from Platzman, 1971)

Fig. 5 (long wave ocean spectra, ( from Platzman, 1971)

Fig. 2 vertically transverse waves, (from Thompson, 1983)

Page 22

Murty et. al. / INDECS-15

4. Clustering
Astronomers tell us that if the material in the universe were
uniformly distributed, there wouldnt be any galaxies, solar
systems and planets. Therefore, clustering in space and time
is a fundamental property in all fields. In the 1950's the socalled Hurst phenomenon (clustering in time) was recognized
in the field of hydrology. Natural disasters like floods and
droughts appear to cluster in time. Later it was recognized
that clustering occurs in all fields of geophysics. For
example, the recent earthquake in Nepal during April-May
2015 is a good example of clustering in time after a long
break, when the previous disastrous earthquake in that
country happened in 1934 or so.

Figure 7 shows the cyclone tracks across Thailand, also


depicting space clustering.

Fig. 7 the cyclone tracks across Thailand, (from


VONGVISESSOMJAI, Personal communication 1994)

5. Aliasing, Nyquist frequency and sampling


The more general phenomenon of clustering is referred to as
the Joseph effect (Mandelbrot and wallis, 1968). A somewhat
related NOAH effect refers to the fact that when extreme
events happen, they are indeed very extreme, such as in the
proverb," when it rains, it pours". The Eden effect suggests
that everything is possible, if you wait long enough in space
and time. In time, we can wait long enough, but earth has
finite dimensions. The names of these effects are of Biblical
origin.

If any periodic process (solid curve fig 8 ) is not sampled


frequently enough, the process will be misrepresented as
showed in the doted curve in figure 8

Figure 6 shows the areas of cyclogenesis in the Indian


Ocean. Clustering in space can be easily seen.

Fig. 8 the phenomenon of aliasing , (Modified from Courtney ,


2015)

In view of the clustering discussed above, it is not advisable


to use a uniform step in space and time for sampling during
post event field surveys, following natural disasters. As is it
shown in figure 8, inadequate sampling leads to wrong
periodicities dominating the signal, and totally contaminating
it. Below, three examples of aliasing are given.
A) Earth rotates from west to east and the sun appears to
move from east to west in the sky with 24 hours between two
consecutive sun rises. If we take a photo of the sky every 23
hours, the sun would appear to move from west to east with
552 hours (24x23) between two sun rises.

Fig. 6 cyclogenesis in the Indian ocean , ( from RAMAGE, 1971)

Page 23

Murty et. al. / INDECS-15

B) For a diurnal tide, the period between two successive high


waters (HW) or between two successive low waters (LW) is
approximately 24 hours. If the ocean level is sampled every
23 hours, we get a periodicity of 552 hours.
C) For a semi-diurnal tide, the period is approximately 12
hours. If the sampling of the water level is done every 11
hours, a period of 132 hours (12x11) will be obtained.
Fig. 9 Sins of omission and commission, ( from C. Starr, 1969)

To avoid temporal aliasing, one can use the Nyquist


frequency criterion, which states that the sampling frequency
must be at least twice or greater than the highest frequency of
interest. To put this in periodicity terms, suppose we are
interested in some physical process with a dominant period of
ten minutes. This means we must sample the process at least
every five minutes, but in practice, we should sample it more
frequently, about every one-minute.

6. Sins of omission and commission


In mathematical modeling, sins of omission refer to the fact
that the model error increases if enough detail is not included,
which is understandable. Sins of commission refer to the fact
that the model error also increases if too much unnecessary
detail is included. This is not obvious, and needs further
explanation (figure 9)

gave the following values in units of centimeters per Second


Square: 672.31, 672.32, 672.30. This instrument is very
precise because it is reproducing almost the same results, but
is terribly inaccurate because the correct value is 980.
B) A new thermometer to measure the temperature of the
human body gave the following values: 98.14, 98.15, 98.13
degrees Celsius. Again, the instrument is very precise, but is
very inaccurate because the correct value should be 37.6.
One can ask the question, if an instrument can be accurate,
but not precise. Generally, this is not possible because by
chance, an occasional reading could be correct, but the results
cannot be reproduced.

8.
An example is the early attempt in numerical weather
forecasting. In the 1920's in L.F. Richardson in U.K.
attempted to predict weather through numerical integration
and failed. His failure was due to the fact that he included
too much unnecessary detail. More specifically, his equations
included not only the slow moving Rossby waves ( relevant
for weather forecasting in mid-latitudes ), but also the fast
moving surface gravity waves, which are irrelevant for
weather forecasting. These gravity waves contaminated the
solutions to such a degree, that his model failed. Later in the
late 1940's and early 1950's, this problem was solved through
eliminating the gravity waves from the equations.

7.

Solids versus fluids

The question is often asked: how come we can predict solar


and lunar eclipses hundreds of years ahead, but we cannot
accurately predict tomorrow's weather? This can be answered
through the Cauchy-Stokes decomposition. Solids and fluids
in principle, undergo (Figure 10) translation, rotation,
dilation and deformation, except the fact that deformation is
negligible in solids under normal conditions. However, fluids
have no shape, and they take the shape of the container in
which they are placed. This is the fundamental difference
between solids and fluids.

Precision versus accuracy

The difference between precision and accuracy is generally


not well understood. It is generally assumed that they are one
and the same, which is not correct. Observational records
may be precise, but are not necessarily accurate. Couples of
examples to show that they are different are given below.
A) Suppose a new device to measure the gravity of the earth
is developed. Three measurements using this new instrument

Fig. 10 Cauchy-Stokes decomposition

Page 24

Murty et. al. / INDECS-15

While solid bodies obey the Newton's second law of motion,


for fluids, several additional terms have to be included in the
equations.

9.

Irregular motion, nonlinear problems and step


function

Some examples of irregular movement are the movement


of oil slicks and icebergs. We know how to deal with
irregular motion to some extent. In principle, any irregular
line is made up of a combination of sine waves with different
periods, wave lengths and phases (Fourier theorem). For
nonlinear problems, we cannot simply extrapolate over the
whole range of occurrences. For example, small earthquakes
do not generate small tsunamis; rather, the earthquake has to
have a magnitude over a critical threshold, before a tsunami
can be generated. Hence this is more a step function type of
problem.

10.

11.

Coastal inundation from storm surges

The amplitude of a storm surge is directly proportional to the


square of the wind speed and inversely proportional to the
water depth (Murty, 1984, Goennert et al, 2001). Hence, on
shallow continental shelves, the surge will grow in
amplitude. Figure 11 shows the wide continental shelves in
the Bay of Bengal and the Arabian Sea where significant
surges occur.

Probability versus determinism

In nature, everything is probabilistic, at least to some


degree. Einstein stated that God does not play dice with
nature. However, later he changed his mind and developed
the Bose-Einstein statistics in Quantum Mechanics. Because
of errors in observational data and shortcomings in our
mathematical techniques, science can give only probabilistic
forecasts, which the public does not like. Suppose the
weather forecast for tomorrow calls for a 30% probability of
rain. Then either one can increase the probability to 100%
and take an umbrella or decrease it to zero and do not take an
umbrella.
Similarly, suppose the forecast calls for colder weather with a
probability of 70%. Again, one can increase the probability to
100% and wear a sweater or decrease it to zero and not wear
a sweater; these are not life and death issues.

Fig. 11 continental shelves of the Bay of Bengal and Arabian Sea

Figure 12 shows the surge build up on the coast of


Bangladesh during November 1970. In the deep water, the
surge (gravity wave) is travelling much faster than the
weather system over the bay. However, closer to the coast
where the water is shallow, the speeds of the gravity wave
and the weather system roughly match, therefore resonant
transfer of energy occurs from the weather system to the
gravity wave, and the surge amplitude increases.

What about a prediction that there is a 50% chance of a 5


meter storm surge or a tsunami hitting the coastline. Either
one can increase it 100% and evacuate thousands of
people or decrease it to zero and do no evacuation. Either
decision is fraught with danger. Supposed you did massive
evacuation and nothing much happened, then you wasted
people's time and inconvenienced them, and businesses lost
lot of money because they were shut down. On the other
hand if you did not evacuate and several people got killed,
there is a lot of explaining to do why you did not evacuate
Fig. 12 buildup of storm surge (m) in the cost of Bangladesh NOV
1970

Page 25

Murty et. al. / INDECS-15

Sometimes, even if there is no weather system close by,


coastal inundation can occur as it is shown in figure 13, (the
Kallakkadal on the coast of Kerala in India). Long period
swells generated by extra-tropical cyclones traversing
Antarctica can derive energy from an opposing current
through Reynolds eddy stresses and get amplified, hence
flooding the coast.

Figures 15, 16 and 17 show the storm surge area to the right
side of the cyclone tracks in the Bay of Bengal and the
Arabian Sea.

Fig. 15 storm surge effected area in the bay of Bengal coast


1949 , ( from Rao 1964)

Fig. 13 costal inundation due to remote forcing on the west


coast of India , (from M. Baba, 2005, Personal
communication)

As is shown in figure 14 schematically, in the northern


hemisphere on the right side of the cyclone track, the winds
blow towards the coast, whereas on the left side, they blow
away from the coast. Hence, the maximum surge occurs on
the right side of the track in the northern hemisphere. In the
southern hemisphere, the opposite situation occurs.

Fig. 16 storm surge effected area in the bay of Bengal coast


1952 , ( from Rao 1964 )

Fig. 14 Schematic representation of a hurricane in the


Northern Hemisphere
Fig. 17 storm surge effected area in the Arabian Sea 1964

Page 26

Murty et. al. / INDECS-15

Figure 18 shows the different areas affected by the super


cyclone of October 1999, on the coast of Odisha, in India.
There are four distinct regions, one area that is affected only
by the strong winds, one affected only by river flooding, one
affected only by the storm surge and one affected by both
storm surge and river flooding.

Fig. 20 Maximum storm surge plus tide amplitude (m) along the
cost of Odisha, India
Fig. 18 Regions affected by the super cyclone in October
1999, in Odisha, India

Figures 19 and 20 respectively show the maximum surge


amplitudes and surge plus tide on the same coast for the same
cyclone.

12.

Coastal inundation from tsunamis

A typical tide gauge record of a tsunami is shown in figure


21. One can see the forerunner, the main tsunami waves
(including the initial withdrawal of the ocean) and the
secondary undulations.
It can be seen in figure 22 that most of the tsunami energy is
directed perpendicular to the fault.

Fig. 19 Maximum storm surge amplitude (m) along the cost of Odisha,
India

Page 27

Murty et. al. / INDECS-15

Fig. 21 Typical tide gage record of a tsunami

Fig. 22 Tsunami energy is most directly perpendicular to the fault, (from Lomnitz and
Nilsen-Hofseth, 2005)

Page 28

Murty et. al. / INDECS-15

Figures 23 and 24 respectively show the tsunami travel times


(hours) in the Indian and Pacific oceans. Being much larger,
the travel times in the Pacific Ocean are greater.

Figure 25 shows quarter wave resonance amplification. The


highest positive amplitude occurs at quarter wave length.
Hence if the dimension of the coastal bay matches I/4 of the
wave length of the incoming tsunami, then it will be
amplified.

Fig. 25 Quarter wave resonance amplification of a tsunami

Fig. 23 Tsunami travel time chart for the Indian Ocean

During the 1964 Alaska earthquake tsunami, due to quarter


wave resonance amplification, a tsunami height of about 0.5
meters in the ocean got amplified to about 5.8 m at the head
of the Alberni Inlet on the coast of British Columbia, in
Canada (Figure 26).

Fig. 24 Tsunami travel time chart for the Pacific Ocean

Fig. 26 Alberni inlet on the cost of British Columbia, Canada

Page 29

Murty et. al. / INDECS-15

Sometimes tsunami energy can enter a harbor through a


narrow connecting channel, but cannot get out, which is
similar to Helmholtz resonance (Figure 27).

Fig. 27 Helmholtz resonance in a harbor

Figures 30 and 31 respectively show tsunami waves hitting


the Kerala coast after getting reflected from the
Lakshadweepa Islands and the Somalia coast in Africa.

Fig. 30 Tsunami waves heating the cost of Kerala, India after


getting reflected from the Lakshadweepa Islands

Figures 28 and 29 respectively show the Indian Ocean


tsunami of 26th December 2004 at 2 hours 50 minutes and at
4 hours after generation. One can clearly see the reflected
waves.

Fig. 28 The Indian Ocean tsunam0i 26th of December 2004, at 2


hours and 15 min (from Z. Kowalik, 2005, Personal
communication)

Fig. 31 Tsunami waves striking the cost of Karela after getting


reflected from Somalia in Africa

Figure 32 shows the tsunami amplitudes in the Andaman and


Nicobar Islands of India during the Indian Ocean tsunami of
December 26th 2004.

Fig. 29 The Indian Ocean tsunami 26th of December 2004, at 4


hours (from Z. Kowalik, 2005, Personal communication)

Page 30

Murty et. al. / INDECS-15


= latitude; = angular velocity of earths rotation
The pendulum day depends upon the latitude of the
location. The Pendulam days for the northern tip of the
Andaman Islands and the southern tip of the Nicobar
Islands are shown below:

Southern extreme = 6 N latitude

Length of pendulum day = 231 hours

Northern extreme = 14 N latitude

Length of pendulum day = 100 hours

Fig. 32 Run-up Levels in Andaman & Nicobar Islands

There are two types of oscillations in the oceans. Oscillations


of the First Class (OFC) are Gravoid modes, and Oscillations
of the Second Class (OSC) are Elastoid-inertia modes. Both
are separated by the so-called Pendulam day.

2
2T p
sin
T = period of revolution of a Foucault Pendulum;
p

Long gravity energy can get trapped around islands at least


for one Pendulam day, in reality it could be much longer than
that.

13. Mathematical modeling.


Mathematical modelling could be challenging if the islands
near the coast change their shape due to geological
instability, as it can be seen in Figure 33, for the coast of
Bangladesh.
Figure 34 shows the coast of the state of Tamil Nadu, in
India.

Fig. 33 The changing islands shapes next the cost of Bangladesh

Page 31

Murty et. al. / INDECS-15

Fig. 34 the cost of Tamil Nadu, India (from S.M. Ramasamy, 2005, Personal communication)

The coast of Tamil Nadu is not just the outer line, one should
include all the indentations due to rivers, estuaries, canals,
back waters, lagoons, bays and gulfs. In earlier days, finitedifference models (f-d) with rectangular telescopic grids were
used as is shown in figure 35 for the Bay of Fundy region, in
Canada.

At present one uses much more sophisticated finite-element


models (f-e) with irregular triangular grids for better
resolution, as is shown in figure 36 for the east coast of North
America.

Fig. 35 A telescoping rectangular grid for the Bay of Fundy region


of Canada

Fig. 36 An irregular triangular finite element grid for


the east cost of the USA

Page 32

Murty et. al. / INDECS-15

The computation of coastal inundation from tsunamis and


storm surges is extremely difficult because of multi-pli
connected regions in a hydrodynamic sense, as is shown in
(figure 37).

Figure 39 compares the coastal length for land area for some
countries, while figure 40 compares the population protected
per km of coastal defense. For countries like India, with a
large population, expenditure on coastal defenses make
sense, while for countries like Canada and Australia with
large areas, but small populations, coastal protection (except
in some selected urban regions) could be prohibitively
expensive.
100 91.3

km/1000km2

80
60
40
20

Because of multiple paths, the problem is mathematically ill


posed, either there are too many boundary conditions or too
few, never the exact number needed. Traditionally, Euler
methods were in use because of their ease, but they are not
suitable for coastal inundation calculations. On the other
hand, Lagrangian methods, where we follow the particles, are
better suited. An example is the SPH (Smoothed Particle
Hydrodynamics) method which is being successfully used for
computation of coastal inundation from Tsunamis (Nistor)

14. Coastal Protection

24.1

17

10 6.8 6.2
3.2 2.2 2.1 1.9

Fig. 39 Compression of costal length for land area

200 170
160
120
60
80
38 36
13 0.9 0.8
40
0
P/L

Fig. 37 MUTIPLI CONNECTED REGIONS

51.4

For tsunamis in the Bay of Bengal, there are at least two


sources, the Sumatra area and the Andaman-Nicobar Islands.
For the Arabian Sea, the Makran subduction zone is relevant
Fig. 40 Population protected per km of costal defense

There is also a very practical problem. Figure 41 shows the


coast of Norway. Depending upon the scale used, the length
of the coastline could by orders of magnitude. For erecting
coastal engineering protection works, one has to consider the

Page 33
Fig. 38 The Makran subduction zone in the Arabian Sea and the
Sumatra tsunami genic region in the Bay of Bengal

Murty et. al. / INDECS-15

famous Civil Engineering Scale (CES), defined as the finest


scale at which engineering construction can be done.

[8] Mandelbrot BB, Wallis JR. Noah, Joseph and operational


hydrology, Water Ressources Research, 1968; 4(1) : 1-10.

Fig. 41 The cost of Norway

References
[1] Thompson PD. A history of numerical weather predication in the
United States. Belletin of the American Meteorological Society,
1983; 60 : 755-769.
[2] Starr C. Social benefit versus technological risk. Science, 1969;
165 : 1232-1238.
[3] Ramage CS. Monsoon Meteorology. Academic Press. New
York, 1971.
[4] Platzman GW. Ocean tides and related waves. Lectures for the
American Mathematical Society. Proceedings of the Summer
seminars on mathematical problems in the Geophysical
sciences, held at the Rensselear Polytechnique, Troy, New
York, 1970.
[5] Courteney D. Introduction To Spectrum Analysis . Experimental
Musical Instruments . Page 18-22 Nicasio, CA , 1992.
[6] Rao NSB. On some aspects of local and tropical storms in the
Indian area. PH.D. thesis, Department of Civil Engineering,
Jadavapur University, 1968.

[7] Lomnitz C, Nilsen-Hofseth S. Seismic monitoring of the Indian


Ocean Tsunami, EOS, Transactions of the American
Geographical Union, 2005; 86(7) :66-70.

Page 34

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

ASSESSING SUSTAINABILITY OF GROUND IMPROVEMENT METHODS:


QUANTITATIVE TRIPLE BOTTOM LINE FRAMEWORK AND CASE STUDY
Krishna R. Reddy 1 and Rajiv K. Giri 2
1

Professor and 2Graduate Research Assistant, University of Illinois at Chicago, Department of Civil
and Materials Engineering, 842 West Taylor Street, Chicago, IL 60607, USA
kreddy@uic.edu, giri2@uic.edu

Abstract
This study examined the overall sustainability of two materials, hydrated lime and Class C fly-ash, used to stabilize the
glacial till subgrade of a new flexible pavement roadway in a rural area outside of Omaha, Nebraska. Two pavement
designs were developed reflecting the improved bearing capacity of the two subgrade stabilizations materials. Design
analyses show that the hydrated lime stabilized subgrade pavement required a thinner wearing surface and base
compared to a Class C fly-ash stabilized subgrade pavement. Sustainability evaluations for both materials were carried
out based on the three pillars of sustainability economic, environmental, and social. The life cycle assessments (LCA)
of both pavement design scenarios were performed using SimaPro 8.1 in order to study the environmental impacts. Nine
different impact categories: (1) ozone depletion, (2) global warming, (3) smog, (4) acidification, (5) eutrophication, (6)
carcinogenics, (7) non-carcinogenics, (8) respiratory effects, and (9) ecotoxicity, were considered to assess
environmental impacts. The study showed that the hydrated lime scenario outperformed the Class C fly-ash scenario in
all with the exception of global warming impacts. Both the economic and social sustainability assessments showed the
hydrated lime scenario to have the lowest impact compared to the Class C fly-ash scenario. When considering the
overall sustainability of a hydrated lime treated subgrade pavement design versus a Class C fly-ash treated subgrade
pavement design, hydrated lime was determined to be more sustainable. Although Class C fly-ash is a recycled material,
hydrated lime still proved to have a lower environmental, social, and economic impact mainly due to the reduced
thickness of the wearing surface and base pavement layers (i.e. less material had to be produced, transported, installed,
maintained, and disposed). This was particularly important for the asphalt wearing surface. The production of asphalt
had the largest environmental impact across all environmental impact sub-categories. It was recommended that there is
need to reduce the asphalt wearing surface where possible in flexible pavements by increasing strength of other
pavement layers including the subgrade.
Keywords: Sustainability, Ground Improvement, Subgrade Stabilization, Life Cycle Assessment, Hydrated Lime, Fly-ash
study focused on two subgrade stabilization materials Hydrated
Lime and Class C Fly-ash (see Figures 2 and 3).

1. Introduction
Currently, several civil engineering infrastructure systems such as
subgrade pavement systems are designed and built mainly based on
evaluating their long-term stability, direct costs, and system
preference. Overall sustainability assessment of these pavement
systems from environmental, social and economic considerations is
not performed. Various life cycle stages of infrastructure systems,
such as material extractions, construction, environmental impacts,
and maintenance and demolition/disposal efforts must be
incorporated, while keeping in mind the overall sustainability of
different civil engineering infrastructure systems.
In this study, firstly, the technical designs are carried out for a new
flexible pavement roadway near Omaha, Nebraska. The pavement
is being constructed on top of glacial till (unstratified soil deposited
by a glacier) with poor load bearing characteristics (see Figure 1).
As was the case in this study, the poor bearing capacity of the
glacial till subgrade necessitated thicker pavement layers to
adequately distribute traffic loads. Alternatively, subgrade
stabilization can improve the bearing capacity of the underlying
soil, reducing the thickness of the overlying pavement layers. This

Fig. 1. Glacial till soil at the proposed site

Both Hydrated Lime and Class C Fly-ash can stabilize a glacial till
subgrade; however, the overall sustainability of these materials is
not necessarily equal. The life cycle of these products (raw

* Corresponding

author. Tel.: +13129964755


E-mail: kreddy@uic.edu

Page 35

Reddy and Giri / INDECS-15

materials extraction, manufacturing, application, and end-of-life


stages) and their impact on the pavement materials have the
propensity for differing economic, environmental, and social
impacts. Selecting Hydrated Lime or Class C Fly-ash as a subgrade
stabilization method for the study roadway was dependent on
which performs better in a sustainability context.

2. Sustainability and Triple Bottom Line Framework


The most commonly used definition of sustainability was given by
1987 UN World Commission on Environment and Development
as: Development that meets the needs of the present without
compromising the ability of future generations to meet their own
needs. However, it is worth mentioning that the concept of
sustainability or sustainable development means different things to
different people. For example, economists are concerned with
growth, efficiency and resource use. Similarly, sociologists focus
on human needs (i.e. equity, empowerment, social cohesion,
cultural identity), whereas, environmentalists want to preserve the
integrity of environment, live within the carrying capacity of the
earth, and deal with the pollutants. Sustainability evaluation of any
given activity, product, or structural system should be based on the
triple bottom line concept of environmental, economic and social
considerations.
2.1. Environmental Sustainability

Fig. 2. Hydrated lime treated soil

Fig. 3. Class C fly-ash soil

Lime powder (calcium hydroxide) is produced through a process of


crushing limestone and heating to extreme high temperatures.
Hydrated lime is lime that has been slaked (i.e. water added) for
easier application [1]. Lime is inherently hydrophilic and reduces
the shrink-swell of treated soils. Lime particles also bind together
in a cementatious nature improving the overall strength of the
treated soil [2].
Fly-ash is a by-product of burning coal in electric and steam plants.
Fly-ash mainly consists of fine oxides of silicon, aluminum, iron,
and calcium [1]. Class C fly-ash has a high content of lime
(between 15 and 30 percent); therefore, it exhibits mainly similar
characteristics to hydrated lime [3].
These two additives are commonly used in highway road
construction to improve the properties of the subgrade, such as the
reduction of the plasticity index, the swelling potential and longterm strength gain. They are also used in cases where the subgrade
creates poor working conditions, especially when using large
machinery to transport materials on site. With proper compaction
and application of these agents, the load-bearing capacity of the
subgrade can be greatly improved.
The main objective of this study was to assess the overall
sustainability, including the life cycle impacts, of hydrated lime
and Class C fly-ash to determine which should be used in the
subgrade stabilization of roadway outside Omaha, Nebraska. The
life cycle assessment (LCA) is modeled using SimaPro 8.1. The
scope of the LCA includes the entire life cycles of hydrated lime
and Class C fly-ash from raw-material extraction, to
manufacturing, to application, and end of life.
In addition, social sustainability tools and analyses were employed
to investigate the social acceptance of each selected subgrade
pavement. It is important to note that social sustainability is based
on individuals perception and can be subjective. Cost analyses
were also carried out to look into the economic viability for all
considered pavement systems.

Environmental assessment invariably focuses on diversity and


interdependence within living systems, the goods and services
produces by the worlds ecosystems, and the impacts of human
wastes. According to SETAC (1999) (Society of Environmental
Toxicology and Chemistry), Life Cycle Assessment is a process to
evaluate the environmental burdens associated with a product,
process, or activity by identifying and quantifying energy and
materials used and wastes released to the environment; to assess
the impact of those energy and materials used and releases to the
environment; and to identify and evaluate opportunities to affect
environmental improvements [4]. The assessment includes the
entire life cycle of the product, process or activity, encompassing,
extracting and processing raw materials; manufacturing,
transportation and distribution; use, re-use, maintenance; recycling,
and final disposal.
In this study, life cycle assessment (LCA) is conducted using
SimaPro 8.1 software that allows users to assess the life cycle of a
product or a system for environmental impacts, from raw material
extraction through material processing, manufacture, distribution,
use, repair and maintenance, and disposal or recycling [5]. LCA
compiles an inventory of relevant energy and material inputs, and
uses this data to evaluate the potential impacts. In SimaPro,
environmental impact assessment can be performed using different
methodologies, of which Eco-Indicator 99 method is commonly
adopted. Eco-indicator 99 defines three types of environmental
damage: Human Health, Ecosystem Quality, and Resources. Each
damage category consists of a number of impact categories all
measured in the same units (kPt). The kPt is a standard eco-indictor
unit that is used to compare different products or activities
affecting environment, and it defined as total environmental load of
a project/number of inhabitants [5]. This facilitates interpretation of
the results, allowing analysis of the data separately for each
damage category without applying any subjective weighting.
2.2 Economic Sustainability
Economic evaluation determines the flow of financial capital, and
the facilitation of commerce. Economic sustainability is performed
by considering all the financial life cycle costs that incur for the
manufacture and transportation of materials during the life cycle of

Page 36

Reddy and Giri / INDECS-15

different subgrade pavement structures. The quantities of materials


required for construction of pavements are calculated along with
the transportation and labor costs, and the corresponding life cycle
costs are determined.
2.3 Social Sustainability
Social sustainability occurs when the formal and informal
processes, systems, structures, and relationships actively support
the capacity of current and future generations to create healthy and
liveable communities. Socially sustainable communities are
equitable, diverse, connected and democratic and provide a good
quality of life (The Western Australia Council of Social Services,
WACOSS). Social dimension of sustainability has received little
attention. Social issues vary geographically, and developing any
unified assessment methods is challenging. Social impacts are
invariably evaluated based on interactions between institutes/firms
and people, functions expressive of human values, aspiration and
well-being, and ethical issues.

3. Case Study

soils. The quantity of lime used to stabilize most soils is usually


between 5% and 10% [1].
Fly-ash is a by-product of coal combustion process in electric
power generating plants. The quantity of fly-ash used to stabilize
most soils is usually between 5% and 15% [1].
3.1.4 Pavement Design Concept
In general, there are two types of pavement structures: flexible
pavements and rigid pavements. Flexible pavements are
constructed with an asphaltic concrete wearing surface and a base
and subbase layer of varying materials. A rigid pavement is
constructed with a layer of Portland cement concrete slab and
subbase of varying types of aggregate [6]. This study is based on a
flexible pavement design.
The basic equation for flexible pavement design is given in the
AASHTO Guide for Design of Pavement Structures [7]. This
equation allows engineers to determine the structural number
(thickness index) for given axle loadings, reliability, overall
standard deviation, change in serviceability, and the soil resilient
modulus.

3.1 Technical Design


3.1.1 Flexible Pavements
Flexible pavements consist of four layers wearing surface, base,
subbase, and subgrade (see Figure 4).
Descriptions of the variables, that are part of the basic equation for
flexible-pavement design (AASHTO Equation), are listed in Table
1. Assumed values and a rationale for why they were selected are
listed in Table 2.
Table 1. AASHTO equation variables

Fig. 4. Flexible pavement system

The wearing surface is made of asphaltic cement. The base layer is


typically made of coarse aggregate materials. In high traffic
volume situations aggregate combined with a cementitious material
or concrete are also used. The subbase layer consists of smaller
aggregates such as crushed stone or gravel. The subgrade layer is
the soil.

Variable

Variable Name

Description

Typical
Values

W18

18-kipequivalent
single-axle load.

The equivalent pavement


impact as an 18-kip
single-axle load.

Varies

ZR

Reliability (zstatistic from the


standard normal
curve).

S0

Overall standard
deviation of
traffic

SN

Structural
Number

PSI

Loss in
Serviceability
form the time
the pavement is

3.1.2 Need for Subgrade Stabilization


Flexible pavement layers must provide the following: shear
strength, stiffness, resistance to moisture, stability, and durability.
Often, in-situ soils do not meet these requirements. Improvement of
soil bearing characteristics by using additives is one solution to
poor soil quality.
3.1.3 Materials for Subgrade Stabilization
Lime is formed by the decomposition of limestone at elevated
temperatures. The addition of lime helps increase the strength of

Estimates the likelihood


that the pavement will
perform at or above the
terminal serviceability
index (TSI the point at
which pavement needs
rehabilitation or
replacement)
Accounts for variability
in design assumptions,
materials, and
construction practices.
The structural
requirement needed to
sustain the designs traffic
loadings.
Loss of serviceability loss
over the life of the
pavement.

90% or
higher for
highways
50% or
higher for
roadways

Between
0.30 to 0.50

Varies
About 1.2
for heavy
traffic loads
and 2.7 for

Page 37

Reddy and Giri / INDECS-15

new until it
reaches its TSI

Soil resilient
modulus of the
subgrade in
lb/in2

MR

Reflects the properties of


the soil under repeated
load. California Bearing
Ratio (CBR) is more
widely used. CBR is the
ratio of load-bearing
capacity of the soil to the
load-bearing capacity of
the aggregate multiplied
by 100.

Pavement
Layer
Wearing
surface
Determined
by field tests
Base

Subbase

Table 2. AASHTO equation assumptions


Variable
W18

Value
Calculated

Reliability (ZR)

-1.645

Overall Standard
Deviation (S0)

0.45

Rationale
Assuming 95%
Reliability. Typical
for interstate
highways.
Typical values
between 0.3 and
0.50.

Source
Principles of
Highway
Engineering
and Traffic
Analysis, 4th
Edition - Page
109

Calculated

4.2

Determined by
AASHTO Road
Test conditions

Terminal
Serviceability
Index (TSI)

2.5

Recommended by
AASHTO for
design of highways

AASHTO
Guide for
Design of
Pavement
Structures,
1993 - Page I-8

Soil Resilient
Modulus (MR)

Calculated

Structural
Number (SN)
Initial Design
Serviceability
Index (PSI)

Table 4. Structural layer coefficients

light traffic
loads.

3.1.5 Structural Number


The structural number, SN, is the structural requirement needed to
sustain the designs traffic loadings. The structural number is
expressed with the following equation:

Table 3.
Variable
1
2
3
D1
D2
D3
M2
M3

Structural number equation variables

Description
Structural-layer coefficient of the
wearing surface
Structural-layer coefficient of the base
Structural-layer coefficient of the
subbase
Thickness of the wearing surface
Thickness of the base
Thickness of the subbase
Drainage coefficients of the base
Drainage coefficients of subbase

Coefficient

Sand-mix asphaltic concrete


Hot-mix asphaltic concrete
Crushed stone
Dense-graded crushed stone
Soil cement
Emulsion/aggregate-bituminous
Portland cement/aggregate
Lime-pozzolan/aggregate
Hot-mix asphaltic concrete
Crushed stone

0.35
0.44
0.14
0.18
0.20
0.30
0.40
0.40
0.40
0.11

3.1.6 Design Alternatives Considered


The following three design alternatives were considered for this
study:

Baseline Design 1: Flexible-Pavement for Untreated Soil

Design Option No. 1: Flexible-Pavement for Lime Treated


Soil

Design Option No. 2: Flexible-Pavement for Class C Fly-Ash


Treated Soil
Baseline design was created only for comparison purpose. Design
options 1 & 2 were compared with this baseline design, and effects
of subgrade stabilizations were evaluated.
3.1.7 Design Parameters
As shown in Table 5, a variety of materials can be used for base
and subbase of flexible pavements. The wearing surface of flexible
pavement is always hot-mix asphaltic concrete. AASHTOs Guide
[7] for Design of Pavement Structures suggests crushed stone and
sandy gravel as typical materials for base and subbase layers,
respectively. A crushed stone base and sandy gravel subbase were
selected for this study. Table 5 presents the details of each
pavement component including the layer coefficient necessary for
the structural number calculation.

(2)
This equation relates individual material types and thicknesses to
the structural number. Descriptions of the variables, that are part of
the SN equation, are listed in Table 3. Typical values for the
structural-layer coefficients are presented in Table 4.

Pavement component

Table 5. Pavement component design selections


Layer
Wearing
Surface
Base

Material
Hot-mix
asphaltic
concrete
Crushed
stone
Sandy
gravel

Coefficient
()
0.44
0.14

Rationale

Source

Average
value used
in the
AASHTO
Road Test

AASHTO
Guide for
Design of
Pavement
Structures,
1993 - Page
I-6

Typical Values

Subbase

See Table 5

3.1.8 Other Structural Number Equation Assumptions

2 to 4 inches
4 to 10 inches
4 to 10 inches
1.0 represents good
drainage
characteristics

The drainage coefficients of the base (M2) and subbase (M3) are
both assumed to be one, which represents a material with good
drainage characteristics. According to Mannering et al. [1], the
thickness of the pavement is largely dependent on the bearing
capacity of the subgrade; however, there is flexibility in the
thickness between layers. For example, the wearing layer thickness
can be minimized by increasing the thickness of the base or

0.11

Page 38

Reddy and Giri / INDECS-15

subbase. AASHTO typically recommends a wearing layer surface


between 2 and 4 inches, a base of 4 to 10 inches, and a subbase of 4
to 10 inches. In general, this study attempted to minimize the size
of the wearing surface layer and keep the base and subbase layers
within approximately 10 inches of thickness.

Table 7.

Layer

Roadway layer depths and volumes

Material

3.1.9 Subgrade Design Assumptions

(a)

The subgrade soil is glacial till - unstratified soil deposited by a


glacier; consists of clay, silt, sand, gravel, and boulders. The glacial
till resilient modulus (MR) equals 4,000 lb/in2 [8]. The optimum
percentage of hydrated lime to treat the soil is 5% (Nebraska
Department of Roads Pozzolan Stabilized Subgrades, 2007 - Page
24). The resilient modulus (MR) of the glacial till subgrade treated
with hydrated lime is 25,265 lb/in2 [8]. The application depth is
assumed to be 12 in.
The optimum percentage of Class C fly-ash is 12% (Nebraska
Department of Roads Pozzolan Stabilized Subgrades, 2007 - Page
24). The resilient modulus (MR) of the glacial till subgrade treated
with Class C fly-ash is 20,546 lb/in2 [8]. The application depth is
assumed to be 12 in.

Volume of
Material Needed

Baseline glacial till soil

Wearing
Surface

Hot-mix
asphaltic
concrete

7 in

36,960 ft3

Base

Crushed stone

15 in

79,200 ft3

Subbase

Sandy gravel

19.5 in

102,960 ft3

Wearing
Surface

Hot-mix
asphaltic
concrete

3.5 in

18,480 ft3

Base

Crushed stone

9 in

47,520 ft3

Subbase

Sandy gravel

11 in

58,080 ft3

Wearing
Surface

Hot-mix
asphaltic
concrete

4 in

21,120 ft3

Base

Crushed stone

10 in

52,800 ft3

Subbase

Sandy gravel

10.5 in

55,440 ft3

(b)

(c)

3.1.10 Functional Unit


The study roadway was a two-lane (i.e. one lane in each direction)
rural highway. The functional unit for the purpose of this study was
1 mile (5,280 ft) of roadway 12 ft in width. The roadway width
encompassed 10 ft of travel lane and 2ft of paved shoulder. As
discussed previously, the excavation depth was 12 in as
recommended by the Nebraska Department of Roads. The analysis
period of this study was 30 years, which is considered by
AASHTO [7] to be a typical design life of a rural highway
roadway.

Thickness

Hydrated lime treated subgrade

Class C fly-ash treated subgrade

3.1.11 Traffic Assumptions


All traffic assumptions are summarized in Table 6. Traffic density
is used to determine the 18-kip ESAL over the 30 year lifespan of
the roadway.
Table 6. Traffic assumptions
Type
Cars, Pickups, Light
Vans

Number of Passes

Axle Weights

30,000

Two 2,000 lb single axles

Single-Unit Truck

1,000

Tractor Semi-Trailer
Truck

350

8,000 lb steering, single


axle
22,000 lb drive, single axle
10,000 lb steering, single
axle
16,000 lb drive, tandem
axle
44,000 lb trailer, triple axle

Fig. 5. Pavement cross-section with untreated soil (glacial till)

3.1.12 Design Results and Analysis


Resulting pavement thickness for the baseline, hydrated lime
treated soil, and Class C fly-ash treated soil can be found in Table
7. Detailed pavement cross sections are shown in Figures 5 and 6.

Page 39

Reddy and Giri / INDECS-15


Approximately 3,168 ft3 of hydrated lime or 7,603.2 ft3 of Class C
fly-ash is needed to stabilize 1 mile of roadway at 12 ft wide as was
calculated in the design stage. The volumes of other materials
required are shown in Table 9.
3.2 Environmental Sustainability Assessment or Life Cycle
Assessment
Environmental sustainability / life cycle assessment was modeled
in SimaPro 8.1, which is a life cycle assessment software
developed by PRe Consultants. Details of this assessment are
explained below.
3.2.1 Functional Unit
The functional unit of this study is 1 mile (5,280 ft) of roadway 12
ft in width. The roadway width encompasses 10 ft of travel lane
and 2 ft of paved shoulder. As discussed previously, the excavation
depth is 12 in as recommended by the Nebraska Department of
Roads.
3.2.2 Life-cycle Stages
Fig. 6. Pavement cross-sections for (a) Class C Fly-ash treated soil, and
(b) Lime treated soil

Comparison of two pavement designs with stabilized subgrade to


the baseline pavement design shows that soil stabilization improves
soil properties and significantly reduces thicknesses of pavement
layers. The asphalt wearing surface is 50% thinner in the hydrated
lime scenario and 43% thinner in the Class C fly-ash scenario. The
crushed stone base is 40% thinner in the hydrated lime scenario and
33% thinner in the Class C fly-ash scenario. Finally, the sandy
gravel subbase is 44% and 46% thinner in the hydrated lime
scenario and the Class C fly-ash scenario, respectively. Side by
side comparisons of pavement thicknesses are show in Table 8.
Table 8. Pavement design summary

Layer

Wearing
Surface
Base
SubBase

Material
Hot-mix
asphaltic
concrete
Crushed stone
Sandy gravel
Table 9.

Layer

Wearing
Surface
Base
SubBase

Lime
Stabilization
Pavement
Design

Class C Fly-ash
Stabilization
Pavement Design

7.0

3.5

4.0

15.0

9.0

10.0

19.5

11.0

10.5

Volume of materials needed for each design

Material
Hot-mix
asphaltic
concrete
Crushed
stone
Sandy
gravel

Baseline
Pavement
Design

Baseline
Pavement
Design

Lime
Stabilization
Pavement
Design

Class C Fly-ash
Stabilization
Pavement Design

36,960 ft3

18,480 ft3

21,120 ft3

79,200 ft3

47,520 ft3

52,800 ft3

102,960 ft3

58,080 ft3

55,400 ft3

This study considers the entire life cycles of (1) a pavement with a
hydrated lime treated subgrade and (2) a pavement with a Class C
fly-ash treated subgrade, including raw material extraction,
manufacturing, application, and end-of-life.
Hydrated lime is a derivative of limestone. Limestone is mined at a
limestone quarry where it is crushed and transported to a lime kiln.
At the lime kiln, the limestone is heated to 1,000C in a process
known as calcination. The resulting substance has a powder
consistency and is known as lime. Water is added to the lime at a 1
to 2 ratio, hydrating the lime for the purpose of easier application to
the soil subgrade. Hydrated lime is transported to the site via truck
and is applied and mixed with 12 inches of soil (glacial till in this
study). After mixing, the soil is compacted with a vibrating roller.
The treated soil is allowed to sit for several days after which the
other layers of the roadway can be constructed. At the end of life of
the pavement, the treated subgrade layer is left in place.
Class-C fly ash is a waste product of coal fired power plants. ClassC fly ash has high lime content, giving it similar properties as
hydrated lime. Class-C fly ash is transported from a supplier
directly to the site via truck. The Class-C fly ash is spread and then
mixed into the subgrade soil. After mixing, the subgrade soil is
compacted with a vibrating roller. At the end of life of the
pavement, the treated subgrade is left in place.
The life cycles of asphalt (wearing surface), crushed stone (base
layer), and gravel (subbase) are the same for the hydrated lime
scenario and the Class C fly-ash scenario, although the volumes of
materials are different.
Both crushed stone and gravel are mined and crushed at the mine
and then transported directly to the site. From the supplier, both
crushed stone and gravel are transported, via truck, to the site. At
the site, the crushed stone and gravel are spread and compacted.
For disposal, both crushed stone and gravel are transported and
landfilled at an inert waste landfill. Asphalts two major
constituents are a binder (asphalt) and mineral aggregate. Like

Page 40

Reddy and Giri / INDECS-15

crushed stone and gravel, mineral aggregate is mined and crushed


and then transported to asphalt plant.
Asphalt is a derivative of the petroleum distillation process.
Petroleum is pumped from naturally forming underground deposits.
The petroleum is transported via tanker truck, tanker train, tanker
ship, or pipeline to an oil refinery where it is distilled. The asphalt
distillation product is sent to an asphalt plant where is heated to
approximately 300F and mixed with mineral aggregate to form
asphaltic concrete. The asphaltic concrete is transported to the
pavement site where it is spread on the base layer surface and then
compacted with a roller.

SimaPro contained material assemblies for the other materials and


are listed in Table 11.
Table 11. SimaPro material assemblies
Design Material

SimaPro Material/Assembly

Hot-mix Asphaltic
Concrete

Asphaltic Concrete, at plant


Crushed stone 16/32, open
pit mining, production mix,
at plant
Gravel, crushed, at mine
Lime, hydrated, loose, at
plant
Fly Ash

Crushed Stone
Sandy Gravel

Every 10 years, twice over the 30 year lifespan of the roadway, the
asphalt wearing surface is removed and replaced. The replaced
asphalt follows the same life cycle as the initial roadway
construction. After the 30 year lifespan of the roadway has elapsed,
the entire roadway materials are excavated and transported to a
landfill for disposal.
3.2.3 Transportation Assumptions
Project materials were sourced from suppliers within a reasonable
distance of Omaha, Nebraska as shown in Table 10.
Table 10. Distance to material suppliers
Supplier
OMNI Engineering - 14012
Giles Road Omaha, NE
68138
Schildburg Construction
Company - 30565 Lytle
Ave, Malvern, IA 51551
Western Sand & Gravel
Co. - 330 County Road B
Ashland, NE 68003
Stalp Ready Mix - 696
River Rd Wisner, NE,
68791-3101
Nebraska Ash Lincoln, NE

Material

Distance to
Site (miles)

Hot-mix asphaltic
concrete (wearing
surface)

13.1

Crushed stone (base


layer)

44.5

Sandy gravel (subbase


layer)

30.5

Hydrated lime (subgrade


treatment)

84.9

Class C Fly-Ash
(subgrade treatment)

57.1

Hydrated Lime
Fly Ash

Notes
Modified Stonemastic asphaltic
concrete

Dummy Variable

3.2.5 Construction Assumptions


SimaPro did not contain construction processes for spreading and
compacting the roadway layers. SimaPros Operation, lorry, >32t
(i.e. a truck with a load of greater than 32 tons) was used a proxy
for spreading and compaction. The 32 ton truck was assumed to
have similar emissions burden as the spreaders, compactors, and
rollers used in roadway construction. The process is in units of
miles traveled. Given a 1-mile functional unit and 12-foot roadway
width, it was assumed that spreading or compacting the entire 1mile segment would require two passes of the truck (i.e. one pass
covers a 1-mile by six-foot wide section; therefore, 2 passes would
be needed to encompass the entire 1-mile by 12-foot wide roadway
section). For base, subbase, and subgrade layers, additional passes
were included to account for additional compaction needs. A
summary of construction processes are listed in Table 12.
Table 12. Construction process - SimaPro proxy
Layer
Asphaltic concrete
Asphaltic concrete
Crushed Stone
Crushed Stone
Gravel

3.2.4 Material Assumptions


SimaPro did not include a material/assembly for hot-mix asphaltic
concrete. The closest equivalent provided is stone-mastic asphalt.
Stone-mastic asphalt and asphaltic concrete differ in their volume
of bitumen binder. Stone-mastic asphalt contains approximately
double the volume of bitumen binder. For purposes of this study,
an asphaltic concrete material assembly was created by duplicating
the components of SimaPros stone-mastic asphalt material
assembly, but reducing the bitumen percentage by 50%.
Class C Fly-Ash was assumed not to have a production impact.
Class C Fly-Ash is obtained as a waste product of coal fired power
plants. Coal is burned to produce electricity, not to produce fly ash.
As such, coal mining, transportation, and burning need not be
considered as part of the life-cycle analysis. For purposes of
integrating Fly-Ash into SimaPro, a dummy material assembly
variable was created. Transportation cost between the Fly-Ash
supplier and the site were still considered.

Gravel
Subgrade
Subgrade

Construction Process
Spreading
Rolling compaction
Spreading
Vibrating roller
compaction
Spreading
Vibrating roller
compaction
Spreading and mixing
Vibrating roller
compaction

32 Ton Truck Passes


2
2
2
4
2
4
2
8

Excavation of soil, prior to pavement construction, was expressed


as a hydraulic digger process in SimaPro. The excavated
overburden was transported to a closed landfill located at 21.9
miles from the site. Although the transportation to the landfill was
included, a landfill process for the excavated material was not. The
disposal of the overburden, either as a reuse product or daily cover
for the landfill, was outside the system boundary for this study.
3.2.6 Maintenance Assumptions
As previously discussed, the life span of this roadway is designed
for 30 years before it requires a complete rebuild. Over the course
of the life of the roadway, it was assumed that the roadway would

Page 41

Reddy and Giri / INDECS-15

need to be resurfaced (i.e. replacing the wearing surface layer)


every 10 years (twice during the 30 year lifespan), which is in
accordance with AASHTO specifications.

providing a specific assessment of impacts depending on their


respective focus. For example, ozone depletion highlights
processes that produce ozone-depleting substances.

An additional life cycle (one for each scenario) encompassed the


materials/assembly, processes, and waste/disposal of the
maintenance phase of the roadway. The material/assembly of the
asphalt layer was referenced to the new-construction
material/assembly as it was assumed that the entire wearing would
be replaced. Next, an excavation process was included to account
for the volume of asphalt being removed during maintenance.
Finally, a waste/disposal scenario was developed for the landfill of
the old asphalt.

As evident in all categories, except for the Carcinogenics and


Ecotoxicity categories, most of the impact results from the
production of asphalt, mainly due to use of fossil fuels burned to
produce asphalt. No recycling was assumed to take place for the
replacement of old asphalt layers during maintenance, which
increased the impact of using an asphalt layer drastically.

3.2.7 End of Life Assumptions


The entire pavement layer (wearing surface, base, and subbase)
was assumed to be removed at the roadways end of life. Both
treated subgrades were assumed to be left onsite. In practice, the
removal of the base and subbase layers depends on their condition
and can be done on some sections of the roadway and not others.
For simplicity, both the base and subbase were completely
removed.
All materials (asphalt, crushed stone, and gravel) were assumed to
be sent to an inert material landfill for disposal. In practice, much
of this material can be recycled and used elsewhere. Again, the
decision to landfill all material was done to be conservative.

In the case of the Carcinogenics and Ecotoxicity category,


impacts are due to transportation needed for all materials to the
site, and for all other machinery used during construction. Since
these categories look at the direct impacts of pollution on
ecosystems, including people, vehicles show the largest impacts
under these scopes.
From the environment sustainability assessment, the pavement
using hydrated lime to stabilize the subgrade would be the best
option. The decision is primarily due to a lesser amount of asphalt
being used in the pavement using hydrated lime. As discussed
previously, the use of asphalt had the greatest environmental
impact due to the consumption of fossil fuels needed to produce
asphalt and its components. However, to decide which pavement
alternative will ultimately be the most feasible and most sustainable
option, we must not only consider the environmental impacts
associated with the production of each pavement, but also
economic and social impacts.

3.2.8 LCA using SimaPro


SimaPro contains a variety of impact assessment methods. Life
cycle impacts were modeled using the US Environmental
Protection Agencys Tool for the Reduction and Assessment of
Chemical and other Environmental Impacts (TRACI). The impact
categories of TRACI include ecotoxicity, respiratory effects, noncarcinogenics, carcinogenics, eutrophication, acidification, smog,
global warming, and ozone depletion.
TRACI was selected over other impact assessment methods
because, it is developed by the U.S. Environmental Protection
Agency specifically for the US using input parameters consistent
with US locations. Other impact assessment methods were
developed for European locations, which are assumed to be less
appropriate for an Omaha based construction site.
3.2.9 Results and Discussion
After inputting all processes and assemblies into SimaPro, a
comparison of the impacts associated with the life cycles of the
pavement with a fly-ash treated subgrade and the pavement with a
lime-treated subgrade was produced. Table 13 and Figure 7 clearly
show that the pavement using hydrated lime has the lowest overall
impact. There are a total of nine different impact categories, each

Table 13. Side by side comparison of the impacts in each category


specified resulting from the use of each stabilizing agent according to
the TRACI 2 V4.00 method of LCA

Impact
category
Ozone
depletion
Global
warming
Smog

Unit

Life Cycle of
Pavement w/ Fly
Ash Treated
Subgrade (One
Life Cycle)

Life Cycle of
Pavement w/
Lime Treated
Subgrade (One
Life Cycle)

kg CFC-11 eq

0.320867451

0.317153*

kg CO2 eq

1018089.064*

1058809

kg O3 eq

85589.34311

78390.71*

Acidification

mol H+ eq

220110.0066

205261.9*

Eutrophication

kg N eq

467.677518

442.0943*

Carcinogenics
Non
carcinogenics
Respiratory
effects
Ecotoxicity

CTUh

0.003441995

0.002953*

CTUh

0.116117269

0.108*

kg PM10 eq

646.0902891

633.281*

CTUe

722176.2769

624749.2*

*denotes the lower impact in each category when comparing


both alternatives

Page 42

Reddy and Giri / INDECS-15

Figure 7. Comparison of environmental impacts based on LCA

3.3 Economic Sustainability Assessment


The purpose of this assessment was to compare the economic
impact of constructing a roadway with a subgrade treated with
hydrated lime compared to a roadway with a subgrade treated
with Class C fly-ash. This assessment is mainly a comparison
of material costs, although excavation of soil and removal of
the wearing surface during maintenance phases is also included
as both are greatly influenced by material quantities.
Construction costs (i.e. rolling, compacting, etc.) are not
included as these costs are shared between the two treatments.
Material costs are assumed to include transportation to site.
Hydrated lime has the highest cost/unit at $132.31/ton. Class C
Fly-Ash has a cost of $30.85/t. The difference is mainly
attributable to the material source (i.e. hydrated lime must be
manufactured whereas Class C Fly-Ash is a waste byproduct).
A summary of material cost is included in Table 14.
Table 14. Cost of materials used
Material

Cost Per Unit

Hydrated Lime

$132.31/ton

Class C Fly-Ash

$30.85/ton

Hot Mix Asphalt


Crushed Stone
Sandy Gravel

$58.00/ton
$5.55/ton
$5.33/ton

Excavation of Soil

$6.22/yd3

Price Source
Nebraska Department of
Roads
Nebraska Department of
Roads
A.E. Stone Incorporated
US Geological Survey
US Geological Survey
Nebraska Department of
Roads

The quantities of hydrated lime and Class C fly-ash are small


relative to the other materials. This is particularly significant
for expensive hydrated lime, which overall cost is only
$19,700 per functional unit. Although Class C fly-ash was
cheaper at $12,548 per functional unit, use of Class C fly-ash
required more hot mix asphalt and crushed stone. At
significantly greater material quantities, both hot mix asphalt
and crushed stone had a much larger monetary impact.

and $160,495 for the Class C fly-ash scenario, a difference of


$5,406. In per mile terms (double the functional unit) as is
appropriate for this two-lane design, the cost is $310,177 for
the hydrated lime scenario and $320,999 for the Class C flyash scenario, a difference of $10,812. A summary of these
findings can be found in Table 15.
Table 15. Cost of pavement layers and excavation
Material Type

Material Needed

Cost

(a) Hydrated lime subgrade treatment


Hydrated Lime
148.9 ton
$19,700
Hot Mix Asphalt
1,293.88 ton
$75,044
Crushed Stone
2,376.00 ton
$13,186
Sandy Gravel
3,484.51 ton
$18,572
Excavation of Soil
124,080 ft3
$28,584
Total
$155,088
(b) Class C fly-ash subgrade treatment
Class C Fly-Ash
406.77 ton
$12,548
Hot Mix Asphalt
1,478.72 ton
$85,765
Crushed Stone
2,640.00 ton
$14,652
Sandy Gravel
3,326.12 ton
$17,728
Excavation of Soil
129,360 ft3
$29,800
Total
$160,495

This roadway has the asphalt wearing surface replaced in year


10 and year 20 of operation, representing the significant bulk
of maintenance work on the roadway. The total cost of
maintenance in terms of only removing the old asphalt and the
material cost of the new asphalt cost approximately $158,604
per functional unit for the hydrated lime scenario and $181,261
per functional unit for the Class C fly-ash scenario. Per lane
mile, these numbers double to $317,208 for the hydrated lime
scenario and $362,523 for the Class C fly-ash scenario. The
overall economic impact, per lane mile, between the two
scenarios, across the life of the pavements (i.e. the material
cost, excavation cost, and asphalt maintenance cost) is
$627,386 for the hydrated lime scenario and $683,514 for the
Class C fly-ash scenario. In terms of economic impact, the
hydrated lime has the lower impact.

The total cost for materials and excavation of soil for one
functional unit was $155,088 for the hydrated lime scenario

Page 43

Reddy and Giri / INDECS-15

social matrix scoring is given from 0 to 4, where 0 highest


impact, 1-substantial, 2-moderate, 3-minimum, and 4-no
impact. The social indicators and scoring is subjective, but
provide the basis for assessment of social issues that otherwise
are often ignored. The social matrix used in this study
demonstrates an approach to assess the sustainability. Based on
the results (Table 16), hydrated lime was found to be more
socially viable than Class C Fly-ash.

3.4 Social Sustainability Assessment


Though social indicators, metrics, and quantifying tools are not
well established, Benoit-Norris et al. [9] presented a
framework to quantify different social aspects. In the present
study, several relevant social indicators are identified (Table
16) for a general assessment of potential impacts on society by
the both pavements throughout their entire life cycle. The

Table 16. Social sustainability matrix for hydrated lime and Class C fly-ash treated soil pavements

Life stage

Jobs /

Demog

Health

Govern

Cultural

Public

Aesthetic

Quality

Product

Income

raphy

& safety

ance

aspects

services

quality

of life

value

Total

Resource

Hydrated Lime

27

extraction

Fly-Ash

18

Product

Hydrated Lime

26

manufacture

Fly-Ash

18

Product

Hydrated Lime

25

delivery

Fly-Ash

17

Hydrated Lime

25

Fly-Ash

22

Recycling,

Hydrated Lime

26

disposal

Fly-Ash

17

Hydrated Lime

10

11

14

14

13

11

129/180

Fly-Ash

14

14

15

15

16

15

16

13

14

92/180

Product Use

Total

4. Conclusions
The design stage of this study produced two roadway designs
appropriate both to handle the assumed traffic volumes and to
demonstrate the bearing capacity of a subgrade treated with
hydrated lime and a subgrade treated with Class C fly-ash.
Both treated subgrades provided much stronger bearing
capacity over the baseline untreated soil. The SimaPro
environmental LCA, the economic impact assessment, and the
social impact assessment all showed the hydrated lime
pavement design to have the lowest impacts. As evidenced by
the environmental and economic assessments, the thickness of
the asphalt wearing surface was the most important factor in
determining impact. The hydrated lime treated subgrade
pavement had a 0.5 inch thinner asphalt wearing surface over
its Class C fly-ash counterpart. Given a holistic environmental,
economic, and social approach to sustainability, it is
recommended that the subgrade of the new roadway be treated
with hydrated lime. The hydrated lime treatment was shown to
be more sustainable than its Class C fly-ash counterpart.

References

[3] Headwaters Resources. Fly Ash - Types and Benefits,


2005. Retrieved from
http://www.flyash.com/data/upimages/press/TB.1%20Fly
%20Ash%20-%20Types%20&%20Benefits.pdf
[4] Consoli F, Allen D, Boustead I, Fava J, Franklin W,
Jensen AA, de Oude N, Parrish R, Perriman R,
Postlethwaite P, Quay B, Sguin J, Vigon B. Guidelines
for Life-Cycle Assessment: A Code of Practice, Society
of Environmental Toxicology and Chemistry (SETAC),
Brussels, Belgium and Pensacola, USA, 1993.
[5] Goedkoop M, Schryver AD, Oele M, Durksz S, Roest D:
Introduction to LCA with Simapro 7: Pr Consultants,
Netherlands, 2010.
[6] Mannering F, Washburn S, Kilareski W: Principles of
Highway Engineering and Traffic Analysis Fourth
Edition: John Wiley & Sons, 2009.
[7] American Association of State Highway and
Transportation Officials, AASHTO Guide for Design of
Pavement Structures, 1993.
[8] Nebraska Department of Roads. Pozzolan Stabilized
Subgrades, 2007. Retrieved from
http://nlc1.nlc.state.ne.us/epubs/R6000/B016.01162007.pdf
[9] Benoit-Norris C, Vickery-Niederman D, Valdivia G,
Franze J, Traverso M, Ciroth A, Mazijn B. The Intl J of
Life Cycle Assessment, 2011; 16: 682-690.

[1] Hensley T, Jensen W, Berryman C. Pozzolan Stabilized


Subgrades. Nebraska Department of Roads, 2007.
[2] Texas Department of Transportation. Guidelines for
Modification and Stabilization of Soils and Base for Use
in Pavement Structures, 2005.

Page 44

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

TRANSPORT OF HEAVY METAL IONS THROUGH SOILS


P V Sivapullaiah
Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore 560012,
India, Email: siva@civil.iisc.ernet.in.

Abstract
Predicting the rates contaminant transport through soils is essential to assess the risk of soil contamination and
subsequent groundwater. It is also essential for the design of landfill liner systems. The contaminant migration rates vary
with type of the soil, type of migrating ion and hydro-geological regime. The transport parameters required in the
governing transport equation that describes the transport processes need to be accurately defined. The laboratory column
experiments, which can be used to estimate the transport parameters of chemicals species migrating through waste
contaminant barriers. There are various techniques available to model the contaminant transport through soil based on
the experimental data. The commonly used test to find the contaminant transport parameters is column test which
effectively simulates the field conditions. The chemical species of interest is passed through the soil sample and the
effluent concentrations are measured. From this effluent concentration profile, the transport parameters are predicted by
trial and error method. The contaminant transport equation which is a partial differential equation can be solved by
various analytical and numerical methods. For less permeable soils where the flow is very less the soil sample is
sectioned and the pore water concentrations are measured at different depths. From this depth vs. concentration profile,
the contaminant transport parameters will be estimated by comparing the theoretical curve with experimental curve. For
the remediation of contaminated soil, the column leach tests are used in which the distilled water with or without
additives is passed through the contaminated soil, and the effluent concentrations are measured with time. From the
number of pore volumes vs. relative concentration profile, the quantity and type of fluid required for removal of
contaminants from the soil will be assessed.

Keywords: Contaminant Migration, Iron, Liners, Remediation, Zinc


________________________________________________________________________________________________
1. INTRODUCTION
Heavy metals are released frequently from different
sources, for example, material, stockpiling batteries,
electro plating commercial ventures and so on. Strong
waste landfills constitute a potential real danger to the
ground water due to these substantial metals.
Heavy metals, for example, iron are ordinarily found in
a few sorts of waste and consequently display in
landfill leachates and ground water. In Indian urban
areas, open, uncontrolled and ineffectively oversaw
dumping is generally worked on, offering ascent to
genuine natural issues. Strong waste in urban
communities and towns are straightforwardly arranged
off ashore in an unacceptable way and thus there is a
risk to ground water quality and brought about ground
water defilement. Likewise the iron is drained from the
waste dumped from landfills and centralizations of iron
are normally found in lechates.

Author. Tel.: +91-80-2293 2672


Fax: 91-80-23600404; E-mail: siva@civil.iisc.ernet.in

Page 45

Sivapullaiah P.V / INDECS-15

Consequently in numerous real urban communities


heavy metals are quickly getting into the
groundwater. Metals added to soil will regularly be
held at the dirt surface. High amassing of broke
down iron results in poor taste, ugly water that
stains both pipes installations and dress.
Convergences of iron as low as 0.3 milligrams for
every liter will store rosy cocoa recolors on
apparatuses, utensils, and garments, all of which
can be hard to uproot. Ferric iron stores inside
eroded channels can break free and create corroded
faucet water. Iron microorganisms gives water an
obnoxious taste and reasons yellow stains on
clothing. This bacterium can likewise stop up water
frameworks, attachment channels, or encompass
pump screens, bringing about lavish repairs.
2. LEACHING OF HEAVY METALS FROM
WASTE DUMPS AND LANDFILLS
Bangalore did not have any scientific treatment
facilities for solid waste generated by municipal
and industries around Bangalore. Only recently
landfills are being constructed. This has led to
development of several illegal and unauthorized
dump sites in Bangalore. Location of unauthorized
dump sites in and around Bangalore city is as
shown in Fig 1.

Fig. 1 Satellite images of built up area along with


dump sites in Greater Bangalore Boundary

3. SOLID WASTE LANDFILL


The design of Modern Secured Landfill is shown in
Fig. 2. Liner system plays an important role in
containing the leachates generated within the
landfill.
3.1 Barrier System

Calcium bentonite also can be used as additive in


soil liner.
Adsorption of soluble metallic species by clays,
oxides and other colloidal matter appears to be an
important means of controlling the trace soluble
metal concentration in the system. Liners for
landfills should be such that they are capable of
minimization of leachates containing metal ions
such as iron. The retention mechanisms for metals
added to soil include adsorption of the metal by the
soil solid surfaces and precipitation. The retention
of cationic metals by soil has been correlated with
such soil properties such as surface area, cation
exchange capacity, organic matter content, clay
content, iron and manganese oxide content, and
carbonate content. Anion retention has been
correlated with iron and manganese oxide content.
It is reported that in addition to soil properties, the
other parameters like type of metal, its
concentration, the presence of competing ions play
an important role in the retention of heavy metal
ions.
3.2 Soil Liner to retard the movement of heavy
metal in the leachate
Though composite liners are preferred for bottom
and side liners, soil liners are preferred because of
their universial availability and cost. The
requirements of clay liners are
1) Liners should possess high sorption capacity
2) The clay barrier shall be compatible with the
leachate to be contained.
3) From a chemical flux point of view, 10-8 cm/s is
preferable since diffusion often becomes the
dominant migration mechanism.
4) Can plug the pore space in natural soil such that
large channels for contaminate transport are
effectively eliminated
3.2.1 Compacted Natural clay as liner
Clay is the most important component of soil liner
because it ensures low hydraulic conductivity. The
soil should have 20% silt and clay; plasticity index
should be greater than 10%.The material should not
contain soil particles larger than 1 to 2 inches in
diameter which may form a permeable layer. Clays
possess sorptive and or attenuative capacity and
reduce the capacity of contaminants in the leachate.
But some soils do not provide an impermeable
boundary. To achieve this, bentonite is amended to
natural soils and compacted.
Considering all above factors into an account soil
bentonite liner is suggested in this work to control
migration of iron ion.

A relatively small amount of bentonite can lower


the hydraulic conductivity to considerable extent.
Addition of 4-5% sodium bentonite soil drops the
hydraulic conductivity from 10-4 to 10-7 cm/sec.

Page 46

Sivapullaiah P.V / INDECS-15

Table 1 Mavallipura site Features [1]

OPTIONAL PROTECTIVE
SOIL COVER

Features
Latitude and Longitude

TOP FML LINER

HAZARDOUS WASTE LANDFILL

LOWER FML LINER


MIN 3-FOOT THICK
COMPACTED SOIL
COMPONENT OF
COMPOSITE LINER

DRAIN PIPE

Mean elevation of the


site
Land area
Land use
Nearest highway

LEACHATE COLLECTION AND


REMOVAL SYSTEM ( 12-INCH
THICK GRANULAR MATERIAL)

Fig. 2 Modern Secured Landfill with composite liner


system

Access Road
Water bodies and dams

3.3 Handling of Landfill in Bangalore


One of the Landfill is situated at, Hesaragatta zone,
Bangalore North, Karnataka state. This site has
been used as processing site for the municipal solid
waste generated from the Bangalore city. Fig. 2
shows the google map with area map of
Mavallipura landfill dump site. This landfill is not
use because of problems. Leachate generated from
the landfill is not contained and hence migration of
the laeachate with many pollutants including heavy
metals are contaminating the nearby land and water
bodies. The features of the site are given in Table
1.
3.4 Calculation of Migration Rates of ions
Migration of ions such as iron from leachates from
waste dump site or unengineered landfills is a
common source for iron ions in the ground water.
Soil liners are generally used to minimize
migrations of contaminants including metal ions.
Soils generally have large sorption capacity and
reduce the migration of contaminants. These liners
are preferred because of their low cost, large
leachate attenuation capacity and resistance to
damage and puncture.

Reserve forests,
Ecological zones,
Monuments, railway
station, major
settlement
Socio economic
Minor settlement
Airport

Details
Latitude 1350 m North
Longitude 7736 East
Ranging from 51.38m to
38.65m above MSL
40.49 hectares
Barren
7.5km away from the
Nation highway No.7
connecting Mangalore to
Chennai
Approach road to the site
is well developed
Hessargatta water tank5.5km
None within 10 km

Agriculture based
Mavallipura village
within 3km
Bangalore airport more
than 30km
Deccan aviation centre at
8km

3.5 Contaminant Transport Modeling


Contaminant transport through compacted soil is
controlled by a variety of Physical, Chemical and
Biological processes. The physical processes
include diffusion, advection and dispersion. The
chemical process usually includes sorption,
dissolution, complexation, hydrolysis/substitution
and oxidation. The biological process includes
decay etc. In order to predict the transport of Iron
pollutant specie, the transport parameters involved
in the governing set of equations that describes the
transport processes need to be accurately defined.
The laboratory column experiments, which can be
used to estimate the transport parameters of
chemicals species migrating through waste
contaminant barriers [2].

Page 47

Sivapullaiah P.V / INDECS-15

travel at an average rate equal to the seepage


velocity of the pore fluid.
3.6.2 Diffusion
Diffusion is the transport process in which a
chemical or chemical species migrate in response
to a gradient in its concentration.
3.6.3 Dispersion
Dispersion refers to the spreading and mixing
caused by the variation in velocity with which
water moves.
3.7 Determination of diffusion coefficients by
column experiment

Fig. 3 Area map of Mavallipura site


(courstey:google earth
images,https://maps.google.com/map)

3.6 Contaminant transport processes through soil


liner
It is impossible to practically provide complete
contamination of leachate generated in a waste
Landfill facility. Hence the design of a suitable
(either as a primary barrier for municipal waste site
or as a backup barrier for hazardous waste) barrier
requires an estimation of potential rate of
contaminant transport through clay liner of landfill
is useful in the design of secure waste storage
facility. These processes depends on various
parameters such as the type of soil, type of
chemical, time, temperature, pH of the system,
density, porosity, permeability, hydraulic gradient
etc. and can be assessed using two important
parameters i.e., Dispersion coefficient and
Distribution coefficient. These two parameters are
known as contaminant transport parameters, which
are essential for modelling the contaminant
transport and for the design of soil liner system.
The transport parameters of different chemical
species are determined by conducting column
experiments for the liner material with the species
of interest. The determination of transport
parameters is a time consuming process as the
column experiments will take a long time to get
breakthrough. The analytical and numerical
methods, which are used to determine these
parameters from the column test, require several
trial calculations and time marching procedures.
Hence, it is necessary to find a suitable technique
to estimate the contaminant transport parameters,
which saves time and effort.
The following physical and chemical processes
usually govern the contaminant transport [3].
3.6.1 Advection
Advection (convection) is the process by which
solutes are transported along with the flowing fluid
or solvent. Due to advection, non-reactive solutes

For predicting the effective diffusion coefficients


of ionic species of interest using column
experiments, a modified form of above equation is
used. The column experiment results are commonly
reported in terms of number of pore volumes that
have passed through packed soil column wherein
one pore volume, is the cross sectional area of
column(A) times its length(L)
times the
porosity(n) i.e.,Aln.
Total number of pore volum U is the total
discharge divided by value of one pore volume

vnAt vt

ALn
L

(1)
Thus with these definitions, the one dimensional
approximate Ogata-Banks equation can be
rearranged as:

C 1
1U
[erfc (
)]
C0 2
2 UDe / vL

J 0.84

(U 1)
[
]
U 1/2

J 0.16 [

(U 1)
]
U 1/2

(2)

C
when
0.84
Co

when

(3)

C
0.16
Co

vL
[J 0.84 J 0.16 ]2
8
3.8 Retardation factor
De

(4)

(5)

The migration of contaminants caused by the


movement of chemicals dissolved in the water is
described by the retardation factor. Attenuation of
leachate contaminant species is one of important
criteria for the choice of landfill liner [4]. The
retardation factor for any particular chemical
includes all the interactions between the many
chemical species and the solid surfaces of the
porous media. These interactions tend to retard the
migration of the chemical behaviors need to be
known.
Retardation factors, Rf, can be determined in flow
experiment where Rf for particular species is the

Page 48

Sivapullaiah P.V / INDECS-15

ratio of solution velocity to the contaminant


velocity. The retardation factor for that species is
given by
Rf = Vgw/Vsp Where Vgw is the velocity of the
water and Vsp is the velocity of the contaminant
species.

divided into number of layers and the


concentrations at different depths and different
time periods are calculated simultaneously without
any dependence on the concentrations of previous
time periods.
4.3 Finite Difference Method

4. MATHEMATICAL FORMULATION
The contaminant migration can be modeled using
one-dimensional Advection-Dispersion equation
(ADE) which is given below:

c D 2 C
C
R

VS
2
t
Z
Z

(6)

Where:
R=Retardation factor =1 + K/n
If there is no adsorption (K=0), then R=1
= dry density
K=distribution coefficient
n= porosity of the transport medium
C= concentration of the solute
t= time
z= distance/ depth
v = seepage velocity
The most commonly used solution methods of
ADE are given below:
1. Analytical solutions
2. Finite-Layer technique
3. Finite Difference technique
4. Finite Element technique
4.1 Analytical solutions to One Dimensional
Contaminant Transport Equations
Analytical methods are widely used for calculating
the contaminant migration parameters. The best
known analytical solution for concentration C at
time t and depth x beneath the surface of a barrier,
which is assumed to be infinitely deep and subject
to a constant surface concentration, Co can be
calculated. [5]
The relative concentration C/Co is calculated by
R f x v st
C 1
erfc
2 ( D etR
Co 2
f

R v sx
exp f
D e erfc
)

R f x v st

2 ( D etR
f

(7)

4.2 Finite-Layer technique


In situations where soil properties can be assumed
the same at any horizontal location within the
barrier layer, the advection-dispersion equation can
be considerably simplified by introducing Laplace
transform (for one-dimensional problems) and
Fourier transform (for two or three dimensional
problems). The transformed equations can then be
readily solved. POLLUTE v7 was used to find the
transport parameters using Finite Layer technique.
In this technique the thickness of soil sample is

A finite-difference model is constructed by


dividing the model domain into square or
rectangular regions called blocks or cells. The
concentrations are computed at discrete points
within the model called nodes. The network of cells
and nodes called the grid or mesh. The soil profile
is discretized into n depth intervals, Z, such that
Z = iZ and i = 0.n. Time is discretized into
intervals t such that t = j t. Then the derivatives
in the ADE are written as discrete differences for
all the time steps to get set of algebraic equations.
By solving these set of equations, the
concentrations can be determined at different
depths and time periods. Three implicit schemes
BTCS, Upwind and Crank- Nicolson can be used to
find the transport parameters.

4.4 Finite Element Method


Most analytical solutions fail to provide acceptable
results in the case of complex boundary conditions
and for heterogeneous and anisotropic aquifers. In
such cases, numerical solutions will provide a
better alternative to the modelling of contaminant
transport through porous media.
Finite element methods have the advantage that it
is easier to discretize complex two and three
dimensional problems. The flow domain is
discretized and the concentrations at different nodal
points are determined. HYDRUS software can be
used to model the contaminant transport with Finite
Element Method.
5. COLUMN TEST
To determine the Seepage velocity (vs), Dispersion
coefficient (D) and Distribution coefficient (K), a
simple column test apparatus can be used to
simulate one dimensional contaminant migration
through soil by advection and diffusion. The
apparatus consists of an overhead tank which is
mounted on a stand fixed to the wall and the soil
column is connected to the overhead tank through a
plastic pipe to allow seepage of solution through
soil sample. The effluent is collected in the effluent
tank placed below the soil column and the
concentration profile with time is prepared. By
adjusting the diffusion coefficient D and the
distribution coefficient K, the theoretical curve may

Page 49

Sivapullaiah P.V / INDECS-15

be matched to the observed concentration profile


(Fig 3) to give inferred values of both D and K.
The values of D and K deduced in this way can be
checked by comparing the calculated and observed
variations of contaminant concentration with time
in the source leachate.
Batch tests can be
conducted to know the ranges of values for
distribution coefficient. For impermeable soils
where the flow was very less, the contaminant
transport parameters can be determined from the
depth vs concentration profile. The soil sample is
sectioned into equal parts and the concentrations
are determined at different depths. This profile is
matched with the theoretical profile and the
transport parameters are determined (Fig 4).
Effluent Concentration (mg/l)

100
80
60

Porosity(n)=0.4
predicted
experimental

40
20
0
0

10

20

30

40

50

60

70

80

Time (hours)
Fig. 4 Variation of concentration with time

The viable option is to decontaminate the soil by


insitu flushing. The nature of flushing fluid
assumes importance. Flushing is economical and
competitive.
Two
important aspects are
permeability of the soil and solubility of
contaminant in the flushing fluid. There is a need to
maintain slightly acidic condition for the flushing
fluid and temperatures higher than 30OC. Other
alternative is flush the soil with water containing
EDTA chelate (ethylenediaminetetraacetic acid
(Na2EDTA), sodium metabisulfite (Na2S2O5))
solution of 0.1 M (the former can be more
effective). The advantage is that other trace
elements present in the soil also can be removed.
The only important issue is the concentration as
excess would have its own implications. It depends
on the utilization of the soil. The ground water
extraction and it use needs to be kept in mind. The
removal rates depend on the nature of fluid which
are being finalised after carrying out soil column
tests which are underway. The information about
the time period for flushing out can be obtained.
Removal Efficiencies of zinc by flushing with
different fluids have to establish in the laboratory
before attempting the same in the field.. Basically,
the removal rates of metal ions from soils depends
on rates of migration of ions by advection and
diffusion coefficient of zinc from the soil. These
are being determined through column tests.

Concentration (mg/l)

100
80
60

Porosity(n)=0.61
Time = 10 days
predicted
experimental

40
20
0
0

10

Depth (cm)
Fig. 5 Variation of Concentration of iron with Depth

5.1 Contaminant Transport rates for the removal


of metal ions
5.1.1 Removal of Zinc from Contaminated Soil
A case of salt induced heaving of floors and
pavement has been reported in Hindustan zinc
limited at Chanderia, Chattisgarh. It was thus
necessary to remove accumulated zinc salt form
soil. The removal rates and the nature of pore fluid
has been selected based on column test.

To find a suitable remediation technique or to find


the volume of fluid required for soil flushing
Column leach test can be used. The undisturbed
sample of contaminated soil is placed in the
column and the distilled water with or without is
passed through the soil sample. The effluent
concentrations are measured with time and also the
volume of effluent with time is noted. As the time
increases the concentration decreases due to
removal of contaminants from the soil.
A contaminated soil was collected from the waste
disposal site of Hindustan Zinc Ltd., Jaipur. The
initial concentration of this soil was determined
from batch leach test. The batch leach tests were
conducted with three fluids i.e., distilled water, 0.1
N HCL and 0.1 N HCL + 0.1 N EDTA. The soil
tested with distilled water alone has shown an
initial concentration of zero which represents the
presence of crystalline form of zinc in the soil. The
initial concentration of soil with 0.1 N HCL + 0.1
N EDTA is higher than that of 0.1 N HCL and
hence it is taken as the initial concentration of this
soil.
The column leach tests were conducted with 0.1 N
HCL and 0.1 N HCL + 0.1 N EDTA as source
solutions and it was found that the soil flushing
with 0.1 N HCL + 0.1 N EDTA is more faster than

Page 50

Sivapullaiah P.V / INDECS-15

that of 0.1 N HCL. The results are as shown in Figs


5 & 6. Initial Concentration from Batch Leach Test
C0 = 4626 ppm.

1
0.8

0.1 N HCL + 0.1 N EDTA

C/C0

0.6
0.4
0.2
0
0.0

100.0

200.0

300.0

400.0

No of Pore volumes
Fig 6. Column Leach Test with 0.1 N HCL + 0.1 N EDTA
1
0.8

7. REFERENCES
[1] Ramachandra, T.V. Management of
Municipal Solid Waste, Capital Publishing
Company, New Delhi .2006.
[2] Folkes, D. J. Fifth Canadian geotechnical
colloquium. Control of contaminant
migration by the use of liners. Canadian.
Geotechnical Journal. 1982; 19: 320344.
[3] Freeze, R. A. and Cherry J. A. Groundwater.
Prentice Hall, Englewood Cliffs. NJ 1979.
604.
[4] Drury, D. Hydraulic considerations for choice
of landfill liner. Proc., Conference
Organized by British Geotechnical Society
and Cardiff School of Engineering,
University of Wales, Cardiff. 1997; 16-18th
September, 312-318.
[5] Ogata A. Theory of dispersion in granular
medium.
Professional
Paper,
U.S.
Geological Survey, 1970. 411-I.

0.1 N HCL

C/C0

0.6
0.4
0.2
0
0.0

10.0

20.0

No of Pore volumes
Fig. 7 Column Leach Test with 0.1 N HCL

30.0

6. CONCLUSIONS
The following are important conclusion from this
presentation:
1. Transport parameters for the different ions
through soils can be obtained by matching the
experimental breakthrough curves with
theoretical curves obtained with assumed
values of contaminant transport parameters.
2. Both Finite Difference method with excel
spread sheet and Finite layer technique Using
Pollute v7 soft ware have been used to
generate theoretical breakthrough curves and
the parameters obtained by both the methods
are almost same.
3. Out of three schemes tested using Finite
different method (BTCS, Upwind and CrankNicolson), Crank-Nicolson method is the most
accurate method
4. Density of soil has significant effect on
contaminant transport parameters. With the
increase in density, Dispersion coefficient
decreases
and
Distribution
coefficient
increases.
5. The importance of mathematical modelling of
contaminant transport has been demonstrated
for the removal of zinc from contaminated soil.

Page 51

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

GEOSYNTHETICS FOR GROUND IMPROVEMENT


Sushovan Dutta1, Maheboobsab B. Nadaf 2, B. S. Asha3 and J. N. Mandal 4
1

Indian Institute of Technology Bombay, Powai, Mumbai, 400076, India, sushovan@iitb.ac.in


Indian Institute of Technology Bombay, Powai, Mumbai, 400076, India, Maheboob_nadaf@iitb.ac.in
3
Datta Meghe College of Engineering, Airoli, Navi Mumbai, 400 708, India, bsasha1967@gmail.com
4
Indian Institute of Technology Bombay, Powai, Mumbai, 400076, India, cejnm@civil.iitb.ac.in
2

Abstract
This paper depicts a brief overview on some preferable ground improvement techniques, prefabricated vertical drains
(PVDs), encased stone columns, geocell mattress and geofoam. When the subsoil on site is very weak to support the
structure, i.e., the soil possess low bearing capacity and suffer excessive settlement, it is necessary to improve the weak
soil by increasing its bearing capacity and reducing the settlement to the desired extent. Particularly, in urban areas
where the land is very expensive and the saturated soil is very poor to support the structure, the ground improvement
becomes compulsory. The prefabricated vertical drains and encased stone columns are installed inside the ground to a
certain depth, while the geocell mattress and geofoam are applied over the ground. The PVDs are installed to achieve a
rapid consolidation. The encased stone columns help in consolidation as well as enhance the ground stiffness. The
geocell mattress and geofoam are applied as reinforcements for ground modification.
Keywords: Ground Improvement, PVD, Encased Stone Column, Geocell, Geofoam

1. Introduction
With the rapid dwindling of good sites for construction activity
of challenging infrastructure projects, the need is being
increasingly felt for utilizing very low and marginal load
bearing sites. The need is even greater in high cost
metropolitan and coastal areas where demand for the
construction outstrips the land availability. Consequently,
many important and major projects such as airways, highways,
railway embankments, large buildings, container yards and
transport terminals have necessarily to be located on areas with
soft alluvial and marine clay deposits of considerable
thickness. Due to the very low shearing strength and high
compressibility of soil deposits, the safe bearing capacity is
very low and the settlements under nominal structural loads
tend to be excessive. Hence, it is required to improve the
ground by technically feasible and economically viable
methods for wider applications in various projects. Therefore,
techniques have to be adopted in such areas to achieve one or
more of the following objectives:
Reduction of post-construction settlements to tolerable
values
Enhancement of shearing strength (and hence, bearing
capacity) of soft soils
Control on rate of loading consistent with the rate of gain
of shear strength
Geosynthetic is an emerging, vast and potential field in many
civil engineering projects. Geosynthetics are recognized in civil
engineering community as cost effective, proven and reliable

materials to carry out civil engineering jobs in better, faster and


economical way. The applications of geosynthetics are vast
because they have versatility in functioning majorly as
reinforcements, separators, filters, drains and liquid barriers in
soil. With the advent of geosynthetics, the revolution of
utilizing them in different ground improvement systems for
weak or soft soils is picking up at an unprecedented pace. The
demand is increasing for the construction of different types of
projects on soft soil. With the technical and economic
feasibility as the major priority in any upcoming projects, the
geosynthetics in ground improvement methods are the norm of
the day.
In the absence of proper ground improvement, the following
may occur.
Excessive/differential settlement in structures
Decrease in structures life
Increase in maintenance cost
Several available ground improvement techniques can
generally be categorized as,
Removal and Replacement
Consolidation techniques: Sand drains, stone columns,
encased sand and stone columns, polymer prefabricated
vertical drains (PPVDs) with full surcharge, PPVDs and
vacuum consolidation with partial surcharge, natural
prefabricated vertical drain (NPVDs) with full surcharge,
NPVDs combined with stone column or sand column,
Thermo-PVD, solar powered drain, electro-osmotic
consolidation, vacuum dewatering and dynamic compaction

* Sushovan

Dutta. Tel.: +91 7506485429


E-mail: sushovan.@iitb.ac.in

Page 52

Dutta et. al. / INDECS-15


Geosynthetics reinforcement techniques: Encased sand
column, encased stone column, mechanically stabilized
reinforced soil wall, geocell or geoweb, geofoam
Densification: Piles (RCC, wooden, steel, and composite),
dynamic compaction, vibro compaction, compaction
grouting
Chemical stabilizations: Lime column, deep soil mixing, jet
grouting, injection grouting
Electro-kinetic stabilizations
The various development works on ground improvements
have been carried out by many researchers and scientists such
as Kjellman (1948), Richart (1959), Dastidar et al. (1969),
Hansbo (1979), Sengupta (1980), Hansbo et al. (1981),
McGown and Hughes (1981), Giroud (1982), Kremer et al.
(1983), Van Zanten (1986), Chen and Chen (1986), Koerner
(1990), Holtz et al. (1991), Mandal and Shiv (1992), Mandal
and Reddy (1992), Bergado et al. (1993, 2001 and 2008),
Ramaswamy (1994), Rawes (1997), Rao and Balan (2000), Bo
et al. (2003 and 2007), Mandal and Kamble (1998),
Rujikiatkamjorn and Indraratna (2006), Liang and Xu (2010),
Indraratna et al.(2008, 2010 and 2011) [1-29].
This paper provides a basic layout on some preferable ground
improvement techniques, consolidation (prefabricated vertical
drains), encased stone column, geocell mattress and geofoam.

2.

Correct

Choice

of

Ground

Improvement

Techniques
The choice of appropriate technique for ground improvement
depends on several factors such as the type of soil and the type
of improvement required i.e., increase in bearing capacity or
decrease in settlement or both needs to be considered. The time
and cost are also very important factors. The long time
consolidation and installation/construction time is not suitable
for any project. Application of huge and heavy equipments
may increase the cost. The risk of non-performance or cost of
time leads to expensive solutions. Sometimes, the advantage of
area of application of the technique can be taken to reduce cost
and time. The embankment can be designed for achieving 80%
consolidation and the remaining 20% consolidation can take
place during the construction of different pavement layers. The
availability of labor and materials also affect the time and cost.
The suitability of technology depends on the availability
of resources of the country. Countries such as India,
Bangladesh, Pakistan, Myanmar, Thailand, China etc. have
abundant natural materials like jute, coir, sisal and bamboo.
The geotextiles made from such natural materials are called as
limited life geotextiles (LLGS) by Sarsby (1997) [30]. Their
technical applications can be made wherever feasible. Many of
the polymer based geosynthetics can be replaced by degradable
and eco-friendly natural geotextiles depending on the areas of
application. Mandal (1987) [31] recommended the use of jute
and coir for ground improvement systems. They have many
good technical properties:

Strength

Extensibility or stiffness

Flexibility

Durability
Natural fibers have been used in the construction industry since
fifth or fourth millennia BC. The villagers used mud-clay
reinforced with straw to build the dwelling units for their
shelter. Some vegetable natural fibers have low tensile strength

and poor durability. However, the tamarisk branches were used


with clay and gravel for the construction of the Great Wall of
China in 200 BC. The ropes were used for many centuries to
lift the heavy loads at dock sides and mine industries.
Currently, renewed natural fibers can be used by automobile
industry. The weight reduction of the door panels of cars are
about 20 % using flax-sisal fiber mat embedded in an epoxy
resin matrix. Natural fibers can be utilized in several
applications:
Usable for short period of time
Unpaved temporary access road (separation)
Embankment construction on soft clay (basal reinforcement)
Ground improvement (time dependent due to consolidation
and drainage)
Currently, the most promising fibers, jute and coir are
economically viable and vegetable fiber fabrics. Those have
high water absorption capacity as well as lower impact on
environment than manmade fiber does. Indigenous natural
geotextiles have certain major advantages and play significant
role for developing countries like India:
The raw materials are environment friendly, renewable
resources and available abundantly (India is the second
largest jute producer in the world), thus making sustainable
construction industry
Natural, bio-degradable, nourishes soil and return to the
ground without pollution
Low unit cost (Production and conversion cost low),
provides competitive cost and economical
Create new market to agricultural products (supports agro
industry) and jobs for local people
Increase range of crops
Farmers can grow and sell
New markets are primarily close to the production point
Money is attracted to rural areas and region without
incurring major expenditure on transportation
Helps in bringing down costly imports

3. Consolidation Technique
When the subsoil on site is very weak to support the structure,
the engineers have to face problems in most of the cases. This
means that the soil possess low bearing capacity and excessive
settlement. Therefore, it is necessary to improve the weak soil
by increasing bearing capacity and reducing the settlement to
the desired extent, particularly, in urban areas where the land is
very expensive and the saturated soil is very poor to support
the structure on it. The conventional system of ground
improvement is the installation of sand drains as shown in
Figure 1.

Fig. 1. Sand drain

Page 53

Dutta et. al. / INDECS-15

Since 1930, sand drains are used in various projects. They are
typically 200 mm to 600 mm diameter vertical drains and
spaced at 1.5 m to 6 m interval. The conventional vertical sand
drain technique is time consuming and very costly. The
availability of good quality sand is becoming scarce in current
years. Recently, the prefabricated vertical band drains are
replacing the sand drains. Compared to sand drain, the polymer
based prefabricated vertical drain (PPVD) is 100 mm in width
and 3-5 mm in thickness. They are placed typically at 1 to 2 m
spacing. The prefabricated vertical drains have many
advantages with respect to the sand drains. The band shaped
prefabricated vertical drains are easy to install in the soil as
well as causes least disturbance to the soil during installation.
The discharge capacity of band drain is more and consolidation
rate is higher than that of sand drain. In general, following are
the major benefits of using the prefabricated vertical drains.

Fig. 3. Different types of prefabricated vertical drains

- They decrease the required time to consolidate the soil


- They decrease the required surcharge for the compression of
the soil, and
- They increase the shear strength due to consolidation of the
soft soil
Fig.4. Drainage core wrapped with the geotextile material

The prefabricated vertical band drain is made of plastic fluted


or nub bed cores wrapped by the nonwoven spun bonded
geosynthetic filter. The prefabricated vertical drains are used to
accelerate consolidation of soft cohesive soils and dissipate
excess pore water pressure faster below the embankment as
shown in the Figure 2.

Fig. 5. Different types of natural prefabricated vertical drains


made of jute and coir geotextiles (Asha and Mandal, 2012)

Fig. 2. Geosynthetic embankment with prefabricated vertical


drains

Different types and shapes of prefabricated vertical drains are


shown in Figure 3. The drainage core is wrapped with
geotextile as shown in Figure 4. The main function of
geotextile is to filter the soil water in to the core, and the core
transports the water through it to the top of the ground surface.
Natural geotextile materials such as jute and coir can be used to
make prefabricated vertical drains, especially in country like
India, where they are available in abundance. The newly
developed and fabricated four different types of band shaped
PVDs, made of single-layer woven and non-woven jute sheath
wrapped around core of coir ropes or coir mats, designated as
Natural Prefabricated Vertical Drains (NPVDs 1, 2, 3 and 4)
are shown in the Figure 5. Physical properties of the fabricated
NPVDs along with a PPVD are given in Table 1.
The NPVDs are more appropriate low cost alternative choice to
PPVDs, especially in developing countries like India because
of following salient factors:

Asha and Mandal (2011) [32] have carried out the absorption
capacity and discharge capacity tests as control tests on
NPVDs. The test results are compared with the tests conducted
on a commercially available polymer-based prefabricated
vertical drain (PPVD). From the graph of absorption capacity
with time as shown in Figure 6, they observed that NPVD 2,
made from nonwoven jute sheath wrapped around coir ropes,
has the greatest water absorption capacity, and the PPVD has
the lowest capacity. NPVDs 1, 3 and 4 have almost the same
absorbency, as all of them are made of same woven jute filter
sheath. A minimum two hours of soaking is sufficient to
saturate the NPVDs. The discharge capacity of PVDs is greatly
affected by PVD configuration, confining stress due to
surrounding soil and hydraulic gradient within the soil. The
variation in structure and in the mechanism of flow through the
PPVD and the NPVDs needs to be considered whenever flow
comparisons are made. The discharge capacity tests carried out
on NPVDs and PPVD show that the discharge capacity
decreases as the compressive stress increases at all hydraulic
gradients for the PPVD and all NPVDs. As shown in Figure 7,
discharge capacity is greater in the PPVD than in the NPVDs.
NPVD 4, which is made of woven jute sheath and a corrugated
coir mat core, has a greater discharge capacity than NPVDs 1,
2 and 3 at the normal stresses and hydraulic gradients studied.

Technically feasible
More economical
Low energy utilization

Page 54

Dutta et. al. / INDECS-15

Table 1. Physical properties of NPVDs and PPVD (Asha and Mandal, 2012)
Test property/Unit

Type of PVD
NPVD 1

NPVD 2

NPVD 3

NPVD 4

PPVD

Sheath Type

Woven jute

Non-woven jute

Woven jute

Woven jute

Core type

Coir strands

Coir strands

Flat coir mat

85 -90

85 -90

85 -90

Corrugated coir
mat
85 -90

Non-woven
Polypropylene
Corrugated and
studded polypropylene
100

Width (mm)
Thickness at 2 kPa (mm)

11

12 -12.5

16 16.5

184-185

160-165

270-280

310-325

85

Tensile strength (kN/ 85mm wide drain)

6.2

2.25

5.75

5.75

Elongation at break (%)

10

25

10

10

35

2.8E-06

0.9E-06

2.5E-06

4.25E-06

13.5E-06

Weight per metre (gm)

Discharge capacity (m /s) at hydraulic


gradient, i = 0.5 under 250 kPa stress
3

is made of nonwoven jute sheath wrapped around coir ropes.


Although the nonwoven jute sheath has good water absorption
capacity, it gets compressed more easily than woven jute
sheath as the compressive stress increases. They also observed
that the flow conditions through all the NPVDs are linear at all
the hydraulic gradients as shown in Figure 8, satisfying the
theoretical requirement of PVDs; whereas for the PPVD, at all
stress levels, the curves are more or less linear up to a
hydraulic gradient i = 0.5 and beyond which the curves become
nonlinear, indicating turbulent flow.

Fig. 6. Variation in absorption capacity with time for NPVDs 1, 2,


3 and 4 and PPVD (Asha and Mandal, 2012)

Fig.7. Discharge capacity variation with normal compressive stress


at different hydraulic gradients for NPVDs and PPVD (Asha and
Mandal, 2012)

The reason for greater discharge capacity of NPVD 4 is that the


corrugations in the coir mat allow water to flow easily as well
as support the jute sheath better than other core configurations.
The lowest discharge capacity was observed in NPVD 2, which

Fig. 8. Discharge capacity variation with hydraulic gradient under


different normal compressive stresses, for PPVD and NPVDs
(Asha and Mandal, 2012)

Asha and Mandal (2011) have conducted laboratory large scale


consolidation tests on marine clay provided without and with
PPVD and NPVDs to observe consolidation efficiency of
NPVDs. As shown in Figure 9, the efficiency of NPVDs to
accelerate the consolidation process in marine clay soil can be
observed clearly from the figure. The respective settlement in
marine clay with PVDs at the end of 7 days under 10 kPa stress

Page 55

Dutta et. al. / INDECS-15

and 16 days under 50 kPa stress is more than the settlement in


marine clay without PVD indicates that in general PVDs aid in
accelerating consolidation process in soil. Among all the
different NPVDs, the settlement measured and rate of
settlement in marine clay with NPVD 4 at the end of 10 kPa
and 50 kPa stress condition is nearly identical to the settlement
measured in marine clay with PPVD.

[D] Capacity of the stone column resulting from increase in


resistance offered by the confining effect due peripheral
geosynthetics encasement.
A schematic view of geosynthetic encased stone column is
shown in Figure 10.

Fig. 10. Geosynthetics encased stone column

4.1. Encasements

Fig. 9. Laboratory large scale consolidation test (a) Stages of


loading, (b) Measured settlement vs. time in marine clay without
and with different PVDs (Asha and Mandal, 2011)

4. Geosynthetic Encased Stone Columns


Greenwood (1970) [33] reported that stone columns develop
their load carrying capacity through bulging and thereby
inducing near-passive pressure conditions in the surrounding
soil. Bergado et al. (1990) [34] found from field studies that
installation of granular piles increased the bearing capacity by
as much as four times and increased the factor of safety of
slopes by approximately 25%. They also reported improved
performance of stone columns compared to prefabricated
vertical drains. Malarvizhi and Ilamparuthi (2004) [35]
reported improved performance of geosynthetic-encased stone
columns based on small-scale laboratory tests on end bearing
as well as floating columns. Ayadat and Hanna (2005) [36]
performed experimental investigation regarding load carrying
capacity and settlement of stone columns encapsulated in
geogrid textile material and concluded that the ultimate
carrying capacity of a stone column increases with an increase
in the stiffness of the geofabric material used to encapsulate the
stone column.
Load carrying capacity of the ground treated with geosynthetic
encased stone column may be obtained by summing up
contribution of each of the following components under wide
spread load such as tanks, embankments, pavements etc.
[A] Capacity of the stone column resulting from the resistance
offered by the surrounding soil against its lateral deformation
i.e., bulging under axial load.
[B] Capacity of the stone column resulting from the bearing
support provided by the intervening soil.
[C] Capacity of the stone column resulting from increase in
resistance offered by the surrounding soil due to surcharge over
it.

Mainly, the encasements for stone columns are made of


polyester geogrid reinforcement. Different types of
encasements can be prepared using geosynthetic
reinforcements. As bamboo is easily available and cheaper in
India with respect to polyester geogrid and also environment
friendly, natural bamboo geogrid encasements can be prepared
to encapsulate the stone columns. Narrow bamboo sticks of 10
mm width with proper finishing were collected to prepare
bamboo encasements (Dutta et al., 2012) [37]. Figures 11 (a),
(b) and (c) show bamboo made encasement, bamboo made
encasement wrapped with jute geotextile and a polyester
geogrid encasement respectively. Wide width tensile strength
test on bamboo geogrid (200 mm width x 100 mm gauge
length) as per ASTM D4595 [38] is shown in Figure 12. Basic
properties of the bamboo geogrid and polyester geogrid are
reported in Table 2.

Fig. 11. (a) Bamboo encasement (b) bamboo made encasement


wrapped with jute geotextile (c) Polyester geogrid encasement
(Dutta et al., 2012)

Fig. 12. Wide width bamboo grid specimens (a) before tensile test
(b) after tensile test (ASTM D4595-11)

Page 56

Dutta et. al. / INDECS-15

Table 2. Properties of geogrid


Material
Bamboo
geogrid

Polyester
geogrid

Properties
Mesh size

Value
5 mm x5 mm

Ultimate tensile strength

110 kN/m

Ultimate stiffness

2200 kN/m

Mesh size
Ultimate tensile strength

5 mm x5 mm
32 kN/m

Ultimate stiffness

160 kN/m

4.2. Effect of Encasement Stiffness on Radial Deformation


From the finite element analysis in PLAXIS 2D, radial
deformation of stone column without and with full length
encasements were evaluated along its depth. Two full length
encasements, one representing the bamboo encasement of
stiffness 2200 kN/m and another representing the polyester
geogrid encasement of stiffness 160 kN/m, were modelled as
linear elastic geogrid. Radial deformation of stone column
along its depth is shown in Figure 8.38. The stone column was
installed in very soft clay at the center of the tank with
dimensions of the tank considered were 850 mm diameter and
500 mm height. End bearing stone column of 100 mm diameter
and 500 mm height with surrounding clay was loaded with a
200 mm diameter rigid plate so as to obtain uniform settlement.
The plate diameter was chosen such so as to simulate the single
stone column with its surrounding influence zone. The tank
was modeled large enough to avoid the boundary effect. 15node triangular elements were used to model the deformations
and stresses in the soil. In the analysis, un-drained MohrCoulomb model was used for soft clay and drained MohrCoulomb model was used for stone column. Properties of clay
and stone for the finite element analysis are reported in Table
3. Short term plastic analysis was used as calculation
procedure. The axisymmetric model of encased stone column
and generated very fine mesh are shown in Figures 13 (a) and
(b) respectively.

(a)

(b)
Fig. 13. (a) Axisymmetric model of encased stone column (b) Fine
mesh generation
Ra dia l deforma tion (mm)
0

12

16

50
100

Table 3 Properties of clay and stone

150

Properties
Clay

Stone

Elastic modulus, E' (kPa)

4000

50000

Poissons ratio, '

0.4

0.3

Cohesion, cu (kPa)

10

Angle of internal friction,

45

It can be observed form Figure 14 that radial deformation of


stone column gets reduced when it is encased with geogrid
reinforcements. More uniform and minimum radial
deformation is obtained with increasing stiffness of the
encasements. Ordinary stone column (OSC) fails by bulging
within almost 2D length of the column with a significant radial
deformation of around 13 mm. Using Polyester geogrid and
Bamboo encasement radial deformation reduces to 5 mm and
1.4 mm respectively. Geogrid encasements provide excess
lateral confining pressure to the columns and prevent its radial
deformation. As the stiffness increases, more hoop tension gets
generated in the encasement and it provides more confining
pressure to the stone column.

Height (mm)

Parameters

200
250
300
350
400
450

OSC
ESC (160 kN/m)
ESC (2200 kN/m)

500

Fig. 14. Radial deformation of stone column without and with


encasement (OSC = Ordinary stone column; ESC = Encased stone
column)

5. Ground Improvement Using Geocell


In India, half of the subcontinents consist of black cotton soils
which are highly plastic and swelling in nature. It is a very
serious problem for the engineers to construct embankments or
reinforced soil retaining walls on swelling soils. The
conventional methods are excavation and replacement with
good quality filling materials or piling which is not economical
and practical. Alternatively, geocell mattress can be used as an
effective ground improvement technique.

Page 57

Dutta et. al. / INDECS-15

Geocell is a three dimensional honeycomb structures made of a


series of interlocking cells. A jute geocell mattress is shown in
Figure 15.

(2) The geocell-reinforced base acts as a mattress to restrain the


soil from moving upward outside the loading area.
The reinforcement mechanism of geocell mattress compared to
unreinforced case [40] is illustrated in Figure 16.

Fig. 15. Jute geocell mattress

Geocell may be made of

Geogrid
Non-woven and woven geotextile
Plastic and
Geofoam

The geometry of cell may be triangular, square,


rectangular, and hexagonal. It provides very good confining
effects. The deployment of geocell reinforced mattress ensure
stiff platform and drastically improve the bearing capacity of
weak soil. Mhaiskar and Mandal (1996) [39] conducted plate
load test on soft saturated marine clay subgrade to find out the
efficiency of geocell, effect of its cell geometry as well as
effect of relative density of the backfill material. They reported
considerable improvement in load carrying capacity as well as
reduction in settlement when the clay is reinforced with
geocell. It was found that the ultimate bearing capacity as well
as bearing capacity ratio of marine clay increases with
decreasing width to height ratio of geocell. A threedimensional finite element analysis was also carried out using
'ANSYS' to validate the experimental results. The finite
element results were found in accord with the experimental
results.
5.1. Reinforcement Mechanism of Geocell Mattress
Unlike the unreinforced base, geocell-reinforced base can
provide lateral and vertical confinement, tensioned membrane
effect, and wider stress distribution. As the geocell is a three
dimensional structure, it can provide lateral confinement to soil
particles within cells. Geocell mattress can provide the lateral
and vertical confinement in following ways:
Lateral confinement:
(1) As the load over the mattress increases, hoop stress is
generated in the cell wall resulting in the lateral confinement to
the infill material.
(2) Lateral confinement from the adjacent cells to prevent the
lateral expansion of any cell.

Fig. 16. Unreinforced and geocell-reinforced bases (Pokharel et al.,


2010)

5.2. Design Considerations of Geocell Mattress


The reinforcing effect can be increased by increasing the height
of cell, width of mattress and decreasing the equivalent
diameter of cell.
Effect of height:
As the height of cell increases, it will provide more frictional
resistance to the infill soil though after a certain height, no
further improvement may be obtained due to the local buckling
failure of the cell. Height of the cell should be designed based
on the footing width.
Effect of width:
As the width of cell increases, it will distribute the load over a
larger area improving the bearing capacity of foundation soil
though after a certain width no further improvement may be
obtained at a smaller settlement. At larger settlement, geocell
mattress as a whole will act as a tensile membrane as well as
foundation soil will provide more shear resistance at the
interface and as a consequence more improvement will be
observed compared to unreinforced case. The mattress width
should be designed based on the footing width.
Effect of equivalent diameter of the single cells:
As the equivalent diameter of the cell decreases, it will provide
more confinement to the infill soil in a unit volume as well as
the rigidity of the mattress in a unit volume will increase due to
the presence of more connections. Both the mechanism
contributes to the improvement in load carrying capacity
compared to the larger diameter cells.
Effect of tensile stiffness:

Vertical confinement:
(1) The friction between the infill material and the geocell wall,
and

As the tensile stiffness of the cells increases, the improvement


in load carrying capacity will be more. The junction strength of
geocells should be properly designed.

Page 58

Dutta et. al. / INDECS-15

Effect of infill soil:


If the infill soil used is frictional in nature with no apparent
cohesion, it will create higher shear resistance in the cell wallinfill interface compared to the infill soil with apparent
cohesion.
Depth of embedment:
The depth of embedment of the mattress should be properly
designed. From the literature review, it is found to be 10% of
the width of the footing. Otherwise the soil above the mattress
may fail by shear due to the presence of stiffer mattress
beneath it.
Effect of basal reinforcement:
(a)

If the geocell mattress is placed with planar geogrid


reinforcement at base, maximum benefit can be obtained.
6. Ground Improvement Using Geofoam
The geofoam is a super light material which is available in the
form of blocks or cellular honeycomb form. The size of the
block is about 600 mm by 1200 mm by 2400 mm. The block
size can vary from country to country. The other sizes can be
made to minimize the geofoam by cutting and trimming on the
site. Weight of the block varies from 11 kg/m3 to 40 kg/m3. Its
density is very less (about 100 times less) compared to
conventional fill material in geotechnical engineering.
Geofoams have been used in many structures around the world
for the last forty-five years. The life time of geofoam is about
70 to 100 years. Figure 17 depicts the development of geofoam
in geosynthetic family. Formation of geofoam and Scanning
Electron Microscopy (SEM) picture of a geofoam bead are
shown in Figures 18(a) and (b) respectively. Figure 19 shows
the preparation of a modified geofoam made of geofoam beads,
fly ash and cement mix with appropriate proportion of water.
Classification of geofoam is presented in Figure 20.

(b)
Fig. 18. (a) Formation of geofoam (b) Scanning Electron
Microscopy (SEM) picture of a geofoam bead

(a)

(b)

Fig. 19. (a) Mix of geofoam beads, fly ash, cement and water in
proper proportion (b) Prepared modified geofoam

Fig. 17. Development of geofoam in geosynthetic family

Fig. 20. Classification of geofoam

Geofoam may have wide applications in the field of Civil


Engineering. As for example, applications of geofoam as

Page 59

Dutta et. al. / INDECS-15

compressible inclusion behind reinforced concrete retaining


wall and as filling material beneath the pavement are pictured
in Figures 21 and 22 respectively.

References
[1] Kjellman, W. (1948). Accelerating consolidation of fine
grain soils by means of cardboard wicks, 2nd
International Conference on Soil Mechanics and
Foundation Engineering, Vol. 2, pp. 302-305.
[2] Richart, Jr., F. E. (1959). Review of the theories for sand
drains, Transactions of the American Society of Civil
Engineers, 124 (1), pp.709-736.
[3] Dastidar, A. G., Gupta, S. and Ghosh, T. K. (1969).
Application of sand wick in housing project,
International conference on soil mechanics and
foundation engineering, Mexico, pp. 59-64.
[4] Hansbo, S. (1979). Consolidation of clay by band shaped
prefabricated drains, Ground Engineering, 12 (5), July,
pp. 16-25.

Fig. 21. Placement of Geofoam as compressible inclusion behind


reinforced concrete retaining wall

[5] Sengupta, D. P. (1980). Sub-strata treatment for a building


using rope drains, Indian geotechnical journal, 10 (4),
pp. 322-332.
[6] Hansbo, S., Jamiolkowski, M. and Kok, L. (1981).
Consolidation by vertical drains, Geotechnique, 31, pp.
45-66.
[7] McGown, A. and Hughes, F. H. (1981). Practical aspects
of the design and the installation of deep vertical drains,
Geotechnique, 43, pp. 3-17.
[8] Giroud, J. P. (1982) Filtration criteria for geotextiles,
Proceedings of second International conference on
geotextiles, Las Vegas, Nevada, USA, August 1-6, 1982,
IFAI, pp. 103-108.

Fig. 22. Geofoam as filling material beneath the pavement

7. Conclusion
The correct choices of attractive, affordable and suitable
ground improvement techniques are mandatory. Practically
natural prefabricated vertical drain, encased stone column,
geocell and geofoam are ideal choices from economic point of
view in developing countries like India.
Natural prefabricated vertical drain (NPVD) is a low cost
alternative system to the artificial PPVD. NPVD is advocated
because it is more appropriate choice, technically feasible,
superior and more economical, low energy utilization,
especially in developing countries like India. A natural bamboo
encasement, a natural jute geocell mattress and a modified
geofoam are also advocated in this study.
With the advent of geosynthetics, the revolution of different
ground improvement systems for weak or soft soils are picking
up at an unprecedented pace. The technical and economical
viability must be the first priority in any upcoming projects.

[9] Kremer, R. H. J., Oosteveen, J. P., Van, A. F., Eele, W., de


Jager, W. F. J. and Meyvogel, I. J. (1983). The quality of
the vertical drainage, Proceedings of 8 european
conference on soil mechanic and foundation engineering,
Helsinky, Vol. 2, pp. 721-726.
[10] Van Zanten, R. V. (1986). Geotextiles and Geomembrane
in Civil Engineering, Balkema, Rotterdam, 658 p.
[11] Chen, R. H. and Chen, C. N. (1986). Permeability
characteristics of prefabricated vertical drain, Third
international conference on geotextile, Vienna, Austria,
Vol. 2, pp. 785-790.
[12] Koerner, R. M. (1990). Designing with geosynthetics,
Fourth edition, Prentice Hall, New Jersy, USA, 761 p.
[13] Holtz, R. D., Jamiolkowsky, M. B., Lanceiotta, R. and
Pedroni, R. (1991). Prefabricated vertical drain: Design
and performance, CIRIA Ground Engineering report on
ground improvement, Butterworth, 131 p.
[14] Mandal, J. N. and Shiv, A. (1992). Computer aided
design
of
prefabricated
geocomposite
drain,
International journal of construction and building
material, 6 (4), pp. 210-225.
[15] Mandal, J. N. and Reddy, L. G. (1992). Computer aided
design of prefabricated geotextile strip drain, Procedding
Indian geotechnical conference on geotechnique today,
IGC-92, Vol. 1, pp.199-201.
[16] Bergado, D. T., Alfaro, M. C. and Balasubramaniam, A. S.
(1993). Improvement of soft Bangkok clay using vertical
drains, Geotextiles and Geomembranes, 12, pp. 615-663.

Page 60

Dutta et. al. / INDECS-15

[17] Bergado, B. T., Balasubramaniam, A. S., Fannin, Jr. and


Holtz, R. D. (2001). Prefabricated vertical drains (PVDs)
in soft Bangkok clay: a case study of the new Bangkok
International Airport Project, Canadian Geotech, 39, pp.
304-315.

[18] Bergado, B. T., Balasubramaniam, A. S., Fannin, R. J. and


Holtz, R. D. (2008). Innovations and performances of
PVD soft ground improvement: a case of the Second
Bangkok International Airport Project, IGC-2008,
Bangalore, pp. 110-135.
[19] Ramaswamy, S. D. (1994). Potential of jute assembled
soil stabilizers, Proc. 2nd Int. workshop on geotextiles,
New Delhi, pp. 77-85.
[20] Rawes, B. C. (1997). Critical parameters for specification
of prefabricated vertical drains, Geosynthetics
International, Vol. 4, No. 1, pp. 51-64.

[31] Mandal, J. N. (1987). Geotextiles in India, Geotextiles


and Geomembranes, Vol. 6, pp. 253-274.
[32] Asha, B. S. and Mandal J. N. (2011) Innovative ground
improvement technique using natural prefabricated
vertical drain Proceedings of the international
conference on advances in geotechnical engineering,
ICAGE-2011, 7-9th Nov.2011, Perth, Australia, 519-524.
[33] Greenwood, D. A. (1970). Mechanical improvement of
soils below ground surface, Ground Engineering
Proceedings Conference Organised by the Institution of
Civil Engineers, London, June 1970, pp. 11-22.
[34] Bergado, D. T., Singh, N., Sim, S. H., Panichayatum, B.,
Sampaco, C.L. and Balasubramaniam, A. S. (1990).
Improvement of soft Bangkok clay using vertical
geotextile band drains compared with granular piles,
International Journal of Geotextiles and Geomembranes,
9(3), pp. 203-231.

[21] Rao, G. V. and Balan, K. (2000). Natural fiber PVDs


discharge capacity, Eds. G. V. Rao and K. Balan, Coir
geotextiles and emerging trends.

[35] Malarvizhi, S. N. and Ilamparuthi, K. (2004). Load


versus settlement of clay bed stabilized with stone and
reinforced stone columns, Proceedings of GeoAsia-2004,
Seoul, Korea, pp. 322-329.

[22] Bo, M. W., Chu, J., Low, B. K. and Choa, V. (2003). Soil
improvement: Prefabricated Vertical Drain Techniques,
Thomson Learning, A division of Thomson Asia Pvt Ltd.,
Singapore.

[36] Ayadat, T. and Hanna, A. M. (2005). Encapsulated stone


columns as a soil improvement technique for collapsible
soil, J. Ground Improvement, 9(4):137147.

[23] Bo, M. W., Arulrajah, A. and Nikraz, H. (2007).


Preloading and prefabricated vertical drains design for
foreshore land reclamation projects: a case study,
Ground Improvement, 11, pp. 167-76.

[37] Dutta, S., Padade, A. H. and Mandal, J. N. (2012)


Experimental Study on Natural Bamboo Geogrid
Encased Stone Column. 5th Asian Regional Conference
on Geosynthetics, Geosynthetics Asia- 2012, 13 to 1
December 2012, Bangkok, Thailand, 417-426.

[24] Mandal, J. N. and Kamble, S. M. (1998). Geosynthetics


encased stone column, Proc. Geosynthetics-99
Conference, April 28-30, Boston, USA, 1:101-113.

[38] ASTM D4595 (2011). Standard Test Method for Tensile


Properties of Geotextiles by the Wide-Width Strip
Method, ASTM, Philadelphia, PA, U.S.A.

[25] Rujikiatkamjorn, C. and Indraratna, B. (2006). Threedimensional analysis of soft soil consolidation improved
prefabricated vertical drains, Geotechnique, 8, pp. 199206.

[39] Mhaiskar, S. Y. and Mandal, J. N. (1996). Investigations


on soft clay subgrade strengthening using geocells,
International journal of construction and building
materials, 10 (4), pp. 281-286.

[26] Liang, R. Y. and Xu, S. (2010). Innovative soft clay


improvements using vacuum and dynamic compaction,
Proc. Indian geotechnical conference, Geotrendz, IIT
Bombay, Mumbai, Vol. 3, pp.134 -136.

[40] Pokharel, S. K., Han, J., Leshchinsky, D., Parsons, R. L.,


and Halahmi, I. (2010). Investigation of factors
influencing behavior of single geocell-reinforced bases
under static loading, J. Geotextiles and Geomembranes,
28(6), pp. 570-578

[27] Indraratna, B., Aljorani, A., Rujikiatkamjorn, C. (2008).


Analytical and numerical modeling of consolidation by
vertical drain beneath a circular embankment,
Geomechanics, Vol. 8, pp. 199-206.
[28] Indraratna, B., Rujikiatkamjorn, C., Ewers, B. and Adams,
M. (2010a). Class A prediction of the behaviour of soft
estuarine soil foundation stabilized by short vertical drains
beneath a rail track, Journal of Geotechnical and
Geoenvironmental Engineering, 136(5), pp. 686-696.
[29]

Indraratna, B., Rujikiatkamjorn, C., Geng, X.,


Ameratunga, J. and Boyle, P. (2011). Performance and
prediction of Vacuum combined surcharge consolidation
at port of Brisbane, Coastal and Marine Geotechnics:
Foundations for Trade, Australian Geomechanics Society
Sydney Chapter Symposium, pp. 45-60.

[30] Sarsby, R. W. (1997). Limited Life Geosynthetics in


Civil Engineering, The Textile Institute, Wood Head
Publishers Limited, Cambridge, England, pp. 244-255.

Page 61

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15) ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SEISMIC BEHAVIOR OF RC FRAMES WITH VARIOUS MASONRY IN-FILL


CONFIGURATIONS
S SURESH BABU1, SIDRAMAPPA V. ITTI2, R AND SRINIVASAN3
1 Department of Civil Engineering, Adhiyamaan College of Engineering

Hosur, Tamilnadu -635109, India. Email: sunisurp@gmail.com


2 K.L.E. College of Engineering & Technology, Chikodi : 591 201,Belgaum, Karnataka India. Email: ittisv@gmail.com
3 Department of Civil Engineering, Adhiyamaan College of Engineering, Hosur,Tamilnadu-635109, India,
Email:bksrinivasan8@gmail.com

Abstract
This paper enlightens about the response or behavior of 3D RC frames with various configurations & amount of
masonry in-fill, under dynamic loading conditions. A1:3 scale, two bay three storey models of a typical office building
were constructed with various locations and percentage of masonry in-fill. Components of structure ie, structural subassemblages of columns, beams and slabs were casted from the same materials. A series of various ground motion tests
were performed on the building model using servo controlled hydraulic actuator. The dynamic characteristics, such as
frequency, acceleration and amplitude were identified at each frequency level. The effect of de-bonding of masonry infill with RC frames with increase in base acceleration was also observed in this study. Analytical models were developed
to predict the natural characteristics and dynamic properties of structure. In modal analysis, masonry infill was modeled
as equivalent diagonal strut and thickness of strut were taken as equal to infill thickness. Material properties and
boundary conditions were assigned suitably for infill.
Keywords: 3-D Reinforced Concrete Frames, Masonry Infill, Dynamic Loads.

1. Introduction
Masonry in-fill panels have been used in reinforced
concrete frame structures as partition walls. However,
because of the complexity of the problem, their interactions
with the RC frame often are neglected in the nonlinear
analysis of building structures. Such an assumption may lead
to an inaccuracy in predicting the response of a structure,
especially when it is subjected to strong lateral loads, such
as earthquake loads and wind loads. Due to this, aseismic
design procedures demand their considerations for
determining the ultimate capacity of the structure.
Furthermore, the study of their influence is also important to
assess the seismic behavior of existing buildings.
Accordingly, the effect of in-fill walls on the behavior of
frames is widely recognized and has been a subject to
numerous investigations in the last decades.
The masonry in-fills are in-variably constructed
after the basic frame work of the beams, columns and slabs
have gained sufficient strength. As a result, the bond of

masonry in-fill with the RC frame work is negligible at the


sides and top surface of the wall. Therefore, they are
classified as non-structural elements and the structures are

analyzed and designed by considering them only as dead


mass, while neglecting any kind of structural interaction of
such panels. This assumption of neglecting the effects of
masonry in-fill is reasonable and justifiable for the structure
under gravity loads. However, the same is not true for the
structures with masonry in-fill when it is subjected to lateral
loads like seismic loads and wind loads. Under seismic
loads, the stiffness additions due to masonry in-fills modify
the behavior of structure significantly by altering its
frequency. Many National and International Standards have
recommended different empirical formulations to tackle the
above problem by dividing it into two parts (i) Bare framed
structure and (ii) Framed structures with masonry in-fill.
They do not give due consideration to the location and
percentage of in-fill in the structures. Extensive researches

* S.SURESH

BABU . Tel.: +919994227629


E-mail: sunisurp@gmail.com

Page 62

Suresh babu et. al. / INDECS-15

have been carried out worldwide in the last five decades.


Many methods have been developed for the dynamic
analysis of RC frames with masonry in-fill incorporating
one or more simplifying assumptions regarding its stiffness.

The general layout of the idealized three-storey


prototype office building is shown in Fig.1 (a) and
1(b).Scaled down model (1:3) was employed for testing.

Un-Reinforced Masonry (URM) infill panels have a


very high initial in-plane lateral stiffness and low
deformability. Therefore, under seismic loads, the existence
of URM walls change the whole lateral force transfer
mechanism of the structure from a predominant frame action
to predominant truss action [3]
It is well understood that, considering the URM walls just as
dead mass is grossly incorrect, since it always over-predicts
the time period of structure because no consideration is
given to the stiffness of the wall panels [8]. This generally
under-predicts the forces attracted by the structure, and thus
making the structures unsafe. Yet, to avoid such ambiguities,
and to make lives of designers, various standards
recommend certain expressions to empirically evaluate the
time period of the structures. However, these experiments do
not give considerations to the amount and location of infill.
Most of the codes provide expressions which incorporate
only height and width of the structure as variables. Few
other codes, recommend more detailed formulations giving
due considerations to the walls in first storey only. The
argument is that the amount of infill panels in the first storey
greatly influences, while those in the upper stories simply
adds to the mass of frames, and its contribution to the
overall stiffness is considerably less. In the present study, an
attempt has been made to understand the behavior of RC
frames with various percentages and location of masonry infills under dynamic loading.
2. EXPERIMENTAL INVESTIGATIONS
Fig 1: Plan & Elevation
2.1 Prototype Structure
A typical three-storey moment resisting reinforced
concrete framed structure used for offices was designed. The
structure was considered as the representative of low-rise
buildings constructed in the zone-V. The prototype structure
was selected such that dynamic test could be performed on a
scaled model structure, using Servo Controlled Hydraulic
Actuator.
The structure was assumed to be built on
stiff soil/rock conditions such that, it not necessary to
consider soil-interaction or differential settlements. The
experimental tests were performed on 3D frames with
different configurations and percentage of in-fills. Bay width
of 3.6m and storey height of 3m has been considered for this
study. In this work, experiments on 3D frames were
performed and utilized to bring out the effectiveness of infill
at different locations and also to verify the numerical
designs obtained using a further simplified strut model. To
achieve this, dynamic tests were performed on 2-bay 3storey frames.

2.2 SCALE DOWN MODEL


It has now become usual to employ
models rather as part of general research and development
programme than with mere aim of solving specific problems
of construction. Models of RC Structures must accurately
reproduce the behavior of the prototype through all the
stages of loading up to the point of rupture, including the
type of failure. The modeling of RC frame with all the
physical properties presents many difficulties, mainly
because of its extreme inelastic property in both
compression and tension. In this study, materials used in
the construction of the model are identical to the materials in
the prototype structure. Therefore, the scale factors were
appropriately developed based on the principles of modeling
the same acceleration and material.

Page 63

Suresh babu et. al. / INDECS-15

2.3 Mass Similitude


Table 1 shows the scale factors for various
parameters for modeling dynamic behavior of structures
with constant accelerations, since gravitational acceleration
cannot be varied. The following table gives the scaling
factors for various quantities for 1: 3 scaling model.
Table 1: Similitude Relations
Description
S.No

Same Acceleration &


Material
Required

Provided

Geometric length

Elastic modulus

Acceleration

Forces

Stress

Strain

Area

Volume

27

27

Mass

27

10

Time

1.73

1.73

11

Critical Damping

While using scale factor of 1:3 the stiffness is


reduced by 1/9, where as the mass is reduced by 1/27. But to
attain the stiffness and mass similitude, these factors should
be same. By increasing the mass, the mass similarity can be
attained.In order to reproduce inertial effects, additional
masses were provided at appropriate points on the model
and were loosely attached, so that the stiffness and strength
of members cannot be altered.
Therefore to make the
reduction factor for mass as 1/9, the mass of model needs to
be added by 3 times the actual mass. The required additional
weights were attached only at the storey level of slabs for
convenience and these were not distributed to all the
members of the model. But, since the mass of the storey
slabs, beams and superimposed loads are substantially
greater than the mass of the columns, the errors introduced
are insignificant for the mass similitude.
2.4 Model Structure
The scale down models consist of columns of cross
section 116 mmx116 mm & beam cross section
100mmx110mm.The concrete mix was prepared using
ordinary Portland cement, fine sand and crushed gravel
having a ratio as per mix design of M25 concrete from
IS:10262-1982. Cement, sand and stone aggregates were
measured individually using weighing balance and machine

mixed. As per IS:516-1959,the representative samples of


100mmx100mmx100 mm size cubes were casted at each
stage and tested for evaluation of compressive strength.
Beams reinforcement has consisted of 2nosof 8mm diameter
MS bar throughout the member length. At the beam column
junction, both top and bottom bars of the beam were
provided with adequate development length. Shear
reinforcement has consisted of 6mm MS bars having a
spacing of 100mm centre to centre. Hoops were also
provided at junctions of beam-column. Column reinforced
bars were 4nosof 8mm diameter. The lateral ties 5mm MS
bars were placed at 100mm c/c. Masonry were constructed,
and tests were performed with infill.
3. FE Analysis of 3D RC Frames
FE analysis was performed by modeling the URM
panels as simplified strut models. The geometry of 3D
models was developed as per the dimensions with
appropriate boundary conditions, and finally the modal
analysis was carried out. Fig 2(a) & 2(b) shows the 3D
model with equivalent diagonal strut. The model was
analyzed using staad.pro software.
4. Hydraulic Actuator Tests on RC Frames:
Controlled specimens of 3D RC frames with
various configurations of masonry in-fill were tested by
applying dynamic loading with broad band excitation prior
to testing, the casted RC frames are carefully positioned for
actuator loadings. The frame mounted with the hydraulic
actuator is shown in Fig 3. During pre-testing, the frame
structure was thoroughly checked for cracks or damage after
positioning it with actuator. Accelerometers were placed and
their positioning are established based on controllability and
observability theory.
The sine sweep tests were carried out for a
frequency range of 1 to 50Hz, with acceleration levels of
0.50m/s2, 1m/s2 and 1.50m/s2. The response acceleration
graph obtained from sine sweep tests for bare frame and
masonry in-filled frame are shown in Fig3. Maximum
response was obtained separately at every frequency in the
test range. The graphs and de-bonding of masonry in-fill
with RC frames obtained from experimental investigations
are shown in Fig4 (a), (b) & Fig (5) respectively. The
analysis and test results are shown in table 2.
The maximum response acceleration for RC frame
without in-fill and with in-fill are given in table 3 & 4
respectively. A1 accelerometer is located at top, at which
high
magnification
is
recorded
compared
to
A2accelerometer at the bottom. Magnification of
acceleration increases as the de -bonding between masonry
in-fill and RC frame increases. Thus, the natural frequency
and damping was reduced due to the decrease in stiffness of
masonry in-fill.

Page 64

Suresh babu et. al. / INDECS-15

Table3. Maximum response accelerations in the


In-plane direction for frame without In-fill
Accelerometer
locations
Input accelerations(m/s2)
0.5
1
1.5
A1
6.54
7.21
7.88
A2
6.89
7.93
8.21
A3
7.68
8.61
8.98

Fig.2 (a) 3D FEM model with diagonal strut

Fig 3(a) 3D RC frame with infill mounted with SCHA

Fig.2 (b)3D FEM model with diagonal strut

Fig.3(b)3D RC frame without infill mounted with SCHA

Page 65

Suresh babu et. al. / INDECS-15

Fig.4 Positioning of Accelerometers

Fig 5(a): Debonding at column

Fig 5(b): Debonding at column


Fig 4(a)Graph showing response Acceleration in the
In-plane direction for Bare Frame (Freq 1Hz)

Fig 4(b)Graph showing response Acceleration in the In-plane


direction for Masonry In-filled Frame (Freq 1Hz)

Page 66

Suresh babu et. al. / INDECS-15

Frame with
infill
Standard

Indian
Code

Frame
without infill

Inplane

Outofplane

Inplane

Outofplane

5.58

5.65

5.71

5.71

5. CONCLUSIONS
The conclusions attained after the analysis are:
The masonry in-fill, although do not interfere in the
vertical load resisting system for the RC frame
structures, but they significantly affect the lateral
load-resisting system.

FE
Analyses

10.47

4.69

5.67

9.86

Euro Code

4.11

5.65

5.714

5.714

Fig 6(a) Debonding at Beam-Column Junction

Fig 6 (b) Debonding at Beam

The natural frequency of the structure with


complete in-fill is significantly higher than the
natural frequency of the bare framed structure.
The effect of de-bonding of masonry in-fill with
RC frames as the base acceleration is increased is
clearly seen in the test results of the 3D RC frames.
Due to the de-bonding effect the natural
frequencies and damping reduces whereas the
magnification factor of the output acceleration
increases.
The natural frequencies and damping values in 3D
RC frames will be high in the presence of masonry
in-fill. Hence the role of masonry infill in resisting
the lateral forces like earthquake and wind is
significant and has to be accounted during
designing of the structures.

REFERENCES
1. A. DAmbrisi M. De Stefano S. Viti (2008),
Seismic Performance Of Irregular 3D RC Frames,
The 14th World Conference on Earthquake
Engineering. (doi:10.6088/ijcser.201203013053)
2.

ChetanK, R. RameshBabu (2010), Influence of


masonry infill on fundamental natural frequency of
2D Rc frames, Journal of structural Engineering,
Vol-37,
No-2.
(http://basharesearch.com/WCSET2014/wcset2014
004.pdf)

3.

C V R Murty and S K Jain (2000), Beneficial


influence of masonry infills on seismic
performance of RC frame buildings,Proc. 12th
World Conference on Earthquake Engineering,
New
Zealand,
Paper
No.
1790.(http://www.iitk.ac.in/nicee/wcee/article/1790
.pdf)

Table 4: Comparison of Natural Frequencies [Hz]


obtained from Indian Code ,Euro code & FE Analyses.

Page 67

Suresh babu et. al. / INDECS-15

4.

Dhanasekhar M & AW Page (1987), The Influence


Of Brick Masonry Infill Properties On The
Behaviour
Of
In-Filled
Frames.ICE
Proceedings,593 605(doi.10.1680/iicep.1986.463)

5.

Diptesh Das and C.V.R. Murty (2004), Brick


Masonry In-fill in Seismic Design of RC
buildings,Indian
concrete
journal,39-42.(
http://www.iitk.ac.in/nicee/RP/2004_Cost_Implicat
ions_ICJ.pdf)

6.

Han-Seon Lee and Sung-Woo Woo (2002), Earth


Quake engineering & Structural Design, 353378.(Text book)

7.

HaroonRasheedTamboli
And Umesh.n.Karadi
(2012), Seismic Analysis Of RC Frame Structure
With And Without Masonry Infill Walls, Indian
Journal
Of
Natural
Sciences,
11371194(http://www.tnsroindia.org.in/JOURNAL/ISS
UE%2014%20final.pdf)

8.

Hemanath B. Kaushik, Durgesh C. Rai and


SudhirK.Jain (2006), Code approaches to seismic
design of masonry infilled reinforced concrete
frames: a state-of-the-art review Earthquake
Engineering Practice, (Vol 1, Issue 3), NICEE, IIT
Kanpur.( DOI: 10.1193/1.2360907)

9.

KulkarniP.B, PoojaRaut, Nikhil Agrawal (2013),


Analysis of Masonry Infilled R.C.Frame with &
without Opening Including Soft Storey by using
Equivalent Diagonal Strut Method, International
Journal of Scientific and Research Publications,
Volume
3,
Issue
9,
September
2013(http://www.ijsrp.org/research-paper0913/ijsrp-p2109.pdf)

(http://conf.ncree.org.tw/Proceedings/i0951012/dat
a/pdf/4icee-0064.pdf)
12. Putul Hauldar, Yogendra Singh, D. K. Paul(2012),
Effect of URM infills on seismic vulnerability of
Indian code designed RC frame buildings,
Earthquake Engineering and engineering
vibration, June 2012, vol 11, Issue 2, pp 233241.(doi10.1007/s11803-012-0113-5)

10. SuyamburajaArulselvan, K. Subramanian, E.B.


PerumalPillai and A.R. Santhakumar, (2007). RC
Infilled Frame-RC Plane Frame Interactions for
Seismic Resistance. Journal of Applied Sciences, 7:
942-95(doi: 10.3923/jas.2007.942.950)
11. Yogendra Singh &Dipankar Das (2006), Effect Of
URM Infills On Seismic Performance Of RC
Frame Buildings, International Conference on
Earthquake
Engineering,
Paper
no
64.

Page 68

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

COMPRESSIBILITY AND CHEMICAL COMPATIBILITY OF


CLAYEY SOIL-BENTONITE BACKFILLS
FOR SLURRY-TRENCH CUTOFF WALLS
Y. J. Du 1, R. D. Fan 2, S. Y. Liu 3, Krishna R. Reddy 4
1

Institute of Geotechnical Engineering, Southeast University, Nanjing, China


Email: duyanjun@seu.edu.cn
2
Institute of Geotechnical Engineering, Southeast University, Nanjing, China
Email:ridong.van@gmail.com
3
Institute of Geotechnical Engineering, Southeast University, Nanjing, China
Email: liusy@seu.edu.cn
4
Department of Civil and Materials Engineering, University of Illinois at Chicago, Illinois, USA
Email: kreddy@uic.edu
Abstract
Locally available clayey soil and high-quality calcium-bentonite can be considered as an alternative backfill for soilbentonite vertical cutoff walls when suitable sandy soil and sodium-bentonite is not available at some sites. However,
studies on the chemical compatibility of clayey soil-calcium bentonite backfills when exposed to heavy metal
contaminants are very limited. The objective of this study is to evaluate the compressibility and hydraulic conductivity
of clayey soil/Ca-bentonite backfills when exposed to heavy metal contaminants through a series of oedometer tests.
Kaolin is used as the control clayey soil and it is amended with 10% Ca-bentonite (by dry weight basis), to prepare the
kaolin-bentonite backfills. Lead nitrate (Pb(NO3)2) is selected as a representative of the group of heavy metals. The
backfills are thoroughly mixed with Pb(NO3)2 solution with lead concentration of 60, 120, and 600 mmol/L and left for
two weeks before testing. The results reveal that the lead concentration and initial water content significantly affect the
compressibility of the backfills. The void ratio at an effective vertical compression stress of 1 kPa is a useful
characteristic parameter to predict compression index for various soil-bentontie backfills in this study and reported in
previous studies. The lead-contaminated backfills shows a maximum twentyfold increase in hydraulic conductivity when
compared with the clean backfill. The hydraulic conductivity valises of the uncontaminated backfills and backfills that
exposed to 60 mmol/L Pb(NO3)2 solution are generally less than the typical regulatory limit of 10-9 m/s; whereas the
hydraulic conductivity corresponding to void ratio higher than 0.9 may fail to meet such typical regulatory limit when
the backfill exposed to Pb(NO3)2 solutions with concentration higher than 120 mmol/L. Two empirical methods, based
on the framework of Kozeny-Carman equation and Sivapullaiah et al.s method, are assessed to predict the hydraulic
conductivity of the backfills. The predicted hydraulic conductivity values using these methods are found to fall in the
range of 1/3 to 3 times those obtained from the oedometer tests.
Keywords: Compression, Chemical Compatibility, Cutoff wall, Lead; Soil-bentonite

1. Introduction
The contamination of subsurface soils and groundwater with high
amounts of heavy metals, resulting from improper past disposal
practices and accidental spills is a critical problem around the
world [1-3]. Soil-bentonite vertical cutoff walls installed with the
slurry trenching technology are used widely as in-situ barriers to
control the migration of subsurface contaminated groundwater in
the United States, Canada and Japan [4]. Soil-bentonite vertical
cutoff walls are often preferred in these countries because they
possess relatively low hydraulic conductivity (typically lower than
10-9) and are generally cost-effective [4].
The influences of the fines content (FC), bentonite content (BC),
gradation of sand and amendment type and content (e.g., zeolite
and activated carbon) on the compressibility and hydraulic
conductivity (k) as well as the relationship between the

compressibility and the lateral deflection of sandy soil/Nabentonite (sandy SB) and sand-clay backfills have been well
understood in previous studies [5-14]. Recently, clayey soil with
high-quality Ca-bentonite are considered as alternative to make up
soil-bentonite backfills when high-quality natural Na-bentonite
(e.g., Wyoming bentonite) is scarce, while Ca-bentonite is
abundant at some sites (e.g., mainland of China and India).
Previous studies show that the clayey soil/high-quality Cabentonite
(clayey
SB)
backfills
perform
competitive
compressibility and hydraulic conductivity when compared with
conventional sandy soil/Na-bentonite backfills [12].
The chemical compatibility of engineered barriers (e.g., bentonite
filter cakes, sandy SB backfill, geosynthetic clay liners and
compacted sand-bentonite mixtures), usually defined as the degree
of change in engineering properties, particularly, the deterioration
of hydraulic conductivity when engineered barriers are exposed to
chemical liquids of engineered barriers, is an important
consideration in geoenvironmental applications [15]. The chemical

* Y.

J. Du. Tel.: +86-25-83793729


Fax: +86-25-83795086; E-mail: duyanjun@seu.edu.cn

Page 69

Author et. al. / INDECS-15

compatibility reflects the suitability of a particular engineered


barrier material for a particular contaminated site [16].
Considerable changes in liquid limit and compression index may
occur over time when engineered barriers are exposed to salts,
heavy metals or organic solutions. In particular, a considerable
increase (e.g., 4 to 10-fold) in hydraulic conductivity may occur,
and it even could fail to meet the typical regulatory limit of 10 -9
m/s when sandy SB backfills and Na-bentonite are exposed to the
chemical liquids [5, 17-20]. The changes in the engineering
properties can mainly be attributed to: (1) the squeeze of diffuse
double layer of Montmorillonitic soil particles; (2) surface charges
and structural characterization for kaolinite soils; (3) the
dissolution of carbonate bonds or cementation between clay
particles; and (4) the complex geochemical and mineralogical
changes [21-27]. By date, studies on the impacts of lead (Pb) on the
compressibility and hydraulic conductivity of clayey soil/Cabentonite slurry-trench wall backfills are non-existent.
In this study, a series of Atterberg limit and oedometer tests are
conducted to: (1) investigate the chemical compatibility in terms of
liquid limit, compressibility and hydraulic conductivity of clayey
soil/Ca-bentonite backfills by using Pb-contaminated liquids with
various concentrations; (2) identify the effects of Pb concentration
and effective vertical consolidation stress levels on the hydraulic
conductivity of the clayey soil/Ca-bentonite backfills; and (3)
examine two proposed empirical methods to predict the hydraulic
conductivity for the Pb-contaminated clayey SB backfills, based on
the Sivapullaiah et al. (2000)s method [28] and framework of
Kozeny-Carman equation.

2. Materials and methods


2.1. Constituent soils
The clayey soil-bentonite (hereinafter referred to clayey SB)
backfills are comprised of commercial kaolin and high-quality Cabentonite provided by MUFENF mineral processing plant in
Zhenjiang City, China. Kaolin is a good control soil for laboratory
tests as the base component of the backfills in order to investigate
the effects of HM contaminants and bentonite content on the
compressibility and hydraulic conductivity. This is due to the facts
that: (1) it is one of the most common minerals found in natural
clays; (2) it has a low organic content, and a consistent and uniform
mineralogy; and, (3) as a typical nonswelling soils, it has a
relatively lower liquid limit and activity, while hydraulic
conductivity for kaolin is nearly 10 to 1000 times higher than that
for bentonite in general .
Table 1 shows the physico-chemical properties of the kaolin and
bentonite clays used for this study. ASTM standards were used to
evaluate these physico-chemical properties, expect that the specific
surface area was determined by the Ethylene Glycol Monoethyl
Ether (EGME) method suggested by Cerato and Lutenegger [30].
The testing methods Based on the x-ray diffraction analysis, the
basal spacing (001) of the montmorillonite in bentonite is identified
as 15.48 , indicating that the bentonite used in this study is Cabentonite.
2.2. Testing Liquids
The chemical liquids of preparing specimens in this study consisted
of deionized water (DIW) and lead nitrate (Pb(NO3)2) solutions.

The Pb(NO3)2 solutions are used as a representative of the group of


HM contaminants. The Pb(NO3)2 solutions were prepared by
dissolving the Pb(NO3)2 solids in powder form (analytical reagent
grade) in DIW with initial Pb concentration (C0) of 60, 120, and
600 mmol/L. These concentrations is in the same range of those
selected in previous studies [15, 20, 31]. The DIW used in this
study (EC = 3.3 S/cm; pH = 6.8) is classified as Type IV per
ASTM D1193 (EC < 5.0 S/cm; pH = 5 to 8). The measured lead
concentration, pH and electrical conductivity (EC) of the Pb(NO3)2
solutions is presented in Table 2. The lead concentration was
measured by an atomic absorption spectrometer (iCE 3300,
Thermo Fisher Scientific Inc.).
Table 1. Properties of constituent soils used in this study

Property
Specific gravity
Liquid limit (%)
Plastic limit (%)
Classification
Specific surface area (m2/g)
pH

Constituent soil
Kaolin
2.66
29.1
19.5
CL
45.7
8.7

Bentonite
2.73
331.4
88.2
CH
378.5
10.0

Table 2. Relevant properties of Pb(NO3)2 solutions used in this study


Targeted lead
concentration
(mmol/L)
60
120
600

Measured lead
concentration
(mmol/L)
59
121
629

pH
4.53
4.31
4.13

Electrical
conductivity
(mS/cm)
8.9
13.7
36.9

2.3. Backfill preparation


The clayey SB backfills were prepared by thoroughly mixing a
predetermined mass of base mixtures with a predetermined volume
of distilled water or Pb(NO3)2 solution for 30 minutes with soil-tosolution ratio of approximately 1:1.5 using a paddle mixer. The
bentonite content of the clayey SB base mixtures was controlled to
be 10% by dry weight basis, which is defined using Eq. 1. The
bentonite content selected in this study is slightly higher relative to
the Na-bentonite (4 to 7%) widely used in practice [5], but may
encompass the typical range of Ca-bentonite content for potential
use in practice.

BC

mben
mkao +mben

(1)

where mkao and mben are the mass of kaolin and bentonite in the
mixture (on dry mass basis), respectively.
After that, the backfills were left for 15 to 20 days, and they were
mixed for 15 minutes in each day. The backfills were then
centrifuged for 5 min at 3000 rpm to solid-liquid separation. The
pore-fluid was collected for the purpose of measuring Atterberg
limits and adjusting initial water content of specimens for
oedometer tests. This method for preparing contaminated soil
specimen was also used in previous studies [19, 32]; and it
describes the worst operating condition: the clayey SB backfill is
fully penetrated by HM contaminants. The specific gravity, liquid
limit, plastic limit, and pH of the clayey SB backfills were
measured based on ASTM standards; while Methylene Blue-Spot
Method was used to evaluate the specific surface area. It should be

Page 70

Author et. al. / INDECS-15

note that coal oil was substituted for distilled water in specific
gravity measurement. The physico-chemical properties of the
backfills are summarized in Table 3. The backfill ID B10Pbj herein
denotes a specimen spiked with Pb(NO3)2 solution with Pb
concentration of i mmol/L.
Table 3. Properties of the backfills tested in this study
Property
Specific gravity
Liquid limit (%)
Plastic limit (%)
Classification
Specific surface
area (m2/g)
pH

B10Pb0
2.67
53.3
25.9
CH

B10Pb60
2.66
40.8
21.7
CL

B10Pb120
2.69
39.8
21.9
CL

B10Pb600
2.75
38.4
21.2
CL

122.3

95.6

81.6

72.5

8.84

8.34

5.64

4.51

Two groups of clayey SB backfill specimens denoted as Model A


and Model B backfills with different initial water content were
prepared for oedometer tests. The initial water contents for the
Model A specimens were controlled to their corresponding liquid
limit. For the Model B specimens, the initial water contents were
controlled to the liquid limit value of their corresponding clean
(uncontaminated) one. A predetermined mass of the prepared
backfill was placed in a conventional consolidation ring made of
stainless steel with 61.8 mm in diameter and 20 mm in height. The
entrapped air bubbles were minimized by tapping the ring and
backfill at regular time intervals. The specimens were then
immersed in their corresponding pore-fluid that collected after
centrifugation for 48 h to reach saturation. An identical specimen
for all the specimens was prepared simultaneously and then
sacrificed for the measurement of the initial water content
immediately after saturation soaking step. Thus, the specimens in
this study are designated as B10PbiA or B10PbiB to denote a
specimen from Model A or Model B with bentonite content of 10%
that was mixed with i mmol/L Pb(NO3)2 solutions.
Table 4. Initial water content (w0), liquid limit (wL), void ratio at =1
kPa (e1) and compression index (Cc) of the specimens in this
study
Specimen ID
B10Pb0
B10Pb60A
B10Pb120A
B10Pb600A
B10Pb60B
B10Pb120B
B10Pb600B

w0 (%)
51.12
40.33
39.01
54.04
52.06
50.29
51.12

wL (%)
52.05
40.76
39.78
38.40
40.76
39.78
38.40

e1
1.72
1.39
1.31
1.22
1.51
1.48
1.44

Cc
0.41
0.28
0.26
0.26
0.31
0.30
0.29

2.4. Testing methods


The oedometer tests were conducted as per ASTM D 2435 , except
that the initial loading on the specimens was kept at 3.125 kPa.
This relatively low loading was chosen to avoid squeezing the soil
from the gap that exists between the specimen ring and porous
disks [12, 33]. The loading was then doubled for each incremental
step until a maximum loading of 1600 kPa was reached. The
duration of each loading was 24 hours. The top cap, guiding ring,
and base plate of oedometer cells are made of polyvinyl chloride
polymer (PVC) to prevent their chemical reactions with lead in
specimen and liquid in water bath.

At a given average effective vertical compression stress (ave),


defined as the mean value of two successive load increments,
values of hydraulic conductivity corresponding to higher than
6.25 kPa were estimated following Terzaghis one-dimensional
consolidation theory, as expressed by:
k cv mv w

(2)

where k is the hydraulic conductivity (m/s), cv is the coefficient of


consolidation (m2/s) determined using the Taylor (square-root-oftime) method in this study, mv is the coefficient of volume change
(kPa-1), and w is the unit weight of water (kN/m3). This method to
determine k value is extensively accepted [20, 28, 34-37]. The
method is likely to underestimate k value of clayey soils , but it is
used in this study for relative comparison of hydraulic conductivity
of various specimens. In recent, a preliminary comparative
assessment of k values estimated from the Terzaghis theory and
those measured directly from the falling-head method is made by
the authors. k values were directly measured for various types of
backfills, including clayey soil-bentonite, sand-bentonite and sandclay-bentonite backfills, using the falling-head tetsing procedure
recommended by Malusis et al. [8] and Hong et al. [11]. The results
indicate that the values of hydraulic conductivity calculated from
the Terzaghi's theory (ke) are well compared with those measured
using the falling-head method (k), especially for the soil-bentonite
backfills that possess relatively low k [13].

3. Results and discussion


3.1. Properties of the lead-contaminated backfills
The specific gravity, liquid limit, plastic limit, specific surface area,
pH values and soil classification of the backfills in this study are
listed in Table 3. It can be seen from Table 3 that the specific
gravity (Gs) increases with an increase in lead concentration; while
the liquid limit, plasticity index, specific surface area, and pH
values tend to decrease with increasing lead concentration. The
higher specific gravity of lead-contaminated backfills is due to a
much higher Gs value of lead. The liquid limit and plastic limit
values of clean (uncontaminated) backfills is approximately1.3 to
1.4 times and 1.2 times higher than these of lead-contaminated
backfills, respectively. In addition, the soil classification of the
backfills is changed from CH to CL after contaminated by lead
nitrate solutions, as shown in Table 3. The decreasing of liquid
limit, plasticity index, and specific surface area can be attributed to
the squeeze of diffuse double layers around bentonite particles,
which leads to the decrease in water-holding capacity and the
aggregation of clay particles. The result of specific surface area
reflects the decrease in water retention and sorption capacity, in
turn [39]. The pH values of the backfills decrease sharply when
lead concentration is higher than 120 mmol/L. This is due to the
fact that Pb(NO3)2 is a strong acid weak base salt; and therefore,
Pb2+ undergoes hydrolysis to give free hydrogen ions [40].
3.2. Compressibility
Figure 1 shows the void ratio (e) effective vertical compression
stress () compression curves on semi-logarithm scale for the
backfills tested in this study. The e-log() compression curves of
the specimens with an initial water content of their corresponding

Page 71

Author et. al. / INDECS-15

liquid limit (i.e., specimens in Model A) display a noticeable


inverse S shape. This inverse S shaped e-log() compression
curve and remolded yield stress (yr) were also observed in the
remolded natural clayey soils, Na-bentonite slurry, as well as
clayey SB backfills with an initial water content in the range of
approximately 0.5 to 2.0 times their corresponding liquid limits
[12, 33, 41]. This can be attributed to the existence of shear
strength of at relatively lower initial water content, which provides
certain resistance to sustain a low external compression stress, as
suggested in previous studies [33]. Thus, the compression index
(Cc) is determined from the linear portion of the e-log()
compression curve at the post-yield state in this study. The results
from Table 4 shows that compression index increases with an
increase in liquid limit. However, compression index of specimens
in Model is 1.1 to 1.2 times higher than that of specimens in Model
A for a given lead concentration, which indicating the effect of
initial water content on the compression index of leadcontaminated clayey SB backfills is considerable.

which can be generalized using a dimensionless parameter, e00.6eL,


as expressed by Eq. 4. The variations of compression index with
void ratio at = 1 kPa (e1) obtained in this study and these
reported in previous studies, including clayey soil-bentonite , sandbentonite and sand-clay backfills [10-14], is presented in Fig. 3.
The result indicates that the relationship between Cc and e1 for the
specimens tested in this study is very similar to the empirical
equation obtained by Fan et al. [12].

Fig. 2. Relationship between lead concentration (CPb) and compression


index ratio (Ccc/Cc0)

Fig. 1. e-log() compression curves of the backfills

Figure 2 presents variations of compression index ratio with lead


concentration for the specimens tested in this study. The
compression index ratio is defined as the ratio of the Cc obtained
from the lead-contaminated specimens (Ccc) to that based on
uncontaminated specimens (Cc0). The result indicates that when
lead concentration increases to 60 mmol/L, the compression index
ratio decreases sharply to 0.68 0.77; whereas it tends to decrease
slightly or keep remain stable when lead concentration is higher
than 120 mmol/L. Thus, it can be concluded that the compression
index of the clayey SB backfills is significantly affected by lead
contaminants at relatively low range of concentration; while the a
further increase in lead concentration had limited effect on
compression index. In addition, the compression index ratio of
specimens in Model B is approximately 1.1 1.2 times larger than
that of specimens in Model A due to the effect of initial water
content.
Fan et al. [12] suggested that there existed a unique empirical
relationship between compression index and void ratio at = 1
kPa (e1) for various types of soil-bentonite backfills, and the
proposed Cc - e1 relationship can be expressed by a quadratic
equation (Eq. 3). In addition, Fan et al. [12] suggested that the
single parameter, e1, in Eq 3 reflects the effects of initial water
content (i.e., initial void ratio, e0) and liquid limit (i.e., void ratio at
liquid limit, eL) on the compressibility of soil-bentonite backfills,

Fig. 3. Relationship between void ratio at =1 kPa (e1) and


compression index (Cc)

Figure. 4 presents the result of the relationship between e00.6eL and


e1 of the specimens tested in this study as well as results of clean
clayey soil-bentonite backfills and natural clays reported in
previous studies [11-13, 32, 42]. The predicted e1 using the
proposed empirical equation reported by Fan et al. [12] are found
to fall in the range of 0.7 to 1.8 times those evaluated based on elog() compression curve. In addition, multiple linear regression is
used to establish the correlation among e1, e0 and eL, as expressed
by Eq. 5 (R2 = 0.953). The predicted e1 using Eq. 5 falls in the
range of 0.7 to 1.35 times those evaluated ones, indicating a better
result of prediction when compared with predicted e1 values using
Eq. 4. Thus, it is concluded that Eqs. 3 and 5 can be used to
preliminarily estimate Cc value of clean clayey SB backfills and
these exposed to Pb(NO3)2 solutions.

Page 72

Author et. al. / INDECS-15


Cc 0.13e1 0.056e12

e1 1.168 e00.6 eL

(3)
0.772

e1 0.72eL +0.93e0 -0.78

initial water content has limited effects on the hydraulic


conductivity of lead-contaminated specimens tested in this study.

(4)
(5)

where e1 is the void ratio at = 1 kPa; e0 is the initial void ratio;


and eL is the void ratio at liquid limit.

Fig. 5. Variations in coefficient of consolidation (cv) with average


effective vertical compression stress (ave) for the backfills

Fig. 4. Relationship between predicted void ratio at =1 kPa (e1,p) and


void ratio at =1 kPa obtained from e-log() compression curve (e1)

Figures 5 and 6 presents the variations in coefficient of


consolidation (cv) as well as coefficient of volume change (mv),
with average effective vertical compression stress on duallogarithmic scale. The cv value is evaluated based on the Taylors
(square-root-of-time) methods. The results indicate that the cv value
tends to increase noticeably with an increase in average effective
vertical compression stress; while the mv value tends to decrease
considerably with increasing average effective vertical
compression stress for all the backfills. This result is similar to the
trend from uncontaminated sandy SB backfills [10] and zeoliteamended clayey-bentonite backfills [13]. The effect of lead
concentration on both cv and mv values for the lead-contaminated
backfills is insignificant. In addition, it is evident that the
coefficient of consolidation of uncontaminated backfills is
considerably larger than that of lead-contaminated ones. However,
the resulting cv values of the backfills in Model B that exposed to
120 and 600 mmol/L Pb(NO3)2 solutions tend to higher than others.

Fig. 6. Variations in coefficient of volume change (mv) with average


effective vertical compression stress (ave) for the backfills

3.3. Hydraulic conductivity


Figure 7 presents the variations of hydraulic conductivity (k) with
void ration (e) on semi-logarithmic scale. It is can be seen from
Fig. 7 that both e - log(k) relationships are approximately linear.
The hydraulic conductivity of lead-contaminated specimens is
approximately 4 to 20 times higher than that of uncontaminated
specimen. The hydraulic conductivity values for uncontaminated
specimens and specimens that exposed to 60 mmol/L Pb(NO3)2
solution are generally less than the typical regulatory limit of 10 -9
m/s; whereas the hydraulic conductivity of the specimens exposed
to Pb(NO3)2 solutions with lead concentration higher than 120
mmol/L fail to meet such typical regulatory limit until void ratio is
lower than approximately 0.9. In addition, the result indicates that

Fig. 7. Relationship between hydraulic conductivity (k) and void ratio


(e) on a semi-logarithm scale for the backfills in this study

Page 73

Author et. al. / INDECS-15

To better understand the effect of lead concentration on hydraulic


conductivity of the specimens tested in this study, the relationships
between slat concentration (C) and hydraulic conductivity
corresponding to void ratio of 0.79 to 0.94 are presented in Fig. 8.
In addition, for the purpose of comparing the results obtained in
this study with that of sandy SB backfills, the results reported by
Malusis et al. [15] are presented in Fig. 8, in which sandy SB
backfills were permeated with 10 to 1000 mmol/L calcium chloride
(CaCl2). The designations of SW101, NG, and MSB in Fig.
8 are the same meanings reported by Malusis et al. [15] . The
reason for choosing k values corresponding to e = 0.79 0.94 is
because this range of void ratio are available from all specimens
tested in this study; and therefore, it allows for a comparison
among the different specimens. The result indicates that the
hydraulic conductivity of the clayey SB backfills as well as the
sandy SB backfills reported by Malusis et al. [15] tends to increase
significantly once they are exposed to salt solutions; and the k-C
relationship remains stable when cation concentration is relatively
high. Thus, there exists a threshold concentration, beyond which
the effect of salt concentration on the hydraulic conductivity is
insignificant. The critical concentration for the backfills with
Pb(NO3)2 solution tested in this study as well as sandy SB backfills
with CaCl2 solution reported by Malusis et al. [15] is
approximately in the range of 50 to120 mmol/L.

Fig. 8. Relationship between metal concentration (C) and hydraulic


conductivity (k) corresponding to void ratio of 0.79 to 0.94

Figure 9 presents the variations of hydraulic conductivity ratio with


average effective compression stress on dual-logarithmic scale to
investigate the effect of stress state on the hydraulic conductivity.
The hydraulic conductivity ratio is defined as the ratio of the k
obtained from the lead-contaminated specimens (kc) to that based
on uncontaminated specimens (kw) for a given bentonite content
and a similar void ratio, and it is denoted by kc/kw. The hydraulic
conductivity ratio tends to increase with an increase in lead
concentration; whereas the increase in hydraulic conductivity ratio
tends to diminish with increasing average effective compression
stress, which is more noticeable when backfill is exposed to
Pb(NO3)2 solution with relatively high lead concentration
(approximately 120 and 600 mmol/L). This is due to the fact that
the decrease in void ratio is likely to increase the overall sizes of
the individual pores within backfill, thereby increasing the
tortuosity of seepage path. Thus, the increase in stress state can
enhance the chemical compatibility (i.e., the decrease in hydraulic

conductivity ratio) of clay SB backfills. However, earth pressure in


a soil-bentonite backfill is unlikely to thoroughly eliminate the
negative effect of HMs contaminants on the hydraulic conductivity
considering that the lateral earth pressure might be lower than 90
kPa [7].

Fig. 9. Relationship between average effective compression stress (ave)


and hydraulic conductivity ratio of (kC/k0) of the backfills

When the bentonite is contaminated by the Pb(NO3)2 solution, the


swell potential (i.e., repulsive stress between clay particles) of
bentonite significantly reduces due to the exchange of readily
exchangeable cations with Pb ions and consequent contraction of
diffuse double layer of bentonite particles. The degree of
contraction of the diffuse double layer or reduction in swell
potential of the bentonite increases with increasing Pb
concentration . In addition, the reduced soil pH, i.e., increased
concentration of hydrogen (H+) ions, with respect to the elevated
Pb concentration is also attributable to the contraction of the
diffuse double layer based on the Gouy-Chapman theory.
Nevertheless, it is noted that the limited decrease in the pH of
backfills with respect to Pb concentration (see Table 3) may have
insignificant effect on the hydraulic conductivity of backfills unless
the pH value is less than 2.0 [43].
The engineering behavior of metal-treated kaolin is dominantly
controlled by its fabric [23]. The influence of metal concentration
on the properties of kaolin crucially depends on the relationship
between pH at isoelectric point for edge (IEP edge, typical value is
5.8 for commercial kaolin) and soil pH. When the soil pH is >
IEPedge and the metal concentration is lower than the "convergence
concentration", a deflocculated fabric forms, leading to a slight
reduction in the hydraulic conductivity [27, 42, 44]. When the
metal concentration is higher than the convergence concentration,
the face-to-face (FF) aggregation or FF aggregate in edge-to-face
(EF) floc would form regardless of the soil pH, resulting in an
increased hydraulic conductivity as compared to the clean kaolin.
The Pb concentrations of the backfills presented in this study are
deemed higher than the convergence concentration for the two
valence-metals concentration (e.g., 2 mmol/L for CaCl2 solution
[27]). Therefore, hydraulic conductivity of the kaolin clay
composed in the backfills would decrease with the increase in the
Pb concentration, which is similar to that for bentonite as
mentioned above.
The hydraulic conductivity values in this study are estimated
indirectly based on the oedometer test results. Direct measurement

Page 74

Author et. al. / INDECS-15

based on either the rigid-wall permeameter or the flexible-wall


permeameter tests is recommended. On-going additional research
is aimed at addressing all of these limitations.

3.4. Predictive method for k


Several methods have been reported in published literature for
predicting hydraulic conductivity of clays, and most of them are
originated from Kozeny-Carman (KC) equation, Taylor (1948)s
method and Nagarajs method [45]. In this study, two empirical
equations are assessed to predict the k of thePb-contaminated
clayey SB backfills based also on Sivapullaiah et al. (2000)s
method [28] and the framework of KC equation. The Sivapullaiah
et al. (2000)s method is initially used to predict k of sandbentonite mixtures, while the KC equation has been used to predict
hydraulic conductivity for most saturated soils including sandy
soils and natural clays [46, 47]. Recently, an empirical equation
based on the framework of KC equation is proposed to predict k of
clean (uncontaminated) clayey SB backfills [12].
Sivapullaiah et al. (2000)s method is based on the assumption that
e-log (k) relationship is linear. The result of e-log (k) relationship
for the backfills in this study is therefore expressed as:
e S k log k I k

S k 1.9 wL 0.152

(7)

I k 23.5wL 2.96

(8)

e 23.5wL 2.96
1.9 wL 0.152

e3
log k 1.18log 6
11.49
wL 1 e

(11)

where wL is in %.

(6)

where Sk and Ik are the slope and intercept of the equation,


respectively; and k is the hydraulic conductivity (m/s)
corresponding to a given e value. As presented in Fig. 10(a) and
10(b), both the Sk-wL and Ik-wL relationships are approximately
linear, as expressed by Eqs. 7 and 8. The R2 values for Eqs. 7 and 8
are 0.951 and 0.971, respectively. When Eqs. 7 and 8 are
substituted in Eq. 6, it yields Eq. 9, which predicts the hydraulic
conductivity of the backfills in this study.

log kp

study can be described using a unique linear function, which is


quite similar to the proposed equation obtained from
uncontaminated clayey soil-bentonite backfills reported by Fan et
al. [12] . The log(k) - log[e3/(wL6(1+e))] relationship determined
using a Least-Square-Root method is expressed by Eq. 11 with R2
value of 0.953.

Fig. 10. Relationship between liquid limit (wL) and slope (Sk) and
intercept (Ik) form the e-log(k) relationship expressed using Eq. 6: (a)
slope (Sk) and (b) intercept (Ik)

(9)

where kp is the predicted hydraulic conductivity in m/s and wL is in


%.
The KC equation is commonly used to estimate hydraulic
conductivity and it is expressed by:

k Cs

w
e3

2
w SSA 1 e

(10)

where Cs is the shape coefficient that reflects the pore shape and
tortuosity of the channels, w is the unit weight of water, w is the
dynamic viscosity of water, SSA is the specific surface area, and e
is the void ratio. Fan et al. [12] suggested that the SSA value in the
original form of KC equation could be replaced by a function of wL
(wL3) considering that wL is a basic soil parameter to be determined
using conventional laboratory testing methods. In addition, Sanzeni
et al. [47] suggested that an inaccurate SSA value would lead to a
significant discrepancy in the predicted k of clayey soils.
Figure 11 presents the k - e3/(wL6(1+e)) relationship on duallogarithmic scale. The results show that data of log(k) log[e3/(wL6(1+e))] relationship for all specimens tested in this

Fig. 11. Relationship between hydraulic conductivity (k) and


dimensionless parameter e3/[wL6(1+e)] based on the original form of
Kozeny-Carman equation on dual-logarithm scale

The predicted hydraulic conductivity (kp) values using Eqs. 9 and


11 are compared with those estimated from the oedometer tests, as
shown in Fig. 12. It can be seen from Fig. 12 that the kp values are
generally in the range of 1/3 to 3 times the k measured during the
oedometer tests. In addition, the results imply that liquid limit is an
important property that reflects the change in k value due to the
impact of heavy metal contaminants. Therefore, it is concluded that
both proposed method are reasonably suitable to predict k for leadcontaminated clayey soil-bentonite backfills tested in this study.

Page 75

Author et. al. / INDECS-15

diffuse double layer of bentonite and the aggregation of


kaolin particles caused by the lead contamination.

Acknowledgments
The authors are grateful for the financial support of the National
Natural Science Foundation of China (Grant No. 51278100,
41330641 and 41472258), Natural Science Foundation of Jiangsu
Province (BK2012022), Key Laboratory of Geotechnical and
Underground Engineering Foundation (Tongji University) (Grant
No. KLE-TJGE-B1202), and Scientific Research Foundation of
Graduate School of Southeast University (Grant No. YBJJ1343).

References
Fig. 12. Relationship between hydraulic conductivity (k) evaluated
from oedometer tests and predicted hydraulic conductivity (kP) using
Eqs. 9 and 11

3. Conclusions
Clayey soil/Ca-bentonite backfill is an alternative to make up
soil-bentonite backfills for slurry cutoff walls when high-quality
natural Na-bentonite is scarce. However, the chemical
compatibility of clayey soil/Ca-bentonite backfills when
contaminated by lead remains unknown. This study investigated
the chemical compatibility of clayey soil/Ca-bentonite backfills in
terms of changes in liquid limit, compressibility and hydraulic
conductivity with respect to lead concentration. Based on the
results presented in this study, the following conclusions can be
drawn:
1.
The liquid limit, plastic limit, specific surface area and pH
decreased with increasing lead concentration. The soil
classification of the backfills is shifted from high-plasticity
clay (CH) to low-plasticity clay (CL) after Pb-contamination.
2.
The compression index decreased considerably with
increasing lead concentration. The unique relationship
between compression index (Cc) and void ratio at 1 kPa (e1)
for the lead-contaminated and clean backfills was expressed
by: Cc = 0.125e1 + 0.055e12.
3.
The hydraulic conductivity for the clayey soil/Ca-bentonite
backfills were significantly affected by lead concentration
and bentonite content. The hydraulic conductivity increased
significantly with increasing lead concentration; it even
exceeded the typical regulatory limit of 10-9 m/s when the Pb
concentration was relatively high. The hydraulic conductivity
ratio is found to increase with increasing bentonite content at
a given lead concentration.
4.
Two empirical equations, namely the Sivapullaiah et al.s
(2000) method and the proposed method based on the
framework of KozenyCarman equation, were applied to
predict the hydraulic conductivity values for the leadcontaminated clayey soil/Ca-bentonite backfills. The results
indicate that the predicted hydraulic conductivity values
using the two proposed equations were within a range of 1/2
to 2 times those obtained from the oedometer tests.
5.
The changes in the liquid limit, compression index and
hydraulic conductivity of the backfills with respect to the lead
concentration are mainly attributed to the contraction of

[1] Du YJ, Wei ML, JIN F, Liu ZB. Stressstrain relation and
strength characteristics of cement treated zinc-contaminated
clay. Engineering Geology, 2013; 167: 20-26.
[2] Du YJ, Wei ML, Reddy KR, JIN F, Wu HL Liu ZB. New
phosphate-based binder for stabilization of soils contaminated
with heavy metals: Leaching, strength and microstructure
characterization. Journal of Environmental Management,
2014; 146: 179-188.
[3] Du YJ, Wei ML, Reddy KR, Liu ZP, Jin F. Effect of acid rain
pH on leaching behavior of cement stabilized leadcontaminated soil. Journal of hazardous materials, 2014; 271:
131-140.
[4] Sharma HD, Reddy KR. Geoenvironmental engineering: site
remediation, waste containment, and emerging waste
management technologies. John Wiley & Sons, Inc., 2004.
[5] Evans JC, Costa MJ, Cooley B. The state-of-stress in soilbentonite slurry trench cutoff walls. Proceedings of the
Geoenvironment 2000: Characterization, Containment,
Remediation, and Performance in Environmental Geotechnics.
New Orleans, Louisiana, United States, 1995.
[6] Filz GM. Consolidation stresses in soil-bentonite backfilled
trenches. Environmental geotechnics: proceedings of the
second International Congress on Environmental Geotechnics.
Osaka, Japan, 1996.
[7] Ruffing DG, Evans JC, Malusis MA. Prediction of earth
pressures in soil-bentonite cutoff walls. GeoFlorida 2010:
Advances in Analysis, Modeling & Design. Orlando, Florida,
United States, 2010.
[8] Malusis MA, Barben EJ, Evans JC. Hydraulic conductivity
and compressibility of soil-bentonite backfill amended with
activated
carbon.
Journal
of
Geotechnical
and
Geoenvironmental Engineering, ASCE, 2009; 135: 664-672.
[9] Castelbaum D, Shackelford CD. Hydraulic conductivity of
bentonite slurry mixed sands. Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 2009; 135: 1941-1956.
[10] Yeo S-S, Shackelford CD, Evans JC. Consolidation and
hydraulic conductivity of nine model soil-bentonite backfills.
Journal of Geotechnical and Geoenvironmental Engineering,
ASCE, 2005; 131: 1189-1198.

Page 76

Author et. al. / INDECS-15

[11] Hong CS, Shackelford CD, Malusis MA. Consolidation and


hydraulic conductivity of zeolite amended soil-bentonite
backfills. Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, 2011; 138: 15-25.
[12] Fan RD, Du YJ, Reddy KR, Liu SY, Yang YL.
Compressibility and hydraulic conductivity of clayey soil
mixed with calcium bentonite for slurry wall backfill: Initial
assessment. Applied Clay Science, 2014; 101: 119-127.
[13] Du YJ, Fan RD, Liu SY, Reddy KR, Jin F. Workability,
compressibility and hydraulic conductivity of zeolite-amended
clayey soil/calcium-bentonite backfills for slurry-trench cutoff
walls. Engineering Geology, 2015; accepted.
[14] FAN R D, DU Y J, LIU S Y, Chen ZB. Compressibility and
hydraulic conductivity of sand/clay-bentonite backfills.
Proceedings of the Geoshanghai2014. Shanghai, China, 2014.
[15] Malusis MA, Mckeehan MD. Chemical compatibility of
model soil-bentonite backfill containing multiswellable
bentonite. Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, 2013; 139: 189-198.
[16] Garvin SL, Hayles CS. The chemical compatibility of cementbentonite cut-off wall material. Construction and Building
Materials, 1999; 13: 329-341.
[17] Jo HY, Katsumi T, Benson C H, Edil, TB. Hydraulic
conductivity and swelling of nonprehydrated GCLs permeated
with single-species salt solutions. Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 2001; 127: 557-567.
[18] Lo IMC, Yang X. Use of organoclay as secondary
containment for gasoline storage tanks. Journal of
environmental engineering, 2001; 127: 154-161.

[26] Gratchev I, Towhata I. Compressibility of natural soils


subjected to long-term acidic contamination. Environmental
Earth Sciences, 2011; 64: 193-200.
[27] Li Z, Katsumi T, Inui T, Takai A. Fabric effect on hydraulic
conductivity of kaolin under different chemical and
biochemical conditions. Soils and Foundations, 2013; 53: 680691.
[28] Sivapullaiah P, Sridharan A, Stalin V. Hydraulic conductivity
of bentonite-sand mixtures. Canadian Geotechnical Journal,
2000; 37: 406-413.
[29] Mitchell J, Soga K. Fundamentals of soil behaviour (3rd ed).
John Wiley & Sons, Inc, 2005.
[30] Cerato AB, Luteneggerl AJ. Determination of surface area of
fine-grained soils by the ethylene glycol monoethyl ether
(EGME) method. Geotechnical Testing Journal, ASTM, 2002;
25: 315-321.
[31] Lee J-M, Shackelford C D, Benson C H, Jo H-Y, Edil, TB..
Correlating index properties and hydraulic conductivity of
geosynthetic clay liners. Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 2005; 131: 1319-1329.
[32] Yukselen-Aksoy Y, Kaya A, ren AH. Seawater effect on
consistency limits and compressibility characteristics of clays.
Engineering Geology, 2008; 102: 54-61.
[33] Hong ZS, Yin J, Cui YJ. Compression behaviour of
reconstituted soils at high initial water contents. Gotechnique,
2010; 60: 691-700.
[34] Chai J, Miura N, Zhu H. Compression and consolidation
characteristics of structured natural clay. Canadian
Geotechnical Journal, 2004; 41: 1250-1258.

[19] Mishra AK, Ohtsubo M, Li LY, Higashi, T, Park J. Effect of


salt of various concentrations on liquid limit, and hydraulic
conductivity
of
different
soil-bentonite
mixtures.
Environmental Geology, 2009; 57: 1145-1153.

[35] Horpibulsuk S, Shibuya S, Fuenkajorn K, Wanchai K.


Assessment of engineering properties of Bangkok clay.
Canadian Geotechnical Journal, 2007; 44: 173-187.

[20] Yong RN, Ouhadi VR, Goodarzi AR. Effect of Cu2+ ions and
buffering capacity on smectite microstructure and
performance. Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, 2009; 135: 1981-1985.

[36] Mishra A K, Ohtsubo M, Li L, Higashi T. Controlling factors


of the swelling of various bentonites and their correlations
with the hydraulic conductivity of soil-bentonite mixtures.
Applied Clay Science, 2011; 52: 78-84.

[21] Yanful EK, Shikatani KS, Quirt DH. Hydraulic conductivity


of natural soils permeated with acid mine drainage. Canadian
Geotechnical Journal, 1995; 32: 624-646.

[37] Watabe Y, Yamada K, Saitoh K. Hydraulic conductivity and


compressibility of mixtures of Nagoya clay with sand or
bentonite. Gotechnique, 2011; 61: 211-219.

[22] Kashir M, Yanful E. Hydraulic conductivity of bentonite


permeated with acid mine drainage. Canadian Geotechnical
Journal, 2001; 38: 1034-1048.

[38] Chapuis RP. Predicting the saturated hydraulic conductivity of


soils: a review. Bulletin of Engineering Geology and the
Environment, 2012; 71: 401-434.

[23] Wang YH, Siu WK. Structure characteristics and mechanical


properties of kaolinite soils. I. Surface charges and structural
characterizations. Canadian Geotechnical Journal, 2006; 43:
587-600.

[39] Yukselen Y, Kaya A. Comparison of methods for determining


specific surface area of soils. Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 2006; 7: 931-936.

[24] Shaw S A, Jim Hendry M. Geochemical and mineralogical


impacts of H2SO4 on clays between pH 5.0 and -3.0. Applied
Geochemistry, 2009; 24: 333-345.
[25] Montoro M A, Francisca F M. Soil permeability Controlled by
particle-fluid interaction. Geotechnical and Geological
Engineering, 2010; 28: 851-864.

[40] Du Y J, Hayashi S. A study on sorption properties of Cd2+ on


Ariake clay for evaluating its potential use as a landfill barrier
material. Applied Clay Science, 2006; 32: 14-24.
[41] Sreedharan V, Puvvadi S. Compressibility behaviour of
bentonite and organically modified bentonite slurry.
Gotechnique, 2013; 63: 876-879.

Page 77

Author et. al. / INDECS-15

[42] Horpibulsuk S, Yangsukkaseam N, Chinkulkijniwat A, Du YJ.


Compressibility and permeability of Bangkok clay compared
with kaolinite and bentonite. Applied Clay Science, 2011; 52:
150-159.
[43] D'Appolonia D, Ryan CR. Soil-Bentonite slurry trench cut-off
walls. Geotechnical Exhibition and Technical Conferenc.
Chicago, Illinois, 1979.
[44] Palomino AM, Santamarina JC. Fabric map for kaolinite:
Effects of pH and ionic concentration on behavior. Clays and
Clay Minerals, 2005; 53: 209-222.

[45] Nagaraj TS, Miura N. Soft clay behaviour: Analysis and


assessment. The Netherlands: A A Balkema Publishers, 2001.
[46] Chapuis RP, Aubertin M. On the use of the Kozeny-Carman
equation to predict the hydraulic conductivity of soils.
Canadian Geotechnical Journal, 2003; 40: 616-628.
[47] Sanzeni A, Colleselli F, Grazioli D. Specific surface and
hydraulic conductivity of fine-grained soils. Journal of
Geotechnical and Geoenvironmental Engineering, ASCE,
2013; 139: 1828-1832.

Page 78

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

STABILITY ASPECTS PERFORMANCE OF NON-PERFORATED VS PERFORATED


QUARTER CIRCLE BREAKWATERS A STUDY

Prof. Arkal Vittal Hegde


Department of Applied Mechanics and Hydraulics, NITK, Surathkal. Mangaluru, 575025, India,
Abstract

Quarter circle breakwater (QCB) is one of the few new types of breakwaters developed in recent
times. It is a composite breakwater made up of a quarter circle shaped caisson with concrete base
slab standing on a rubble mound base. These breakwaters are technically advanced, economically
viable and aesthetically pleasing. The paper discusses the experiments conducted on non-perforated
QCB as well as seaside perforated QCB with spacing to diameter ratio of perforations of 8, in order
to determine the critical (minimum) weight required to resist the sliding forces caused by wave
action. Critical weight required to prevent sliding of the structure for different wave and structure
specific parameters was expressed in terms of dimensionless stability parameter, and it is a function
of incident wave steepness and depth parameter. The investigations were carried out for the
breakwater models, and the variations were recorded graphically using these non-dimensional
parameters, obtained by applying Buckinghams theorem. The wave climate off the Mangaluru
coast of Karnataka state in India was considered for wave loading on the structure. Incident wave
heights used in the flume varied from 3 to 18 cm, wave periods ranged from 1.4 s to 2.5 s, water
depths of 35 cm, 40 cm and 45 cm were used in the experiments. The adopted model scale was of
1:30. It was found that non-dimensional stability parameter decreases as incident wave steepness
increases for both non-perforated and seaside perforated breakwaters for all ranges of d/gT2 values.
Also, it was found that critical weight required for sliding stability is more in non-perforated QBW
as compared to seaside perforated QCB. Typically, there was a decrease of 12 to 20% in W/Hi2
values for perforated structures, depending upon the d/gT2 value.
Keywords: Quarter circle breakwater, Incident wave steepness, Dimensionless stability parameter, Dimensionless depth
parameter, Buckingham theorem.
1. Introduction
More than 40% of global population lives in the coastal areas
because of their socio-economic features. The protection of coastal
area is a continuous challenge for the coastal engineers and coastal
protection works such as the seawalls, offshore breakwaters, groins
and beach nourishment are being designed and installed to
overcome the coastal problems. As all of these structures are not
completely successful in fulfilling their structural and functional
requirements, research is going on in the domain of coastal defense
structures with the development of breakwaters, like berm
breakwaters, floating pipe breakwaters, semicircular and quarter
circle breakwaters and other types of protection structures.
Emerged perforated quarter circle breakwater (EPQCB) is a surface
piercing precast reinforced concrete caisson having perforations on
its seaside circular face with a bottom slab resting on a low-mound
rubble base. The concept of perforated seaside face was derived to
absorb most of the incident wave energy and the vertical wall on
rear side is used to avoid wave transmission to the lee side. Figs. 1

and 2 show emerged, non-perforated and perforated quarter circle


breakwater sections respectively with a free board, which means
crest level of the structure is raised above the still water level and
further, in the present studies there is no wave overtopping allowed
over QCB crest. In other words, the crest is so elevated as to cause
no wave overtopping. The paper discusses the sliding stability of
the breakwater, which is a function of incident wave height (Hi),
water depth (d), radius of the caisson (R), ratio of spacing to
diameter of perforations (S/D), height of the structure (ht) and
wave period (T).To ensure publication quality and uniformity, the
following requirements have been prepared to assist authors in
preparing papers for the Conference. If these requirements are not
followed, papers will be sent back for revision and re-submission.

2. Literature review
Qie et al. (2013) have conducted studies on partial coefficients for
design of quarter circle caisson breakwater for the purposes of safe
design of the structure. Jiang et al. (2008) have conducted twodimensional (vertical) wave numerical model and also physical
model studies to research the performances of quarter circle
breakwater (QCB) by comparing the hydraulic behavior of

Page 79

Arkal Vittal Hegde. / INDECS-15

Fig. 1 (a)

Fig.1 (b)
Fig. 1 (a) Cross section of emerged non-perforated QBW model, (b) Cross section of EPQCB model (all dimensions are in m)

Fig. 2 Longitudinal section of wave flume used with details

Page 80

Arkal Vittal Hegde/ INDECS-15


the increase in force is stabilized by increasing the weight of

3. RESULTS AND DISCUSSION


The experiments were conducted in the two dimensional
monochromatic wave flume of the department in order to
determine the critical weight required to resist the sliding of the
non-overtopping EPQCB model. Two models were studied, one

breakwater by adding additional weight into the caisson. In all of


the cases, it is found that the non-dimensional stability parameter
(W/Hi2) decreases with the increase of incident wave steepness
(Hi/gT2).

non-perforated and other seaside perforated model with S/D=8.

From Figures 5 and 6, it may be seen that as the dimensionless

Variations in critical weight required for sliding stability with

depth parameter d/gT2 increases, the value of W/Hi2 also increases

different wave specific and structure specific parameters were

for the given wave steepness. This is because higher the water

studied, and were recorded graphically using non-dimensional

depth, greater is the area of the QBW model structure exposed to

parameters

using

wave action, and hence, the increase in d/gT2 imparts more force

Buckinghams theorem. Effect of perforations on sliding stability

and therefore increase in W/Hi2 means more structure weight is

was studied by comparing the results of two models. Non-

required for stability in larger depths.

obtained

from

dimensional

analysis

dimensional stability parameter W/Hi2 was used to represent the


stability of the structure. The equations for the best-fit plots of
W/Hi2 Vs wave steepness Hi/gT2 was found out by least square
regression. It was found that logarithmic fit gave the maximum
value of R2 in all the cases. The results obtained from the
experimental work, and discussions on same are presented in

Influence of perforations
The effect of perforations plays a crucial role in reducing pressures,
especially closer to SWL [Isaacson and Subbiah (1991);
Mallayachari and Sundar (1995)]. Generally, the value of W/H i2
decreases with the increase in percentage of perforations because,

subsequent sections.

the increase in percentage of perforations on the seaside of the


Influence of incident wave steepness and depth parameter

QCB model generates more turbulence inside the caisson chamber,


which causes more energy dissipation. Increase in percentage of

The graphs of non-dimensional stability parameter (W/Hi2)

perforations, generates less force of impact on the QCB caisson,

against the incident wave steepness (Hi/gT2) are plotted for

hence the critical weight required to resist sliding would be

different ranges of values of dimensionless depth parameter

reduced. The variation of stability parameter with incident wave

(d/gT2), for a constant radius to breakwater height ratio (R/ht) of

steepness for each range of d/gT2 values for a non-perforated QCB

0.902. Figures 3 and 4 respectively represent the variation of

model and a seaside perforated QCB model with S/D=8, is

W/Hi2 with incident wave steepness Hi/gT2 for all four ranges of

discussed here below.

d/gT2 values, for constant R/ht=0.902 for an emerged nonperforated QCB and an emerged perforated QCB with S/D=8.

It may be seen that the non-dimensional stability parameter W/Hi2


decreases with increase in the percentage of perforations. The

From Figures 3 and 4 above, it may be noted that for all the ranges

percentage reduction in W/Hi2 with perforations called as

of depth parameter values for the two models, as the incident wave

%[W/Hi2]red for all the ranges of d/gT2 is calculated from equation

steepness

parameter

(1) in order to investigate the effect of perforations. Equation (1) is

decreases, i.e. critical weight required for sliding stability

valid for the range of values of variables used in the present

decreases. This is due to fact that long period waves (low

experiments.

increases,

the

dimensionless

stability

steepness) exert more force on the caisson, demanding higher


critical weight, and short period (steep) waves transfer less force,

%[W/Hi2]red = {([W/Hi2]non-perforated - [W/Hi2]S/D=8) / ([W/Hi2]non-

hence low critical weight is required. The disturbance caused by

perforated)}*100

.......(1)

* Corresponding

author. Tel.: +1234567890


Fax: +9876543210; E-mail: Author.Name@iasks.org
2010 International Association for Sharing Knowledge and Sustainability.

Page 81

Arkal Vittal Hegde. / INDECS-15

Fig. 3 Variation of stability parameter with incident wave steepness and depth parameter for a constant R/ht=0.902 for a nonperforated QBW model

Page 82

Arkal Vittal Hegde. / INDECS-15

Fig. 4 Variation of stability parameter with incident wave steepness for a constant R/ht=0.902 for seaside perforated QBW
model with S/D=8

Page 83

Arkal Vittal Hegde. / INDECS-15

Fig. 5 Variation of stability parameter with incident wave steepness and depth parameter for a constant R/ht of 0.902 for a seaside
perforated QBW model of S/D ratio 8

Fig. 6 Variation of stability parameter with incident wave steepness and depth parameter for a constant R/ht of 0.902 for a non-perforated
QBW model

Table 1 below shows the percentage reduction in stability


parameter for seaside perforated QCB compared to nonperforated QCB, with non-perforated model as reference, for
all the ranges of d/gT2.

Table 1. Variation of %[W/Hi2]red for perforated model with S/D=8 for different ranges of d/gT2
d/gT2

0.005 0.010

0.010 0.015

0.015 0.020

0.020 0.024

%[W/Hi2]red for

12.8-12.9

16.80-26.10

15.67-15.06

20.38 18.36

Perforated QBW with


S/D=8

Page 84

Arkal Vittal Hegde. / INDECS-15

4. CONCLUSIONS
Based on the results obtained and discussion carried out,
following conclusions have been drawn:

Non-dimensional stability parameter W/Hi2


decreases as incident wave steepness increases
for both non-perforated and seaside perforated
breakwaters for all the ranges of d/gT2 values.

Mallayachari, V. and Sundar, V. Standing wave pressures


due to regular and random waves on a vertical wall. Ocean
Eng., 1995, 22: 859-879.
Jiang, X., Gu H., and Li, Y. Numerical simulation on
hydraulic performances of quarter circular breakwater.
China Ocean Engineering, 2008, 22 (4), 585-594.

Critical weight required for sliding stability is


found to be more for non-perforated QCB
compared to seaside perforated QCB. Typically,
there is a decrease of 12 to 20% in W/Hi2
parameter, depending upon the d/gT2 values.

Qie, L., Xiang, Z., Xuelian J., and Yinan, Q. Research on


Partial Coefficients for Design of Quarter-circular Caisson
Breakwater, J. Marine Sci. Appl., 2013, 12: 65-71.

W/Hi2 is found to increase with increase in


d/gT2 for both breakwaters for the given wave
steepness.

5. REFERENCES
Isaacson, M. Measurement of regular wave reflection.
Journal of Waterway, Port, Coastal, and Ocean Engineering,
1991, 117(6): 553-569.
Isaacson, M. and K. Subbiah. Numerical simulation of
random wave forces near the free surface. J. Offshore Mech.
Arctic Eng., 1991, 113: 14-42.

Page 85

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

ROLE OF STRUCTURAL ENGINEERS IN


SUSTAINABLE CONSTRUCTION
Ranjith Dissanayake1 and Chaminda Bandara2
1

University of Peradeniya, Peradeniya 22400, Sri Lank, ranjith@fulbrightmail.org


2
University of Peradeniya, Peradeniya 22400, Sri Lanka,
chamindasbandara@yahoo.com

Abstract
Bridges risk facing unexpected loadings due to floods, landslides, tsunamis, earthquakes, blasts and various accidents.
Displacement of bridges due to erosion and collapsing of abutments, lateral pressure generated by exceeding water
levels and floating debris etc are common. Even though there are provisions for accidental loading used during bridge
designs, managing the risk of damages completely is not possible (e.g. the tsunami damages in Sri Lanka in December
2002, Japan in March 2011).Building new bridges generally consumes more money and time than repairing and
retrofitting of damaged bridges. Therefore, the latter can be considered more sustainable. However, proper methods are
necessary to assess the level of damages and to verify the fitness of such bridges prior to repair and retrofitting. In the
assessment, there are two important criteria to consider. One is the amount of fatigue damage to the bridge due to usual
past vehicle loading and hence the remaining fatigue life of the bridge. The other is the magnitude of damage caused to
the bridge by the unexpected loading. In order to determine the fitness of a damaged bridge for reuse, condition surveys
are usually carried out. Using a validated numerical model of the damaged bridge (a finite element model, FEM),
efficient decisions are possible as such models can be loaded with past loading histories as well as future expected
loadings and then the stresses and deflection can be obtained from the FEM. Then the fitness of the damaged bridge can
be verified and elements which need to be replaced or retrofitted can be determined. The case study is about a 34 m
long, 5.2 m wide, single spanned, double lattice girded, wrought iron railway bridge, located at Puttalam (Bridge No. 02
on the railway track between the Puttalam Cement Factory and Limestone Quarry, used for transporting limestone)
which was built about 40 years ago and damaged and displaced from its abutments by floods. The bridge was then
placed on temporary timber abutments for several years.With the increase in cement production, the owners of the
bridge wanted to use heavier locomotives on this railway track and also increase the number of trips. Therefore, there
was a need for an assessment of bridges on this track in order to determine whether the bridges can be used further or
should be demolished and new bridges built in its place.In order to do the assessment, a condition survey was carried out
on all the bridges on the track. One of the bridge was found weak as this had been damaged by floods. Then an analysis
was done by modeling the bridge (FEM) by using general purpose SAP 2000 program and validating the FEM by using
results of a field loading test (the bridge was temporarily erected on timber abutments for several years). Both static and
dynamic loading tests were carried out using an M2 locomotive with 6 numbers of 13.16 ton axles for 5 different loading
cases to measure the displacement, strain and acceleration at pre-determined (critical) members of the bridge. The future
fatigue life of the bridge was estimated using the prescribed last method. The future life was found as 30 years with a
factor of safety of 3. Further, using the validated model, the ability of the bridge for higher loading situations was
confirmed. The cost, estimated for retrofitting work and constructing new reinforced concrete abutments was much less
than that for constructing a new bridge. Therefore it was decided that rehabilitation of the bridge with necessary
retrofitting work is more sustainable than demolishing it and constructing a new one. The bridge is now in use after
being repaired, retrofitted and placed on new abutments.
Keywords: Damage assessment, Steel bridges, Retrofitting, Sustainability, Fatigue damage

1. Introduction
Bridges risk facing unexpected loadings due to floods,
landslides, tsunamis, earthquakes, blasts and various
accidents. Displacement of bridges due to erosion and

collapsing of abutments, lateral pressure generated by


exceeding water levels and floating debris etc are
common. Even though there are provisions for accidental
loading used during bridge designs, managing the risk of

Page 1

Author et. al. / INDECS-15

damages completely is not possible (e.g. the tsunami


damages in Sri Lanka in December 2002, Japan in
March 2011).

on this track in order to determine whether the bridges


can be used further or should be demolished and new
bridges built in its place.

Building new bridges generally consumes more money


and time than repairing and retrofitting of damaged
bridges. Therefore, the latter can be considered more
sustainable. However, proper methods are necessary to
assess the level of damages and to verify the fitness of
such bridges prior to repair and retrofitting. In the
assessment, there are two important criteria to consider.
One is the amount of fatigue damage to the bridge due to
usual past vehicle loading and hence the remaining
fatigue life of the bridge. The other is the magnitude of
damage caused to the bridge by the unexpected loading.

In order to do the assessment, a condition survey was


carried out on all the bridges on the track. One of the
bridge was found weak as this had been damaged by
floods. Then an analysis was done by modeling the
bridge (FEM) by using general purpose SAP 2000
program and validating the FEM by using results of a
field loading test (the bridge was temporarily erected on
timber abutments for several years). Both static and
dynamic loading tests were carried out using an M2
locomotive with 6 numbers of 13.16 ton axles for 5
different loading cases to measure the displacement,
strain and acceleration at pre-determined (critical)
members of the bridge. The future fatigue life of the
bridge was estimated using the prescribed last method.
The future life was found as 30 years with a factor of
safety of 3. Further, using the validated model, the
ability of the bridge for higher loading situations was
confirmed.

BS 5400: Part 10: 1980 [6] and District line fatigue of


riveted under bridges, Infrastructures Consultancy
Services (Engineering Services Group), London
Underground Limited, June 1998 [5] provide two
methods used to determine the remaining fatigue life of
bridges. However, using such codes without due care
and modifications for local conditions may provide
misleading results. The procedure introduced by
Siriwardane & Dissanayake et al., [3] is one of the
procedures developed (using wrought iron railway
bridges in Sri Lanka) based on past stress histories of
critical bridge elements using present day measured
strains and applying damage indicator based sequential
law to estimate the fatigue life.
In order to determine the fitness of a damaged bridge for
reuse, condition surveys are usually carried out. Using a
validated numerical model of the damaged bridge (a
finite element model, FEM), efficient decisions are
possible as such models can be loaded with past loading
histories as well as future expected loadings and then the
stresses and deflection can be obtained from the FEM.
Then the fitness of the damaged bridge can be verified
and elements which need to be replaced or retrofitted can
be determined.
2. The Case Study
The case study is about a 34m long, 5.2m wide, single
spanned, double lattice girded, wrought iron railway
bridge,
located at Puttalam (Bridge No. 02 on the railway track
between the Puttalam Cement Factory and Limestone
Quarry, used for transporting limestone) which was built
about 40 years ago and damaged and displaced from its
abutments by floods. The bridge was then placed on
temporary timber abutments for several years.

The cost, estimated for retrofitting work and constructing


new reinforced concrete abutments was much less than
that for constructing a new bridge.
Therefore it was decided that rehabilitation of the bridge
with necessary retrofitting work is more sustainable than
demolishing it and constructing a new one. The bridge is
now in use after being repaired, retrofitted and placed on
new abutments.
3. Condition Assessment
3.1. Condition Survey

Several deficiencies were identified during visual


inspections such as corroded places in the bridge deck
and in load carrying members, missing bracings, missing
timber sleepers, improper arrangement of sleepers,
improper alignment of the rail track on the bridge deck,
increased thicknesses of the members due to oxides,
missing rivets, replacements of rivets by nuts and bolts
and etc., as shown in Fig.1. According to the information
received about the flood damage, the bridge had got
displaced gradually due to erosion of abutments. The
bridge had not been loaded at the time when the
displacement occurred. There were no signs of damages
to any element due to this incident.

With the increase in cement production, the owners of


the bridge wanted to use heavier locomotives on this
railway track and also increase the number of trips.
Therefore, there was a need for an assessment of bridges

Page 87

Author et. al. / INDECS-15

The bridge and individual elements were measured


including the sizes of webs and flanges and their current
thicknesses, thicknesses of gusset plates, sizes of rivets
and bolts etc., (main girders of the bridge are of equal
span, each 33.65 m end to end, cross girders are placed
at the panel points of the main girders and also at the
centers of each panel, there are two longitudinal girders
connected between cross girders in each panel and there
are bracings between main girders, the rail is of gauge
1.75m and etc.).
Fig. 1. Corroded load carrying elements

3.2. Modelling the bridge

The bridge was modeled using SAP 2000 as a three


dimensional frame element model. Effective measured
thicknesses were applied for the members of the model
as much as possible. Riveted connections were assumed
as fully fixed and rotational stiffness behavior with a
magnitude of 18,200 kNm/rad was assumed at
connections of cross girders and bottom chord of main
truss girder [3].
3.3. Onsite loading tests

Fig. 2. Defective connections

Fig. 3. Missing elements and rivets

A series of tests were carried out on site in order to


determine the behavior of the bridge under loading,
dynamic factors and for the validation of the finite
element model. An M2 engine (6 Nos., of 13.16 ton
axels) was used for loading the bridge at 5 separate
loading cases; the center of the engine stopped at 1/4 of
the length of the bridge, stopped at mid span of the
bridge, stopped at 3/4 of the length of the bridge, the
train traveling at a speed of 10 km/hr and the train
traveling at a speed of 20 km/hr. In all the loading cases,
displacement measurements (horizontal and vertical) of
the mid span of the main girder were measured using 2
displacement gauges, strain measurements at four
selected locations were obtained by using strain gauges
and acceleration at the mid span using an acceleration
gauge.

3.4. Validation of the model

Values obtained at the field loading test for horizontal


and vertical displacement of the main girders and strains
at 4 critical elements were compared with that obtained
from the FEM for similar loading cases. As the readings
from onsite displacement and strain gauges and the
numerical results obtained from the model are quite
close, it was concluded that the FEM would behave in a
similar way to the actual bridge. Therefore the model
was used to find numerical results of anticipated loads of
the bridge.
\

3.5. Checking the bridge for anticipated higher loading

Fig. 4. Temporary abutment

The FEM was loaded with an engine of weight 100 tons


(with 6 axels) which is the anticipated engine which will
be used on the bridge in the future. Then the strains,
Page 88

Author et. al. / INDECS-15

stresses and displacements at critical members


(elements) were observed. Fig.2 shows the finite element
model with critical members and Table 1 shows the
stresses developed in these critical members against the
yield strength and hence the factors of safety of each
member. As the factors of safety of all the critical
members are within reasonable limits, it was decided
that the bridge is safe for 100 ton engines subject to
further analyses and improvements as mentioned in the
next sections.

stresses were multiplied by a dynamic factor of 1.3 (for


diesel engines) and load factors corresponding to
weights of all the types of trains traveling on this bridge
[3]. The past life span of the bridge was divided into 2
periods (2 loading blocks), loading block 1 from 1970 to
2000 (diesel engines M2, WDM 6 of 75 tons and
carriages of 25 tons each) and loading block 2 from 2000
to 2010 (diesel engine M2 & carriages) depending on the
trains that traveled on the bridge assuming a constant
sequence of trains. The loading of the bridge in the
future (from 2010 upwards) was predicted with M6
engines (100 tons, 16.7 tons/axel, 6 axels), WDM6
engines and carriages traveling per a given train
timetable. Using the reservoir counting method, the
stress ranges were determined.
4.2 Damage indicator based sequential law, Siriwardane,
Dissanayake et al (2009)

Fig. 5. Critical elements of the bridge

Table 1. Factor of safety for critical members

Critical
member

Yield stress /
MPa

Developed
stress/ Mpa

Factor of
Safety

276

245

97.953

2.501

291

245

96.781

2.531

60

245

68.916

3.555

106

245

60.515

4.049

159

245

96.313

2.544

163

245

64.968

3.771

4. Fatigue life of the bridge


4.1 Fatigue life evaluation

In order to avoid errors that can occur when using BS


5400 Part 10: 1980 which uses the Miners rule which
does not take the effects of the sequence of loading in to
account, the damage indicator based sequential law that
capture the loading sequence effect of variable amplitude
loads more precisely was used in the life evaluation
method developed by Siriwardane, Dissanayake et al. In
this method, a full range S-N curve (Wohler curve) for
wrought iron is used and the equation of the S-N curve is
given by;

N 1050
20
N 1000000000

0.202

(1)

For a considered stress range i, having a number of


cycles ni, the damage indicator Di can be obtained from
the equation;
Di = (i(eq)-i)/(u-i)

(2)

where, u is the ultimate tensile strength amplitude (or


range) for rotating bending test-based S-N curves and
i(eq) is the ithlevel damage stress amplitude obtained
from the Wohler curve for (Ni ni) number of cycles.

In load carrying structural elements, repeated


applications of stresses that are not sufficient to cause
failures by a single application create cracks which
propagate gradually until failure of the element. These
failures are called fatigue failures [6].

The damage indicator Di is transformed to the next stress


range i+1 having ni+1 number of cycles using the
equation;

Determination of fatigue life of a bridge has a risk of


errors as life evaluation methods are based on various
assumptions and statistics. This risk can be mitigated by
estimating the life using the most appropriate method.

where, '(i+1)eq is the damage equivalent stress amplitude


or stress range at the level i+1.

The stresses obtained from the validated FEM for critical


members were used for the life evaluation. These

'(i+1)eq = Di(u-i+1)+i+1

(3)

The number of cycles N(i+1)R that corresponds to '(i+1)eq


is obtained from the Wohler curve and the residual life at
the level i+1, N(i+1)R is obtained from the equation;

Page 89

Author et. al. / INDECS-15


N(i+1)R = N(i+1)R ni+1

(4)

maintenance. Fig.3 shows the bridge after improvement


work.

The stress amplitude (i+1)eq that corresponds N(i+1)R is


obtained from the Wohler curve. The Damage indicator
of the (i+1)th level stress range Di+1 can then be
determined from the equation;
Di+1 = (i+1(eq)-i+1)/(u-i+1)

(5)

Applying the same procedure for all the stress levels, the
damage indicator for the past (loading blocks 1 and 2)
can be obtained and then this indicator is used to
calculate the future fatigue life for the expected loading
block.
4.3 Fatigue life

The fatigue lives obtained for three critical elements


using the above method is given in Table 2. The most
critical element with respect to its fatigue life is the
middle cross girder which has a fatigue life of 93 years.
The bridge is expected to be reused for another 30 years.
Therefore, evaluation verifies that the bridge is capable
of being reused for 30 years with a factor of safety of 3.1
(93 years / 30 years).
Table 2. Future fatigue life of critical elements
Fig. 6. Bridge after improvement work

Member

Future Fatigue Life (Years)

Main girder

1059

6. Verification tests

Diagonal truss member

94

Middle cross girder

93

Verification tests are important for rehabilitated bridges


in order to verify the structural soundness. Such
verification tests can be done by loading the bridge and
measuring its deflections and strains of critical elements.
Regular inspections together with verification tests
minimize the risk of collapsing the bridge during its
commissioning.

5. Improvements
As per the assessment results, the following
improvements were recommended. ( a ). Placing the
bridge on newly built reinforced concrete abutments, ( b
). Making the railway track straight near the bridge to
reduce lateral forces (stresses), ( c ). Cleaning the steel
structure completely using sand blasting techniques, ( d )
rectifying defects, retrofitting or replacing defective parts
(these improvements increase the fatigue life of these
elements as well as the stiffness of the bridge), ( e ).
Applying a corrosive resistant paint, ( f ). Placing new
steel sections for missing elements and well tight high
strength bolts for all missing rivets, ( g ). Aligning the
railway track as smooth as possible to reduce vibration, (
h ). Placing sleepers at regular intervals, ( i ). Providing a
proper river bank protection and ( j ). Regular

Further, there are changes to material properties of


bridge elements due to various actions during its
rehabilitation (and retrofitting). The bridge discussed in
this paper was moved from its previous temporary
abutment to new abutments without dismantling it.
(Note: the bridge is assembled by rivets. Elements
dismantling and re-joining by new rivets is not easy).
Verification tests are useful after such work for
identifying the status.
6.1 Verification test procedure
A series of loading tests on the bridge was carried out
using an M2 engine (6 numbers of 13.16 ton axles)
under different static and dynamic loading conditions as
mentioned below. Deflections and strains at critical
locations of the bridge were measured.

Page 90

Author et. al. / INDECS-15

The loading cases used are; case 1: the center of the


engine was stopped at 1/4 of the length of the bridge and
measurements were taken, case 2: the center of the
engine was stopped at the mid span of the bridge and
measurements were taken, case 3: the engine was
allowed to travel at a speed of 10 km/hr and
measurements were taken and case 4: the train was
allowed to travel at a speed of 20 km/hr and
measurements were taken.
In order to take measurements, two displacement gauges
were fixed at the mid span of a main girder; one below
the flange of the main girder, fixed vertically to measure
the vertical displacement and the other touching the web
of the main girder, fixed horizontally to measure the
horizontal displacement. Strain gauges were fixed at
three pre-defined locations on critical elements to
measure the strains for all four loading cases.
6.2 Analysis of verification test results
Maximum vertical and horizontal displacements of the
bridge at static loading were observed as 6.40mm and
0.16mm respectively. Further, at dynamic loading, the
maximum vertical and horizontal displacements in the
recent test were 6.55mm and 0.45mm respectively.
When comparing these new results with the
displacement values obtained before improvement
works, a clear reduction in the displacement
measurements were seen. Due to the retrofitting works
(especially due to replacement of weak elements by new
elements and by strengthening connections with the use
of bolts where they were missing), the stiffness of the
structure has improved.
For the 3 elements in which the strains were measured
and stresses were calculated, it was found that the
maximum stresses happen when the engine moves at
20km/hr (loading case 4). The factors of safety
calculated for these 3 members for loading case 4 are
2.40, 2.28 & 18.04 and they are reasonably safe values.
As the verification tests have shown improvements with
regard to deflections of the bridge and reasonable factors
of safety for the three critical elements with regard to
strains and stresses, it was concluded that the
improvement works are successful.
7. Conclusions
Constructing a new bridge usually consumes more time
and money than rehabilitating and retrofitting an existing
bridge. There are ways to assess the possibility of
reusing old and/or damaged bridges. This paper explains
one such assessment method applied to a damaged steel

railway bridge using onsite non-destructive tests, finite


element models and principals of fatigue life. Further,
the bridge so assessed and retrofitted was subjected to
verification tests which have proved that the method
used is successful.
Most of the old highway steel bridges and railway
bridges in Sri Lanka are nearing the end of their design
lives. Knowing the future life spans of these bridges is
very important because, then the relevant authorities can
plan for their rehabilitation or replacement in advance,
thereby contributing to sustainable development. The
method described in this paper can be applied practically
in Sri Lanka to carry out life assessments of steel
bridges.
Acknowledgements
The authors acknowledge the collaboration of the team
who worked on the investigation of railway bridge No.
2, Puttalam, especially the assistance of Technical
Officers, Mr. J.D. Gamlathge, Mr. C.W. Jayawardane
and the kind support given by the Authority of the
railway track.
References
[1] Dissanayake, P.B.R.: Risk Assessment and
Retrofitting Methodology, Assessment Report of
Bridge No.02, Puttalam, 2011.
[2] Dissanayaka,
P.B.R.,
Bandara
A.M.A.C.S.,
Rathnayaka R.M.S.U.P.: Investigation Report of
Bridge No.02, Puttalam, 2012.
[3] Siriwardane, S.C., Ohga, M., Dissanayake, R., and
Tatsumasa, K.: Remaining Fatigue Life Estimation
of Existing Railway Bridges. New York, Nova
Science Publishers Inc., 2009.
[4] Siriwardane, S.A.S.C., Ohga, M., Dissanayake,
P.B.R., and Kaita, T. : Structural Appraisal based
Different Approach to Estimate the Remaining
Fatigue Life of Railway Bridges, Structural Health
Monitoring, Vol. 9 (4), pp.323-339, 2010.
[5] Infrastructures Consultancy Services, Engineering
Services Group, Civil Infrastructure Design, District
line fatigue of riveted under bridges, London
Underground Limited, London, 1998.
[6] Code of Practice for Fatigue, Steel Concrete and
Composite Bridges, BS 5400, Part 10: 1980, British
Standard Institute, London.

Page 91

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SHORTTERM SHORELINE CHANGES ALONG


KARNATAKA COAST, INDIA: A GEOSPATIAL APPROACH
ATEETH SHETTY1 *, K.S. JAYAPPA2 , PRAVEEN G. DESHBHANDARI3 ,
A.S. RAJAWAT4 , RATHEESH RAMAKRISHNAN5
1

Department of Marine Geology, Mangalore University, Mangalagangothri, Mangalore,


Karnataka, India 574 199, athith89@gmail.com
2
Department of Marine Geology, Mangalore University, Mangalagangothri, Mangalore,
Karnataka, India 574 199, ksjayappa@yahoo.com
3
Department of Marine Geology, Mangalore University, Mangalagangothri, Mangalore,
Karnataka, India 574 199, praveendeshbhandari@gmail.com
4
Geo-Sciences Division, Space Applications Centre, Indian Space Research Organisation,
Ahmedabad, India 38 0015, asrajawat@sac.isro.gov.in
5
Geo-Sciences Division, Space Applications Centre, Indian Space Research Organisation,
Ahmedabad, India 38 0015, ratheeshr@sac.isro.gov.in

Abstract
The shoreline is one of the most rapidly changing coastal landforms. Accurate detection and frequent monitoring of
shorelines are very essential to understand the coastal processes and dynamics of various coastal features. In the present
study, an attempt is made to understand short-term shoreline changes during 2014 along the Karnataka coast which
extends from Devbagh in the north to Talapady in the south covering a total length of ~ 320 km. Landsat-8 Operational
Land Imager (OLI) satellite images during pre-monsoon, immediately after the monsoon and post-monsoon periods have
been analyzed using Remote Sensing and GIS techniques to demarcate shoreline positions. Beach profiles are carried out
to assess the impact of natural (including sea-level rise), anthropogenic and hydrological factors on coastal morphology
of this region. The study reveals that erosion is found in almost all the beaches of Karnataka during south-west monsoon
whereas after the monsoon they get accreted gradually till the end of fare weather season. Thus Remote Sensing and GIS
techniques along with field investigations can provide sustainable approach to the coastal zone management.
Keywords: Shoreline Changes, Remote Sensing, GIS, Beach Profiles, Coastal Zone Management

1. Introduction
Understanding the shoreline positions and erosion/accretion trend
through time are of elemental importance to coastal scientists,
engineers and managers [2], [9]. Shoreline position measurements
for various time periods can be used to derive quantitative
estimates of rate of progradation / recession. Coastal processes in
the study area are controlled by the natural processes - waves,
littoral currents, offshore relief and river mouth / sea-level changes
and anthropogenic activities, such as construction of coastal
structures, sand mining and dredging of navigation channels [21],
[22]. The RS and GIS applications have proved effective in
delineation of coastal configuration and coastal landforms,
detection of shoreline positions, estimation of shoreline and
landform changes, extraction of shallow water bathymetry [17],
[21], [22], [24], [26], [31], [35], [36], [39], [40]. Beach profiling in
any given location (i.e., cross sectional contour of a beach,
perpendicular to the shoreline) is the simplest method to quantify
parameters of beach morphology and reconstruct the trends of
beach evolution [7]. The profile surveys are used to establish direct

measures of sediment movement, to evaluate shoreline changes and


compute beach volume changes along and across the shore [14].
Beach volumes tend to increase over time at a changing shoreline,
leading to progradation of the foreshore. An envelope of profile
variation is important in calculating long-term trends, where
repeated surveys are necessary to separate gradual changes from
short-term fluctuations [20], [30], [38]. Calculation of cross section
areas and estimation of accreted / eroded volume of sand on the
beach is possible by monitoring beach profiles throughout the year
and comparing the maximum and minimum profiles [5]. Beach
profiling contributes to a better understanding of coastal processes
by providing quantitative information on the spatial / temporal
evolution of beaches, which is crucial for coastal management [12].
Response of beach profile configuration to natural and
anthropogenic changes is useful in selecting the most appropriate
engineering design required to mitigate coastal erosion and
accretion alongshore [19]. In the present study, beach profiling and
shoreline change analysis was carried out during pre-monsoon
(March), immediately after the monsoon (October) and postmonsoon (December) periods of 2014 along Karnataka coast.

Corresponding author. Tel.: +91 9036238769


E-mail: athith89@gmail.com

Page 92

Ateeth et al. / INDECS-15

2. Study Area
Karnataka coast measuring ~320 km length consists of three
districts - Dakshina Kannada, Udupi and Uttara Kannada and lies
between 12o45' - 15o00' N latitude and 74o00' E - 75o00' E
longitude (Fig. 1). It is a wave dominated coast with complex
geomorphology. The northern part of the coast having headland,
pocket beaches and southern part of the coast is characterized by
spits and straight sandy beaches. This coast is influenced by two
monsoons: (1) the southwest (summer) monsoon (June-September)
which is stronger than (2) the northeast (winter) monsoon
(October-November). During southwest monsoon, the coastal
current is stronger and sets-in clockwise direction, while during the
northeast monsoon, it is in the counter clockwise direction. As a
result, littoral current in the study area are directed towards south
during November-April when waves approach from WNW and
NW directions and towards north from May-October when waves
approach from SW, WSW and W directions [16], [18], [27].

Elevations of beaches have been measured along these profile lines


at an interval of 3 m or less whenever any distinct morphological
features such as cliffs, ridges, erosional scarps etc. were present on
the beach. Dumpy level method, using survey telescope, a levelling
staff and a measuring tape, was employed to fix the elevation of
reference points. A simple, rapid and low cost method suggested by
[11] was used to measure the profiles in the present study with the
help of only two wooden rods and a measuring tape. Wooden rods
are 5 ft long and the major markings are made at an interval of 1 ft
and they are further divided into one inch intervals throughout. A
minor elaboration is a small wooden pad nailed to one end of each
rod to prevent the rod from sinking into loose sand. The profile
survey data were then plotted and analysed on graph software for
seasonal volume change studies.

3.2. Shoreline Change


Landsat - 8 (Operational Land Imager) satellite images (Table 1)
acquired during March, October and December 2014 have been
selected in order to understand the shoreline change caused by
natural processes and / or anthropogenic activities.
Table 1. Details of Satellite data
Satellite
(Sensor)

Dates of
acquisition

Tidal
Spatial
height (m) resolution

Path

Row

Landsat 8
(OLI)

13-03-2014

0.50

146

50, 51

07-10-2014

0.78

10-12-2014

0.27

30.0 m

Satellite images were geometrically corrected using Ground


Control Points (GCP) with Root Mean Square (RMS) error of < 0.3
pixel using Erdas Imagine v2013 software. In order to understand
the shoreline change, vector layers were created from the above
mentioned satellite data products using ArcGIS v10.1 software.
Permanent vegetation was considered as shoreline and erosion /
accretion analysis was carried out for all the locations and periods.
The overlaid image divided automatically into different polygons at
the intersect of the input and overlaid polygons. Accordingly, the
polygons developed outside the input theme were identified as
accretion and polygons developed inside the input theme was
identified as erosion.

4. Results and Discussion


4.1. Beach Morphological Change

Fig. 1. Study area showing the locations of beach profiles


and shoreline positions

3. Materials and Methods


3.1. Beach Morphological Change
Beach profiling was carried out during different seasons at 25
selected beach locations on the coastline of Karnataka during
March, October and December, 2014 to understand the spatial and
temporal beach morphological changes. Reference points were
fixed at the backshore where negligible change is observed by
wave or wind action. Profile surveys were conducted perpendicular
to beach from reference point to low water line or mean sea level.

Beach profiling is the traditional and simplest method with highest


accuracy to understand the morphological evolution of beaches.
This method can be extensively used for high resolution regional
scale coastal management and planning. It helps in estimating the
cross section area and accreted / eroded volume of sediment on the
beach throughout the year by surveying the profiles over a period.
The results based on these surveys are illustrated in this section and
an attempt has been made to infer the seasonal changes (Fig. 2).
Stable landmarks were selected during first survey (March, 2014)
and the same were considered as bench marks for the next two
surveys. In beach volume analysis, the true volumes above and
below the master profile was calculated using Graph software.
Beach sediment volume, volume difference (survey-to-survey and
baseline-to-survey) at all the locations are given in Table 2.
Photographs of the changes in pre-monsoon, monsoon (October)
and post-monsoon seasons at selected locations of Karnataka coast
are shown in the figure 3.

Page 93

Ateeth et al. / INDECS-15

Page 94

Ateeth et al. / INDECS-15

Page 95

Ateeth et al. / INDECS-15

Fig. 2. Changes in Shoreline and Beach Profiles at selected locations of Karnataka coast
during pre-monsoon, immediately after the monsoon and post-monsoon periods

Page 96

Ateeth et al. / INDECS-15

Fig. 3. Changes in pre-monsoon (a), immediately after the monsoon (b) and post-monsoon seasons (c) at selected locations of Karnataka coast

Page 97

Ateeth et al. / INDECS-15


Table 2. Beach volume changes (in m3) along selected locations of
Karnataka coast
Location

Mar
2014

Oct
2014

Dec
2014

Mar to
Oct 14

Oct to
Dec 14

Mar to
Dec 14

Talapady

19.92

9.55

23.61

-10.37

+14.06

+3.69

Someshwara

20.49

48.96

63.52

+28.47

+14.56

+43.03

Ullal

30.89

4.00

26.53

-26.89

+22.53

-4.36

Bengre

355.60

212.86

259.53

-142.74

+46.67

-96.07

Thanirbhavi

242.54

179.63

268.47

-62.91

+88.84

+25.93

Panambur

205.64

100.09

150.16

-105.55

+50.07

-55.48

Surathkal

155.76

82.12

108.37

-73.64

+26.25

-47.39

60.34

72.63

85.01

+12.29

+12.38

+24.67

Sasihitlu

111.89

99.43

89.64

-12.46

-9.79

-22.25

Hejamady

150.68

147.29

168.26

-3.39

+20.97

+17.58

Yermal

128.36

102.03

130.27

-26.33

+28.24

+1.91

Kaup

216.49

77.60

132.38

-138.89

+54.78

-84.11

Malpe

150.23

72.96

86.54

-77.27

+13.58

-63.69

Kodi Bengre

28.97

1.61

7.71

-27.36

+6.10

-21.26

Kota

41.07

28.31

34.19

-12.76

+5.88

-6.88

Maravanthe

58.90

56.20

62.26

-2.70

+6.06

+3.36

177.14

160.88

195.52

-16.26

+34.64

+18.38

67.98

40.43

35.16

-27.55

-5.27

-32.82

Manki

253.60

211.44

235.46

-42.16

+24.02

-18.14

Apsarakonda

271.54

338.72

289.85

+67.18

-48.87

+18.31

31.84

39.69

45.24

+7.85

+5.55

+13.40

8.30

6.81

3.32

-1.49

-3.49

-4.98

Om beach

12.93

17.05

23.70

+4.12

+6.65

+10.77

Karwar

74.95

67.23

69.60

-7.72

+2.37

-5.35

Devbagh

50.62

29.30

37.18

-21.32

+7.88

-13.44

2926.67

2206.82

2631.48

-719.85

424.66

-295.19

Mukka

Nesthar
Murdeshwara

Vannalli
Tadadi

Total

+ Accretion; - Erosion

The results based on beach profiles, suggest more sediment


volume during pre-monsoon (2,926.67 m3) and comparatively less
during October (2,206.82 m3) of 2014. It can be attributed to
calm waves which carry the sediment from offshore during the
pre-monsoon period and the beaches are built up and they
become narrow as the sediment is taken offshore by high waves.
During March 2014, high beach volume was estimated at Bengre
(355.60 m3) and low at Tadadi (8.30 m3), whereas during October
2014 high beach volume was estimated at Apsarakonda (338.72
m3) and low at Kodi Bengre (1.61 m3). During December 2014
high beach volume was estimated at Apsarakonda (289.85 m3)
and low at Tadadi (3.32 m3). Comparison of March and October
profiles revealed maximum accretion of 67.18 m3 at Apsarakonda
and maximum erosion of 142.74 m3 at Bengre. Whereas October
and December profiles revealed maximum accretion of 88.84 m3
at Thanirbhavi and maximum erosion of 48.87 m3 at
Apsarakonda and comparison of pre-monsoon and post-monsoon
profiles revealed maximum accretion of 43.03 m3 at Someshwara
and maximum erosion of 96.07 m3 at Bengre (Table 2). The
results of the profile change analysis revealed that Ullal, Bengre,
Panambur, Surathkal, Sasihitlu, Kaup, Malpe, Kodi Bengre,
Kota, Murdeshwara, Manki, Tadadi, Karwar and Devbagh
beaches were eroding and the remaining beaches were accreting
during the study period. Spatial variations in erosion / accretion
pattern of beaches are the result of temporal variability in the
intensity and reversibility of wave directions and associated
longshore currents, wind induced circulation, coastline
orientation and existing coastal protection structures [8], [13],
[29]. The information on the state of the current erosion and the
possible prediction of its impact is important in coastal
monitoring [25]. The study related to the response of beach
profile configuration to natural and anthropogenic changes is
helpful in selecting the suitable engineering design to mitigate
coastal erosion and accretion [19]. Analysis of beach profiles is
helpful in understanding the variability in cross-shore elevation
patterns and volume change that occur along a profile line. Beach
response to coastal processes can be interpreted from geometric
and volumetric comparison of beach profile sets [14]. To protect
the eroding beaches in the study area, seawalls have been
constructed. However, most of these structures failed to serve the
purpose and led to severe erosion at some places and increase in
beach slope in front of seawalls as well as on adjacent beaches
[18], [23].

4.2. Shoreline change


Geospatial techniques and multi-spectral satellite images have
been utilized to demarcate the shoreline positions of different
periods. Shoreline positions of different time periods have been
digitized accurately by taking permanent vegetation into
consideration. The vector layers were superimposed and overlay
analyses of shoreline changes were carried out to estimate the
area of erosion and accretion between different periods. The
results based on multi-spectral satellite images, suggest more
changes in shoreline positions between pre- and monsoon
(341,255 m2) and comparatively less during the pre- and postmonsoon seasons (133,880 m2) of 2014. Between March and
October, high erosion was noticed at Maravanthe (26,839 m2)
and low at Vanalli (2,392 m2), and during October-December,
high accretion was noticed at Karwar (25,601 m2) and low at
Talapady (2,225 m2). Comparison of pre- and post-monsoon
shorelines revealed the maximum accretion of 9,718 m2 at
Bengre and maximum erosion of 34,188 m2 at Apsarkonda
(Table 3). The results of the shoreline change analysis revealed
that at Bengre, Thanirbhavi, Surathkal and Karwar the shoreline

Page 98

Ateeth et al. / INDECS-15

Mar to Oct
2014

Oct to Dec
2014

Mar to Dec
2014

Talapady

-10557

+2225

-8332

Someshwara

-5690

+3481

-2208

Ullal

-11282

+4892

-6390

Bengre

-7680

+17399

+9718

Thanirbhavi

-15388

+10742

+2061

Panambur

-15549

+15522

+869

Surathkal

-12388

+12397

+4

Mukka

-12034

+9884

-1946

larger on gentler slopes. Considerable interannual and interdecadal


sea-level changes at a coast are forced primarily by large-scale
winds [4], [15], [34], large changes in salinity [33], [34] as well as
steric oscillations, which are due to the changes in specific
volume [32]. [3] has developed a model which estimates the
shoreline recession with respect to rise in sea level. If this model is
considered for Karnataka coast, it suggests every mm rise of sealevel will result in a shoreline retreat of about one meter [10].
Erosion / accretion is a cyclic phenomenon, which is normal along
this coast. Beaches along the barrier spit are subjected to erosion
due to migration of rivers mouth. Both natural processes (waves,
littoral currents, river mouth and sea-level changes and variation in
offshore relief) and anthropogenic activities (construction of
coastal structures, sand mining and dredging of navigation
channels) are responsible for shoreline changes [6], [10], [18], [21].
Shoreline configuration and beach slope influence the natural
sediment movement in addition to other nearshore parameters [28].
Increase in sediment transport rates corresponds with areas of
shoreline erosion, while decrease in sediment transport rates
alongshore corresponds with shoreline accretion.

Sasihitlu

-11207

+7427

-1334

5. Summary and Conclusion

Hejamady

-13895

+7427

-6481

Yermal

-16331

+14795

-1577

Kaup

-11409

+7228

-4193

Malpe

-15560

+1824

-13696

Kodi Bengre

-17634

+2402

-15238

Kota

-18828

+10085

-8787

Maravanthe

-26839

+8810

-18029

Nesthar

-3078

+8023

-4954

Murdeshwara

-10129

+5786

-4344

Manki

-12760

+9945

-2848

Apsarakonda

-26529

+22351

-34188

Vannalli

-2392

+10510

-8117

Tadadi

-3407

+6355

-2931

Om beach

-11723

+9621

-2378

Karwar

-25035

+25601

+1518

Devbagh

-23931

+23882

-79

-341255

+258614

-133880

was prograding and at the remaining beaches it was retreating


during the study period.
Table 3. Erosion and accretion (in m2) along selected locations of
Karnataka coast
Location

Total

+ Prograding; - Retreating

Monsoonal rainfall is one of the important factors that control the


coastal processes and determines the freshwater discharges through
river systems along this coast. A strong relationship was reported
between the variability of rainfall and sediment transport, where
high sediment discharges are recorded with high rainfall [1], [37].
Further, high wave activity during intensive monsoon makes the
sea rough, and erodes the sediment along the coast, resulting in
change of the shoreline configuration. Whereas, low rainfall results
in reduction of sediment supply to the coastal region. In addition,
the larger and strong waves and undertow processes continuously
disturb and erode the nearshore bed during monsoon season and
therefore sediments are triggered into suspension and transported.
Sea-level rises have direct impact on the shoreline changes which
correspond to higher shift to the zone of wave action on the beach.
This would be reflected in a shoreline recession which will be

The results based on beach profiles, suggest more sediment volume


during pre-monsoon (2,927 m3) and comparatively less during the
October (2,207 m3) of 2014. Results of beach profile analysis
revealed that Ullal, Bengre, Panambur, Surathkal, Sasihitlu, Kaup,
Malpe, Kodi Bengre, Kota, Murdeshwara, Manki, Tadadi, Karwar
and Devbagh beaches were eroding and the remaining beaches
were accreting during the study period. During October, width of
the beaches reduced greatly and minimum volume of sediment was
stored in them. The sediment volume was high during the premonsoon season compared to the post-monsoon season. Although
cyclic in nature, the large-scale erosion takes place during SW
monsoon and later in post-monsoon period the lost sediment is
regained but due to human innervations related actives reduce the
ratio of sediment gain and loss. The results based on multi-spectral
satellite images, suggest more changes in shoreline positions
between pre-monsoon and immediately after the monsoon (341,255
m2) and comparatively less during the pre- and post-monsoon
seasons (133,880 m2). The results of the shoreline change analysis
revealed that at Bengre, Thanirbhavi, Surathkal and Karwar the
shoreline was prograding and at the remaining beaches it was
retreating during the study period. Both natural processes and
anthropogenic activities are responsible for erosion / accretion of
the beaches in the study area.

Acknowledgement
The First author (A S) is thankful to the Space Applications Centre,
ISRO, Ahmedabad for the award of fellowship under MOP-3.

References
[1]Avinash K, Jena B, Vinaya MS and Jayappa KS. Regionally
tuned algorithm to study the seasonal variation of suspended
sediment concentration using IRS-P4 Ocean Colour Monitor
data. The Egyptian Journal of Remote Sensing and Space
Science, 2012; 15(1): 6781.
[2]Boak EH, Turner IL. Shoreline definition and detection: A
review. J Coastal Research, 2005; 21(4): 688-703.
[3]Bruun P. Sea level rise as a cause of shore erosion. J Waterways
and Harbors Div, 1962; 88(1): 117-130.
[4]Clarke AJ, Liu X. Interannual sea level in the northern and
eastern Indian Ocean. J Geophys Res, 1994; 99: 1224-1235.

Page 99

Ateeth et al. / INDECS-15

[5]Cooper NJ, Leggett DJ, Lowe JP. Beach-profile measurement,


theory and analysis: practical guidance and applied case
studies. Water Environ J, 2000; 14(2): 7988.
[6]Deepika B, Avinash K, Jayappa KS. Shoreline change rate
estimation and its forecast: remote sensing, geographical
information system and statistics-based approach. Int J Environ
Sci Technology, 2013; 11(2): 395416.
[7]Delgado I, Lloyd G. A simple low cost method for one person
beach profiling. J Coastal Res 2004; 20(4): 1246-1252.
[8]Di Stefano A, De Pietro R, Monaco C, Zanini A. Anthropogenic
influence on coastal evolution: a case history from the Catania
Gulf shoreline (eastern Sicily, Italy). Ocean Coast Manage,
2013; 80: 133-148.
[9]Douglas BC, Crowell M. Long-term shoreline position
prediction and error propagation. J Coastal Res, 2000; 16(1):
145-152.
[10]Dwarakish GS, Vinay SA, Usha Natesan, Toshiyuki Asano,
Taro Kakinuma, Katta Venkataramana B, Jagadeesha Pai,
Babita MK. Coastal vulnerability assessment of the future sea
level rise in Udupi coastal zone of Karnataka state, west coast
of India. Ocean Coast Manage, 2009; 52: 467-478.
[11]Emery KO. A simple method of measuring beach profiles.
Limnology and Oceanography, 1961; 6: 90-93.
[12]Ferreira MA, Soares L, Andrade F. Educating citizens about
their coastal environments: beach profiling in the Coast watch
project. J Coast Conserv, 2012; 16: 567574.
[13]Frihy OE, Debes EA, El Sayed WR. Processes reshaping the
Nile delta promontories of Egypt: pre- and post-protection.
Geomorphology, 2003; 53: 263-279.
[14]Gorman L, Morang A, Larson R. Monitoring the coastal
environment; Part IV: mapping, shoreline changes, and
bathymetric analysis. J Coastal Res, 1998; 14(1): 61-92.
[15]Han W, Webster PJ. Forcing mechanisms of sea level
interannual variability in the Bay of Bengal. J Phys Oceanogr,
2002; 32: 216-239.
[16]Hariharan V, Reddy MPM, Kurian NP. Littoral and rip
currents and beach profiles off Someshwar. Ind Geogr J, 1978;
53: 14-20.
[17]Jantunen H, Raitala J. Locating shoreline changes in the
Porttipahta (Finland) water reservoir by using multitemporal
Landsat data. Photogrammetria, 1984; 39: 1-12.
[18]Jayappa KS, Vijaya Kumar GT, Subrahmanya KR. Influence
of coastal structures on beach morphology and shoreline in
Southern Karnataka, India. J Coastal Res, 2003; 19: 389-408.

[23]Kumar Avinash, Narayana AC, Jayappa KS. Shoreline changes


and morphology of spits along southern Karnataka, west coast
of India: A remote sensing and statistics-based approach.
Geomorphology, 2010b; 120: 133-152.
[24]Lafon V, Froidefonda JM, Lahetb F, Castaing P. SPOT shallow
water bathymetry of a moderately turbid tidal inlet based on
field measurements. Remote Sens Environ, 2002; 81: 136-138.
[25]Li R, Keong CW, Ramcharan E, Kjerfve B, Willis D. A coastal
GIS for shoreline monitoring and management - case study in
Malaysia. Surv Land Inform Syst, 1998; 58(3): 157-166.
[26]Maiti S, Bhattacharya AK. Shoreline change analysis and its
application to prediction: a remote sensing and statistics based
approach. Mar Geol, 2009; 257:11-23.
[27]Narayana AC, Manojkumar P, Tatavarti R. Beach dynamics
related to the mudbank along the southwest coast of India. In:
McAnally WH, Mehta AJ (Ed), Coastal and Estuarine Fine
Sediment Processes, Elsevier Science BV, 2001, pp. 495-507.
[28]Prasad KVSR, Arun Kumar SVV, Ch. Venkata Ramu,
Sreenivas P. Significance of nearshore wave parameters in
identifying vulnerable zones during storm and normal
conditions along Visakhapatnam coast, India. Nat Hazards,
2009; 49: 347-360.
[29]Rajith K, Kurian NP, Thomas KV, Prakash TN, Hameed TSS.
Erosion and accretion of a placer mining beach of SW Indian
coast. Mar Geod, 2008; 31:128-142.
[30]Reeve D, Chadwick A, Fleming C: Coastal engineering,
processes, theory and design practice. Oxford: Spon Press ltd.,
2004
[31]Ryu JH, Won JS, Min KD. Waterline extraction from Landsat
TM data in a tidal flat: a case study in Gosmo Bay, Korea.
Remote Sens Environ, 2002; 83: 442-456.
[32]Shankar D. Seasonal cycle of sea level and currents along the
coast of India. Curr Sci, 2000; 78: 279-288.
[33]Shankar D, Shetye SR. Are interdecadal sea level changes
along the Indian coast influenced by variability of monsoon
rainfall? J Geophys Res, 1999; 104: 2603126041.
[34]Shankar D, Shetye SR. Why is mean sea level along the Indian
coast higher in the Bay of Bengal than in the Arabian Sea?
Geophys Res, 2001; 28: 563565.
[35]Siddiqui MN, Maajid S. Monitoring of geomorphological
changes for planning reclamation work in coastal area of
Karachi, Pakistan. Adv Space Res, 2004; 33: 1200-1205.
[36]Singh A. Digital change detection techniques using remotely
sensed data. Int J Remote Sens, 1989; 10: 989-1003.

[19]Kaiser MFM, Frihy OE. Validity of the equilibrium beach


profiles: Nile Delta Coastal Zone, Egypt. Geomorphology,
2009; 107: 25-31.

[37]Syvitski JPM, Morehead MD. Estimating river-sediment


discharge to the ocean: application to the Eel margin, northern
California. Mar Geol, 1999; 154: 13-28.

[20]Komar PD. Beaches processes and sedimentation, 2nd edn.


Prentice-Hall, Englewood Cliffs, 1998, pp. 545

[38]Thomas T, Phillips MR, Williams AT, Jenkins RE. Medium


time-scale behaviour of adjacent embayed beaches: influence of
low energy external forcing. Appl Geogr, 2012; 32: 265-280.

[21]Kumar A, Jayappa KS. Long and short-term shoreline changes


along Mangalore coast, India. Int. J. Environ Res, 2009; 3: 177188.
[22]Kumar Avinash, Jayappa KS, Deepika B. Application of
remote sensing and GIS in change detection of the Netravati
and Gurpur river channels, Karnataka, India. Geocarto Int.,
2010a; 25: 397-425.

[39]White K, El Asmar HM. Monitoring changing position of


coastlines using Thematic Mapper imagery, an example from
the Nile Delta. Geomorphology, 1999; 29: 93-105.
[40]Yamano H, Shimazaki H, Matsunaga T, Ishoda A, McClennen
C, Yokoki H, Fujita K, Osawa Y, Kayanne H. Evaluation of
various satellite sensors for waterline extraction in a coral reef
environment: Majuro Atoll, Marshall Islands. Geomorphology,
2006; 82: 398-411.

Page 100

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

TRAFFIC CONGESTION STUDY FOR COIMBATORE DISTRICT


USING GIS
Anitha Selvasofia.S.D 1, Prince Arulraj.G 2
1

SNS College of Technology, Coimbatore, 641035, TamilNadu, sofiasamuel2912@gmail.com


SNS College of Technology, Coimbatore, 641035, TamilNadu, princearulraj@yahoo.com

Abstract
The urbanization in developing countries indicates that more people live in cities than before. The trend of urbanization,
population increase and increase in the number of registered vehicles induces pressure on traffic movements and makes
living in urban area more difficult. Traffic congestion is one of the major problems faced by many transportation
decision makers for urban areas. The problem has many impacts on social, economic and development aspects of urban
areas. Hence the solution to this problem is not straight forward. It requires a lot of efforts, expertise, time and money.
General congestion related data collection and congestion management measure is labour intensive and a heavy
investment is needed for these mitigation measures. Latest technologies like GIS and GPS will help to analyze the live
traffic situation and suggest the cost effective measures to mitigate the congestion. An attempt has been made to use
GPS and GIS effectively to find a solution for traffic congestion problem. In this paper a comprehensive GIS based
interactive Traffic Congestion Management System has been developed for the Coimbatore national highway
transportation network. National highways NH 47 and NH 209 were selected for modelling. NH47 connects Salem and
Kanyakumari. This NH passes through the cities Salem erode- Coimbatore Palakkad Thrissur Aluva Kochi
Alappuzha Kollam Trivandrum Nagercoil and Kanyakumari. Among these cities, Coimbatore is one of the busiest
and congested city. NH 209 connects Dindigul and Bengaluru. This NH passes through Palani Udumalaipettai
Pollachi Coimbatore Sathyamangalam Bannari Chamrajanagar and Kollegal. In this route methods to reduce the
congestion have been suggested.
Keywords: Congestion Management, GIS, GPS, Modeling, Regression Analysis.
intersections along the road),extreme traffic congestion
sets in .Traffic congestion occurs when a volume of
1. Introduction
The word traffic originally meant trade(as it still does)
and comes from the old Italian verb traffic. Traffic is
formally organized in many jurisdictions , with marked
lanes , junctions , intersections , interchanges ,traffic
signals or signs. Traffic laws are the laws which govern
traffic and regulate vehicles.The laws and the informal
rules have been developed over time to facilitate the
orderly and timely flow of traffic. Events which disturb
the flow and may cause traffic to degenerate in to a
disorganized mess include: road construction, collisions
and debris in the road way on particularly busy
freeways. A minor disruption may persist in a
phenomenon known as traffic waves. A complete
breakdown of organization may result in traffic
congestion. Traffic congestion is a condition on road
networks that occurs as usage increases and is
characterized by slower speeds, longer trip times, and
increased vehicular queuing. When traffic is large
between the interaction vehicles slows the speed of the
traffic stream, resulting in some congestion. As the
demand approaches the capacity of a road ( or of the

traffic or modal split generates demand for space greater


than the available road capacity. This point is commonly
termed as saturation. A Geographic Information System
(GIS) is a system designed to capture, store, manipulate,
analyse, manage, and present all types of spatial or
geographical data. The acronym GIS is sometimes used
for geographical information science or geospatial
information studies to refer to the academic discipline or
career of working with geographic information systems
and is a large domain within the broader academic
discipline of Geo-informatics. In a general sense, the
term describes any information system that integrates
stores, edits, analyses, shares, and displays geographic
information. GIS applications are tools that allow users
to create interactive queries (user-created searches),
analyse spatial information, edit data in maps, and
present the results of all these operations.
Kernel Density calculates the density of features in a
neighbourhood around those features. It can be

* Corresponding

author. Tel.: +91 9894352472


E-mail: sofiasamuel2912@gmail.com

Page 101

Anitha Selvasofia.S.D et. al. / INDECS-15

calculated for both point and line features. Possible uses


include finding density of houses, crime reports or
density of roads or utility lines influencing a town or
wildlife habitat. The population field could be used to
weigh some features more heavily than others,
depending on their meaning, or to allow one point to
represent several

GIS has begun to open new avenues of scientific inquiry


into behaviors and patterns of previously considered
unrelated
realworld

Name
of the
Office

observations. For example, one address might represent


a condominium with six units, or some crimes might be
weighed more severely than others in determining
overall crime levels. For line features a divided highway
probably has more impact than a narrow dirt road and a
high-tension line has more impact than a standard
electric pole.

Cbe
South
Cbe
North
Cbe
Central
Cbe
West
Total

2. Study Area
Annur is a taluk of Coimbatore rural district, located to
the north of the city of Coimbatore on the way to Sathy.
It is located about 31 km north of Coimbatore, in the
Indian state of Tamil Nadu. The latitude and longitude of
Annur are 11o23'00" N and 77o10'00" N. Sathy Road(NH
209) is a busy arterial road in city of Coimbatore, Tamil
Nadu, India. The distance form Coimbatore to Annur is
31 km . Figure 1 shows the boundary of NH47 and
NH209 in Coimbatore. Avinashi is a city that is located
to the east of the city

information.
Fig.2.

Vehicles Registered During the Year 2013


Two
Car
Auto Te
Bu Lor
whee
/Thr mp s
ry
ler
ee
o
whe
eler
1815
2
4400 316
120 21 940
1745
0
4247 178
20
39 36
1285
4
3354 342
364 31 125
1144
9
1416 179
130 8
194
5990 1341
129
5
7
1015 634 99 5
The methodology adopted is shown in

Fig. 1. Boundary of NH47 and NH209 in Coimbatore.


of Coimbatore on the way to Erode. The latitude and
longitude of Avinashi is 11o19'528" N and 77o26'861" E
.NH 47,is a national highway which is a busy road from
Coimbatore to Avinashi. The distance from Coimbatore
to Avinashi is 42 km.
2.1. Methodology
GIS uses spatial-temporal location as the key
index variable for all other information. Just as a
relational database containing text or numbers can relate
many different tables using common key index
variables, GIS can relate unrelated information by using
location as the key index variable. Any variable that can
be located spatially, and increasingly also temporally,
can be referenced using GIS. Locations or extents in
Earth spacetime may be recorded as dates/times of
occurrence, and x, y, and z co-ordinates representing,
longitude, latitude, and elevation, respectively. These
GIS coordinates may represent other quantified systems
of temporal-spatial reference. This key characteristic of

Fig. 2. Methodology
2.1.1. Data Collection

In order to study the existing traffic conditions from


Coimbatore to Avinasi and Coimbatore to Annur, the
following data were collected and used

Traffic volume data


Roadway geometric data

Page 102

Anitha Selvasofia.S.D et. al. / INDECS-15

Traffic Survey
Latitude and longitude

4.30P.M. In NH 47 Lakshmi mills was found to be the


zone with heavy traffic. In NH 209 Ganaphathy was
found to be the place with maximum traffic.

2.1.1.1 TRAFFIC VOLUME DATA


The details of the vehicles registered in the past two
years in the RTO office are given in Table1 and Table2.
Table 1.Number of vehicles registered in the year
2013

Table 2.Number of vehicles registered in the year


2014

Name of
the
Office

Cbe
South
Cbe
North
Cbe
Central
Cbe
West
Total

Vehicles Registered During the Year 2014


Two Car
Auto Te
Bu Lor
whe
/Thr mp s
ry
eler
ee
o
whe
eler
2269 5500 395
150 26 117
0
5
2181 5309 223
25
49 45
2
1606 4192 428
454 39 156
7
1431 1770 224
163 10 242
1
7488 1677 1270 792 12 161
0
1
4
8

2.1.1.2 ROADWAY GEOMETRIC DATA


In geographical information system, the land area is
digitized with toposheets as base map. The collected
toposheets as follows

Fig. 3. Toposheet of Coimbatore


2.1.1.3 TRAFFIC SURVEY
Traffic Survey has been carried out during peak
hour and non-peak hours. The peak hour timing are
8.00-10.00A.M,1.00-2.00P.M and 4.30-6.00P.M.The
non-peak hour timing are 10.00-1.00P.M and 2.00-

Table 3. Vehicle Count per day


Vehicle Count per day near Lakshmi mills
(NH209)
S.No Time Two
Four
Heavy
Wheeler Wheeler Vehicles
1.
8.00
1483
135
64
2.
8.30
1635
150
71
3.
9.00
1539
140
60
4.
9.30
1203
150
106
5.
10.00
1364
107
100
6.
10.30
1409
130
72
7.
11.00
1358
100
50
8.
11.30
1186
130
80
9.
12.00
1396
129
80
10. 12.30
1428
150
50
11.
1.00
1603
180
67
12.
1.30
1749
120
80
13.
2.00
1697
133
70
14.
2.30
1284
140
78
15.
3.00
1303
140
100
16.
3.30
1386
152
100
17.
4.00
1562
138
70
18.
4.30
1324
130
70
19.
5.00
1195
152
90
20.
5.30
1458
160
80
21.
6.00
1704
141
100
22.
6.30
1642
128
80
3. REGRESSION ANALYSIS
One of the most powerful uses of regression
analysis is to predict the values of the dependent
variable, sometimes for a future time period, but more
generally, for any given set of values of the explanatory
variables. This makes it a very powerful tool for policy
analysis since a variety of scenarios can be simulated
and compared relatively easily and quickly. For the NH
47 the dependent and independent variables considered
are as follows:
Dependent variable - Time
Independent variables Number of two wheelers,
four wheelers, heavy vehicles.
The mathematical model obtained was,
Time(Peak) = 8.494 0.005*no of two wheelers +
0.27*no of four wheelers + 0.29*heavy vehicles
(1)
Time(Non Peak) = 8.528 -0.001
*no of two wheelers +0.072* no of wheelers
- 0.118* no of heavy vehicles
(2)
For the NH 209 the mathematical model obtained was,
Time(Peak) = 33.360 0.009 * no of two wheelers
0.067 * no of four wheelers 0.066
Page 103

Anitha Selvasofia.S.D et. al. / INDECS-15

* no of heavy vehicles
Time(Non Peak) = 5.202 0.007 *
no of two wheelers 0.011 * no of four wheelers
0.115 * no of heavy vehicles

(3)

can be
Fig.6. Geo referenced Toposheet

(4)
3.2 CREATING THEMATIC MAP
The thematic maps were created. One map was
created for NH47 and the other was created for NH209.
The Thematic map for NH47 contains the road between
Coimbatore and Annur Boundary of the highway and
important places along the road. The Thematic map for
NH209 contains the road between Coimbatore and
Avinashi Boundary of the highway and important places
along the road

Fig. 4. Graph showing the Vehicle movement


From this analysis it was found that,
1) A correlation coefficient (R) value of 0.891 was
obtained, which is acceptable and it indicates a
fair positive correlation.
2) An increase in time will increase the number of
vehicles. So, when time is increased the number
of vehicle also increases and hence, Time and
vehicle are directly proportional. The model
developed also establishes the same.

Fig.5. Bar chart showing the Predicted Registered


Vehicle
Thus the mathematical model developed was
acceptable. Hence, this model can be used to predict
future number of vehicle in the NH.
3.1 ANALYSIS USING GIS
The Survey of India topographical map at a scale of
1:25000 was scanned as the raster input. Existing maps

Fig. 6. Line Kernel density


Kernel Density calculates the density of features in a
neighbourhood around those features. It can be
calculated for both point and line features. Possible uses
include finding density of houses, crime reports or
density of roads or utility lines influencing a town or
wildlife habitat. The population field could be used to
weigh some features more heavily than others.Kernel
density was applied for NH209 and NH47.The kernel
density for NH47 varies from 0-100%. Kernel Density
between 0-30% indicates that there is no Traffic
congestion and it falls under `non-peak hour while the
range of 31-100% falls under high Traffic congestion
and it falls under peak hour.
.The kernel density for NH209 varies from 0-100%.
Kernel Density between 0-30% indicates that there is no
Traffic congestion and it falls under non-peak hour while
the range of 31-100% falls under high Traffic congestion
and it falls under peak hour.

4. Conclusion
`
In developing countries like India, cities are growing
very fast. Coimbatore is a cosmopolitan city which gives
opportunity for educational activities, job and tourist,
due to the vast development of technology. Population
of Coimbatore
increases rapidly, which leads to
increase in number of vehicles .The vehicle population
in Coimbatore is increasing at alarming rate. The
number of registered vehicles also increases rapidly
Page 104

Anitha Selvasofia.S.D et. al. / INDECS-15


number registered vehicles for the year 2013 two
wheeler is 59905 and four wheeler is 14432, heavy
vehicles is 2028 and for 2014 two wheeler is 74880 and
four wheeler is 18041, heavy vehicles is 2534.Based on
anlyses in SPSS it is found that the time increases
number of vehicles also increases. Also with use of GIS
kernel density establish the traffic congestions in NH
roads. This way GIS can be used all the ways to
carefully analyse all the criteria and successfully give
good solution that is acceptable to everyone .the study
presented above has been made only for two NH roads.
Congestion management activities include.
Redesigning traffic signal timing to improve
progressive traffic movements along arterial
road ways
Road widening Projects to improve level of
service
All these approaches and others require a system wide
traffic and road way data in order to formulate effective
management strategies.
References
[1] Deepthi Jayan.K , B.Ganeshkumar Identification of
Accident Hot Spots: A GIS Based Implementation
for Kannur District, Kerala,International journal of
geomatics and geo science ,volume 1,ISSN 09764380, (2010).
[2] Gopala Raju SSSV.,Assessment of Nois level due to
vehicular
traffic
at
Warangal
city,
India,International Journal of Environment and
Pollution,Vol 30. No.1, pp.137-153, 2007 .
[3] Lim Yu Liang, Law Teik HUA and Dadang
Mohamad
MASOEM,
Traffic
Accident
Application Using Geographic Information System,
Journal of the Eastern Asia Society for
Transportation Studies, Vol. 6, pp. 3574 - 3589,
2005.
[4] Sarin, S.M.,Road Traffic Safety in Indian Issues and
Challenges Ahead, Indian Highways, Vol. 26,
No.6., pp 26-38,1998.
[5] Ortuzar, J. De and Willumsen, L.G. "Transport
Modeling". J. Wiley & Sons Ltd., USA, 1995.
[6] Banik, B. K. "Evaluation of Traffic Congestion in
Sylhet city and a Development of Mathematical
Model". B.Sc. Engg. Thesis, Department of
Civiland Environmental Engineering, Shah Jalal
University
of
Science
and
Technology,
Sylhet,Bangladesh,2005.

Page 105

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

PRIORITIZATION OF EROSION POTENTIAL AREA THROUGH


MORPHOMETRIC ANALYSIS IN VENKATAPURA WATERSHED,
KARNATAKA: A RS AND GIS APPROACH
Photo of First
Author

Praveen G. Deshbhandari , C. Krishnaiah

in
JPEG Format

Department of Marine Geology, Mangalore University Mangalagangothri, 574 199, India,


praveendeshbhandari@gmail.com, krishnaiah56@yahoo.com

Photo of Second
Author (if any)
in
Abstract

JPEG Format

Soil erosion is one of the most significant and widespread form of soil degradation that has environmental and economic
impacts. The geomorphological characteristics of a watershed are commonly used for solving various hydrological
problems of the un-gauged watersheds in inadequate data situations. Therefore, in this study to find out the most
vulnerable sub-watershed to soil erosion using morphometric analysis and prioritization were carried out on Sixteen subwatersheds of Venkatapura watershed, Karnataka. An evaluation of quantitative geomorphic parameters was made to
study basin characteristics and its influence on erosion in micro watersheds. A Remote Sensing and GIS based
conceptual method was used in this study, based on results sub-watersheds have been classified into three categories as
high, medium and low. The highest priority zone consists of seven sub-watersheds, medium of seven and low of two sub
watersheds. High priority indicates that watersheds are much more susceptible to soil erosion hence it should be provide
with immediate soil resource management measures.
Keywords: Morphometric analysis, Watershed, Erosion, Prioritization, GIS
study an attempt has been made to prioritize erosion prone subwatersheds on the basis of morphometric analysis

1. Introduction
Soil erosion is the biggest threat to the conservation of soil and
water resources. A heavy rainfall is the main cause of erosion
during monsoon months. Economically, soil erosion affects the
productivity of land and hence conservation strategies assume lead
roles in the development programs. [6] The amount of soil erosion
is mainly affected by vegetation cover, topographic features,
climatic variables, and soil characteristics.
Soil erosion and degradation of land resources are significant
problems in a large number of countries. In order to ensure
sustainable management of natural resources and to implement
necessary soil conservation techniques, the areas with soil erosion
risk should be determined and classified according to potential
erosion risk levels. Monitoring soil Erosion in all watersheds by
installing gauzes is difficult. In these situations geomorphologic
characteristics of a watershed are commonly used to solve
hydrological problems.
The watershed management planning highlights the management
techniques to control erosion in the watershed area.[8] Drainage
analysis is very important for watershed planning since it gives an
idea about the basin characteristics in terms of slope, topography,
soil condition, runoff characteristics, surface water potential etc.
Remote sensing and GIS have been used as tools in watershed
development, management and studies on prioritization of subwatersheds. Several recent and past studies, the watersheds are
prioritized on various basis such as morphometry, Universal Soil
Loss Equation (USLE), Sediment Yield Index (SYI), land use/land
cover etc. [1], [3], [4], [5], [7], [9], [11], [22], [28], [29] In this

2. Study area
The Venkatapura watershed is one of the west flowing river basins.
It is located between1358'48" to 1408'46" N latitude and
7428'48" to 7444'35"E longitude in the southern part of Uttara
Kannada and western part of Shivamogga districts of Karnataka
state. It originates in the Western Ghat, flowing for about 45 Km
with a catchment of 335 km and confluence into Arabian Sea near
Venkatapura. [14] Lithogically the study area comprises laterites ,
schists and metabasalt. The tributaries of river exhibit dendritic
drainage pattern. The study are receives an average annual rainfall
is ~2750 mm.

Fig. 1. Location map of Venkatapura Watershed

Corresponding author. Tel.: +8762201544


E-mail: praveendeshbhandari@gmail.com

Page 113

Praveen G. Deshbhandari et. al. / INDECS-15

3. Methodology
Base map of the study area prepared from the Survey of India
(SOI) toposheets. Watershed boundary was delineated and drainage
network was derived from SOI toposheets, which was overlaid on
the Landsat data to update and modify in terms of channel numbers
and lengths. The Hortons law was followed for stream ordering.
According to Hortons low un-branched stream considered as first
order stream, when two first order streams join to form second
order. Two second order steams join together to form third order
and so on.

According to the guidelines of [2] the mean area of watershed is


less than 500 km 2 (_+ 50%). Watershed is further classified into
16 sub-watersheds of fourth order streams. Hence, The subwatersheds were designated as SW1 to SW14. Morphometric
analysis was performed through the measurement of linear aspects,
areal aspects and relief aspects of the watershed using standard
methods. (Table 1) The results of the morphometric parameters are
computed to analyze erosion potential zones in the
study area.

SOI Topo maps

Scanning

Geometric correction

Mosaicing

Geometric correction

Extraction of study area

Drainage layer

Sub- watershed

Morphometric Analysis

Basic
Parameters

Linear
Parameters

Areal
Parameters

Compound Value

PRIORITIZATION

Fig. 2. Flowchart of the methodology used in this study

Page 114

Praveen G. Deshbhandari et. al. / INDECS-15

Fig. 3. Sub-watershed map of Venkatapura Watershed


Table 1. Formulae adopted for computation of morphometric parameters
Serial No.
1

Morphometric Parameter
Stream Order (Nu)

Formula
Hierarchial rank

References

2
3

Stream Length (Lu)


Bifurcation ratio (Rb)

Drainage density (Dd)

Length of the stream


Rb = Nu / (Nu + 1)
Nu = Total no. of stream segments of order 'u'
D = Lu / A
Lu = Total stream length of all orders
A = Area of the basin (km2)

Stream frequency (Fs)

Fs = Nu / A
Nu = Total number of streams of all orders
A = Area of the basin (km2)

[12]

Texture ratio (T)

T = Nu /P
Nu = Total number of streams of all orders
P= Perimeter of the basin

[24]

Circularity ratio (Rc)

Rc = 4* *A/P2
Where, Rc = Circularity ratio;
= Pi value i.e. 3.14;
A=Area of the basin (km2);
P=Perimeter (km)

[17]

Form factor (Ff)

Ff = A/Lb2
A = Area of the basin (km2)
Lb2= Square of the basin length

[13]

Elongation Ratio (Re)

Re = 2 v (A / Pi) / Lb
Where, Re = Elongation Ratio
A = Area of the Basin (km'-)
Pi = 'Pi' value i.e., 3.14
Lb = Basin length

[23]

10

Compactness Coefficient (Cc)

Cc = Pc/P u
Where, P c= Perimeter of watershed;
P = Perimeter of circle of watershed area

[27]

[26]
[13]
[23]
[12]

Page 115

Praveen G. Deshbhandari et. al. / INDECS-15

4.1.3. The basin length (L)


The basin length corresponds to the maximum length of the basin
measured parallel to the main drainage line. The basin length of
sub-watersheds varies from 54.97 km and the values of L are
shown in table 2.

4. Results
The various morphometric parameters in the Venkatapura
watershed were determined and summarized in Table 2 and 3.
4.1. Basic Parameters
The basic parameters such as stream numbers, stream length,
area, perimeter, basin length of the watershed were calculated
using GIS software and are discussed below.

4.1.4. Stream Order (Nu)


The designation of stream orders is the first step in drainage basin
analysis and is based on a hierarchical ranking of streams. In the
present study, ranking of streams has been carried out based on
the [26] method. The order wise Nu of all sub-watersheds are
represented in Table 2.

4.1.1. Perimeter (P)


The perimeter is the total length of the drainage basin boundary.
The perimeter of the sub-watersheds are recorded in Table 2

4.1.5. Stream Length (Lu)


The number of streams of various orders in a sub-watershed was
counted and their lengths were measured. The order wise length
of streams of sub- watersheds are given in Table 2

4.1.2. Area (A)


The computed basin area ranging from 3.96 Km2 to 31.69 Km2.
The area of all sub-watersheds are represented in Table 2

Table 2. Basic morphometric parameters


Subwatershed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Basin
area
(Km2)
31.69
13.05
27.18
7.71
12.79
6.85
14.97
19.01
9.65
4.96
12.74
3.96
33.81
7.42
14.60
5.11

Perimeter
(Km)
30.80
16.29
34.64
14.25
15.97
12.55
17.37
20.21
15.18
9.59
15.47
9.25
27.64
13.72
16.41
9.94

Basin
length
(Km)
11.35
3.95
13.24
4.69
5.29
3.43
5.96
5.01
4.49
3.29
5.23
3.36
7.76
4.95
5.40
3.68

Nu

I
59
25
119
28
93
71
73
82
50
31
89
19
204
40
58
25

Nu
II
12
9
48
7
46
15
15
17
11
8
16
5
39
18
10
6

4.2. Derived Parameters


The derived parameters are computed through the measurement
of linear and aerial aspects of catchment using standard methods
and are discussed below.
4.2.1. Bifurcation ratio (Rbm)
The term bifurcation ratio (Rb) may be defined as the ratio of the
number of the stream segments of given order to the number of
segments of the next higher order. [23], [25] Rbm reflects the
complexity and degree of dissection of a drainage basin. The
lower values of Rbm are characteristics of the sub-watersheds
which have suffered less structural disturbances. [24], [20] Higher
Rbm values indicates potential for flash flooding. [22] SW3 has
highest Rb value (7.15) and whereas SW12 as lowest Rbm
(2.76)value.

4.2.2. Drainage density (Dd)

III
3
2
3
2
6
3
2
4
3
2
3
2
6
2
2
2

IV
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

75
37
171
38
146
90
91
104
65
42
109
27
250
61
71
34

Lu
(Km)
I
29.41
12.78
54.43
13.42
37.10
23.96
41.45
46.78
23.34
12.56
44.08
9.33
91.02
17.31
23.79
13.23

Lu
(Km)
II
15.28
4.52
13.22
4.11
7.75
6.73
8.73
11.63
5.30
3.70
7.53
1.74
23.44
4.70
8.14
2.00

III
8.92
4.34
3.47
1.88
7.60
3.61
2.71
11.14
4.76
2.14
5.16
0.95
17.67
4.09
3.21
1.79

IV
9.53
3.83
15.33
3.46
5.48
2.57
4.89
1.75
2.64
1.33
2.53
2.07
9.37
1.26
4.04
1.44

63.15
25.49
86.46
22.89
57.95
36.89
57.80
71.32
36.05
19.75
59.32
14.11
141.51
27.38
39.20
18.47

Drainage density (Dd) is a measure the total stream length in a


given basin to the total area of the basin. [26] The lower
drainage density of any watershed indicates that it has permeable
subsurface material, good vegetation cover and low relief and
vice versa. [10], [16] The highest drainage density (5.38) was
observed in SW 6 and lowest Dd (1.95)was observed in SW2.
4.2.3. Stream frequency (Fs)
It is the total number of stream segments of all orders per unit
area. [12] Stream frequency is inversely related to permeability,
infiltration capacity and directly related to the relief of
watersheds. [18], [19] Sub-watershed SW1 (2.36) showed the
highest Fs and similarly the sub-watershed SW6 showed the
lowest Fs. (2.36)
4.2.4. Circulatory ratio (Rc)
The circularity ratio is the ratio of the area of the basins to the
area of circle having the same circumference as the perimeter of
the basin. [17], Higher Rc is indicative of circular shape of the

Page 116

Praveen G. Deshbhandari et. al. / INDECS-15

watershed and of the moderate to high relief and permeable


surface 9 [22], The Rc values varies from 0.28 (SW3) to 0.68
(SW15)

According to Horton (1945), T is the total number of stream


segments of all orders per perimeter of that area., It depends upon
a number of natural factors such as climate, rainfall, vegetation,
rock and soil type, infiltration capacity, relief and stage of
development of a basin [24] The dt< 2 indicates very coarse,
between 2 and 4 is related to coarse, between 4 and 6 is moderate,
between 6 and 8 is fine and >8 is very fine drainage texture. The
T values of sub-watersheds varies from 2.27 (SW2) -9.13 (SW5),
which indicates that all textures. Drainage texture is greatly
influenced by infiltration capacity. [13] Regions of low
infiltration capacity will give rise to higher T and thus will lead to
more erosion. [22]

4.2.5. Form factor (Rf)


Form factor is the dimensionless ratio between basin area and the
square of basin length. [12] The value of Rf would always less
then 0.7854 in perfectly circular basin. The highest Rf (0.83) was
observed in SW2 and lowest value (0.15) was observed in SW3.
4.2.6. Elongation ratio (Re)
Elongation ratio (Re) is defined as the ratio of diameter of a circle
having the same area as of the basin and maximum basin length
[23]. Generally it varies from 0.6 to 1.0 depends upon climate
and geology. The sub-watersheds shows lower elongation ratio
values which are high susceptible to erosion. The sub-watershed
elongation ratio varies from 0.44 (SW3) to 1.00(SW2)

4.2.8. Compactness coefficient (Cc)


According to Gravelius (1914), compactness coefficient of a
watershed is the ratio of perimeter of watershed to circumference
of circular area, which equals the area of the watershed. [15] The
Cc of a watershed directly corresponds to the infiltration capacity
of the watershed. [22] The sub-watershed compactness coefficient
ranging
from
0.04
(SW13)
to
0.21.(SW12)

4.2.7. Texture ratio (T)

Table 2. Derived morphometric parameters


Subwatershed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Rb
I/II

II/III

4.91
2.77
2.47
4.00
2.02
4.73
4.86
4.82
4.54
3.87
5.56
3.80
5.23
2.22
5.80

4.00
4.50
16.00
3.50
7.66
5.00
7.50
4.25
3.66
4.00
5.33
2.50
6.50
9.00
5.00

4.16

3.00

III/IV
3.00
2.00
3.00
2.00
6.00
3.00
2.00
4.00
3.00
2.00
3.00
2.00
6.00
2.00
2.00
2.00

Rbm

Dd

Fs

Rc

Rf

Re

3.97
3.09
7.15
3.16
5.22
4.24
4.78
4.35
3.73
3.29
4.63
2.76
5.91
4.40
4.26

1.99
1.95
3.18
2.96
4.53
5.38
3.85
3.75
3.73
3.97
4.65
3.55
4.18
3.68
2.68

2.36
2.83
6.29
4.92
11.41
13.13
6.07
5.45
6.73
8.45
8.55
6.80
7.39
8.21
4.86

2.43
2.27
4.93
2.66
9.13
7.16
5.23
5.14
4.27
4.37
7.04
2.91
9.04
4.44
4.32

0.41
0.61
0.28
0.47
0.63
0.54
0.62
0.58
0.52
0.67
0.66
0.58
0.55
0.49
0.68

0.24
0.83
0.15
0.35
0.45
0.58
0.42
0.75
0.47
0.45
0.46
0.35
0.56
0.30
0.49

0.55
1.0
0.44
0.66
0.76
0.86
0.73
0.98
0.77
0.76
0.76
0.66
0.84
0.62
0.79

0.04
0.08
0.06
0.13
0.08
0.14
0.07
0.06
0.11
0.17
0.08
0.21
0.04
0.14
0.07

3.05

3.61

6.64

3.41

0.64

0.37

0.69

0.17

4.3. Prioritization of Sub-watersheds


For prioritization of sub-watersheds in water resources
management, the morphometric analysis uses some very crucial
linear and shape morphometric parameters. These are termed as
erosion risk assessment parameters. Linear parameters such as
drainage density, stream frequency, mean bifurcation ratio,
drainage texture, length of overland flow have a direct
relationship with erodibility whereas shape parameters such as
elongation ratio, circularity ratio, form factor, basin shape and
compactness coefficient have an inverse relationship with [10],
[21] Hence, ranking of the watersheds has been carried out for
giving highest priority/rank based on highest value in case of
linear parameters and lowest value in case of shape parameters.
[4], For prioritization of sub-watersheds, in case of linear
parameters, the highest value was given ranking of 1, next higher
value was given a ranking of 2 and so on. After the ranking has
been assigned based on each parameter, the rating value for all the

Cc

nine micro-watersheds were averaged so as to arrive at a


compound value. (Cp) In this study The sub-watersheds where
classified as high (4.5-7.625),medium ( 7.626-10.750) and low (
10.751-13.875) priority based on the compound value. According
to results seven sub-watersheds (SW3, SW5, SW6, SW7, SW11,
SW13 and SW13) falls in high priority whereas seven subwatersheds (SW1, SW4, SW8, SW9, SW10, SW12 and SW16)
fall in medium priority and two sub-watersheds (SW2 and SW15)
fall in low priority class. Fig. 4. shows priority of sub-watersheds.

Page 117

Praveen G. Deshbhandari et. al. / INDECS-15

Table 3. Prioritization sub-watersheds using morphological parameters


Subwatershed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Rb
10
14
1
13
3
9
4
7
11
12
5
16
2
6
8
15

Linear Parameters
Dd
Fs
15
16
16
15
12
10
13
13
3
2
1
1
6
11
7
12
8
8
5
4
2
3
11
7
4
6
9
5
14
14
10
9

T
15
16
7
14
1
3
5
6
11
9
4
13
2
8
10
12

Rc
2
10
1
3
12
6
11
9
5
15
14
8
7
4
16
13

Rf
2
16
1
4
8
14
7
15
11
9
10
5
13
3
12
6

Shape Parameters
Re
2
16
1
4
8
14
7
15
11
9
10
5
13
3
12
6

Cc
2
8
3
11
9
13
6
4
10
15
7
16
1
12
5
14

Compound
rank
8.00
13.87
4.50
9.37
5.75
7.62
7.12
9.37
9.37
9.75
6.87
10.12
6.00
6.25
11.37
10.62

Final
Priority
Medium
Low
High
Medium
High
High
High
Medium
Medium
Medium
High
Medium
High
High
Low
Medium

Fig. 4. Prioritzation of Sub-watersheds in Venkatapura Watershed

5. Conclusion
Watershed prioritization is one of the most important aspects
of natural resource management programme. A morphometric
analysis is an efficient technique for prioritization of subwatersheds. The present study demonstrates GIS based
conceptual method for prioritization of sub-watersheds in
Venkatapura Watershed. The results reveals that seven sub-

watersheds (SW3, SW5, SW6, SW7, SW11, SW13 and


SW13) shows very high priority. i.e. which are highly
susceptible to erosion.
The immediate water and
conservation measures can be required in these watersheds

Page 118

Praveen G. Deshbhandari et. al. / INDECS-15

Acknowledgements
The Authors are thankful to chairman of the department and
Co-ordinator, UGC-SAP (DRS-I) programme for extending
laboratory facilities.

References
[1] Aher PD, Adinarayana J, Gorantiwar SD. Quantification
of morphometric characterization and prioritization for
management planning in semi-arid tropics of India: A
remote sensing and GIS approach, 2014:511:850-860
[2] AIS and LUS, Watershed Atlas of India, Department of
Agriculture and Cooperation. All India Soil and Land
Use Survey, IARI Campus, New Delhi. 1990
[3] Akram J., Mohd YK . Rizwan A, Prioritization of Subwatersheds based on Morphometric and Land Use
Analysis using Remote Sensing and GIS Techniques, J.
Indian Soc. Remote Sens. 2009:37:261274
[4] Akram J, Mohd YK, Subah R Watershed prioritization
using morphometric and Land Use Land Cover: a
Remote sensing and GIS based approach. J Geol Soc
India, 2011: 78:6375
[5] Biswas, S, Sudhakar S, and Desai VR. Prioritization of
sub-watersheds based on morphometric analysis of
drainage basin, district Midnapore, West Bengal. J.
Indian Soc. Remote Sensing, 199:27:155-166.
[6] Chaitra R, Chandrakantha G. Soil Erosion Zonation
Mapping in Parts of Hulgi Watershed using Remote
Sensing and GIS Techniques A Case Study Int. J,
Scientific and Engineering Research, 2014:5: 2229-5518
[7] Gajanan KK, Ritesh V, Pawan K. L. Prioritization of
catchments based on soil erosion using remote sensing
and GIS. Environ Monit Assess, 2015: 187: 333
[8] Gajbhiye S, Mishra SK, Ashish P. Prioritizing erosionprone area through morphometric analysis: an RS and
GIS perspective. Appl. Water Sci, 2014: 4:5161[8]
[9] Gajbhiye S, Sharma SK, Meshram C. Prioritization of
Watershed through Sediment Yield Index Using RS and
GIS Approach. International Journal of u- and eService, Science and Technology, 2014:7:47-60
[10] Harlin, JM, Wijeyawickrema C. Irrigation and
groundwater depletion in Caddo county, Oklahoma.
JAWRA Journal of the American Water Resources
Association, 1985: 21:1522.
[11] Hasan RN, Mohammed AA, Hilal AG, Masood AS. Soil
erosion planning using sediment yield index method in

the Nun Nadi watershed, India, International Soil and


Water Conservation Research 2015:3: 8696
[12] Horton RE. Drainage basin characteristics. Transactions
of the American Geophysical Union, 1932: 13:350361.
[13] Horton RE. Erosional development of streams and their
drainage basins: Hydrophysical approach to quantitative
morphology. Geological Society of America Bulletin,
1945:56: 275370.
[14] Karthik B, Ramachandra TV. The spatial variation of
physic-chemical and hydrological parameters with land
use in Venkatapura catchment, Karnataka, Asian Jr of
Microbial Biotech Env. Sc, 2007 :9: 1001-1005
[15] Kuldeep P, Upasana P. Quantitative Morphometric
Analysis of a Watershed of Yamuna Basin, India using
ASTER (DEM) Data and GIS, International J.
Geomatics and Geosciences, 2011:2:248-269
[16] Luo W. (1900). Quantifying groundwater- sapping
landforms with a hypsometric technique. Journal of
Geophysical Research, 105, 16851694.
[17] Miller VC.
A Quantitative geomorphic study of
drainage basin characteristics on the Clinch Mountain
area, Virgina and Tennessee, Proj. Tech Rep 3, NR 389402, 1953
[18] Montgomery DR, Dietrich WE. Source areas, drainage
density and channel initiation. Water Resources
Research, 1989: 25, 19071918.
[19] Montgomery DR, Dietrich WE. Channel initiation and
the problem of landscape scale. Science, 1992: 255,
826830.
[20] Nag S, Chakraborty S. Influence of Rock Types and
structures in the Development of Drainage Network in
Hard rock area. J. Indian Soc. Remote Sensing,
2003:31:.25-35.
[21] Nooka RK, Srivastava YK, Venkateshwara RV,
Amminedu E, Murthy, K.S.R. Check dam positioning
by prioritization of micro-watersheds using SYI model
and morphometric analysis, J. of the Indian Society of
Remote Sensing, 2005:33:25-38
[22]Sadaf A, Gowhar M, Shakil A. Romshoo. Morphometry
and Land Cover Based Multi-Criteria Analysis for
Assessing the Soil Erosion Susceptibility of the Western
Himalayan Watershed. Environmental Mon., and
Assessment. 2014:186:83918412
[23] Schumms SA. Evolution of drainage systems and
slopes in badlands at Perth Amboy, New jersey. Bulletin
of the Geo., Society of America, 1956: 67:597646.

Page 119

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

EVALUATION OF WATER QUALITY INDEX FOR SURFACE AND


SUBSURFACE WATERS IN URBAN AREA OF BANGALORE
Prasad C S M V1 , Dr. Inayathulla M2
1

Research Scholar, Department of Civil Engineering, UVCE,Bangalore University,


Bangalore-560056,India; email: csmvprasad@gmail.com
2
Associate Professor, Department of Civil Engineering, UVCE, Bangalore University,
Bangalore-560056, India; email: drinayath@gmail.com

Abstract
The world's fresh water resources are unequally distributed both in time and in space. Until recently water
resource management focused on reallocating water to when and where it was required, a supply-side or fragmented
approach. Nowadays there are signs that water resource availability is dwindling - due to both population growth and
increased per capita water use - and ecosystems are being damaged. To face this challenge a new holistic approach is
needed. This approach includes the integrated or conjunctive use of surface and groundwater resources and takes
account of social, economic and environmental factors. Moreover, it recognizes the importance of water quality issues.
Land use has significant impact on water resources in terms of quantity and quality. The emphasis of
integrating land and water resources management seems inevitable as land use changes not only have a major impact on
water resources but also has a great potential for modifying hydrological cycle within a river basin.
The quality of the available water must be evaluated to see how it fits the intended use. Conceptually, water
quality refers to the characteristics of a water supply that will influence its suitability for a specific use. It is necessary
that the various stretches of water bodies are maintained at the characteristics quality, which will sustain the respective
uses of the stretches. It is possible to accomplish the task by controlling the pollution caused by human activities.
Keywords: Water Quality Index, Pollution, Electrical Conductivity, Total Dissolved Solids, Total Alkalinity
1. Introduction
The Greater Bangalore Metropolitan Area has spatially
grown from 69 sq.km (as in 1949) to 741 sq.km (as in
2012) with a population of 6.8 million. Meeting the
water needs of the growing population along with the
expansion of water supply networks to upcoming areas
of Bangalore city continue to remain as a major
challenge to the authorities. Bangalore North District has
seen sudden changes in terms of land use patterns.
The water table which was 80 feet deep below a few
years back, is today at a depth of 800 feet in some areas.
Indiscriminate discharge of untreated domestic sewage,
industrial effluents, leachates from solid waste dumps,
agricultural runoffs. etc into streams, lakes has resulted
in ground water pollution in Bangalore City and
surroundings.
Bangalore being situated on a ridge, the natural
drainage of Bangalore is divided into three valleys, viz.,
Hebbal Valley, Koramangala and Challaghatta Valley
and Vrishabhavathi Valley. The water demands of
Bangalore since times have been met by Hesaraghatta
lake, Tippagondanahalli reservoir and various stages of
*

Cauvery Water Supply Schemes. The supply of water


from the old reservoirs viz., Hesaraghatta lake and
Tippagondanahalli reservoir have become unreliable
owing to declined flow in Arkavathi river catchment.
Functioning of the Stage IV Phase II of Cauvery Water
Supply Scheme by 2012 will seal the upper limit of
water draft from Cauvery river to Bangalore City, as per
Cauvery Water Tribunal allotment. The current water
demand of Bangalore is estimated as 1219 million liters
per day. The availability of water from all the above
mentioned sources is 923 million litres per day, thus
creating a shortage of 296 million litres per day. The
problem is likely to aggravate more with the merger of
seven city municipality corporations and one town
municipality corporations with Bruhat Bangalore
Mahanagara Palike. Bangalore Water Supply and
Sewerage Board, the agency entrusted to supply water to
Bangalores population is finding it difficult to supply
the declared water supply of 110 LPCD, thus driving
more than 40% of the population to depend on ground
water. The water table which was 80 feet deep below a
few years back, is today at a depth of 800 feet in some
areas. Indiscriminate discharge of untreated domestic

Prasad C S M V, Tel.: +919972835799

E-mail-csmvprasad@gmail.com

Page 113

Prasad C S M V et. al. / INDECS-15

sewage, industrial effluents, leachates from solid waste


dumps, agricultural runoffs. etc into streams, lakes has
resulted in ground water pollution in Bangalore City and
surroundings. The lakes which catered to the water
supply needs of some localities have become useless due
discharge of effluents and many lakes have vanished due
to silting, solid waste dumping and also encroachments.
2. Aims and Objectives
The main aim of this lesson is to discuss the different
types of water qualities like physical, chemical and
microbial aspects and the objectives are
a.To study the physical, chemical and microbial qualities
of the water,
b.To learn the different types of procedures to analyze
the parameters of water qualities and
c.To learn different water quality standards and
guidelines for drinking, agricultural and industrial
purposes.
Water quality is a term used here to express the
suitability of water to sustain various
uses or processes. Any particular use will have certain
requirements for the physical,
chemical or biological characteristics of water; for
example limits on the concentrations of toxic substances
for drinking water use, or restrictions on temperature and
pH ranges for water supporting animals. Consequently,
water quality can be defined by a range of variables
which limit the water. Although many uses have some
common requirements for certain variables, each use will
have its own demands and influences on water quality.

laboratory. The main elements of water quality


monitoring are, therefore, on-site measurements, the
collection and analysis of water samples, the study and
evaluation of the analytical results, and the reporting of
the findings. One purpose of a monitoring programme is,
therefore, to gather sufficient data (by means of regular
or intensive sampling and analysis) to assess spatial and
or temporal variations in water quality.
4. Evaluation of Water Quality Index
4.1 Methodology
Surface and sub-surface water samples were collected
from HEBBAL area, and analysed for selected
parameters. The following methodology was adopted for
the study purpose.
4.2. Classification of Study Area
Surface and sub-surface water samples were collected at
16 different locations of HEBBAL region.
To study the quality of ground water and surface water
samples number of locations are as shown in the Fig.1

3. Water Quality
The composition of surface and underground water is
dependent on natural factors (geological, topographical,
meteorological, hydrological and biological) in the
drainage basin and varies with seasonal differences in
runoff volumes, weather conditions and water levels.
Large natural variations in water quality may, therefore,
be observed even where only a single watercourse is
involved. Human intervention also has significant effects
on water quality.
The effects of faecal pollution vary appreciably in space
and time. Eutrophication results not only from point
sources, such as wastewater discharges with high
nutrient loads (principally nitrogen and phosphorus), but
also from diffuse sources such as run-off from livestock
feedlots or agricultural land fertilized with organic and
inorganic fertilizers. Pollution from diffuse sources, such
as agricultural runoff, or from numerous small inputs
over a wide area, such as faecal pollution from
unsewered settlements, is particularly difficult to control.
The quality of water may be described in terms of the
concentration and state (dissolved or particulate) of some
or all of the organic and inorganic material present in the
water, together with certain physical characteristics of
the water. It is determined by in situ measurements and
by examination of water samples on site or in the

Fig. 1

4.3. Selection of Sampling Points

The already existing tube wells and open wells


were selected for the purpose of collecting
groundwater samples from the study area.

The sampling points were located in areas


where there was no treated water supply and
groundwater is the only source of drinking
water.

Page 114

Prasad C S M V et. al. / INDECS-15

4.4. Analysis and Discussions

K = Proportionality constant derived from the equation

Table 1. Water quality classification based on WQI value.

WQI value

Water quality status

<50
50-100
100-200
200-300
>300

Excellent
Good Water
Poor water
Very poor water
Water unsuitable for
drinking

K = 1/( n

(1.b)

1/Sn)

Sn = Recommended drinking water standard as per B I S.


5. Results and Discussions
The Water Quality Index of the groundwater in study
area of groundwater in study area is established from
important physicochemical parameters in different
seasons. The mean values of various physicochemical

Table 2. Drinking water standards, recommending agencies and unit weights

Sl.No.

Parameters

Standard Permissible Value


(Sn)

Recommended
Agency

1/Sn

Unit Weight

PH

8.5

ICMR/BIS

0.118

0.192

EC

300

ICMR

0.003

0.005

TDS

500

ICMR/BIS

0.002

0.003

Total Alkalinity

120

ICMR

0.008

0.014

Chlorides

250

ICMR

0.004

0.006

Total Hardness

300

ICMR/BIS

0.003

0.005

DO

ICMR/BIS

0.200

0.325

BOD

ICMR

0.200

0.325

Ca

75

ICMR/BIS

0.013

0.022

10

Mg

30

ICMR/BIS

0.033

0.054

11

Sulphate

150

ICMR/BIS

0.007

0.011

12

Nitrate

45

ICMR/BIS

0.022

0.036

4.5. Water Quality Index


WQI is a system of ranking the quality of water in the
environment by numerical means. The index assigns a
number to a body of water to indicate its quality. The
applications of WQI can be summarised as follows:
It can provide a way to summarise overall water
quality conditions in a manner that can be
communicated to general public.
It can tell us whether the overall quality of water
bodies poses a potential threat to various uses of
water.
It can be used as a broad tool to indicate success in
protection and remediation efforts.
Computation of water quality indices: The WQI can be
calculated using the formula as shown here.
WQI= Antilog W n n= 1 log 10 qn

(1)

Wn = K/Sn

(1.a)

Where Wn = Weight age factor computed from

parameters during seasons for the calculation of WQI


are presented in Table 3. Season wise Water Quality
Index calculations are depicted in Table 4, 5 and 6. The
Water Quality Index obtained for the groundwater in
different seasons of study period i.e., rainy season,
winter season and summer season are 163.812, 171.465
and 186.05 respectively, which indicate the poor quality
of water (Yogendra et.al, 2007).
5.1. pH
pH is one of the most important factors that determines
the suitability of water for various purposes. In the
present study pH value ranged between 7.0 and 7.9. The
average pH values of the lake water was7.2 during rainy
season, 7.5 during winter season and 7.7 during summer
season. pH of water was relatively high in the summer
season and low in monsoon and winter season. However,
when the average values for three seasons are taken into
account the water body was found to be slightly alkaline.
This is in accordance with earlier work done by Sisodia
(2006), Yogendra (2007) and Thakor (2011).
Page 115

Prasad C S M V et. al. / INDECS-15


Table 3. Seasonal variations of the physico-chemical parameters of groundwater in study area

Parameter

EC
(-S/cm)

TDS (Mg/l)

T. Alkalinity
(Mg/l)
Chlorides
(Mg/l)
Hardness
(Mg/l)

DO (Mg/l)

BOD (Mg/l)

Ca mg/l

Mg mg/l

Sulphates
mg/l

NO3 mg/l

Sampling Locations
L1

L2

L3

L4

L5

L6

Min.
Value

Rainy

7.1

7.3

7.0

7.6

7.1

7.2

7.0

7.6

7.2

Winter

7.8

7.8

7.6

7.3

7.0

7.2

7.0

7.8

7.5

Summer

7.8

7.9

7.6

7.7

7.4

7.6

7.4

7.9

7.7

Rainy

1192.5

1130.0

1296.0

1083.5

1087.0

1852.5

1083.5

1852.5

1273.6

Winter

1110.3

1080.8

1424.8

1079.8

1043.0

1778.0

1043.0

1778.0

1252.8

Summer

1388.7

1116.3

1549.3

1167.3

1207.7

1933.3

1116.3

1933.3

1393.8

Rainy

535.9

557.8

733.3

531.4

543.5

752.3

531.4

752.3

609.0

Winter

462.6

462.2

585.3

440.1

233.6

897.2

233.6

897.2

513.5

Summer

599.6

626.3

815.7

578.4

421.0

982.1

421.0

982.1

670.5

Rainy

287.5

196.3

267.5

200.0

206.3

410.0

196.3

410.0

241.3

Winter

216.0

199.0

288.5

209.0

211.5

439.0

199.0

439.0

244.5

Summer

194.7

162.7

330.7

168.0

169.3

417.3

162.7

417.3

250.4

Rainy

217.4

232.4

249.9

224.9

219.9

344.0

217.4

344.0

248.1

Winter

250.5

236.5

272.5

229.0

212.5

314.0

212.5

314.0

252.5

Summer

328.0

250.7

289.3

289.3

294.3

367.0

250.7

367.0

303.1

Rainy

237.2

271.2

338.4

224.9

217.8

456.3

217.8

456.3

291.0

Winter

206.6

211.9

380.6

216.5

220.0

440.9

206.6

440.9

279.4

Summer

244.0

198.7

387.7

208.0

203.8

508.7

198.7

508.7

291.8

Rainy

4.8

3.2

3.6

5.0

6.1

4.9

3.2

6.1

4.6

Winter

2.5

2.5

4.3

3.5

5.7

4.3

2.5

5.7

3.8

Summer

3.5

3.0

3.2

3.9

4.2

2.5

2.5

4.2

3.4

Rainy

16.0

16.0

20.0

12.0

16.5

22.0

12.0

22.0

17.1

Winter

18.8

17.5

17.0

13.3

11.3

25.5

11.3

25.5

17.2

Summer

23.3

15.0

17.3

14.3

14.3

26.0

14.3

26.0

18.4

Rainy

32.0

48.3

38.9

45.8

33.9

47.1

32.0

48.3

41.0

Winter

24.3

30.9

51.0

37.3

30.7

76.5

24.3

76.5

41.8

Summer

26.5

22.9

25.6

26.3

48.5

103.1

22.9

103.1

42.2

Rainy

38.4

36.8

46.7

27.0

32.5

59.7

27.0

59.7

40.2

winter

35.6

32.9

61.8

30.1

35.0

60.9

30.1

61.8

42.7

Summer

49.5

34.5

79.0

39.7

36.4

91.3

34.5

91.3

55.0

Rainy

13.2

13.4

24.6

9.7

13.2

15.6

9.7

24.6

14.9

Winter

14.4

10.6

31.5

11.4

11.8

24.7

10.6

31.5

17.4

Summer

13.8

7.4

29.7

9.1

17.2

28.1

7.4

29.7

17.6

Rainy

12.5

12.8

10.1

13.1

11.3

10.9

10.1

13.1

11.8

Winter

15.5

13.6

9.1

12.3

9.0

13.4

9.0

15.5

12.2

Summer

16.3

13.0

11.2

12.7

11.0

12.3

11.0

16.3

12.8

Season

Max.
value

Mean

Page 116

Prasad C S M V et. al. / INDECS-15


Table 4. Calculation of Water Quality index of groundwater in study area in Rainy Season
Rainy Season
Quality
Standard
Observed
Unit Weight
Rating
Sl.No.
Parameters
Value
1/Sn
Values
(Wn)
(Sn)
(Vn)
( n)

PH

8.5

0.118

0.192

7.2

13.333

EC

300

0.003

0.005

1273.6

424.533

TDS

500

0.002

0.003

609

121.800

Total Alkalinity

120

0.008

0.014

241.3

201.083

Chlorides

250

0.004

0.006

248.1

99.240

Total Hardness

300

0.003

0.005

291

97.000

DO

0.200

0.325

4.6

104.167

BOD

0.200

0.325

17.1

342.000

Ca

75

0.013

0.022

41

54.667

10

Mg

30

0.033

0.054

40.2

134.000

11

Sulphate

150

0.007

0.011

14.9

9.933

12

Nitrate

45

0.022

0.036

11.8

26.222

Wn=0.998

Weighted
(Wnqn)
2.555
2.297
0.395
2.720
0.644
0.525
33.820
111.039
1.183
7.251
0.108
0.946

Wnqn=163.48

Water Quality Index = 163.812


Table 5. Calculation of Water Quality of groundwater in study area index in Winter Season
Winter Season
Sl.No.

Parameters

Standard
Value
(Sn)

1/Sn

Unit Weight
(Wn)

Observed
Values
(Vn)

Quality
Rating

( n)

PH

8.5

0.118

0.192

7.5

33.333

EC

300

0.003

0.005

1252.8

417.600

TDS

500

0.002

0.003

513.5

102.700

Total Alkalinity

120

0.008

0.014

244.5

203.750

Chlorides

250

0.004

0.006

252.5

101.000

Total Hardness

300

0.003

0.005

279.4

93.133

DO

0.200

0.325

3.8

112.500

BOD

0.200

0.325

17.2

344.000

Ca

75

0.013

0.022

41.8

55.733

10

Mg

30

0.033

0.054

42.7

142.333

11

Sulphate

150

0.007

0.011

17.4

11.600

12

Nitrate

45

0.022

0.036

12.2

27.111

0.614

Wn=0.998

Weighted
(Wnqn)
6.387
2.260
0.333
2.756
0.656
0.504
36.526
111.688
1.206
7.702
0.126
0.978

Wnqn=171.12

Water Quality Index = 171.465

Page 117

Prasad C S M V et. al. / INDECS-15


Table 6. Calculation of Water Quality index of groundwater in study area in Summer Season
Summer Season
Sl.No.

Parameters

Standard
Value
(Sn)

1/Sn

Unit Weight
(Wn)

Observed
Values
(Vn)

Quality
Rating

( n)

PH

8.5

0.118

0.192

7.7

46.667

EC

300

0.003

0.005

1393.8

464.600

TDS

500

0.002

0.003

670.5

134.100

Total Alkalinity

120

0.008

0.014

250.4

208.667

Chlorides

250

0.004

0.006

303.1

121.240

Total Hardness

300

0.003

0.005

291.8

97.267

DO

0.200

0.325

3.4

116.667

BOD

0.200

0.325

18.4

368.000

Ca

75

0.013

0.022

42.2

56.267

10

Mg

30

0.033

0.054

55

183.333

11

Sulphate

150

0.007

0.011

17.6

11.733

12

Nitrate

45

0.022

0.036

12.8

28.444

0.614

Wn=0.998

Weighted
(Wnqn)
8.942
2.514
0.435
2.823
0.787
0.526
37.879
119.481
1.218
9.921
0.127
1.026
Wnqn=185.68

Water Quality Index = 186.05

Fig. 2 Graphical Representation of Seasonal WQI


The alkaline nature of water was a characteristic
throughout the study period with slight seasonal
variation in the lake. In the present investigation pH
value were within the ICMR standards (7.0- 7.9)
5.2. Electrical Conductivity (EC)
Electrical conductivity is the ability of water to allow
electric current through it and is expressed in -s/cm. In
the present the average EC values of the lake water was
1273.6 mhos/cm during rainy season, 1252.8
mhos/cm during winter season and 1393.8 mhos/cm
during summer season. Season wise it is found to be
high during summer season.

5.3 Total Dissolved Solids (TDS)


The total dissolved solids in Jakkur lake water was 609
mg/L in rainy season, 513.5 mg/L during winter season
and 670.5 mg/L during summer season. The
concentration of TDS is high in summer and rainy
season, which may be due to addition solids from runoff
water, sewage and industrial effluents to the lake.
5.4. Total Alkalinity
Alkalinity value less than 100 mg/L is desirable for
domestic use. According to ICMR the maximum
permissible limit is 120 mg/L. The observed average
Page 118

Prasad C S M V et. al. / INDECS-15


value of total alkalinity was 241.3 mg/L during rainy
season, 244.5 mg/L during winter season and 250.4
mg/L in summer season. Higher values of alkalinity
registered during summer might be due to the presence
of excess of CO2 product as a result of decomposition
process coupled with the mixing of sewage and domestic
waste. The low alkalinity during rainy season may be
due to dilution.

5.9. Calcium (Ca)

5.5. Chloride

5.10. Magnesium (Mg)

Chloride occurs in all types of natural waters. The high


concentration of chloride is considered to be an
indication of pollution due to high organic waste of
animal origin. Chloride value obtained in the study was
248.1 mg/L during rainy season, 252.5 mg/L during
winter season and 303.1 mg/L in summer season. The
chloride in groundwater was found within the acceptable
limit of 250mg/L in rainy season and above the
permissible limit in winter and summer season.

Magnesium occurs in water mainly due to the presence


of olivine, biotite, augite and talc minerals. Permissible
limit of magnesium is 30 mg/L. The observed average
value of magnesium was 40.2 mg/L during rainy season,
42.7 mg/L during winter season and 55.2 mg/L during
summer season. Agricultural runoff might be responsible
for the high levels of Magnesium in lake water.

5.6. Total Hardness

Sulphate ion does not affect the taste of water if present


in low concentration. The sulphate ion concentration was
14.9 mg/L during rainy season, 17.4 mg/L during winter
season and 17.6 mg/L during summer season. The
sulphate in ground water was found within the
acceptable limit of 150 mg/L.

The observed average total hardness value was 291 mg/L


during rainy season, 279.4 mg/L during winter season,
and 291.8 mg/L during summer season. Higher values of
hardness in summer and rainy season can be due to high
rate of evaporation of water and addition of calcium and
magnesium from surface runoff. Thakor (2011) stated
that addition of sewage and human use may be the
reason of elevation of hardness.
5.7. Dissolved Oxygen (DO)
The average dissolved oxygen was 4.6 mg/L during
rainy season, 3.8 mg/L during winter season and 3.4
mg/L during summer season. The maximum dissolved
oxygen in the water in study area was recorded in rainy
season. Thereafter it started declining gradually and in
summer reached the lowest concentration. This can be
attributed to addition of infiltration of effluents
containing oxidizable organic matter and consequent
biodegradation and decay of vegetation at higher
temperature leading to consumption of oxygen from
water.
5.8. Biochemical Oxygen Demand (BOD)
Biochemical oxygen demand is a parameter to assess the
organic load in a waterbody. Many researchers have
recorded higher BOD values in polluted water. The flow
of sewage water, industrial effluents and the agricultural
runoff might be responsible for the high levels of BOD.
The average BOD values of Jakkur lake was 17.1 mg/L
during rainy season, 17.2 mg/L during winter season and
18.4 mg/L during summer season. The values of BOD in
some areas were recorded much above permissible limits
indicating presence of decomposable organic matter in
ground water also.

Calcium occurs in water mainly due to the presence of


limestone, gypsum, dolomite and gypsiferrous minerals
(Srinivas et. al., 2011). Permissible limit of calcium is 75
mg/L. The observed average value of calcium was 41
mg/L during rainy season, 41.8mg/L during winter
season and 42.2 mg/L during summer season.

5.11. Sulphate

5.12. Nitrate
Nitrate is the most important nutrient in an ecosystem.
The source of nitrate is the biological oxidation of
organic nitrogenous substance (Kalavathy et. al., 2011).
Nitrate value obtained in the study was 11.8 mg/L during
rainy season, 12.2 mg/L during winter season and 12.8
mg/L during summer season. In the present study water
samples of all the seasons showed low concentration of
nitrate well below permissible levels as per the
standards.
6. Conclusions
At the outset, high value of WQI (Rainy season 163.812, Winter - 171.465 and in Summer-186.05)
indicates that the water quality is poor and unsuitable for
drinking. Further, the seasonal values of WQI indicate
that in summer season lake water is more affected than
during rainy and winter season. The seasonal variations
of index values are due to variation in physicochemical
characteristics of lake water. The reason for high index
values were continuous discharge of agricultural runoff,
industrial effluents and inadequately treated sewage
effluent flowing into the streams in the study area.
Application of water quality index (WQI) in this study
has been found useful in assessing the overall quality of
water. This method appears to be more systematic and
gives comparative evaluation of the water quality in
different seasons of the year. It is also helpful for public
to understand the quality of water as well as being a
Page 119

Prasad C S M V et. al. / INDECS-15


useful tool in many ways in the field of water quality
management.
7. References
[1]. APHA (1995): Standard Methods (18 Ed.) for
the examination of water and waste water,
APHA, AWWA, WPCE, Washington DC.
[2]. Bhaven N. Tandel, JEM Macwan and Chirag K.
Soni (2011): Assessment of Water Quality Index
of small lake in south Gujarath region, India,
Proceedings of ISEM-2011, Thailand.
[3]. BIS 10500, 1991: Specifications for drinking
water, Indian Standard Institutions (Bureau of
Indian Standards), New Delhi.
[4]. Brown R.M, N.J. Mccleiland, R.A. Deiniger,
M.F.A. Oconnor (1972), Water quality index
crossing the physical barrier, Proc. Int. Conf. on
water pollution research, Jerusalem, 6, 787-797.
[5]. ICMR Manual of standards of quality for
drinking water supplies (1975). ICMR, New
Delhi.
[6]. Kalavathy S., T. Rakesh Sharma and P.
Sureshkumar (2011), Water uality Index River
Cauvery in Tiruchirappalli district, Tamilnadu,
Arch. Environ. Sci., 5, 55-61.
[7]. Kankal N.C, M.M. Indurkar, S.K. Gudadhe and
S.R. Wate (2012), Water Quality Index of

Surface Water Bodies of Gujarat, India, Asian J.


Exp. Sci. Vol. 26, No1; 39-48
[8]. Ramakrishnaiah C.R, C. Sadashivaiah and G.
Ranganna (2009), Assessment of Water Quality
Index for theGroundwater in Tumkur Taluk,
Karnataka State, India; E-Journal of Chemistry,
6(2), 523-530.
[9]. Sisodia Rashmi and Chaturbhuj Moundiotiya
(2006), Assessment of the waterquality index of
Wetland Kalakho lake, Rajasthan, India; Journal
of Environmental hydrology; 14 (23),1-11.
[10].Srinivas P., G.N Pradeep kumar, A. Srinivas
Prasad and T. hemalatha (2011), Generation of
Groundwater Quality Index Map A case study,
Civil and Environmental Research; 1 (2), 9-21.
[11].Thakor F.J, D.K Bhoi, S.N Pandya and Nikitaraj
B. Chauhan (2011), Water Quality Index(WQI)
of Pariyej Lake Dist. Kheda Gujarat, Current
World Environment; 6 (2), 225-231.
[12].WHO (1992). International Standards for
Drinking Water. World Health Organization,
Geneva, Switzerland
[13].Yogendra
K.
and
E.T.
Puttaiah
(2008),Determination of water quality index and
Suitability of an Urban Waterbody, Shimoga
Town, Karnataka, Proceedings of 12th world
Lake Conference 342-346.

Page 120

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

HYDROLOGIC ANALYSIS OF WATERSHED USING


IDF CURVES & RUNOFF MODELS
M Inayathulla1 , M. Raghunath2, C. Rangaraj3& Y A Narayanaswamy 4
1

U.V.C.E Bangalore University. Bangalore- 560056, India, drinayath@gmail.com


Assistant Director, SPFU-TEQIP, DTE Bangalore India, raghu.spfu@hotmail.com
3
Professor in Civil Engineering, Sri SIT, Tumkur, India, drrangarajc@gmail.com
4
SET Jain University, Ramanagara District-562112, India, yanresearchjain@gmail.com
2

ABSTRACT
Quantification of short duration high intensity rainfall is generally done using IDF (IntensityDuration-Frequency) curves, based on historic rainfall data of significant years. Due to non-availability of
short duration rainfall data, an attempt is made to derive short duration empirical reduction formula to
understand urban hydrology. Bangalore is a rapidly growing city in terms of population and intense urban
growth. Today about 70 per cent of the 262 water tanks in 1961 in Bangalore have disappeared leading to
surface flooding. Daily rainfall data of5 stations for the years 1998 to 2011 collected from Indian
Meteorological Department (IMD) were used in the study. The missing rainfall data, during this period was
interpolated by Airthematic_mean method. The IMD empirical reduction formula was used to estimate the short
duration rainfall. The rainfall depth for various return periods were predicted using different probability
distributions and analyzed. The Chi-Square goodness of fit was used, to arrive at the best statistical distribution
among Normal, Log-Normal, Gumbel and Pearson. Chi-Square test showed that log-normal is the best
probability distribution for the 5 stations considered. The IDF curves were plotted for short duration rainfall of
5, 10, 15, 30, 60, 120, 720 and 1440 minutes for a return period of 2, 3, 5, 10, 25, 50, 100 and 200 years for
stations with peak rainfall values.

Key Words: Sub-watersheds-Hydrological studiesIDF Curves, Runoff models.


1. Introduction
With the nonlinear interactions between rainfall
and runoff processes as described by various urban
runoff models, synthetic design storms are required
for the estimation of the complete runoff
hydrographs for urban drainage design surface and
groundwater
management
purposes. Many
frameworks have been conceived in different
countries for the computation of design storms [1],
with varying shapes, storm durations, time to peak,
maximum intensity and total volume of rainfall;
however, none matches every situation, forcing
hydrologists to perform assessment processes
before using a design-storm model at a new site
[2]. Peyron et al. (2005)[3] proposed a procedure
to systematically evaluate design storms models for
specific locations. First, the targeted design storms
for different return periods, based on rainfall data
from Intensity-Duration-Frequency (IDF) curves,
were
constructed.
Then,
these
synthetic
hyetographs were used as input in the model to
obtain the respective runoff values (peak flows and
volumes). Thus, the runoff properties estimated
from the design storms were compared to those
values obtained from observed historical storms to
assess the accuracy of different design storm
models. Based on this approach, the best design
storm can be selected for the design of urban
drainage systems and groundwater recharge
structures.

Keeping the above aspects in view, Suryanagar


sub watershed of Urban Bangalore is taken up for
hydrological studies. These studies are concerned
about water conservation, storage and management
with the motto of recharging groundwater.
2. Study Area
The study area is located between Latitude
124732N and Longitude 774159 E as shown in
figure .1 The study area covers an area of 172.42
km2 and attains maximum elevation 950m and
minimum elevation of 880m. Suryanagara Township
is situated on the Anekal main road, Chandapura
near by cities Benahalii, Attibele, Bangalore.
Suryanagara located at distance of 25 km from
Bangalore. physiography of the area is characterized
by undulating topography with pediplains, pediment
and valley fills. The mean annual total rainfall is
about 920 mm with about 60 rainy days a year over
the last ten years.. The summer temperature ranges
from 17 C to 36 C, while the winter temperature
ranges from 12 C to 25 C. Thus, Bangalore enjoys
a salubrious climate all round the year.
The area of the watershed is obtained from
delineating the toposheets covering 57 H/9 and 57
H/10 of 1:50000 scale by using ARC GIS software.
The area of the watershed is found to be 172.42km2

Page 121

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

THIESSEN POLYGON MAP

Hosahalli
Sarjapura

Jigani

Attibele

Anekal

Source; SOI Toposheets 57H/9


and 57H/10

Legend
0

4
Kilometers

Source; SOI

Stations
0

1,950 3,900

7,800

11,700

15,600
Meters

Boundary
DPR THIESSEN POLYGON

Study area- Suryanagar (Source : SOI Toposheet)

Thiessen polygon showing rain gauge stations

Fig No.1: Location Map of Study Area


3. Intensity-Duration-Frequency-Curves
IDF stands for Intensity-Duration Frequency.
Rainfall intensity is defined as the ratio of the total
amount of rain (rainfall depth) falling during a
given period to the duration of the period It is
expressed in depth units per unit time, usually as
mm per hour (mm/h). The period of time over
which rainfall is measured is called duration. The
number of times, during a specified period of years,
that precipitation of a certain magnitude or greater

occurs or will occur at a station is called frequency.


[4]. The IDF-relationships give an idea about the
frequency or return period of a mean rainfall
intensity or rainfall volume that can be expected
within a certain period, i.e. the storm duration. In
this sense the storm duration is an artificial
parameter that can comprise any part of a rainfall
event. (IDFCURVE, 2012)
4. Frequency Analysis using Frequency Factor

Page 122

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015
The magnitude of xT of a hydrologic event may be
represented as the mean plus the departure xT
of the variate from the mean i.e., x = + x [5]
The departure may be taken as equal to the product
of and a frequency factor KT are functions of the
return period and the type of distribution to be used
in the analysis. The above equation may be
expressed as
= +
which may be

approximated by = +
. [5]. In the event
the variable analysed is y = In(x), then the same
method is applied to the statistics for the logarithms
of the data using
= +
and the required
value of is foundby taking the antilog of [5].
Short duration rainfall using IMD for Hosahalli
raingauge station is tabulated in Table 1

Table 1: Short duration rainfall by using IMD empirical formula for Hosahalli station

Rainfall
(mm)

Year

Duration in
Minutes
1998
82
1999
80.4
2000
120.4
2001
112
2002
25
2003
47
2004
76.8
2005
68.4
2006
45.4
2007
53.4
2008
70.1
2009
95.2
2010
56.2
2011
68.1

10

12.42
12.17
18.23
16.96
3.79
7.12
11.63
10.36
6.87
8.09
10.62
14.42
8.51
10.31

15.64
15.34
22.97
21.37
4.77
8.97
14.65
13.05
8.66
10.19
13.37
18.16
10.72
12.99

in mm where, time t is in hours

15

30

60

17.91
17.56
26.29
24.46
5.46
10.26
16.77
14.94
9.92
11.66
15.31
20.79
12.27
14.87

22.56
22.12
33.13
30.82
6.88
12.93
21.13
18.82
12.49
14.69
19.29
26.20
15.46
18.74

28.43
27.87
41.74
38.83
8.67
16.29
26.63
23.71
15.74
18.51
24.30
33.00
19.48
23.61

5. Normal distribution
Normal probability distribution, also called
Gaussian distribution refers to a family of
distributions that are bell shaped. The PDF for a
normal random variable x is
=

exp

> 0

Where exp is the exponential function with base e


= 2.718. is the mean and the standard deviation.
1/ ( (2)) is a constant factor that makes the area
under the curve of f(x) from - to equal to 1.
The curve of f(x) is symmetric with respect to x =
because the exponent is quadratic. Hence for =
0 it is symmetric with respect to the y-axis x = 0.
The frequency factor for normal distribution is
given by
=
. This is same as the standard
normal variate z i.e., frequency factor
= z.
Figure 2 shows the variation of Normal distribution
for different durations and results are tabulated in
Table2. The scope of this study was to develop IDF
curve and to derive IDF empirical formulae for the
5 stations considered for the study area
Suryanagara Bangalore, so that the estimation of
rainfall depth and intensity for any standard
duration and return period in the study area
considered can be obtained with minimum effort.
And also to estimate the surface runoff for the
study area by using different methods can be
obtained
with
minimum
effort.

120
35.82
35.12
52.59
48.92
10.92
20.53
33.55
29.88
19.83
23.32
30.62
41.58
24.55
29.75

720
65.08
63.81
95.56
88.89
19.84
37.30
60.96
54.29
36.03
42.38
55.64
75.56
44.61
54.05

1440
82.00
80.40
120.40
112.00
25.00
47.00
76.80
68.40
45.40
53.40
70.10
95.20
56.20
68.10

Daily rainfall data for 14 years i.e., 1998 to 2011


was collected for 5 stations in and around
Suryanagara,
Bangalore
from
Indian
Meteorological Department (IMD), Government of
India. The missing rainfall values were calculated
using the airthematic mean method and the IMD
empirical reduction formula was used to estimate
the short duration rainfall. Using different
probability distributions the rainfall depth was
found out for different durations and standard
return period, and subsequently the rainfall
intensity was found out for calculated rainfall
depths. The Chi-Square goodness of fit was used to
arrive at the best statistical distribution among
Normal, Log-Normal, Gumbel and Pearson.IDF
curve was plotted for short duration rainfall of 5,
10, 15, 30, 60, 120, 720 and 1440 minutes for a
return period of 2, 3, 5, 10, 25, 50, 100 and 200
years for station with peak rainfall values. The use
of IDF curves becomes cumbersome and hence a
generalized empirical relationship was developed
through method of least squares. The daily 24 hour
rainfall data for the years 1998 to 2011 was
collected from IMD for 5 stations located in and
around Suryanagara, Bangalore. The five stations
are Anekal, Attibele, Jigani, Sarjapura and Hosalli.
The missing rainfall values for the years 1998 to
2011 were calculated and tabulated in table3.

Page 122

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

0.1

0.06

0.08

0.04

0.06
0.04

0.02

0.02

0
0

10

15

20

Normal For 5 Min

10

15

20

25

Normal For 10 Min

0.08

0.06

0.06

0.04

0.04

0.02

0.02
0
0

10

20

30

10

Normal For 15 Min

20

30

40

Normal For 30 Min

Fig. No2: Variation of Normal Distribution For Different Durations


Table 2 : Estimation of maximum rainfall intensity for various return period by normal distribution for Hosahalli
Normal
Distribution
Duration in
minutes
5
10
15
30
60
120
720
1440
Normal
Distribution
5
10
15
30
60
120
720
1440

Return period T

Mean

Standard
Deviation

10.82
13.63
15.61
19.66
24.77
31.21
56.72
71.46

3.95
4.98
5.69
7.17
9.04
11.39
20.7
26.08

2
Rainfall
Depth
(mm)
13.55
17.07
19.54
24.62
31.02
39.09
71.03
89.49

Return period T
10.82
13.63
15.61
19.66
24.77
31.21
56.72
71.46

3.95
4.98
5.69
7.17
9.04
11.39
20.7
26.08

3
Rainfall
(mm/hr)
162.62
102.44
78.18
49.24
31.02
19.54
5.92
3.73

Rainfall
Depth
(mm)
13.78
17.36
19.87
25.03
31.54
39.74
72.22
90.98

25
14.10
17.77
20.34
25.62
32.29
40.68
73.93
93.14

5
Rainfall
(mm/hr)
165.32
104.15
79.48
50.05
31.54
19.87
6.02
3.79

Rainfall
Depth
(mm)
13.93
17.55
20.10
25.31
31.89
40.19
73.03
92.01

50
169.26
106.62
81.36
51.24
32.29
20.34
6.16
3.88

14.12
17.80
20.37
25.66
32.33
40.74
74.03
93.27

10
Rainfall
(mm/hr)
167.20
105.33
80.38
50.63
31.89
20.09
6.09
3.83

Rainfall
Depth
(mm)
14.04
17.69
20.25
25.51
32.15
40.50
73.61
92.74

Rainfall
(mm/hr)
168.52
106.16
81.01
51.02
32.15
20.25
6.13
3.86

100
169.49
106.77
81.48
51.31
32.33
20.37
6.17
3.89

14.13
17.81
20.38
25.67
32.35
40.77
74.09
93.34

200
169.60
106.84
81.53
51.35
32.35
20.38
6.17
3.89

14.13
17.81
20.38
25.68
32.35
40.77
74.09
93.34

169.61
106.85
81.54
51.36
32.35
20.38
6.17
3.89

Page 123

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Duration in
minutes
5
10
15
30
60
120
720
1440

Table 3: Mean and Standard Deviations of Short Duration Rainfall


Standard
Standard
Standard
Mean
Mean
Mean
Deviation
Deviation
Deviation
HOSAHALLI
ANEKAL
ATTIBELE
10.82
3.95
11.45
3.95
11.13
4.48
13.63
4.98
14.43
4.74
14.02
5.64
15.61
5.69
16.52
5.43
16.05
6.46
19.66
7.17
20.81
6.84
20.22
8.14
24.77
9.04
26.22
8.62
25.48
10.25
31.21
11.39
33.04
10.86
32.10
12.92
56.72
20.7
60.03
19.74
58.33
23.48
71.46
26.08
71.46
25.24
73.49
29.65

5
10
15
30
60
120
720

Standard
Deviation
SARJAPURA
9.68
2.91
12.20
3.66
13.96
4.20
17.59
5.29
22.17
6.66
27.93
8.39
50.75
15.25

1440

63.94

Mean

Duration in
minutes

JIGANI

20.73

6. Rainfall depth and intensity


The short duration rainfall depths were
calculated for the years 1998 to 2011 from IMD
empirical reduction formula. Then the mean and
standard deviations of short durations of 5, 10, 15,
30, 60, 120, 720 and 1440 minutes were estimated.
These estimated mean and standard deviations
were used in Normal, Log-Normal, Gumbel and
Pearson
probability distribution methods to

Standard
Deviation

Mean
15.46
19.48
22.3
28.09
35.39
44.59
81.03

6.84
8.62
9.87
12.44
15.67
19.74
35.88

102.09

45.2

determine the rainfall depths and intensity for


standard return periods of 2, 3, 5, 10, 25, 50, 100
and 200 years for 5 stations. It was found that the
rainfall depths increased with the increasing time
duration. But the rainfall intensity decreased
appreciably with increasing duration. These
distributions were subjected to chi-square goodness
of fit test to find the best distribution. The table4
shows specimen calculations for Hosahalli station

Table 4: Specimen calculations for Hosahalli station


NORMAL
DISTRIBUTION

Duratio
Observe
n in
d values
minutes
Expected
values
5
10
15
30
60
120
720
1440

10.82
13.63
15.61
19.66
24.77
31.21
56.72
71.46

13.97
17.61
20.15
25.39
31.99
40.31
73.25
92.29

Chisquare
values
0.71
0.90
1.02
1.29
1.63
2.05
3.73
4.70

LOG-NORMAL
DISTRIBUTION
Expected
values
13.48
16.99
19.45
24.50
30.87
38.89
70.68
89.05

7. IDF curve
It was found from chi-square test that lognormal distribution gave the best results with
minimum deviations from the observed values.
Hence the IDF curve was plotted from log-normal

Chisquare
values
0.52
0.66
0.76
0.96
1.21
1.52
2.76
3.47

GUMBELS
DISTRIBUTION
Expected
values
17.52
22.08
25.26
31.82
40.11
50.53
91.84
115.71

Chisquare
values
2.56
3.23
3.69
4.65
5.87
7.39
13.43
16.92

PEARSON TYPE
III
DISTRIBUTION
ChiExpected
square
values
values
16.17
1.77
20.37
2.23
23.31
2.54
29.37
3.21
37.06
4.08
46.63
5.10
84.75
9.27
106.77
11.68

values for each station considered. The IDF curve


is plotted with duration in minutes on the abscissa
and rainfall intensity in mm/hr on the ordinate for
standard return periods.

Page 124

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015
Figure 3 represents the rainfall IDF curves for five
stations in the study area i.e., rainfall intensityduration-frequency curve for short durations of 5,
10, 15, 30, 60, 120, 720 and 1440 minutes and
return periods of 2, 3, 5, 10, 25, 50, 100 and 200
years for Log-normal distribution. The use of IDF
curves becomes more cumbersome and hence a
generalized empirical relationship of the form i
x t
was developed for each station, for the
various return period considered. Rainfall IDF
empirical equation constant x and y were calculated
for different return period by the method of leastsquares. IDF empirical equation was formed by
putting the value of x and y in the mentioned
equation format for each return period separately.
Table 5 gives the empirical constant x for 5 stations
for the return periods considered.
It is seen that the empirical constant y remains
constant for all return period and for all stations
with a value of 0.6667 or 2/3. The empirical

Station
2
465
478
483
408
678

Table 5: Empirical constant x for 5 stations for different return periods


Return Periods
3
5
10
25
50
100
474
481
486
489
489
490
487
493
498
500
501
502
494
502
507
510
511
512
415
420
424
426
426
427
695
707
715
720
721
722

Rainfall Intensity(mm/hr)

175
150
125
100
75
50
25
0
1

10

100
1000
Duration (min)

10000

200
490
502
512
427
722

250
225
200
175
150
125
100
75
50
25
0
1

10

HOSAHALLI

Rainfall Intensity(mm/hr)

Rainfall Intensity(mm/hr)

Hosahalli
Anekal
Attibele
Sarjapura
Jigani

constant x varies at lower return periods and tends


to become constant with higher return periods.
These IDF empirical equations will help to estimate
the rainfall intensity for any specific return period
in Urban in a short time and more easily for the
locations considered
8. Soil Conservation Service (SCS) Curve
Number Model:
The SCS developed an index, which is called as the
runoff Curve Number (CN) to represent the
combined hydrologic effect of soil, land use,
agriculture treatment class, hydrologic condition
and antecedent soil moisture. These factors can be
accessed from soil surveys, site investigations and
land use maps, while using the SCS hydrologic
models for design. Fig 4, 5 & 6 show the Curve
Number Map generated, rainfall runoff relationship
and CWC Hydrograph for the study area. Weighted
curve number and runoff estimation is tabulated in
Table 6 & 7.

100

1000

10000

JIGANI

200
175
150
125
100
75
50
25
0
1

10
100
Duration(min)

1000

10000

ANEKAL
Fig. 3: IDF Curves for Rain Gauge Stations

Page 125

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

CURVE NUMBER MAP

Legend
73

CN

Source; Soil map, Land use/Land cover map and


Hydrological soil group map
0

1,450 2,900

5,800

8,700

77

53

81

60

83

66

86

67

88

71

91

72

96

11,600
Meters

Figure 4: Curve Number Map


Table 6. Weighted Curve Number
Watershed

Area (Sq km)

CNI

CNII

CNIII

SURYANAGARA

172.42

58.13

76

88.12

Table 7: Runoff Estimated for Surynagara Catchment


Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Average
Maximum
Minimum

Rainfall(mm)
1387.3
1518.4
1374
886.6
669.3
617.8
1111.2
1204.2
417.2
1076.6
1079.8
1196.5
898.8
903.2
584.4
995.02
1518.4
417.2

Runoff(mm)
498.81
560.48
427.69
313.96
227.69
58.28
251.24
352.60
48.66
259.70
329.57
390.71
209.69
240.18
141.54
287.39
560.48
48.66

Page 126

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

suryanagara watershed

600
y = 0.4259x - 136.37
R = 0.8817

500

300
200
100
0

200

400

600

800

1000

1200

1400

1600

Rainfall(mm)

Figure 5: Rainfall-Runoff relationship of watershed.

Discharges(m3/s)

Runoff (mm)

400

120
110
100
90
80
70
60
50
40
30
20
10
0

tm = 4.84hr
W75=2.45h
r
WR75=1hr
W50=4.02h
r

Trial1
Trial2

WR50=1.52hr

10

12

14

16

Time (hrs)

Fig 6: CWC Unit Hydrograph for the Suryanagara watershed

Page 127

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015
9. CONCLUSIONS;

Among the various available probability


distribution functions Log_ Normal
distribution had the best approximation of
rainfall intensity for various return periods.
Study showed that i = x * (td)-y was the best
form of IDF empirical equation for
Suryanagara, Bangalore.
These IDF equations will help to estimate
the rainfall intensity for any specific return
period in Suryanagara, Bangalore in a short
time and more easily.
The runoff models developed in this study
(ie..Rational, SCS- CN model and Unit
hydrograph method) is useful for designing
surface drain network for recharging ground
water and for surface water management.
References
[1]
Marsalek, J. and Watt, W.E. (1984) Design
Storms for Urban Drainage Design.
Canadian Journal of Civil Engineering
11(3), 574-584.
[2]
Peyron, N., Nguyen, V.T.V. and Rivard, G.
(2005) Un modele optimal de pluie de projet
pour la conception des reseaux de drainage
urbain. Annales du batiment et des travaux
publics, 35-42.
[3]
Peyron, N. (2001) Design Storms for Urban
Runoff Estimation. Project Report, McGill
University, Montral, Canada.
[4]
Food and Agriculture Organization, (2012),
Rainfall
Runoff
Analysis,
Rainfall
Characteristics,
http://www.fao.org/docrep/U3160E/u3160e
05.htm
[5]
Chow V.T., D.R. Maidment and L.W.Mays,
1988, Applied Hydrology, McGraw- Hill,
Chapter 10 Probability, Risk and
Uncertainty Analysis for Hydrologic and
Hydraulic Design: 361

Page 128

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

GROUNDWATER PROSPECTUS MAP FOR KOLAR TALUK SUBWATERSHEDS


M Inayathulla 1*, Y A Narayanaswamy2, Shashishankar.A3 & Chalapathi k4
1

U.V.C.E Bangalore University. Bangalore- 560056, India, drinayath@gmail.com


SET Jain University, Ramanagara District-562112, India, yanresearchjain@gmail.com
3
SET Jain University, Ramanagara District-562112, India, shashianant@gmail.com
4
U.V.C.E Bangalore University. Bangalore- 560056, India, chalapathi10k@gmail.com

ABSTRACT
The present work is an attempt to integrate RS and GIS based analysis and methodology in
groundwater potential zone identification in the Kolar Taluk subwatersheds, Kolar Ditrict, study area. The
information on geology, geomorphology, soil, slope, Lineaments, Drainage network and land use/land cover
was gathered, in addition, GIS platform was used for the integration of various themes. The composite map
generated was further classified according to the spatial variation of the groundwater potential. Six categories
of groundwater potential zones namely poor to nil, moderate to poor, moderate, good to moderate, good and
very good were identified and delineated. The hydrogeomorphological units like pediplain weathered and
alluvial plain and are potential zones for groundwater exploration and development and valley fills associated
with lineaments is highly promising area for ground water recharging. The spatial variation of the potential
indicates that groundwater occurrence is controlled by geology, land use / land cover, slope and landforms.
Key Words: Sub-watersheds-Hydrological studies- -Groundwater-water , Groundwater prospectus map.

1. Introduction
Groundwater is the water that is found underneath
the Earth's surface at profundities where all the
pore (open) spaces in the soil, sediments, or rock
are completely stacked with water. Groundwater of
any structure whether from a shallow well or a
significant well, devises and is refilled (energized)
by precipitation. Groundwater is a piece of the
hydrologic cycle, beginning when a piece of the
precipitation that falls on the Earth's surface sinks
(infiltrates) through the soil and enters (seeps)
diving to wind up groundwater. Groundwater will
at long last come back to the surface, discharging
to streams, springs, lakes, or the oceans, to
complete the hydrologic cycle.
Geophysical methods are conventionally employed
for groundwater prospecting though there are
several methodologies to locate and map the
occurrence and distribution of groundwater. The
advent and development of new technologies, such
as remote sensing with its advantages of spatial,
spectral and temporal availability of data have
proved to be useful for quick and useful baseline
information about the factors controlling the
occurrence and movement of groundwater like
geology, geomorphology, land use/cover, drainage
patterns, lineaments etc. Further, remote sensing

techniques provide a synoptic view of large areas,


facilitating better and quicker assessment,
development and management of water resources
with collateral information.
The integration of remote sensing and GIS has
proven to be an efficient tool in groundwater
studies [1,2] where remote sensing serves as the
preliminary inventory method to understand the
groundwater prospects/conditions and GIS enables
integration and management of multi-thematic
data. In addition, the advantage of using remote
sensing techniques together with GPS in a single
platform and integration of GIS techniques
facilitated better data analysis and their
interpretations.
Krishnamurthy et al.[3] applied digital
enhancement techniques for groundwater in a hard
rock terrain of parts of the Raichur district,
Karnataka. The geological structures were
highlighted by filtering. Band subtraction brought
out the vegetation along valley fills and moistureladen lineaments. Based on the results, a package
was suggested which could be used on an
operational basis for groundwater targeting in
typical hard-rock crystalline formations.
Ramamurthy et al. [4] a study was conducted to
find out the groundwater potential zones in

Page 129

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Varahanadhi Sub basin, Tamil Nadu India.


Thematic maps such as geology, geomorphology,
land use/land cover, lineament, drainage, soil,
rainfall maps are prepared in the Arc GIS. The
thematic maps and data are integrated in to GIS
environment where it is digitized, registered, stored
analyzed and finally overlaid to classify the
groundwater potential zones.
Ramakrishna and D.Nagaraju,[5] utilized IRSLISS-II data along with other data sets extract
information on the hydrogeomoiphic features of a
hard rock terrain in the Piriyapatna area of mysore
district of Karnataka, India. Tattekere watershed is
in general part of hard rock ter-rain, which is
mainly covered by Granite Gneissic Complex.
Based on the integrated thematic maps, weighted
analysis in Arc GIS ground water resource prospect
map of the area has been prepared and discussed.
2. Study area
The study area geographically lies between North
latitude 12 46 to 13 58 and East Longitude 77
21 to 78 35 as shown in figure 3.3 The study
area covers an area of 231..37 km2 and attains
maximum elevation 920m and minimum elevation
of 820m. It is located at distance of 72 km from
Bangalore. It is land locked district and hard rock
terrain of Karnataka in the maiden (plain) region.
The district lies almost in the central part of
peninsular India, which has immense bearing on its
geoclimatic conditions. The topography of the
district is undulating to plain. Kolar district falls in
the Eastern dry agro climatic Zone. The mean
annual total rainfall is about 744 mm with about 60
rainy days a year over the last ten years. The
summer temperature ranges from 20 C to 40 C,
while the winter temperature ranges from 10 C to
22 C. Thus, entire district experiences tropical
climate throughout the year, Kolar district owes its
prosperity and development to the existence of
ancient tanks.
The area of the watershed is
obtained from delineating the toposheets covering
57K/4 on 1:50000 scale by using ARC GIS
software. The area of the watershed is found to be
231.37km2 The Location map of the study area is
shown in Fig: .1
3. Morphometric Analysis

based on a hierarchic ranking of streams. Stream


length measurements and statistical analysis of a
stream length of overland flow is among the most
commonly used attributes in linear aspects.
One of the first attribute to be quantified in
morphometric analysis is the designation of the
stream orders. The channel segments are ordered
numerically from streams head water to point
somewhere down stream. Numerical ordering
begins with the tributaries at the streams headwater
being assigned as 1st Stream segment resulting
from the joining two 1st order segments is given an
order of 2nd Two second order streams form a 3rd
order stream, and so on. The highest order stream is
known as trunk or principal stream through which
all discharge of the watershed passes through the
outlet. In the present study, the highest stream
order obtained is 5th order and hence it is designed
was 5th order and hence it was designed as 5th
order watershed. In Order to know the total number
of streams in each order, the segment of each order
was numbered. The number of stream segments of
any given order will be lesser than the next lower
order but more numerous than the next higher
order. The concept of stream order is used to
calculate other indicators of drainage character of a
watershed. Fig 2 shows the stream order assigned
to Kolar Taluk Sub watershed for the
morphometric analysis.
Plotting the number of streams and stream order
(Fig 3) reveals a good relationship between them.
Table 1&2 shows all the morphometric analysis of
the study area.
Table 1&2 shows the watershed wise
morphometric characteristics and parameters
respectively of Kolar Taluk subwatersheds. A fig 3,
4 & 5 shows the regression of stream order on
number of stream segments of the Kolar Taluk sub
watersheds and the regression of stream order on
mean stream length of the Kolar Taluk sub
watersheds.
The drainage density reflects the land use, affects
the infiltration and watershed response time
between precipitation and discharge. The drainage
density of the area varies from 1.38 to 2.95
km/sq.km indicating that the sub watersheds are
coarse textured.

For morphometric analysis the drainage map from


SOI on 1:50000 has used. The morphometric
analysis was divided into 3 types as linear aspects,
areal aspects and relief aspects. Linear aspects
include the measurements of the linear features of
drainage system such as stream order, stream
length, etc. The allocation of the stream orders is

Page 130

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE,
Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig:1 The Location map of the study area

Fig2: Stream Order Map of Watershed

Page 131

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Table 1 Sub watershed wise morphometric characteristics of Kolar taluk watershed


WS
No

Area
(Sq.km)

Highest
stream
order

Perimeter
(km)

Stream Order

Stream Length (km)

16.63

19.36

14

13.72

6.32

11.74

34.54

28.91

68

17

40.56

14.95

9.25

19.98

33.32

28.21

96

27

55.47

14.06

8.9

19.98

29.75

26.36

56

15

38.82

14.64

5.7

15.95

23.88

22.67

18

15.06

5.29

1.08

15.95

6
7

33.95
31.06

25.56
30.91

4
5

22
20

6
6

2
1

1
1

21.4
19.77

9.98
9.6

13.69
6.7

1.82
8.36

15.86

28.24

25.87

33

25.11

10.48

2.66

1.82

15.86

Watershed

Ruggedness
Number
(Rn)

0.04
0.08
0.06
0.02
0.02
0.02
0.04
0.04

0.01
0.008
0.006
0.00
0.003
0.003
0.006
0.005

0.002
0.0028
0.002
0.001
0.0010
0.001
0.002
0.0015

0.001
0.002
0.002
0.001
0.000
0.000
0.001
0.001

Watershed 2

100
No of streams

100
No of streams

0.49
0.38
0.33
0.45
0.40
0.53
0.78
0.36

Relative
Relief
(Rhp)

0.56
0.52
0.53
0.54
0.58
0.41
0.65
0.53

Relief Ratio
(m)

0.44
0.39
0.36
0.43
0.40
0.46
0.56
0.38

Max. Relief
(km)

2.97
3.42
3.13
3.51
4.24
6.77
4.49
4.45

Form
Factor
(Rf)

5.84
9.57
10.11
8.12
7.74
7.69
6.61
8.87

Circularity
Ratio (Rc)

1.14
2.63
3.96
2.59
1.01
0.93
0.91
1.63

Elongation
Ratio (Re)

1.51
2.45
2.95
2.52
1.57
1.94
1.38
1.98

Width (km)

4.52
2.81
3.73
5.52
5.72
2.24
2.03
3.19

Length (km)

3.75
3.42
3.64
3.91
3.17
2.83
2.89
2.79

Stream
frequency

1
2
3
4
5
6
7
8

Drainage
density (Dd)

WS
No

Stream
Length
Ratio (Rl)

Mean
Bifurcation
Ratio
(Rbm)

Table 2 Sub watershed wise morphometric parameters of Kolar taluk watershed

10

10

1
1

2
Stream order

stream order

Page 132

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Watershed 3

Watershed 4

100
No of streams

10

10

1
1

Watershed 5

Watershed 6

100
No. of Streams

100
No of streams

Stream order

Stream Order

10

10

1
1

Stream order

stream order

Watershed 7

Watershed 8

100
No. of Streams

100
No of streams

No of Streams

100

10

10

1
1

3
Stream order

Stream order

Fig. 3 Regression of stream order on number of streams of all sub watersheds


Page 133

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Watershed 1
Mean Stream Length
(kms)

Mean Stream Length


(Kms)

100.00
10.00
1.00

Watershed 2

10

0.1

0.10

2
Stream order

2
3
stream order

Watershed 4

10
1
0.1
1

Stream Order

Mean stream length (kms)

Mean Stream Length (kms)

Watershed 3
100
10
1
0.1
1

1
0.1
3

Stream order

Mean stream length (kms)

Mean stream length (kms)

10

10
1
0.1
1

1
0.1
5

Mean stream length (kms)

Mean stream length (kms)

10

StreamOrder

Watershed 8

100

Stream order

Watershed 7

Watershed 6

100

Stream order

Watershed 5

100
10
1
0.1
1

Stream order

Fig. 4 Regression of stream order on mean stream length of all sub watersheds

Page 134

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Stream Frequency (no/sqkm)

following thematic maps were used. Lithology,


Slope, Soil, Land use, land covers Lineaments,

y = 2.0088x - 2.3434
R = 0.8898

Hydro-geomorphology, and surface drains (Table 3).

Were prepared as shown in Fig.4a, b, c,d e & f

Table 3: Characteristic Features of Thematic Layers

2
1
0

Thematic
Layers
Geomorphology

Characterstics

Fig. 5 Regression of Drainage density on Stream


frequency
The length of overland flow suggests that the surface

Land use and


land cover
Slope

Land forms, Weathered zones


etc.,
Rock
type,
weathering
character,
thickness
of
weathering, joints, fracture etc.,
Forest area, barren land,
vegetation land etc.,
Slope percent

runoff will reach the stream faster. The values of

Soil

Permeability, Porosity, texture

Lineaments

Faults, fracture etc.,

Drainages

Drain network, drainage density

Drange Density (km/sq km)

Lithology

stream frequency for eight watersheds exhibits positive


correlation with the drainage density values of the area
indicating the increase in stream population with
respect to increase in drainage density. Stream
frequency value for the watershed varies from 0.91 to
3.96 this indicates that the stream frequency is low.
The circularity ratio varies from 0.41 to 0.65 for the
watersheds. Its low, medium and high values are
correlated with youth, mature and old stage of the
cycle of the tributaries in sub watershed of the
region. The elongation ratio ranges from 0.38 to 0.56
which indicates that the watershed is more elongated
to elongated.

Fig 4a Geomorphology Map of the Study Area

4. Groundwater Prospect Map


The groundwater prospect map is a systematic
effort and has been prepared considering major
controlling

factors.

The

map

depicts

hydrogeomorphological aspects, which are essential


as basis for planning and execution of groundwater
mapping.

In order to demarcate the groundwater

potential zones using GIS for the study area the

Fig 4b. Lithology Map of the Study Area

Page 135

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig 4c Lands Use / Land Cover Map of the Study Area

Fig 4e. Lineament map of the study area

Page 136

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig 4f. Lineament density map of the study area

Fig 4i Slope Map of the Study Area

Page 137

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig 4j Soil Map of the Study Area

Fig 4h Drainage map of the study area

Page 138

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig 4i Drainage Density Map Of the Study Area


In the first step of integration,
Geomorphology layer (L1) and Lithology layer (L2)
maps were integrated by intersect option and a new
integrated layer (R1) is generated 183 polygons of
layer L1 (geomorphology layer) and 2 polygon of
layer L2 (Lithology layer) has resulted in the
generation of integrated layer (R1) with 199
polygons. The maximum and minimum WCV are 15
and 40 respectively. In the second step, the integrated
layer R1 layer containing 199 polygons (the resultant
of geomorphology and lithology) was intersected
with the landuse and land cover layer (L3) which has
415 polygons and it resulted in the generation of
integrated layer R2 with 873 polygons. The
minimum and maximum WCV are 20 and 65
respectively. In the third step, integrated layer R2
containing 873 polygons was intersected with the
Slope (L4) which has 55 polygons, which resulted in
the generation of integrated layer R3 containing 1419
polygons. The minimum and maximum WCV are 25
and 90 respectively. In the fourth step, Soil layer
(L5) containing 272 polygons was intersected with
the integrated layer R3 having 1419 polygons, this
resulted in the generation of integrated layer R4

containing 1802 polygons. The minimum and


maximum WCV are 30 and 115 respectively. In the
fifth step, Lineament layer (L6) containing 275
polygons was intersected with the integrated layer R4
having 1802 polygons, which resulted in the
generation of integrated layer R5 containing 2113
polygons. The minimum and maximum WCV are 35
and 140 respectively. In the sixth step, Drainage
layer (L6) containing 443 polygons was intersected
with the integrated layer R5 having 2113 polygons,
which resulted in the generation of integrated layer
R6 containing 2713 polygons. The minimum and
maximum WCV are 35 and 165 respectively.
Theoretically, if WCV of the all the layers
which are integrated, a maximum of 165 and
minimum of 35 WCV must be obtained. But
practically maximum of 155 and minimum of 50
WCV are obtained. This shows that the overlap of
some of higher weights polygons with one another in
the integrated layer. Based on the total weights
obtained by integration the study area has been
delineated into Very Good, Good, Good to Moderate,
Moderate, Moderate to Poor and Poor to Nil
groundwater potential zones (Fig 5, Table 4).

Page 139

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

Fig 5 GROUNDWATER POTENTIAL ZONES MAP OF THE STUDY AREA


Table 4 INTEGRATED GROUNDWATER POTENTIAL ZONES
Higher
Weight
age value

Lower
weight
age value

8.04

155

135

51.28

22.16

135

115

Good to
Moderate

8.09

3.48

115

95

Moderate

114.78

49.61

95

75

Moderate to
Poor

8.41

3.63

75

60

Poor to Nil

30.19

13.04

60

45

Sl.
No

Groundwater
category

Area
Km2

Very Good

18.61

Good

% of
total
area

Ground water structures feasible


Dug wells fitted with low power pumps and
tube wells fitted with hand pumps
Dug well, dug well cum bored well and tube
well fitted with hand pump
This zone is mainly controlled by geologic
structure. Dug well and tube well fitted with
hand pump
Dug well, dug-cum bored well, with tube well
This zone mainly comprises areas where the
recharge condition and the water-yielding
capacity of the underlying materials are
neither suitable nor poor. Dug-cum bored
wells are fitted.
Generally, groundwater structures will not be
successful. Dug well dug-cum-bored well
may be constructed. Surface water should be
harnessed and roof top rainwater harvesting
schemes may be adopted.

Page 140

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

5. Conclusions
From Groundwater Potential map, it is observed that
Kolar Taluk watershed is having Moderate Ground
water prospect zone. Ground water potential maps
thus developed will be useful for planning surface
drainage networks and construction of ground water
recharge structures in very good ground water
potential zones.

References
[1].

Krishnamurthy J., Venkatesa Kumar N.,


Jayaraman V. and Manivel M.(1996): " An
approach to demarcate groundwater potential
zones through Remote Sensing and a
Geographical
Information
System".
International Journal of Remote Sensing. Vol.
17, No. 10, 1996. p 1867-1884.

[2].

Saraf A.K. and Choudhury P.R(1998):


"Integrated remote sensing and GIS for
groundwater exploration and identification of
artificial recharge sites", International Journal
of Remote Sensing, Vol. 19, No. 10, 1998, p
1825-1841

[3].

Krishnamurthy J., Manavalan P. and Saivasan


V., "Application of digital enhancement
techniques for groundwater exploration in a
hard-rock terrain", International Journal of
Remote Sensing, Vol. 13, No.15, 1992, p
2925-2942

[4].

Ramamoorthy. P, Arjun. A, Gobinath. K,


Senthil Kumar. V, Sudhakar . D Geo Spatial
Analysis of Groundwater Potential Zone
Using Remote Sensing and GIS Techniques in
Varahanadhi
Sub
Basin,Tamilnadu,
International Journal Of Science, Engineering
and Technology-2014 www.ijset.In.

[5].

Ramakrishna, D.Nagaraju, Mohammad suban


lone, Siddalingamurthy.S, and Sumithra.S
Groundwater Prospectus Studies of Tattekere
Watershed, Mysore District, Karnataka, India
Using
Remote Sensing And GIS,
International Journal of Remote Sensing &
Geoscience (IJRSG)-2014 www.ijrsg.com.

Page 141

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15) ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

ENVIRONMENTAL CONSERVATION ASSESMENT USED BY GIS AND


REMOTE SENSING -BHARAMASAGAR AT CHITHRADURGA DIST,
KARANATAKA
V. Praveena Kumara*1, Bgayalakshmi*2, Lingananda*3M V.Sivani*4 Shobha udaya*5
*1
MSc Applied Geology,Department of Geology*1*2BSc Geography,*1*2*31sem, , *4*5BTech, 3rd SEM, Dept of
Engineering, Central University of Karnataka, Kadaganchi Village, Dist-.Kalburgi,State- Karnataka, country - India.

Tel: +91 9901744719 E-mail: praveenakumarav@gmail.com

_______________________________________________________________________________________________________
Abstract
This paper describes the studies conducted generation of a digital data base for using spatial data modeling using Remote sensing
and GIS techniques for the Bharamasagara watershed in Chithrdurga District. This digital data base serves as a ready reckoned- tool
for pollution risk assessment and aids in faster decision making process in issues related to development.The studies consisted in
development of a digital data base for the Bharamasagar water shed of 900sq.kmsL depicting Land use, Land cover changes in
classes such as Barren Rocky, Water Bodies, Semi-Evergreen Forest, Evergreen Forest, Litho logy, geo-morphology and
Agricultural Area by calculating the area in hectares of the resulting Land use/Land cover types for each study years and
subsequently comparing the results. Thus the study detects land use changes between the year 2000 to 2009 by using Land and
LISS-3 satellite images.Information on Land use/Land cover in the form of maps and statistical data is very vital for spatial
planning, management and utilization of land. In the study, Remote Sensing and geographic information system (GIS) were used in
order to study Land use/Land cover changes. Land use change may influence many natural phenomena and ecological processes,
including runoff, soil erosion and sedimentation and soil conditions. The Areas are changing due to various human activities, natural
conditions and development activities. According to the user requirements, updating of land use mapping is required to various
departments. Monitoring of Land use/Land cover changes which would help to plan the development activities. Change detection
has shown that the Barren Rocky area increased between 2000 and 2009 by 22.26% from 5825.21 ha to 7121.87 ha, Evergreen
Forest area increased between 2000 and 2009 by 24.83% from 16333.40 ha to 20388.85 ha, Water Body area decreased between
2000 and 2009 by 2.02% from 10822.37 ha to 10603.42 ha,
Keywords: Satellite, Hectors, Land, GIS
___________________________________________________________________________________________________
1.INTRODUCTION
The increasing availability of geospatial data provides an
opportunity for environmental engineers to contribute to the
identification of Potentially Polluting Sites. To adequately
assess the environmental risk of these sites, relevant
information must be collected and converted into a multiscale geodatabase suitable for site inventory and geo-spatial
analysis. However, the successful collection and integration
of data model information requires some effort to normalize
and standardize the data based on recognized international
standards.Local governments need tools that make the most
of available information to target high risk locations for
pollution movement and to evaluate current and future
impacts of on-site systems. Hence this spatial data models

can be used to identify, manage, monitor and plan remedial


measures for the vulnerable site. These data models require
less time, effort and are economical when compared to
conventional survey or non-spatial data modelling methods.
These data can also be used for preparedness planning and
reduce the risk of human beings experiencing the after effects
of the pollution.
2. Content
In classification process, Supervised Classification method in
GRASS was performed based on a set of user-defined

Page 142

V. Praveen kumara, et. al. /INDECS-15

classes, by creating the appropriate user-defined polygon.


The methodology of extracting Land uses / Land cover from
satellite image is shown in fig 1. In supervised classification
process, User-Defined Polygon function reduces the chance
of under estimating class variance since it involved a high
degree of user controls. Training points were repeatedly
selected from the whole study area by drawing a polygon
around training sites of interests. Land use / Land cover
classes of these training points were extracted with respect to
general knowledge obtained from topographic maps and field
visits. The supervised classification was performed using the
maximum likelihood algorithm. To evaluate the accuracy of
the classified image, Accuracy Assessment tool in GRASS
was used. The reference class values were compared with the
classified class in error matrix. Then overall accuracy and
kappa values were computed by using users accuracy (a
measure of commission error) and producers accuracy (a
measure of omission error) of each class. Calculation of the
Area in hectares of the resulting land use/land cover types for
each study year and subsequently comparing the results. The
comparison of the land use/land cover statistics assisted in
identifying the percentage change, trend and rate of change
between 2000 and 2009.
2.1 Data Acquired and Source

For the study, Landsat and LISS-3 satellite images of


Bharamasamdra watershed were acquired for two
Epochs; 2000 and 2009. Landsat satellite image of the
year 2000 was obtained from U.S .Geological Survey
(USGS), LISS-3 satellite image of the year 2009 was
obtained from BHUVAN website.

These class values are taken as the reference points, to


make a comparison with the class values of the
classified images. The overall accuracy and KAPPA
statistics are used to assess classification accuracy
based on error matrix. Overall accuracy is computed
by dividing the total correct value (i.e. sum of the
major diagonal) by the total number of pixels in the
error matrix. Accuracy assessment is performed for
2000 & 2009 LU/LC maps. An overall accuracy of
79.80% for 2000 & 83.65% for 2009 are obtained.
2.4 Change Detection Analysis
The most commonly used Change Detection methods
are,i) Image overlay,ii)Classification comparisons of
land cover statistics or Calculate the area in hectares
of the resulting Land use/Land cover types for each
study years and subsequently comparing the results,iii)
Change vector analysis,iv) Principal component
analysis,v) Image rationing, vi) The differencing of
Normalized Difference Vegetation Index (NDVI).
The method used in this project is classification
comparison of land cover statistics. The comparison of
the Land use/Land cover statistics, assisted in
identifying the increase and decrease in area under
different classes between 2000 and 2009 as shown in
table 2. The change in area can be interpreted with
reference to time of acquisition of image i.e. 2000
image is acquired on December and also 2009 image
on December, which shows a seasonal variation.
2.5 Illustrations

2.2Data types & their Source


The trend of change can then be calculated by dividing
observed change by sum of changes multiplied by
100(Trend) percentage change = observed change *
100 (1)Sum of change In obtaining annual rate of
change, the percentage change is divided by 100 and
multiplied by the number of study year 2000 2009.

2.3 Accuracy Assessment


In order to assess the classification accuracy, 200
points are generated randomly throughout each image
usingthe Add Random Point utility in GRASS A
class value is then entered for each of these points.

Figure: 1 LULC map of the Year 2000

Page 143

V. Praveen kumara, et. al. /INDECS-15

2.6 Tables
SL
No.

Data
Type

Data
product

Resolution

Source

Landsat
image

20-122000

30

USGS

LISS-3

15-122009

23.5

BHUVAN

geologically acceptable setting formunicipal and


waste disposal sites, and also helps users to direct
resources for further evaluation.

3. Conclusion
Land use maps resulted from Landsat and
BHUVAN images and classification shows a
satisfactory comparison. An average accuracy of
79.80% for Landsat and 83.65%for LISS-3 has
been obtained from classification accuracy
assessment. The urban areas and significant open
areas are not identified successfully in the
classified land use maps. Overall this appears to be
a quick and satisfactory way to obtain ground
cover information for a watershed of this size
when sufficient ground data are either not available
or for any reasons are difficult to obtain.
Therefore, remote sensing imagery can be used
successfully in providing up to date information.
The study demonstrated the use of spatial
data modelling for pollution risk assessment by
remote sensing and GIS is an effective method for
pollution vulnerability assessment. The GIS
technology has provided an efficient environment
for analyses and high capabilities of handling
spatial datasets.This spatial data can prove to be a
very valuable tool for those who are in
management position because it gives a very
comprehensive indication of vulnerability to
environmental contamination.

Figure: 2 Depth of Ground Water


When the state or local administrator has limited
resources available to devote to environmental
protection, they are forced to focus these resources
in certain areas. The spatial data model helps
identify areas, which are more or less vulnerable
than others to contamination. This delineation
allows administrators to direct their resources to
those vulnerable areas most critical to
environmental contamination, thereby making use
of most of the limited resources available.

The spatial data thus prepared for


Bharamasamdra Watershed helps the planners in
broadly screening areas for waste disposal sites,
industrial sites etc.The environmental atlas also
helps users to recommend the most hydro-

Page 144

V. Praveen kumara, et. al. /INDECS-15

References
[1] David Keith Todd, Groundwater Hydrology,
John Wiley India Pvt Ltd., second edition, 1980.
[2] Zubair, Ayodeji Opeyemi, Change detection in
land use and land cover using remote sensing data
and GIS, Department of Geography, University of
Ibadan, Ibadan (Masters thiesis- October, 2006)
3] Ravinder Kaur and K.G. Rosin, Ground Water
Vulnerability Assessment Challenges and
Opportunities, 2009

Page 145

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15)ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SCS-CN AND GIS BASED APPROACH FOR ESTIMATION OF SURFACE


RUNOFF IN SITA-SWARNA BASIN, UDUPI DISTRICT
Anil, B1, Shivanna2, M.S.Vinaya3
1, 2

Department of Marine Geology, Mangalagangotri, Mangalore University, Konaje- 574 199, Karnataka
3
Department of Civil Engineering, Adhiyamaan College of Engineering, Hosur-635 109, Tamil Nadu
anilbaindur@gmail.com, shivannag@rediffmail.com, msvinaya@gmail.com

Abstract
Runoff is one of most important hydrological variables that are used in catchment treatment, ground water
recharge study, planning for optimal use of reservoirs, organizing rivers and warning flood. The per capita
availability of land for cultivation has been decreasing over the years. In India, the availability of accurate
information on runoff is scarcely available in few selected sites. Therefore, water and the related land resources
must be developed, utilized and managed in an integrated and comprehensive manner. Advances in
computational power and the growing availability of spatial data have made it possible to accurately predict the
runoff. Remote sensing and GIS techniques are being increasingly used for planning, management and
development of natural resources. The study area, Sita-Swarna basin geographically lies between 13008 and
13035 N latitude and 74041 and 75011 E longitude with an area of 1438.38 Sq. km. The thematic layers such as
land use/land cover and soil maps were derived from remotely sensed data and overlayed through ArcGIS
software to assign the curve number on polygon wise. The daily rainfall data of four rain gauge stations in and
around the basin (2009-2013) was used to estimate the daily runoff from the basin using Soil Conservation
Service - Curve Number (SCS-CN) method. The runoff estimated from the SCS-CN model (SCS, 1972), a
versatile and widely used procedure for runoff estimation to know the variation of runoff potential with different
land use/land cover and with different soil conditions.
Keywords: Sita-Swarna Basin, SCS-CN Method, Runoff, GIS, Remote Sensing
1. INTRODUCTION
In India, the availability of accurate information on
runoff is scarcely available in few selected sites.
Water is one of the most vital requirements
foreconomic and social development. Human
population of the Indian subcontinent is ever
increasingthereby increasing the demand for water
fordomestic, agricultural and industrial use.
However, the quantum of rainfall and surface
wateravailabilityhas remained the same; thus,
resulting inover-exploitation of ground water,
declining watertable levels and deterioration of
water quality. Advances in computational power
and the growing availability of spatial data have
made it possible to accurately predict the runoff.
The possibility of rapidly combining data of
different types in a Geographic Information System
(GIS) has led to significant increase in its use in
hydrological applications. The curve number
method (SCS, 1972), also known as the hydrologic
soil cover complex method, is a versatile and
widely used procedure for runoff estimation. This
method includes several important properties of the
basin namely soil's permeability, land use and
antecedent soil water conditions which are taken
into consideration. Jasrotia et al., 2002used a
mathematical model to estimate rainfall, runoff in

conjunction with remote sensing data and GIS


using SCS CN method and runoff potential map.
The runoff was estimated from SCS curve number
model modified for Indian condition by
conventional data base and GIS for Dikrong river
basin (Ashish Pandey et al., 2004). In Geographic
Information System (GIS) the overlay analysis is a
common, widely used method for analyzing and
evaluating geospatial data. Overlay analysis
employs various thematic layers in a GIS to
determine relationships across the layers and
subsequently enables to derive specific outputs
based the mathematical and scientific approaches.
2. OBJECTIVES
To prepare thematic maps on surface and
sub-surface features viz., Land Use/ Land
Cover, Soil, Drainage.
Determination of Curve Number (CN)
values by integrating the land use/land
cover map and hydrological soil group
(HSG) map using overlay analysis in GIS
domain.
Determination of Antecedent Moisture
Condition (AMC) by total rainfall of 5
days period preceding storm.

Page 146

Anil B, Shivanna, M S Vianya / INDECS-15


Calculation of recharge capacity and
direct runoff using standard SCS-CN
model.
3. STUDY AREA
The study area Sita-Swarna basin falls under Udupi
district, Karnataka between North latitudes
130800 - 133500 & East longitudes
744100- 751100 with an area of 1438.39
sq.km as delineated from the Survey of India (SOI)
toposheet numbers 48K/10, 48K/11, 48K/14,
48K/15, 48K/16, 48O/2, 48O/3 and 48O/4 on 1:
50000 scale. Udupi district is bounded on the north
by Uttara Kannada district, on the east by Western
Ghat, on the south by Dakshina Kannada district
and Arabian Sea on the west. The rivers Sita and
Swarna originate on the Western Ghats and flow
for a considerable distance in the Coastal plain,
before they form a common estuary at about 10km
north of Udupi and discharge an average of 54 m3
s1 of water into the Arabian Sea, of which
significant part is being discharged during the
monsoon. The climate is marked by heavy rainfall,
high humidity and small variation in temperatures.
The average annual rainfall of the area is about
3623 mm. Of this, about 82% of the rainfall is
received during the southwest monsoon and the rest
during northeast and inter-monsoon months. It is
the main source of drinking water to Udupi
District. Several large scale and small scale
industries are located along the coast. Economy of
Sita-Swarna drainage basin is mainly based on
agriculture, aquaculture and medium to small scale
Industries.

Figure 1: Location Map of the Study Area

4. METHODOLOGY

In order to estimate the surface runoff in the study


area, a multi-parametric dataset comprising satellite
data, conventional maps including Survey of India
(SOI) topographic sheets are used. The softwares
used for this study includes ERDAS Imagine 9.2
and ArcGIS 10. The details of the Primary &
Secondary data used for this study are summarized
in the Table 1, Table 2 & Table 3 respectively.
Satellite

Path/Row

Acquisition
Date

IRS-P6
97-64 A
28/10/2013
(Resourcesat-1)
&B
(5.8m)
Table 1: Details of Satellite data used

Swath

70KM

Toposheet No

Scale

Contour
Interval

Source

48K/10, 48K/11,
48K/14, 48K/15,
48K/16, 48O/2, 48O/3,
48O/4

1:50,000

20m

SOI

Table 2: Details of Toposheet used

The methodology adopted for the study is


mentioned in the following Flowchart (Figure2). In
order to prepare various thematic layers of the
study area, toposheets along with the satellite data
was used to update drainage, land use / land cover
and soil map. The map of drainage is shown in the
Figure 3. The rainfall data of four gauge stations
pertinent to Sita-Swarna basin for a period of 2009
to 2013 was used for estimation of runoff. The
runoff estimated from the SCS-CN model (SCS,
1972), a versatile and widely used procedure for
runoff estimation to know the variation of runoff
potential with different land use/land cover and
with different soil conditions.
Existing Maps

Scale

Source

Drainage

1:50,000

SOI

National Bureau of
Soil Survey &
Landuse / Land
1:50,000
Landuse Planning
Cover
(NBSS&LUP,
ICAR)
National Bureau of
Soil Survey &
Soil
1:50,000
Landuse Planning
(NBSS&LUP,
ICAR)
Department of
Rainfall Data
Economics &
(2009-13)
Statistics, Govt. of
Karnataka
Table 3: Details of Secondary data used

4.1 SCS Curve Number Model

Page 147

Anil B, Shivanna, M S Vianya / INDECS-15

Basin wise runoff was calculated by Soil


Conservation Services Curve number (SCS-CN)
method, which is associated with curve number and
curve number depends on land use/ land cover,
hydrological soil group (HSG) and Antecedent
Moisture Content (AMC). HSG are four types as
A, B, C, D from highest to lowest infiltration rate
respectively and AMC as I, II & III condition from
driest to wettest.
INPUT DATA

SOIL MAP

LU/LC MAP

AMC

this study. The runoff curve number, CN, is then


related as,
=

25400

254

Generally, CN has a range from 0 to 100; lower


numbers indicate low runoff potential while
largernumbers are for increasing runoff potential. If
the area consists of patches of land use / land cover
then a weighted curve number (CNw) for the

RAINFALL

am
SOIL VEGETATIN INDEX

CN

WCN

SCS CN MODEL
Figure 3: Drainage Map of the Study Area
RUNOFF
Figure 2: Methodology to estimate surface runoff by SCS CN
Model

basincan be obtained by weighing them in


proportion of the area and given as follows:
CNw = (CNi *Ai) /A

4.2 Runoff Volume


Surface runoff is mainly controlled by the amount
of rainfall, initial abstraction and moisture
retentionof the soil. The relation of runoff with
rainfall is given as follows by (Ponce and Hawkins,
1996).
Q = (P - 0.2S) 2 / (P + 0.8S) &
I=P-Q
Where, I is infiltration (mm), Q is runoff (mm); P
is rainfall (mm); S is the potential maximum soil
moisture retention after runoff begins (mm); Ia is
the initial abstraction (mm), Ia = 0.2S is taken in

4.3 Antecedent Moisture Condition


The Antecedent Moisture Condition (AMC) is the
index of catchment wetness which is determined by
total rainfall of 5 days period preceding storm. The
AMC value is intended to reflect the effect of
infiltration on both the volume and rate of runoff
according to the infiltration curve. An increase in
index means an increase in the runoff potential.
SCS developed three antecedent soil-moisture
conditions and labeled them as I, II, III, according
to soil conditions and rainfall limits for dormant
and growing seasons. Classification of Antecedent
Moisture Condition is shown in Table 4.

AMC
Class

Description of soil condition

Lowest runoff potential. The watershed soils are dry enough for satisfactory cultivation

II

Average condition

III

High runoff potential. The watershed is practically saturated from antecedent rains

Total five day antecedent


rainfall (cm)
Dormant season Growing season
< 1.27

< 3.56

1.27 - 2.79

3.56 - 5.33

> 2.79

>5.33

Page 148

Anil B, Shivanna, M S Vianya / INDECS-15

Table 4:Classification of Antecedent Moisture Conditions (AMC)

4.4 Hydrological Soil Group Classification


The hydrological soil groups as defined by the Soil
Conservation Service (1972) were classified into A,
B, C & D according to their minimum infiltration
rate. Table 5 shows the hydrological soil group
classification. CN values were determined from
hydrological soil group and antecedent moisture
conditions of the basin. The Curve Number values
for AMC-I and AMC-II were obtained from AMCII (Chow et al. 1988) by the method of
conservation. Runoff curve numbers (AMC II) for
hydrologic soil cover complex are shown in Table
6. The Hydrological Soil Group map of SitaSwarna Basin is shown in the Figure 4.

Figure 4: HSG Map of the Study Area

Description

Minimum
Infiltration
rate (mm/hr)

Soils in this group have a low runoff potential (high-infiltration rates) even when thoroughly wetted.
They consist of deep, well to excessively well drained sands or gravels. These soils have a high rate of
water transmission.

7.62 - 11.43

Soils in this group have moderate infiltration rates when thoroughly wettedand consists chiefly of
moderately deep to deep, well-drained to moderatelywell-drained soils with moderately fine to
moderately coarse textures. Thesesoils have a moderate rate of water transmission.

3.81 - 7.62

Soils have slow infiltration rates when thoroughly wetted and consist chieflyof soils with a layer that
impedes the downward movement of water, or soilswith moderately fine-to fine texture. These soils
have a slow rate of watertransmission.

1.27 - 3.81

Soils have a high runoff potential (very slow infiltration rates) whenthoroughly wetted. These soils
consist chiefly of clay soils with high swellingpotential, soils with a permanent high-water table, soils
with a clay layer nearthe surface, and shallow soils over nearly impervious material. These soilshave a
very slow rate of water transmission.

0 - 1.27

Soil
Group
A

Table 5: Hydrologic Soil GroupClassification (Mc. Cuen, 1982)

Sl. No.

Hydrologic Soil Group

Land Use
A

Agricultural land without conservation (Kharif)

72

81

88

91

Double crop

62

71

88

91

Agriculture Plantation

45

53

67

72

Land with scrub

36

60

73

79

Land without scrub (Stony waste/ rock out crops)

45

66

77

83

Forest (degraded)

45

66

77

83

Forest Plantation

25

55

70

77

Grass land/pasture

39

61

74

80

Settlement

57

72

81

86

10

Road / railway line

98

98

98

98

11

River / stream

97

97

97

97

12

Tanks without water

96

96

96

13

Tank with water


100
100
100
Table 6: Runoff curve numbers (AMC II) for hydrological soil cover complex (Source: Chow et al, 1988)

96
100

5.RESULTS AND DISCUSSION

Page 149

Anil B, Shivanna, M S Vianya / INDECS-15

Based on the hydrological soil group, in Sita &


Swarna basin constitute maximum area of
hydrological soil group C (84.91% & 79.58%)
followed by group B (10.23% & 7.72%), group D
(2.6% & 10.36%) and group A (2.26% & 2.34%)
respectively. Fig. 5 shows the LU/LC map of the
Sita-Swarna basin. Similarly, the study area was
identified by thedifferent type of land use / land
cover classes, viz., crop land of about 22.89%&
13.53%, 36.28% & 48.96 of area covered by
agriculture plantation, forest area of about 27.15%
& 23.52%, land with scrub area is occupied by
4.7% & 2.51% and about 8.62% & 11.48% of the
area is occupied by settlements,tanks and rocky
area in both Sita and Swarna basin respectively. In
general, among the different land cover types the
crop land plays the major role for the direct surface
runoff. Fig. 6 shows the curve number map of the
Sita-Swarna basin. From SCS-CN analysis, the
maximum runoff for the Sita basin was estimated
to be 4566.3 mm in the year 2009 and 2797.7 mm
for Swarna basin and minimum runoff of 1539.4
mm for the Sita basin in the year 2013 and 1882.8
mm for Swarna basin in the year 2012. Table 8
shows the annual rainfall and runoff for SitaSwarna basin for the period 2009 to 2013. Fig. 7 &
8 shows the rainfall runoff relationship for SitaSwarna basin. The rainfall and runoff are strongly
correlated with correlation coefficient (r) value
being 0.985 & 0.833.
Type of Land Use

Area (Km2)

Crop Land (Double + Single)


Agriculture Plantation
Forest (Evergreen/Deciduous)
Scrub Forest
Land with Scrub
Land without Scrub
Barren Rocky
Settlement (High + Low Density)

Sita
146.83
232.72
175.67
0.8
26.66
3.5
8.11
31.6

Swarna
79.36
287.16
137.88
0.11
13.55
1.19
7.15
45.58

Water Bodies (Tanks/Streams)

15.59

14.59

Figure 5: LULC Map of the Study Area

Sita
Year

Swarna

2009

Rainfall
(mm)
7074.8

Runoff
(mm)
4566.3

Rainfall
(mm)
5436.6

Runoff
(mm)
2797.7

2010

4192.4

1829.4

4896.4

2111.3

2011

4541.2

2187.9

5337.0

2440.7

2012

3995.7

1950.1

4151.0

1882.8

2013

3985.0

1539.4

4829.0

2156.1

Table 8: Annual Rainfall and Runoff for Sita-Swarna basin

Table 7: Land uses in Sita-Swarna basin

Figure 6: Curve Number Map of the Study Area

Page 150

Anil B, Shivanna, M S Vianya / INDECS-15

REFERENCES

Runoff (mm)

5000

y = 0.9252x - 1987.2
R = 0.9853

4000
3000

1.

2000
1000
0
0

2000

4000
Rainfall (mm)

6000

8000

Figure 7: Rainfall Runoff relationship for Sita basin

Runoff (mm)

3000
2500

y = 0.6298x - 827.14
R = 0.833

2000
1500
1000
3000

4000

5000

6000

Rainfall (mm)

Figure 8: Rainfall Runoff relationship for Swarna basin

6. CONCLISIONS

From the SCS-CN analysis, it is clear that, very


low quantities of the infiltration are predicted, due
to presence of more than70-80% of impermeable
soil in the Sita-Swarna basin.Since Sita-Swarna
basin provides drinking and domestic water to the
entire district, the demand for the water may
increase with time. Theanalysisshows that there
isabout 40-45% of water is lost during the monsoon
due to surface runoff and reaching the Arabian Sea
through streams and rivers. So, keeping future need
and requirement, some kind of interventions are
required at watershed level to store and
rechargegroundwater to overcome water sacristy
problem and long term sustainability of ground
water extraction. The study clear shows that, GIS
based SCS-CN method is playing an important role
in estimation of surface runoff for given
basin/watershed effectively and also these model
parameters can be used in estimation of surface
runoff from the ungauged area. It is concluded that,
the SCS-CN analysis at Micro-watershed level with
updated LU/LC data can further fine-tune the
runoff estimation, which helps in planning and
implementation of development activities at a
larger scale.

Mysooru R. Yadupathi Kutty (2011),


Principles of Hydrology, Publisher I.K.
International Publishing House Pvt. Ltd., pp.
133-158.
2. Somashekar R. K.et al. (2011), Runoff
estimation and morphometric analysis for
Hesaraghatta watershed using IRS-ID LISS III
FCC satellite data, Journal of Indian Society
of Remote Sensing, Vol.39,pp. 95-106.
3. Patil J. P. et al. (2007), Development of a GIS
Interface for Estimation of Runoff from
Watersheds, Springer Science.
4. Gupta P. K. et al. (2008), Predicting the spatiotemporal variation of run-off generation in
India using remotely sensed input and Soil
Conservation Service curve number model,
Current Science, Vol. 95, No. 1.
5. Sindu D. et al. (2013), Estimation of surface
runoff in Nallur Amanikere watershed Using
SCS-CN Method, International Journal of
Research in Engineering and Technology.
6. D. Ramakrishnan et al. (2009), SCS-CN and
GIS-based approach for identifying potential
water harvesting sites in the KaliWatershed,
Mahi River Basin, India, Journal of Earth
System Science, Vol. 4, pp. 355-368.
7. Mishra S. K.et al.(2004),Evaluation of the
SCS-CN-based
model
incorporating
antecedent moisture, Journal of Water
Resources Management, Vol. 18, pp. 567589.
8. K. X. Soulis et al. (2012), SCS-CN parameter
determination using rainfall-runoff data in
heterogeneous watershedsthe two-CN system
approach, Hydrology and Earth System
Sciences, Vol. 16, pp. 1001-1015.
9. Roger Cronshey et al. (1986), Urban
Hydrology for Small Watersheds, Published
by U.S. Department of Agriculture, 2nd
Edition,pp. 13-28.
10. Mc. Cuen, R. H. (1982), A guide to
hydrologic analysis using SCS methods,
Englewood Cliffs:Prentice Hall Inc.

Page 151

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

PERFORMANCE OF GEOPOLYMER CONCRETE IN ACIDIC


ENVIRONMENT
Shankar H. Sanni 1, Dr. R. B. Khadiranaikar 2
1

Basaveshwar Engineering College, Bagalkot, 587 102, India, shsanni@gmail.com


2 Basaveshwar Engineering College, Bagalkot, 587 102, India, rbknaikar@gmail.com

Abstract
The present paper focuses the experimental work done in accessing the durability of geopolymer concrete compared to
conventional concrete in acidic media. The molarity used for the preparation of geopolymer specimens is 12. The grade
chosen for the investigation were M-30 and M-40. The alkaline solution used for present study is the combination of
sodium silicate and sodium hydroxide solution with the ratio of 2.50. The test specimens were 150x150x150 mm cubes,
cured in ambient room temperature. The performance evaluation of the specimens were assessed by immersing GPC and
OPC specimens in 5% sulphuric acid solution separately, periodically monitoring surface deterioration and depth of
dealkalization, changes in weight and strength over a period of 15, 30, 45, 60 and 90 days. The test results indicate that
the geopolymer concrete has an excellent resistance to acid attack when compared to conventional concrete. Thus we
can say that the production of geopolymers have a relative higher strength, excellent volume stability and better
durability.
Keywords: geopolymer concrete, fly ash, molarity, sodium silicate, sodium hydroxide, sulphuric acid

1. Introduction
Construction industry is one of the major users of the natural
resources like cement, sand, rocks, clays and other soils. The
ever increasing unit cost of the usual ingredients of concrete
have forced the construction engineer to think of ways and
means of reducing the unit const of its production. At the same
time, increased industrial activity in the core sectors like
energy, steel and transportation has been responsible for the
production of large amounts like fly ash, blast furnace slag,
silica fume and quarry dust with consequent disposal problem
[1].
The geopolymer technology was first introduced by Davidovits
in 1978. His work considerably shows that the adoption of the
geopolymer technology could reduce the CO2 emission caused
due to cement industries. Davidovits proposed that an alkaline
liquid could be used to react with aluminosilicate in a source
material of geological origin or in by-product materials such as
fly ash to make a binder [2]. Geopolymer is synthesized by
mixing aluminosilicate-reactive material with strong alkaline
solutions, such as sodium hydroxide (NaOH), potassium
hydroxide (KOH), sodium silicate or potassium silicate. The
mixture can be cured at room temperature or temperature cured
[3]. Fly ash is the most common source material for making
geopolymers. Normally, good high-strength geopolymers can
be made from class F fly ash [4]. Alkaline activating solution is
important for dissolving of Si and Al atoms to form
geopolymer precursors and finally alumino-silicate material.
The most commonly used alkaline activators are NaOH and
KOH [5-9].

2. Experimental Investigations
Materials:
The following materials have been used in the experimental
study [12]
a) Fly Ash (Class F) collected form Raichur Thermal
power plant having specific gravity 2.00.
b) Ground granulated blast furnace slag collected from
JSW Steel Ltd., Vidyanagar, Toranagallu, Bellary
having specific gravity 2.90.
c) Fine aggregate: Sand confirming to Zone III of
IS:383-1970 [20] having specific gravity 2.51 and
fineness modulus of 2.70.
d) Coarse aggregate: Crushed granite metal confirming
to IS:383-1970 [20] having specific gravity 2.70 and
fineness modulus of 5.85.
e) Water : Clean Potable water for mixing
f) Alkaline liquids: Specific gravity of
i) Sodium Hydroxide (NaOH) = 1.16
ii) Sodium Silicate (Na2SiO3) = 1.57
Tests were conducted on specimen of standard size as per IS:
516-1959 [21]. For the present investigation two types of
mixes were designed, they are designated with the specific
identification as given in Table 1.
Table 1: Specimen Identification
Type of mix
Geopolymer
concrete

Identification
GPC

Conventional
concrete

OPC

Source Materials used


Fly ash (60%), GGBS
(40%), CA, FA, Alkaline
solutions
Cement, CA, FA, Water

Corresponding author. Tel.: +91 9448440295


Fax: +9876543210; E-mail: shsanni@gmail.com

Page 152

Shankar H. Sanni et. al. / INDECS-15

Mix design of geopolymer concrete


In the design of geopolymer concrete mix, coarse and fine
aggregates together were taken as 77% of entire mixture by
mass. This value is similar to that used in OPC concrete in
which it will be in the range of 75 to 80% of the entire mixture
by mass. Fine aggregate was taken as 30% of the total
aggregates. The density of geopolymer concrete is taken
similar to that of OPC as 2400 kg/m3 [10]. The details of mix
design and its proportions for different grades of GPC are
given in Table 2.

used for the preparation of alkaline solutions) and dosage of


super plasticizer was added to the mix according to the mix
design details. The fly ash and alkaline activator were mixed
together in the mixer until homogeneous pate was obtained.
This mixing process can be handled within 5 minutes for each
mixture with different ratios of alkaline solution. After casting
the specimens, they were kept in rest period for two days and
then they were demoulded. The demoulded specimens were
kept for ambient air curing.

Alkaline Solution
In geopolymerization, alkaline solution plays an important role.
The most common alkaline solution used in geopolymerization
is a combination of sodium hydroxide (NaOH) or potassium
hydroxide (KOH) and sodium silicate (Na2SiO3) or potassium
silicate (K2SiO3). In this study, a combination of sodium
hydroxide and sodium silicate was choosen as the alkaline
liquid. Sodium based solutions were choosen because they are
cheaper than Potassium based solutions. Generally sodium
hydroxide and sodium silicate are readily available in market in
the form of pellets and gel (liquid). The pellets of NaOH are
dissolved in one liter of water for the required concentration.
When sodium hydroxide and sodium silicate solutions mixed
together polymerization will take place liberating large amount
of heat, which indicates that the alkaline liquid must be used
after 24 hours as binding agent.

Fig. 1 Alkaline solution

Table 2: GPC mix design details for 12 Molarity with FA and


GGBS
Materials
Coarse
aggregates

20 mm
14 mm
7 mm
Na2SiO3/ NaOH
Fine sand
Fly ash
GGBS
NaOH solution
Na2SiO3 solution
Super Plasticizer
Extra water

Mass (kg/m3)
M-30
M-40
277.20
277.20
369.60
369.60
646.80
646.80
2.50
2.50
554.40
554.40
228.41
152.28
236.52
157.68
48.95
45.06
122.36
112.65
5.70
5.91
38.06
39.42

Mixing, Casting, Compaction


geopolymer concrete

and

Curing

Fig. 2 Slump cone test

Acid attack test on concrete specimens

of

GPC can be manufactured by adopting the conventional


techniques used in the manufacture of Portland cement
concrete. In the laboratory, the fly ash and the aggregates were
first mixed together dry on pan for about three minutes. The
liquid component of the mixture is then added to the dry
materials and the mixing continued usually for another four
minutes (Fig.1).
The addition of sodium silicate is to enhance the process of
geopolymerization [11]. For the present study, concentration of
NaOH solution is taken as 12M with Na2SiO3 / NaOH as 2.5
for all the grades of GPC mixes. The workability of the fresh
concrete was measured by means of conventional slump test
(Fig. 2). In order to improve the workability, superplasticizer
Conplast SP-430 with a dosage of 1.5% by mass of the fly ash
was added to the mixture. Extra water (other than the water

To access the quality of geopolymer concrete mixes they were


subjected to aggressive media like immersing the specimen in
acidic environment. The acidic environment was selected
mainly, as we are aware of the conventional concrete which
will disintegrate badly. Hence a comparison will be studied
with geopolymer concrete mixes with conventional concrete.

Preparation of the solution


For the present study it was aimed to prepare the sulphuric acid
with 5% concentration. The concentrated sulphuric acid with
98% purity and density of 1.84 g/ml was selected for the
preparation. For this following formula was used
S = D1 * [(A1/A2) -1]
Where,
A1= Concentration of commercially available concentrated
sulphuric acid
D1= Density of concentrated sulphuric acid of concentration
C1, gm/ml
A2 = Concentration of desired diluted sulphuric acid for testing
S = Volume of water added per ml of sulphuric acid of
concentration, A1 to get diluted sulphuric acid of concentration
A2. From this calculation, we need about 11 ml of concentrated
sulphuric acid to mix with 989 ml of distilled water to get 1 liter
of 5% acid solution for the investigation.

Page 153

Shankar H. Sanni et. al. / INDECS-15

Results and Discussions


Workability
Fresh GPC mixes were found to be highly viscous and
cohesive with medium to high slump. The workability of the
geopolymer concrete decreases with increase in the grade of
the concrete, this is because of the decrease in the ratio of water
to geopolymer solids by mass. For a given geopolymer
concrete, the total mass of water in the mixture is taken as the
sum of the mass of water in the sodium silicate solution, the
mass of water in the sodium hydroxide solution and the mass
of extra water added to the mixture. The mass of geopolymer
solids is the sum of the mass of fly ash, the mass of sodium
hydroxide flake and the mass of sodium silicate solids (the
mass of Na2O and SiO2 in sodium silicate solution). The test
data shown in Figure are somewhat analogous to the wellknown effect of water-to-cement ratio on the compressive
strength of OPC concrete, although the chemical processes
involved in the formation of the binders of both these types of
concretes are entirely different.

The compressive strength lost during the period of exposure in


acidic media for the mixes are presented in Fig. 5 to 6. The
GPC mix retained the characteristic strength with both the
grades of concrete, but there was reduction in characteristic
strength for OPC mixes in range of 12.5% to 30% for both the
grades of concrete. The geopolymer concrete mixes are high
resistant to sulphuric acid attack because they do not have free
lime content in the matrix composition. On contrary, the OPC
specimens do contain the free lime content in the matrix, which
leads to the hydration reactions of Portland cement and this
free lime and also the C-S-H gel of the matrix are easily
attacked by acids. The disintegrated GPC and OPC specimens
after 90 days of exposure are presented in Fig. 3 to 4. The
action of acids on cement paste leads to a conversion of all the
calcium compounds, unreacted residue of C3S and C2S in
cement grains, calcium hydroxide, calcium silicate hydrate and
calcium acuminate hydrate to the calcium salt of the attacking
acid. As a result of the conversion, the binding capacity of the
hardened cement is destroyed. The observations made in the
present investigation match with the findings of the earlier
research [14, 15, 16].

Specimens in the acidic media


The cube specimens of GPC and OPC mixes were submerged
in acid environment of known concentration. To perform the
acid studies, immersion techniques was adopted. After 28 days
of casting, 150 mm cube specimens were immersed in 5%
sulphuric acid solution kept in a plastic tubs such that there was
a minimum of 30 mm depth of acid over the top surface of
specimens. To maintain the uniformity of acidic solution, it
was stirred regularly at least twice in a day. Care was taken to
maintain the concentration of solution, which was replaced at
regular intervals. The pH of the solution was monitored
periodically using pH meter. The results were summarized
after 15, 30, 45, 60 and 90 days of curing period..
Fig. 5 Compressive strength with days of exposure (M30)

Visual appearance
There was no noticeable change in the colour of the
geopolymer concrete specimens. Even after the exposure for 90
days these specimens remain structurally intact though surface
turned little softer and at the verge of slippage of aggregates,
but there was no significant change in mass and shape.
Whereas, in case of OPC specimens, a deposition of a white
layered gypsum crystals was observed on the exposed surface.
These specimens were in highly deteriorated condition with
noticeable bulging. The surfaces were rough and yellowish in
colour. (Fig. 3 and 4)

Fig. 6 Compressive strength with days of exposure (M40)

Conclusions

Fig. 3 GPC specimens after


after 90 days of exposure

Compressive strength

Fig. 4 OPC specimens


90 days of exposure

Based on the experimental investigations carried out, it can be


concluded that geopolymeric materials perform much better in
acidic environment compared to Portland cement The better
performance of geopolymeric materials than that of Portland
cement in acidic environment might be attributed to the lower
calcium content of the source material as a main possible
factor, since geopolymer concrete does not rely on lime like
Portland cement concrete.

Page 154

Shankar H. Sanni et. al. / INDECS-15

References
[1] Narasimhan, M. C. Patil, B. T, and Shankar H. Sanni
Performance of Concrete with Quarry Dust as
fine
aggregate An Experimental Study, Civil Engineering
and Construction Review, September, 1999, pp. 19-24.
[2] Rangan, B. V. Studies on low-calcium fly ash based
Geopolymer concrete, Indian Concrete Institute, 2006, pp.
9-17.
[3] Davidovits, J. Geopolymer chemistry and application,
Institute Geopolymer, France, 2008, pp. 585.
[4] Schmucker, M and MacKenzine, KJD., Microstructure of
sodium
polysialate
siloxogeopolymer,
Ceramic
International, 31, 2004, pp. 433-437.
[5] Fenandez-Jimenez, A and Palomo, A., Characteristics of
fly ashes, Potential reactivity as alkaline cements, Fuel,
2003, pp. 2259-2265.
[6] Davidovits, J., Chemistry of Geopolymeric systems
Terminology, 99 International Conference, Saint-Quentin,
France, 30 June-2 July 1999.
[7] Fernandez-Jimenez, A., Palomo, J. and Puertas, F., Alkali
activated slag mortars, mechanical strength behavior,
Cement and Concrete Research, 29, 1999, pp. 1323-1329.
[8] Hua Xu, van Deventer, J.S.J., The Geopolymerisation of
Alumino-Silicate Minerals, International Journal of
Mineral Processing, 59(3), 2000, pp. 247-266.
[9] Hardjito, D., Wallah, S.E., Sumajouw, DMJ and Rangan,
B.V., On the development of fly ash based geopolymer
concrete, ACI Materials Journal, 101(52), 2004, pp. 467472.
[10] Rangan, B. V., (2008), Mix design and production of fly
ash based geopolymer concrete, The Indian Concrete
Journal, 82(5), pp. 7-14.
[11] R. B. Khadiranaikar and Shankar H. Sanni, Variation of
alkaline solutions on mechanical properties of geopolymer
concrete, ICI Journal, 15(1), 2014, 24-31.
[12] Shankar H. Sanni, Experiemental Investigations on
properties of geopolymer concrete, Ph.D thesis submitted
to Visvesvarya Technological University, Belagavi, 2015.
[13] M. S. Shetty, Concrete Technology, (S. Chand and
Company Ltd., New Delhi, 2002).
[14] Suresh Thokchom, Partha Ghosh and Somnath Ghosh,
Resistance of fly ash based geopolymer mortars in
sulphuric acid, ARPN Journal of Engineering and Applied
sciences, Vol. 4, No. 1, 2009, pp. 65-70.
[15] Rajamane, N.P, Nataraja, M.C, Lakshmanan, N and
Dattatreya, J. K., Sulphuric acid resistance of geopolymer

concrete, Proceedings of Asian Conference on concrete


ACECON 2010, IIT Madras, Dec. 2010, pp. 789-794.
[16] Dinakar, P., Babu, K. G. and Manu Santhanam, Durability
properties of high volume fly ash self concretes, Cement
and Concrete Composites, Vol. 30, No. 10, 2008, pp. 880886.
[17] M. S. Shetty, Concrete Technology, S. Chand and
Company Ltd., New Delhi, 2002
[18] IS: 2386 (Part-IV)-1963, Methods of test for aggregates
for concrete-mechanical properties, Bureau of Indian
standards, New Delhi.
[19] IS: 456-2000, Code of practice for plain and reinforced
concrete, Bureau of Indian standards, New Delhi.
[20] IS: 383-1970, Specification for coarse and fine aggregates
from natural sources for concrete, Bureau of Indian
standards, New Delhi.
[21] IS: 516-1959, Methods of test for strength of concrete,
Bureau of Indian standards, New Delhi.
[22] IS:3812-2003, Specifications for fly ash for use as
pozzolana and admixture, Bureau of Indian standards,
New Delhi.

Page 155

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

LIGHT WEIGHT FLYASH AGGREGATE CONCRETE AN ENVIRONMENTAL FRIENDLY APPROACH

GEENA GEORGE 1, Dr. ASHA.K 2

Department of Civil Engineering , East Point College of Engg & Technology, Bangalore .India.
geenasajith@gmail.com
2

Department of Civil Engineering , MVJ College of Engineering, Bangalore. India,


ashasavi@gmail.com

Abstract
The environmental impacts of crushed stone aggregates become a source of increasing concern in most parts of the
Country. Pollution hazards, noise, dust, blasting vibrations, loss of forests and spoiling of natural environment are the
bad impacts caused due to extraction of aggregates. Landslides of weak and steep hill slopes are induced due to
unplanned exploitation of rocks. In a developing country like India, coal is a major source of fuel for production of
electricity which in turn produces large quantity of fly ash which is an environmental hazard. Disposal of such a huge
quantity poses challenging problems, in the form of land usage, health hazards and environmental dangers. Both in
disposal as well as in utilization, utmost care has to be taken to safeguard the interest of human life, wild life and
environment .Under such circumstances both the environmental issues can be brought together to produce a ecofriendly
solution by developing fly ash aggregates which can replace natural aggregates to some extent.The purpose of this paper
is to provide an overview of utilization of Fly ash as an alternative for natural aggregates in the production of light
weight concrete .

Keywords: Fly Ash, Fly ash Aggregates, Pelletization

1. Introduction
Any country's economic & industrial growth depends on the
availability of power. In a developing country like India where the
coal is a major source of fuel for power generation. About 60%
power is produced using coal as fuel. Indian coal is having low
calorific value (3000-3500 K cal.) & very high ash content (3045%) resulting in huge quantity of ash is generated in the coal
based thermal power stations.. With the commissioning of super
thermal power plants and with the increasing use of low grade coal
of high ash content, presently the annual production of Fly Ash in
India is about 220 million tonnes in the year 2011-2012 with 65000
acre of land being occupied by ash ponds and is expected to cross
1000 million tonnes by the year 2030 and pose serious ecological
problems.

In India, only a small quantity of the total ash produced is currently


utilized in use of ash in concrete, brick making, soil-stabilization
treatment and other applications. In growing need for electricity in
India, 70% of power is generated through thermal power plants.
The environmental dreads from these plants include air pollution
due to particulate emission, water pollution and shortage of land
for dumping the fly ash. Further, the poor quality of Indian coal
has high ash content, which worsens the disposal problem. Most
of the ash generated from the power plants is disposed off in the
vicinity of the plant as a waste material covering several hectares of
valuable land. Instead of dumping the fly ash as landfills, fly ash is
widely used as cement replacement material, pavement base,
blocks etc., in these days.

Geena George. Tel.: +919449820116


E-mail: geenasajith@gmail.com

Page 156

Author et. al. / INDECS-15

With increasing concern over the excessive exploitation of natural


aggregates, synthetic lightweight aggregate produced from
environmental waste is a viable new source of structural aggregate
material. Though the artificial aggregate production attained
attention in research field, in India it is not implemented widely.
This may be due to the availability of natural resources, relatively
higher initial cost for manufacturing and the energy required for
curing. There are also methods like cold bonding which does
not require energy for curing of artificial aggregates. This paper
discuss about the use of fly ash in the flyash aggregates in the light
weight concrete.

Impact of Modern Concrete Use


The usage of concrete worldwide is twice as much as steel, wood,
plastics, and aluminum combined. Concretes use in the modern
world is only exceeded by the usage of naturally occurring
water.Concrete is also the bases of a large commercial industry,
with all the positives and negatives that entails. In the United states
alone, concrete production is a $30 billion per year industry,
considering only the value of ready-mixed concrete sold each year.
Given the size of the concrete industry, and fundamental way
concrete is used to shave the infrastructure of the modern world, it
is difficult to overstate the role this materials plays today.
2.1. Literature Review
1.A. Sivakumar and P. Gomathi carried out the research work on
Pelletized fly ash lightweight aggregate concrete. They concluded
that conversion of fly ash with aggregate is technically feasible and
are found to be light weight in nature.
2.Dr.J.B.Behera, Dr.B.D.Nayak, Dr.H.S.Ray and Dr.B.Sarangi
examined the use of sintered fly ash aggregate in concrete as a
partial replacement of granite aggregate. They concluded that in
addition to light weight characteristics, the sintered fly ash concrete
possesses strength and deformation characteristics similar to
concrete with natural granite aggregate.
3.Gao Li-Xiong, Yaoyan and Wang Ling, China building materials
Academy, Beijing found light weight aggregate concrete made with
sintered fly ash aggregate showed good workability, high
compressive strength and low absorption of water.
4. Mehnet Gesoglu, TuranOzturan and ErhanGunegisi found the
compressive strength of concrete ranging from 20 to 50Mpa was
practically produced by using light weight fly ash aggregates. They
also found the increase in splitting tensile strength of concrete due
to increase in aggregate crushing strength of fly ash aggregates.
5.HaydarArsian and Gokhan Baykal investigated the fly ash
aggregates produced from fly ash and cement mixing by
pelletization method and evaluated Engineering properties such as
crushing strength, specific gravity, water absorption, particle size
distribution, surface characteristics and shear strength properties of
the manufactured aggregates experimentally. The experimental

investigation showed that these aggregates are a good alternative


for wide range of civil Engineering applications.
6.Gokhan Baykal and ATA Gurhan Doven studied the properties of
fly ash aggregates made by pelletization process such as unit
weight, specific gravity, water absorption, crushing value test,
California bearing ratio (CBR), Direct shear tests, soundness tests,
compressive and flexural strength and obtained good results when
compared to conventional aggregates in concrete
2.2.Lightweight Aggregates
In conventional concrete, weight of concrete is one of the
parameters to compare with weight of fly ash aggregate concrete.
Normally density of concrete is in the order of 2200 to 2600 kg/m3.
This heavy self weight makes an uneconomical structural material
compared to low self weight of fly ash aggregate concrete. In order
to produce concrete of desired density to suit the required
application, the self weight of structural and non structural
members are to be reduced. Hence economy is achieved in the
design of supporting structural elements which lead to the
development of light weight concrete.
Lightweight aggregate concrete is the concrete made by replacing
the usual material aggregate by lightweight aggregates. Though
lightweight concrete cant always substitute normal concrete for its
strength potential, it has its own advantages like reduced dead load,
and thus economic structures and enhanced seismic resistance, high
sound absorption and good fire resistance. Because of the above
reasons the study on fly ash aggregate concrete is taken in this
research work.
Heat treatment usually results in relatively thick pore walls that
give the aggregates suitable compressive strengths. The porosity of
all these aggregates results from heat treatment, either naturally or
artificially induced. A porous but relatively impermeable texture
not only makes these materials lightweight but also makes them
both thermal and acoustic insulators, desirable for bulk use in floor
and roof fill and with an added advantage of saving weight in
concrete.In this broad range, lightweight aggregate concretes can
be used for bridge decks; concrete walls and floors; concrete
frames using precast, prestressed, or post-tensioned girders; precast
tilt-up panels; poured barge and ship hulls; and lightweight
concrete masonry units. In bulk form, the aggregates are also
sometimes used as railroad ballast, as highway and embankment
fill, and in the surface course of bituminous paving mixtures.
Reducing the dead load of structures can be a major advantage in
construction.

2.2.1. Properties Of Light Weight Aggregtes


The principal properties desired in materials used as lightweight
aggregates

Light weightreduces dead load (weight in place) and facilitates


the physical handling of the materials

Page 157

Author et. al. / INDECS-15

Strengthmaintains the structural integrity of the end


product,such as in concrete products

Thermal and acoustical propertiesnormally the result of air


spaces, voids, or pores in the lightweight aggregate materials

High fire resistancelow probability of physical breakdown at


temperatures below the melting point of the aggregate

Toughnessminimum tendency to crack or break if nailed or


stressed in construction use.

material like cement, lime, bentonite, metakaolin, kaolinite, glass


powder and ceramic powders are added. Characteristics of class-C
and class-F are given in the table:2

2.2.3.1.FLYASH AGGREGATE PRODUCTION


There are two basic methods to produce an aggregate from fly ash:
agglomeration of the fine fly ash particles to form aggregate-size
pellets, and bonding of the material in the pellets. This gives the
pellets strength and other properties necessary to meet the criteria
for a lightweight aggregate.

Pelletizing Process
Other properties that are sometimes considered ,depending on the
application, include the following are low water absorption,
resistance to freezing and thawing , low shrinkage characteristics
and minimum thermal expansion, good bonding with cement ,
chemical inertness, good elastic properties , abrasion resistance.
2.2.2. FLY ASH
Fly ash is a pozzolan. Pozzolan as a siliceous or siliceous and
aluminous material that itself possesses little or no cementitious
value, but that will, in finely divided form and in presence of
moisture, chemically react with calcium hydroxide (CH) at
ordinary temperatures to form compounds having cementitious
properties; there are both natural and artificial pozzolans.
All fly ash exhibit pozzolanic properties to some extent For this
simple reason it is rapidly becoming a common ingredient in
concrete all over the world. Most of the reasons for using fly ash in
any proportion are practical, such as increasing strength and
durability, decreasing heat of hydration, and decreasing
permeability. The use of fly ash as a performance-enhancing
ingredient in concrete is one of the most outstanding examples of
industrial ecology-i.e., making effective use of waste resources,
and ultimately eliminating the concept of waste altogether.
2.2.3. FLY ASH AGGREGATES
The chemical composition and physical characteristics of the fly
ash from a coal-fired furnace are controlled by the type of coal and
the processing conditions of the furnace . However, the pozzolanic
properties of the fly ash mainly depend upon the mineralogy and
particle size of the ash. Fly ash consists mostly of SiO2. The SiO2
can be amorphous (glassy and rounded) and crystalline (sharp and
pointed)1. The largest fraction of fly ash consists of glassy spheres
of two types: solid and hollow (cenospheres). The amorphous
glassy particles are the primary contributor to the pozzolanic
reaction.
The two types of fly ash are specified in ASTM C 618, namely
class-C and class-F classified based on the chemical composition
resulting from the different types of coal burning. Class-C fly ash is
normally produced from the burning of sub-bituminous coal and
lignite and class- C fly ash. The flyash aggregates are porous
material and to improve the strength of the pellet the binder

The desired grain size distribution of an artificial lightweight


aggregate is either crushed or by means of agglomeration process.
Agglomeration (also called "pelletizing') can be performed by two
distinct methods.One is by agitation, where fly ash particles are
introduced onto an inclined rotating disk along with a wetting agent
and an appropriate binder. Balling of the material occurs by the
formation of 'seeds' which ultimately grow into pellets of a certain
maximum size, usually about 3/8 in. to 5/8 in. in diameter.
The other method is the pressure or extrusion method where
agglomeration is accomplished by using a continuous piston-type
press where more or less rectangular or cylindrical pellets are
formed, that are 3/8 in. to 1/2 in. in size. The extrusion process
generally results in a product having a higher density than the
spherical pellets produced by agitation .
The pelletization process is used to manufacture artificial
lightweight coarse aggregate using fly ash. Pelletization depends on
(i) speed of revolution of pelletizer disc, (ii) angle of pelletizer disc,
(iii) moisture content, and (iv) duration of pelletization.The
different types of pelletizer machine were used to make the pellet
such as disc or pan type, drum type, cone type and mixer type.
With disc type pelletizer the pellet size distribution is easier to
control than drum type pelletizer. With mixer type pelletizer, the
small grains are formed initially and are subsequently increased in
particle size by disc type pelletization .The dosage of binding agent
is more important for making flyash balls .
Bonding
The bonding of the green pellets can be accomplished in three
ways: firing, hydrothermal processing, and cold bonding. There are
different methods of firing method. One method is the continuous
rotary kiln whereby the material is heated in a rotating, almost
horizontal kiln to produce the lightweight aggregate product. In
another method utilizes a traveling sintering grate that transports
the material through a sintering oven. The combustible portion is
provided by the organic portion of component materials such as the
carbon fraction of the fly ash as measured by a loss on ignition test.
In the hydrothermal process ,also called cold bonding is
accomplished by adding lime and/or cement to fly ash in order to
achieve a chemical reaction that will bond the materials together.
After the fly ash and lime have reacted, the mixture is pelletized in

Page 158

Author et. al. / INDECS-15

an inclined rotating drum. The pellets are then placed in an


autoclave to harden under pressurized saturated steam conditions.
The product from the cold bonding process is thought to possess
inferior properties to those of the other two processes due to the
lack of firing .
Curing and drying of flyash aggregate
After the formation of flyash aggregate, the fly ash aggregate
which are formed are out from the cement mixer and then it is dried
for about 24 hours. Then the dried fly ash aggregate is then cured
in the water tank for 30 days. Curing of 30 days is done to attain
the strength of the flyash aggregate.

Workability of Concrete
Flyash aggregates usually also have high water absorption, which
can reduce the workability. Therefore, aggregates are usually presoaked to compensate for the reduction in workability; however,
there is probably a tendency to increase the water dosage, which
leads to further reductions in strength.

Drying And Curing Of Fly Ash Aggregates


The fly ash aggregates were taken out from the mixer and allowed
to dry for a day. The dried aggregates were cured in the water tank
for 28 days . The fly ash aggregates are shown in figure:2

Properties of Fly ash Aggregates


The shape and texture of aggregate affects the fresh property of the
concrete. Fly ash aggregate is rounded in shape while natural
gravel is angular in shape . Rounded aggregates promotes
workability of concrete while the angular nature of natural gravel
gives a better bonding property but requires more cement mortar
for better workability .The Physical Properties of Natural granite
Aggregate and Fly ash Coarse Aggregate are given in the table:3

2.2.3.2.Hardening Process of Flyash Aggregates


The hardening of pellets is done by different process such as cold
bonding, sintering and autoclaving. Cold-bonded fly ash aggregates
are hardened by different curing process namely normal water
curing, steam curing and autoclaving. Autoclave and steam curing
method is less effective to improve the properties of aggregate as
compare to normal water curing method. Among accelerated cured
class C fly ash aggregate, autoclaved aggregates has properties
closer to the normal water cured aggregate due to the dense
microstructure formation. The way of curing is important in
enhancing the aggregate strength. Hence, for high-early strength .
autoclaving may be adopted whereas normal curing gives better
results. In sintering method the cold bonded pellets are burned in a
muffle furnace at temperature range of 800 to 1200C. The mineral
particles in the binder fuse together to form the crystalline structure
(CSH) and results in higher strength of the aggregate.

Mix Design
The mix design of lightweight aggregate concrete is different from
the conventional concrete mix design. It is more complex due to
the porous nature of aggregates. Since the flyash aggregates are
porous in nature it requires extra water for good workability. The
mix design concepts are usually based on the production of higher
strength matrix to low water cement ratio for the weaker aggregate.
The gradation of aggregate is important to improve the engineering
properties in the concrete mix for that different aggregate grading
size distributions are required The design of flyash aggregate are
followed in two methods; loose volume calculation and absolute
solid volume calculation. The lightweight aggregate is pre-wetting
before addition of concrete mix.

Fig:1 Flyash aggregates

Microstructural Characteristics Of Flyash Aggregate


Concrete
The durability and strength aspects of concrete affected by its
aggregate and cement paste as well as the interfacial zone between
them.In normal concrete the aggregate-cement paste interface is the
weakest part of the micro-structural system where cracks begins,
strongest component is normal aggregate. But, in Flyash aggregate
concrete is the interaction between the cement paste-aggregate is
complex . This type of aggregate are porous in nature, absorbs
more water which yielded to the surrounding matrix. The porosity
of flyash aggregate can vary between 25 to 75% depending on the
manufacturing process used. For applied micromechanical method
considered the perfect bonding between the aggregate and mortar
.Normally sintered fly ash lightweight aggregate were produced by
heat and polymer treatment so that to improve their strength,
absorption and pozzolanic activity according to their properties of
aggregate by change to the microstructure. SEM analysis to
observe the higher magnification to see more uniform distribution
of small pore size in the sintered fly ash aggregate at the
temperature treated aggregate as 1200 to 1300C .Mechanical
interlocking plays an important role for strengthening the interface

Page 159

Author et. al. / INDECS-15

2.4. Tables
Table 1.Physical Properties of Natural granite Aggregate and Fly ash Coarse Aggregate
Material
Cement
Fine aggregate
Coarse aggregate
Fly ash aggregate
Specific gravity

3.155

2.71

2.77

1.32

Water absorption
(%)

0.2

0.1

7.6

Table 2.Physical and chemical characterization of flyash


Some definition of fly ash
Low calcium fly ash
types according to ASTM
C618

High calcium fly ash

Fly ash from

Bituminous coal and anthracite

Subbituminous and lignite

Reaction characteristic

Pozzolanic

Pozzolanic and hydraulic

Definition by ASTM C 618

Class F

Class C

SiO2 + Al2O3 + Fe2O3 70 %


Free moisture, max: 3,0 %
LOI, max: 6,0 %
SO3, max: 5,0 %
CaO, max: No limit
Amount retained when wet sieved
on 45 m: Max. 34 %

SiO2 + Al2O3 + Fe2O3 50 %


Free moisture, max: 3,0 %
LOI, max: 6,0 %
SO3, max: 5,0 %
CaO, max: No limit (Note: CaO > 10
%) Amount retained when wet sieved
on 45 m: Max. 34 %

Some definition of fly ash


types according to ASTM
C618

Low calcium fly ash

High calcium fly ash

Fly ash from

Bituminous coal and anthracite

Subbituminous and lignite

Reaction characteristic

Pozzolanic

Pozzolanic and hydraulic

Table 3. Physical Properties of Natural granite Aggregate and Fly ash Coarse Aggregate
Properties
Natural granite Aggregate
Fly ash Coarse Aggregate
Shape

Angular

Spherical

Specific gravity

2.66

1.41

Bulk density (Kg/m3)

1715

912

4.75mm to 20mm

4.75mm to 20mm

Water absorption (%)

1.16

21

Crushing value (%)

24.94

30.70

Impact value (%)

23.86

22.52

Size (mm)

Page 160

Author et. al. / INDECS-15

References
3. Conclusion

Fly ash aggregates shows results comparable with natural


gravel and the natural resource is in the side of depletion, fly
ash aggregates can be considered as a replacement material for
coarse aggregate.It also improves the property of concrete as
fly ash is a pozzolanic material. The obtained aggregates can be
considered for various applications like wall panels, masonry
blocks, roof insulation material, structural load bearing
elements etc. Use of sintered fly ash as coarse aggregate can
reduce the cost of construction and it is useful in environmental
point of view.
Concrete with density 2150 kg/m3 can be achieved using fly
ash aggregates while density of normal concrete mix goes up to
2600 kg/m3 .The rounded shape of fly ash aggregate gives
better workability compared to the angular natural gravel. Low
specific gravity compared to natural gravel proves it to be a
light weight aggregate material and fly ash has been consumed
in large volume when it is used as a coarse aggregate
replacement material due to its occupation of large volume in
concrete. This in turn reduces the problem of dumping as
landfills to greater extent.
The water absorption of fly ash aggregate is much higher than
that of natural gravel which is the major disadvantage which
can be eliminated by various treatment methods .

[1] A. Sivakumar and P. Gomathi Pelletized fly ash


lightweight aggregate concrete Journal of Civil Engineering
and Construction Technology Vol. 3(2), pp. 42-48, February
2012
[2] . Manikandan, R., Ramamurthy, K., 2008, Effect of curing
method on characteristics of cold bonded fly ash aggregates,
Cement & Concrete Composites, 30, pp 848853
[3] Mehmet Gesoglu, Turan O zturan, Erhan Gu neyisi,
2007, Effects of fly ash properties on characteristics of
coldbonded fly ash lightweight aggregates, Construction and
Building Materials, 21, pp 18691878.
[4] Geetha, S., Ramamurthy, K., 2010, Reuse potential of low
calcium bottom ash as aggregate through pelletization, Waste
Management, 30, pp 15281535.
[5]
Rajamani N.P, Annie peter J. ,Sabitha D. and
Gopalakrishanan S. (2004). Studies on development on
bonded fly ash aggregates for use as coarse aggregate in
structural grade concretes New building materials
andconstruction world Vol 10,issue 4, Oct (pp 60 to 70).
[6] Dr. J.P. Behera, Dr. H.S. Ray, Dr. B.D. Nayak and Dr. B.
Sarangi. (2004). Light weight concrete with sintered fly ash
aggregates. A study on partial replacement to normal granite
aggregate. Institution of Engineers, India (IE(I)) Journal CV
Vol 85 August2004, PP 84 to 87.

Page 161

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

EXPERIMENTAL INVESTIGATION ON BEHAVIOUR OF STEEL CONCRETE


COMPOSITE COLUMNS
1

R.THIRUMALAI

Asst. Professor, Dept of Civil Engineering,


Adhiyamaan College of Engineering, Hosur,Tamil Nadu India -635 109
thirumalai063@gmail.com

Abstract
Composite columns of the steel and concrete have been used and studied worldwide, but filled tubular columns
need more attention. Extensive research work has been done in Japan in the last 15 years. This project presents an
experimental and analysis study on the behavior of steel concrete composite columns under axial and lateral load.
Totally 22 number of specimens (9 number kept hallow without concrete filled, 9 with concrete filled and 4 under
lateral load).Columns specimens where made with different cross section, various shapes and various thickness of
steel tube. The concrete filled in the steel tubes is M20 grade. And the strength comparison of different cross-section,
shape and thickness columns, comparison of experimental failure loads with the predicated failure loads in accordance
with method described in eurocode-4 ,part 1.1and specimens where analyzed by ANSYS SOFTWARE.
Keywords:Composite, ANSYS,EUROCODE-4.1.

1. Introduction
Concrete filled steel tubular columns have many
excellent structural properties, such as high compressive
strength, large ductility and large energy absorption
capacity. Then, composite tubular columns have been
gradually used widely in the world.The strength of steel and
concrete for building structures is getting higher with the
development of new materials. The cross section with high
strength materials becomes smaller, and consequently a
column becomes more slender.The greatest advantage of
this concept is that the two materials are put to their ultimate
use. The steel jacket confines the concrete and provides the
flexural strength for the system.The presence of the concrete
delays steel problems in compression zones such as local
and overall buckling and also provides stiffness for the
column.The main advantage of CSC columns is the no usage
of longitudinal and transverse reinforcement.
2. Strength by Design Code(EUROCODE-4.1)
EC4 is the most recently completed international standard in
composite construction. EC4 covers concrete - encased and
partially encased steel sections and concrete-filled sections
with or without reinforcement.EC4 considers confinement
effects for circular sections when relative slenderness has
value less than 0.5. EC4 uses limit state concepts to achieve
the aims of serviceability and safety by applying partial
safety factors to load and material properties. It is the only
code that treats the effects of long-term loading separately.
The ultimate axial force of a circular column is,

PP = Aa* 2a + Ac* Ck * 1 + 1* { t / d } * { fy / fck}


Where,

1= 10[

1 {10e /d}.

=
+
{
1
2
20
20 } * {10e / d}.
a = { fy / a }.Ck = { fck / c }.
PP plastic resistance of columns.Aa&Ac- Area of the steel
and concrete.t thickness of steel.d outer diameter of the
columns.fy&fck yield strength of steel and concrete.a& c
- partial safety factor for steel and concrete.

Pcr = 2*{ (EI)e/l2}, (EI)e = Ea Ia + 0.8 EcdIc.


Ecd = { Ecm / c },Ecm = 5000 ( fck)1/2
(EI)e - Effective elastic flexural stiffness of the composite
column,Ia&Ic second moments of area of steel and
concrete.Ea& Ec - modules of elasticity of steel and
concrete.c - partial safety factor concrete is reduced to 1.35
as per eurocode 4.
Non Dimensional Slenderness
= (PP/PCR)(1/2)
Reduction Factor
Xx = 1/ { x + (x2 - 2) (1/2)}
Where ,
x = 0.5 [ 1 + ( - 0.2 ) + ]

Page 162

Author et. al. / INDECS-15

3.Casting and Testing of Circular Steel Concrete

Composite Columns
Casting of circular steel concrete composite column
of various thickness where casted
S.No

Inner Diameter

Thickness

2mm
1

60mm

85mm

110mm

4mm
6mm
2mm
4mm
6mm
2mm
4mm
6mm

Table 1. circular steel concrete composite column of various thickness

3.1. Testing Procedure

Each specimen was tested under axial compression


loading in a loading frame at structural laboratory. The
column of the test assembly was placed in a loading frame
the column was centered accurately using plumb bob to
avoid eccentricity at the both ends of the column mild steel
plate support condition was provided. The axial load is
achieved by using spherical steel ball having 55mm in
diameter and 0.6 kg of mass was used in between the two
mild steel plates to ensure proper verticality of support mild
steel plates are welded and are aligned using 6mm diameter
mild steel smooth bar which are used in on the fur sides of
the plate.
The column is lifted by 2000 KN capacity
hydraulic jack which is kept at the bottom of the testing
platform above which the steel plate bottom support was
placed. Then the test specimen is placed bottom support and
the top support is placed over the top column head and
finally it touches the load cell which is pivoted in the
horizontal element of the loading frame. The column vertical
alignment was checked before start testing. The column was
loaded under axial compressive load (10KN) by hydraulic
jack which is kept at the bottom of the column. The column
is loaded after the failure and buckling of column. For
Lateral load the hydraulic jack is placed in horizontal
direction and connected to the specimen.

levels, temperature distributions, pressure, etc. It permits an


evaluation of a design without having to build and destroy
multiple prototypes in testing. The ANSYS program has a
variety of design analysis applications, ranging from such
everyday items as dishwashers, cookware, automobiles,
running shoes and beverage cans to such highly
sophisticated systems as aircraft, nuclear reactor
containment buildings, bridges, farm machinery, X-ray
equipment
and
orbiting
satellites
ANSYS/LS-Dy is a tightly integrated program that
combines industry-leading finite element technology with
explicit nonlinear dynamic solution capabilities.
4.1.Results and Discussion

For axial loading a set of detailed experiments have


been conducted on hollow and filled steel tubular columns
under axial loading. The hollow section is failed due to
buckling around the circumference of the specimen.
For lateral loading, Support conditions of the
composite column is one end fixed at bottom and other is
free at top. Lateral load is applied at free end with intensity
of 50 KN. For C1 maximum displacement occurs at free end
with 0.0625 mm. Minimum displacement occurs at 100 mm
at 0.0069 mm.
For C2 maximum displacement occurs at free end
with 0.1234 mm and minimum displacement occurs at 100
mm at 0.0137 mm. For S1Maximum displacement occurs at
free end with 0.1234 mm and Minimum displacement occurs
at 100 mm at 0.0137 mm. For S2 maximum displacement
occurs at free end with 0.039 mm and Minimum
displacement occurs at 100 mm at 0.0043 mm.

3.2. Software analysis (ANSYS)

ANSYS, Inc. has developed product lines that allow you to


make the most of your investment and choose which product
works best in your environment. ANSYS is a finite element
analysis (FEA) code widely used in the computer-aided
engineering (CAE) field. ANSYS software allows engineers
to construct computer models of structures, machine
components or systems; apply operating loads and other
design criteria; and study physical responses, such as stress

Fig 1. Cross section and plan of


circular composite column

Fig 2. cross section and plan


of square composite column

Page 163

Author et. al. / INDECS-15

COMPRESSSIVE STRENGTH (kN)


O.D
mm

S.NO

L/D

T
mm

(AC)
mm2

(Aa)
mm2

Concrete filled column


EUROCODE-4

TEST
VALUE

HOLLOW
COLUMN

37

62

17

2327.4

191.6

64.68

90

64

16.5

2327.4

389.5

107.8

127.8

66

16

2327.4

593.7

176.2

190

66

87

12.1

5674.50

270.17

108.1

120

67.2

89

12

5674.50

546.6

143.7

170

81.3

91

11.6

5674.50

829.3

363.30

400

90.3

112

9.4

9503.31

348.7

348.53

390

87.1

114

9.2

9503.31

703.7

437.71

470

175.1

116

9.1

9503.31

1064.9

527.51

630

266.2

Table 2. Compressive Strength For Columns


Load
Sno

Shape

Circle

Square

Size

As per software analysis

Thickness

intensity

of steel wall

at free

Minimum

Maximum

end (kN)

displacement(mm)

displacement(mm)

100mm

50

0.0069

0.0624

(dia)

50

0.0069

0.0624

100*100

50

0.0137

0.1234

mm

50

0.0043

0.039

Table 3. Lateral load as per software results

4.2. Strength comparison between the thickness of steel

200
150
100
50
0

190
127.8
90.1
49
37

65.5

hollow
filled

THICKNESS (mm)
Fig 3 : For 60mm Diameter

LOAD (KN)

LOAD(KN)

250

210

200
150
100

170
120
67.2

81.1

91

50

Hollow
Filled

0
2

THICKNESS (mm)
Fig 4 : For 85mm Diameter

Page 164

700
600
500
400
300
200
100
0
116

THICKNESS (mm)

112

eurocode
-4
102

Filled

89

Hollow

66

630
700
600
470
500
390
400
266.2
300
175.1
200
87.1
100
0
2
4
6

LOAD(KN)

LOAD(KN)

Author et. al. / INDECS-15

DIAMETER ( mm)

Fig .5 : For 110 mm Diameter

Fig .6 : Comparison of EUROCODE 4 VALUE Vs


TEST VALUE

0.07

DISPLACEMENT

0.06
0.05
0.04
0.03
0.02

C1

0.01

C2

0
0 to 10

10 to
20

20 to
30

30 to
40

40 to
50

50 to
60

60 to
70

70 to
80

80 to
90

90 to
100

C1

0.0069 0.0138 0.0208 0.0277 0.0346 0.0416 0.0485 0.0555 0.0624

C2

0.0069 0.0139 0.0208 0.0278 0.0347 0.0417 0.0486 0.0556 0.0625

10 CM INTERVALS
Fig. 7 : Displacement Of Circular Column 50 kN Load Intensity

DISPLACEMENT

0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
0 to 10 10 to 20 20 to 30 30 to 40 40 to 50 50 to 60 60 to 70 70 to 80 80 to 90

90 to
100

S1

0.0137

0.0274

0.0411

0.0548

0.0685

0.0822

0.096

0.1097

0.1234

S2

0.0043

0.0086

0.013

0.0173

0.0216

0.026

0.0303

0.0347

0.039

10 CM INTERVAL
S1

S2

Fig.8. Displacement of Square Column 50 Kn Load Intensity

Page 165

Author et. al. / INDECS-15

5. Conclusion

A set of detailed experiments have been


conducted on hollow and filled steel tubular
columns. The hollow section is failed due to
buckling around the circumference of the
specimen.

But the filled column gives the strength,


2to4 times greater than hollow columns.

As the thickness of the steel wall increases,


strength of the columns increases.

The experimental test value satisfies the


eruocode-4 theoretical value of steel
concrete composite columns

Displacement of 4 mm thick circular


column is less than 0.16% of 3 mm thick
circular column.Displacement of 4mm thick
square column is less than 68.39% of 3 mm
thick square column.

Displacement of square column is greater


than 49.35% of circular column under
lateral load. Performance of circular column
under lateral load is good compared with
square column.

References :
1.Workshop
on
steel-concrete
composite
structures in ANNA UNIVERSITY.
2. Indian code IS: 11384: 1985 CODE OF
PRACTICE FOR COMPOSITE CONSTRUCTION
IN STRUCTURAL STEEL AND CONCRETE
3. EUROCODE-4 : DESIGN OF COMPOSITE
STEEL AND CONCRETE STRUCTURE:PART
1.1.
4. Johansson M, Gylltoft K. Behaviour OF
CIRCULAR STEEL-CONCRETE COMPOSITE
STUB
COLUMNS.
JOURNAL
OF
STRUCTURAL ENGINEERING, 2002, 128(8):
1073-1081.
5.Behavioural Studies On Hollow Double Skinned
Steel Concrete Composite Columns
,By
N.Balasubramanian,
R.B.Karthika
And
Dr.R.Thenmozhi, GCT Cbe.
6. Concrete-filled double-skin tubular columns with
external steel rings,C.X. Dong and J.C.M. HO.
7. Behaviour Of Hollow Concrete-filled Steel
Tubular Composite Elements, Artiomas Kuranovas,
Audronis Kazimieras Kvedaras .
8. Plastic mechanism analysis of CHS stub columns
strengthened using CFRP,
M. Elchalakani,M.R. Bambach.
9. New concrete-filled hollow FRP composite
column, Amir Mirmiran.

Page 166

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

EFFECT OF FIBRE REINFORCEMENT ON STRENGTH AND


STIFFNESS IMPROVEMENT OF COHESIVE SOIL
Suchit Kumar Patel1, Baleshwar Singh2
1

Civil Engineering Department, Indian Institute of Technology, Guwahati, 781039, India, p.suchit@iitg.ernet.in
Civil Engineering Department, Indian Institute of Technology, Guwahati, 781039, India, baleshwar@iitg.ernet.in

Abstract
A study has been conducted to investigate and quantify the strength and stiffness improvement behaviour of a cohesive
soil (MI) by reinforcing with glass fibres. The influence of fibre length (10, 20 and 30 mm) and fibre content (0.25, 0.5,
0.75 and 1% by dry weight of soil) on compressive and shear strength was examined by conducting unconfined
compression, triaxial compression and California Bearing Ratio (CBR) tests. The test results show that the compressive
strength, shear strength parameters, stiffness modulus and energy absorption capacity of the fibre-reinforced soil have
significantly improved. Compressive strength of the reinforced soil is found to enhance by two fold along with
accumulation of ductile behaviour during deformation. The maximum improvement in CBR value is found nearly 2.5
times that of the original soil. The maximum improvement in cohesion, friction angle and energy absorption capacity
have occurred for 20 mm long fibre of 0.75% fibre content. The strength improvement ratio of the reinforced soil is
observed to be higher at lower confining pressure irrespective of fibre content and length. The stiffness modulus
improvement is found to be larger at initial strain than at larger axial strain.
Keywords: Fibre reinforcement, unconfined compression strength, shear strength, stiffness modulus, CBR
2. Experimental Programme
1. Introduction
The basic idea of fibre-reinforced soil has been developed from the
protection principle of soils in nature by plant roots [1]. In past
decades, several research works has been carried out to improve the
engineering performance of soils by using fibre-reinforcement
technique. Majority of work started with initial research on fibrereinforced cohesionless soil to understand the mechanisms. Fibrereinforced fine-grained soils have also been studied through
unconfined compression tests [2-4], tensile strength tests [5-6]
California Bearing Ratio tests [4, 7] and triaxial tests [8-11].
Locally available soils are used as foundation material and in
several earth structures such as road pavement, dam and retaining
wall backfill. These applications need strength characterization of
the soil which may not satisfy the required criteria. With this in
view, the soil can be improved by reinforcing it with randomly
distributed discrete fibres as tensile element. Glass fibres have
some advantage over other synthetic and natural fibres as it is more
resistant to ageing, heat and aggressive chemical environment and
has very good tensile strength. It is important to specify the
optimum combination of reinforcement for field applications.
This study investigates the influence of glass fibre inclusion on the
strength and stiffness behaviour of a locally available cohesive soil
under different loading condition. The results of a series of
unconfined compression (UC) tests, triaxial tests (CD) and CBR
tests (soaked and unsoaked) on soil-fibre composite specimens are
presented.

2.1. Materials and Methods


A locally available fine-grained soil classified as MI was used. The
plastic and liquid limits of the soil were 25% and 46%. Glass fibres
of 0.15 mm diameter were used as reinforcing material. Three
different fibre lengths (L = 10, 20, 30 mm) and different
percentages of fibre content (fc = 0.25, 0.5, 0.75, and 1% by dry
weight of the soil) were chosen to prepare soil-fibre specimens at
the optimum moisture content and maximum dry density of the
soil. Unconfined compressive test on the specimens was conducted
at an axial strain rate of 1.2 mm/min. Triaxial testing of specimens
was carried out at four different confining pressures ranging from
100 to 400 kPa with a shearing strain rate of 0.12 mm/min. The
deviatoric stress was applied up to 20% axial strain level or up to
failure of specimen whichever occurred earlier. CBR test of the
specimens was done under both unsoaked and soaked conditions.

3. Results and Discussion


3.1. Unconfined Compression Tests
3.1.1. Axial Stress-Strain Response
Axial stress-strain response showing the effect of fibre content and
fibre length is presented in Figs. 1 and 2 respectively. It is clear
from these figures that fibre has very significant effect on stressstrain response. The peak strength of the soil mix increases with
both fibre content and fibre length up to certain limit. It is also seen

Suchit Kumar Patel. Tel.: +9508996424


E-mail: p.suchit@iitg.ernet.in

Page 167

Patel and Singh / INDECS-15

that the fibre-reinforced soil exhibits smaller loss of post-peak


strength, and the loss still becomes less pronounced for higher fibre
content. Even after peak, the reinforcement supports the soil and
reduces the post-peak loss in strength. The results indicate that the
optimum fibre content depends on the fibre length. For L = 20 mm,
the strength improves up to 0.75% fibre content (Fig. 1). For
various fibre contents, the strength is found to improve up to 20
mm fibre length (Fig. 2).
300

the addition of fibres improved the deformation behaviour along


with the strength. The fibres have imparted ductility to the natural
soil which can sustain more deformation.
Table 1. Peak UC strength and axial strain of fibre-reinforced soil

L (mm)

fc (%)

Peak stress
(kPa)

10

0.25
0.5
0.75
1
0.25
0.5
0.75
1
0.25
0.5
0.75
1

137
157
185
205
201
182
241
279
249
187
237
261
247

Axial stress (kPa)

250
200

20

150
fc = 1%
fc = 0.75%

100

fc = 0.5%
fc = 0.25%

50

30

Stress
improvement
(%)
19.5
47.5
67.5
63.5
44.5
103.5
141.5
119.5
49.5
99.5
123.5
109.5

Peak axial
strain (%)
2.7
4.5
5.4
7.5
8.6
5.3
6.1
8.9
10.8
5.9
7.0
10.3
9.7

No fibre
0
0

10

15

20

Axial strain (%)


Fig. 1. Effect of fibre content on stress-strain response (L = 20 mm)

300

Axial stress (kPa)

250
200
150
100

L = 30 mm
L = 20 mm
L = 10 mm
No fibre

50
0
0

10
Axial strain (%)

15

20

Fig. 2. Effect of fibre length on stress-strain response (fc = 0.75%)

Table 1 shows the peak stress, percentage improvement in peak


stress and the failure axial strain of the reinforced soil. For 10 mm
long fibre, the highest strength improvement is from 137 kPa for
the parent soil to 205 kPa i.e. almost 67.5% for 0.75% fibre
content. In case of 20 mm long fibre, the strength has modified to
279 kPa (141.5%) for 0.75% fibre content, while for 30 mm long
fibre the maximum enhancement in strength is 261 kPa (almost
123.5%) for 0.75% fibre.
The strength of soil has improved significantly for all fibre
inclusions, and the improvement is higher for 20 mm fibre for
almost all fibre contents. The failure axial strain of fibre-reinforced
soil increases with both fibre content and fibre length, showing that

Fig. 3. Failure pattern of fibre-reinforced soil (a) unreinforced (b) L =


10 mm, fc = 0.25% (c) L = 10 mm, fc = 0.5% (d) L = 20 mm, fc = 0.25%
(e) L = 20 mm, fc = 0.5% and (f) L= 30 mm, fc = 1%

Fig. 3 indicates typical failure patterns of tested specimens with


different fibre length and content combinations. In unreinforced
specimen, a single major crack can be seen running from top to
bottom (Fig. 3a) indicating sudden failure i.e. brittle behaviour of
soil after its peak strength. With addition of fibres, the specimen
failure pattern has changed. Instead of the single major crack,
multiple small cracks appear up to the half portion of specimen. At

Page 168

Patel and Singh / INDECS-15

Stiffness modulus at any particular strain can be obtained by


dividing the axial stress at that strain with the corresponding strain.
Table 2 shows the stiffness modulus value at strain corresponding
to peak, half-peak and a particular axial strain of 2.7%, which is the
failure strain of plain soil. The initial stiffness or pre-failure
stiffness of soil at smaller strain seems to be improved significantly
as can be noted from the stiffness value at 2.7% strain. Only at
higher fibre dose (fc = 1%) stiffness is found to decrease in more
visible way. The soil reinforced with 20 mm fibre is showing
higher stiffness response for all cases.
Table 2. Stiffness modulus of fibre-reinforced soil
L
(mm)

fc (%)

10

0.25
0.5
0.75
1
0.25
0.5
0.75
1
0.25
0.5
0.75
1

20

30

At Peak
(MPa)

At half-peak
(MPa)

5.18
3.62
3.05
2.75
2.34
4.2
4.03
3.25
2.2
3.17
3.48
2.62
2.55

7.29
5.96
5.93
4.83
3.63
7.29
7.22
5.92
4.17
5.82
5.89
4.81
4.8

At 2.7%
strain
(MPa)
5.18
5.21
5.96
6.15
3.61
6.27
8.41
8.67
6.87
5.81
7.41
7.85
7.24

fc = 1%

fc = 0.25%

No fibre
400
300
200
100
0
0

10
Axial strain (%)

15

20

600

(b)

500
400
300
200
fc = 1%
100

fc = 0.5%

fc = 0.75%
fc = 0.25%

No fibre
0
0

10
Axial strain (%)

15

20

Fig. 4. Effect of fibre content (a) 3 = 100 kPa, L = 20 mm (b) 3 = 400


kPa, L = 20 mm
Table 3. Peak deviatoric stress and strength ratio
fc (%)

3
(kPa)

100
200
300
400
100
200
300
400
100
200
300
400
100
200
300
400
100
200
300
400

3.2. Triaxial Tests


3.2.1. Deviatoric Stress-Strain Response
Figs. 4 and 5 present the effect of fibre content and fibre length on
deviatoric stress-strain response under low (3 = 100 kPa) and high
(3 = 400 kPa) confining pressure respectively. The stress-strain
response at initial smaller deformation is similar for all soil-fibre
specimens. The deviatoric stress is found to increase with confining
pressure and fibre content. The strength is found to improve up to
0.75% fibre content and 20 mm length. Table 3 lists the peak
deviatoric stress and strength ratio (reinforced peak
stress/unreinforced peak stress) under any particular confinement.
The strength ratio of soil is noted to be the maximum at low
confining pressure (100 kPa) for all fibre combinations which
further decreases with increase in confinement. The maximum
strength ratio is found to be 1.87 for 20 mm fibres of 0.75% content
under 100 kPa confining pressure and the least value is 1.19 for 10
mm fibres under 400 kPa confining pressure.

(a)

fc = 0.75%

fc = 0.5%

500

Deviatoric stress (kPa)

3.1.2. Stiffness Modulus Response

600

Deviatoric stress (kPa)

larger strain, the smaller multiple cracks appear all around the
specimen and it further increases with fibre content and length. At
high fibre content of 1%, no obvious single dominated failure plane
is found (Fig. 3f). The gradual increase in axial strain during
compression leads to network of tiny cracks that form progressive
failure zones with a barrelled failure shape of the specimen. Overall
in specimens with fibre reinforcement, the fibres confine the soil
particles and increase the global stability of the soil by arresting the
deformation behaviour.

0.25

0.5

0.75

10 mm
Peak
Stress
stress
ratio
(kPa)
200
1
230
1
276
1
312
1
243
1.21
279
1.21
333
1.2
372
1.19
285
1.42
324
1.41
382
1.38
414
1.32
319
1.59
354
1.54
395
1.43
435
1.39
282
1.41
318
1.38
371
1.34
396
1.27

20 mm
Peak
Stress
stress
ratio
(kPa)
200
1
230
1
276
1
312
1
272
1.36
310
1.34
370
1.34
396
1.27
318
1.59
357
1.55
411
1.49
428
1.37
374
1.87
420
1.83
467
1.69
497
1.59
338
1.69
386
1.68
432
1.56
468
1.5

30 mm
Peak
Stress
stress
ratio
(kPa)
200
1
230
1
276
1
312
1
261
1.3
297
1.29
346
1.25
380
1.21
298
1.49
338
1.47
389
1.41
417
1.34
339
1.69
367
1.59
404
1.46
436
1.4
286
1.43
330
1.43
346
1.25
380
1.21

Page 169

Patel and Singh / INDECS-15

350

600
500

fc = 1%

300

400
300

fc = 0.75%

fc = 0.5%

fc = 0.25%

No fibre

250
q (kPa)

Deviatoric stress (kPa)

(a)

L = 30 mm
L = 20 mm
L = 10 mm
No fibre

200
150
100

200

50

100

0
0

0
0

10

15

Axial strain (%)

Deviatoric stress (kPa)

L
(mm)
10

400
300

20

L = 30 mm
L = 20 mm
L = 10 mm
No fibre

30

0
5

10

400

500

600

700

Table 4. Shear strength parameters of fibre-reinforced soil

500

300

p (kPa)

(b)

100

200

Fig. 6. Failure envelope of fibre-reinforced soil (L = 20 mm)

600

200

100

20

15

20

fc
(%)
0.25
0.5
0.75
1
0.25
0.5
0.75
1
0.25
0.5
0.75
1

c (kPa)
67.6
82
100
118
103
95
120
141
124
90
109
132
114

Improvement
in cohesion
1.21
1.48
1.74
1.52
1.41
1.77
2.1
1.83
1.33
1.61
1.95
1.69

()
9
10
10.6
9.8
9.4
9.8
10
10.4
10.3
9.7
9.8
9.3
8.7

Improvement
in friction
1.11
1.17
1.09
1.04
1.09
1.11
1.15
1.14
1.07
1.09
1.03
0.97

Axial strain (%)


Fig. 5. Effect of fibre length (a) 3 = 100 kPa, fc = 0.75% (b) 3 = 400
kPa, fc = 0.75%

3.2.2. Shear Strength Parameters


The shear strength parameters (c and ) of the fibre-reinforced soil
have been calculated by plotting p-q plots, where {p = (1 + 3)/2
and q = (1 - 3)/2}. Fig. 6 depicts the failure envelope for 20 mm
long fibre at different fibre content. It can be noticed that the
envelopes are more or less parallel to each other. With addition of
fibres, the intercept of the failure envelope is shifting upwards. The
shear strength parameters determined from the failure envelopes
are summarized in Table 4. It can be noted that the cohesion
induced by fibres improves significantly up to 0.75% fibre content
for all fibre lengths and up to 20 mm fibre length at any particular
fibre content. The friction angle of soil increases slightly for all soil
fibre mix except for 30 mm fibre at 1% content, where it becomes
less than that of plain soil. Maximum improvement in cohesion (c)
parameter is 1.83 times (for 20 mm fibre at 0.75% content) and
friction angle () is 1.17 times (for 10 mm fibre at 0.5% content)
that of plain soil.

3.2.3. Stiffness Modulus Response


Typical stiffness modulus variation of fibre-reinforced soil showing
the effect of fibre content and fibre length is presented in Fig.7a
and Fig. 7b. Stiffness modulus is found to increase with fibre
content, fibre length and confining pressure. Initial stiffness
modulus at smaller strain is found to be higher and it further
decreases nonlinearly with strain indicating stiffness reduction with
strain increment. At higher strain stiffness variation almost become
parallel to each other and effect of fibre length and fibre content
become least significant. Nevertheless the stiffness of reinforced
soil is higher than that of plain soil for all strain levels studied.
Stiffness modulus increases with fibre dose and fibre length and
maximum improvement in stiffness is found for 0.75% fibre
content and 20 mm long fibre for all confining pressures.

Page 170

Patel and Singh / INDECS-15

35

10

(a)

fc = 1%

fc = 0.75%
25

fc = 0.5%

20

fc = 0.25%

EAC (MJ/m3)

Stiffness modulus (MPa)

30

No fibre
15
10

7
6
5
4

5
0
0

10

15

3
0.00

20

0.25

Axial strain (%)

35

0.50

0.75

L = 30 mm
L = 20 mm
L = 10 mm
No fibre

9
8

EAC (MJ/m3)

25

1.00

Fibre content (%)

10

(b)

30

Stiffness modulus (MPa)

(a)

L = 30 mm
L = 20 mm
L = 10 mm

20
15
10
5

(b)

400 kPa
300 kPa
200 kPa
100 kPa

7
6
5
4

0
0

10
Axial strain (%)

15

20

Fig. 7. Stiffness modulus response (3 = 100 kPa): effect of (a) fibre


content and (b) fibre length

3
0.00

0.25

0.50

0.75

1.00

Fibre content (%)


Fig. 8. Energy absorption capacity (EAC) response (a) effect of fibre
content, 3 =400 kPa (b) effect of confining pressure, L = 20 mm

3.3. CBR Tests


3.2.4. Energy Absorption Capacity
3.3.1. CBR Response
The improvement in toughness due to fibre inclusion can be
quantified by comparing the energy absorption capacity (EAC) for
the fibre-reinforced soil with that for the plain soil. The energy
absorption capacity is calculated by taking into consideration the
area under the deviatoric stress-strain curve up to axial strain of 20
%. Figs. 8a and Fig. 8b show the effect of fibre reinforcement
(fibre content and length) and confining pressure. It is observed
that more energy is needed to start deformation of soil and fibre at
higher confining pressure. The maximum increase in EAC is from
5.54 MJ/m3 for plain soil to 8.8 MJ/m3 for 20 mm long and 0.75%
fibre content reinforced specimen.

Fig. 9 presents the effect of fibre content on CBR values for 10 mm


and 20 mm long fibres calculated from load-penetration curves at
2.5 mm and 5 mm penetration depth. The CBR value
corresponding to 5 mm penetration is found to be maximum for all
soil-fibre combinations. After repetition of tests, the same response
was observed. Therefore the CBR value at 5 mm penetration is
considered for further study of strength behaviour. CBR values
calculated at 5 mm penetration are summarized in Table 6 for both
unsoaked and soaked conditions. The CBR values of plain soil are
8.05% and 2.74% for unsoaked and soaked conditions. With fibrereinforcement, the values have improved maximum up to 1.77 and
1.54 times for 20 mm and 0.75% fibre in unsoaked and soaked
conditions.

Page 171

Patel and Singh / INDECS-15

Table 6. Secant modulus of fibre-reinforced soil

24
L
(mm)

20

fc
(%)

CBR (%)

16
10

12
8

L = 20 mm_5 mm
L = 20 mm_2.5 mm
L = 10 mm_5 mm
L = 10 mm_2.5mm

4
0
0.00

0.25

0.50

0.75

20

1.00

Fibre content (%)

30

Fig.9 Effect of fibre content on CBR at different penetration depth

Table 6. CBR values of fibre-reinforced soil


L
(mm)

10

20

30

fc
(%)

0.25
0.5
0.75
1
0.25
0.5
0.75
1
0.25
0.5
0.75
1

Unsoaked
CBR
(%)
8.05
12.95
16.03
18.42
14.15
17.07
19.84
22.31
16.87
13.68
16.61
19.35
14.46

Improvement
(%)
60.86
99.13
128.82
75.77
112.30
146.46
177.14
109.56
69.93
106.33
140.37
79.62

0.25
0.5
0.75
1
0.25
0.5
0.75
1
0.25
0.5
0.75
1

Unsoaked
Secant
modulus
(MPa)
19.63
31.54
40.93
46.63
36.9
41.01
44.2
53.51
41.09
28.18
39.92
46.47
30.36

Improvement
(%)
60.67
108.51
137.54
87.98
108.91
125.16
172.59
109.32
43.55
103.36
136.73
54.66

Soaked
Secant
modulus
(MPa)
6.68
6.69
7.2
9.43
8.4
12.43
15.51
16.61
12.77
8.48
10.11
11.83
11.14

Improvement
(%)
0.15
7.78
41.17
25.75
86.01
132.18
148.65
91.17
26.95
51.35
77.09
66.76

Soaked
CBR
(%)
2.74
3.01
3.32
3.87
3.45
5.11
6.81
6.97
5.24
3.48
4.15
4.85
4.57

Improvement
(%)
9.85
21.17
41.24
25.91
86.49
148.54
154.38
91.24
27
51.46
77
66.79

4. Conclusion
Based on the results obtained from the present study, the following
conclusions can be drawn:
1.

2.

3.
3.3.2. Secant Modulus Response
4.
Secant modulus calculated at 5 mm penetration depth is shown in
Table 7. The modulus of soil has improved significantly with fibre
inclusion. Improvement of modulus is more pronounced with fibre
length and content under both unsoaked and soaked conditions up
to a limit of 20 mm length and 0.75% content. Improvement lies in
range of 43 to 172% for unsoaked soil and 0.15 to 148% for soaked
soil. Maximum improvement of 172% and 148% is for 20 mm long
fibre of 0.75% content for unsoaked and soaked conditions.
Improvement in unsoaked condition is almost 2.7 times (19.63
MPa to 53.51 MPa) and 2.5 times for soaked specimen (6.68 MPa
to 16.61 MPa). Improvement for 10 mm fibre under both condition
is smaller compare to other fibre length with very small
improvement under soaked condition for lower percentage (0.25 to
0.5%) it is below 10% indicating almost no improvement.

5.

Glass fibre reinforcement improves the deformation behaviour


of the cohesive soil by increasing failure strain and changes
the brittle behaviour of parent soil to more ductile behaviour
by arresting the failure pattern. The maximum improvement in
unconfined compressive strength is 141.5% for 20 mm long
fibre of 0.75% content.
Stiffness modulus of reinforced soil at failure strain level of
plain soil improves from 5.18 MPa for unreinforced soil to
8.16 MPa for soil reinforced with 20 mm fibre of 0.75%
content.
Cohesion is induced by the fibres and maximum improvement
is for 0.75% fibre for all fibre length used. Friction angle of
soil improves marginally up to 0.5% fibre content.
CBR value of soaked soil is reduced by almost by one third
than that of unsoaked soil.
Fibre-reinforced soil shows greater CBR values with
maximum improvement of 177% and 154% for 20 mm fibre
of 0.75% content in unsoaked and soaked cases.

References
[1] Waldron LJ. The shear resistance of root-permeated
homogeneous and stratified soil. Soil Science Society of
America Proceedings, 1977; 41: 843-849.
[2]

Freitag DR. Soil randomly reinforced with fibres. Journal of


Geotechnical Engineering, 1986; 112, 8: 823-826.

Page 172

Patel and Singh / INDECS-15

[3] Al Wahab RM, El-Kedrah MM. Using fibres to reduce tension


cracks and shrink/swell in compacted clays. Geoenvironment
2000, Geotechnical Special Publication No. 46, Y. B. Acar and
D. E. Daniel (eds.), ASCE, Vol. 1, 791805.
[4] Nataraj MS, McManis KL. Strength and deformation properties
of soils reinforced with fibrillated fibres. Geosynthetic
International Journal, 1997; 4, 1: 65-79.

[10] Estabragh AR, Bordbar AT, Javadi AA. Mechanical behaviour


of a clay soil reinforced with nylon fibres. Journal of Geotech
and Geological Engineering, 2011; 29: 899-908.
[11] Ozkul ZH, Baykal G. Shear strength of clay with rubber fibre
inclusions, Geosynthetics International, 2006; 13, 5: 173-180.

[5] Maher MH, Ho YC, Mechanical properties of kaolinite/fibre


soil composite. Journal of Geotechnical Engineering, 1994;
120, 8: 1381-1393.
[6] Zeigler S, Leshchinsky D, Ling HI, Perry EB. Effect of short
polymeric fibres on crack development in clays. Soils and
Foundations, 1998; 38,1: 247253.
[7] Hazirbaba K, Gullu H. California Bearing Ratio improvement
and freezethaw performance of fine-grained soils treated with
geofibre and synthetic fluid. Cold Regions Science and
Technology, 2010; 63: 5060.
[8]. Prabakar J, Sridhar RS. Effect of random inclusion of sisal fibre
on strength behaviour of soil. Construction and Building
Materials, 2010; 16, 123-131.
[9] Sivakumar Babu GL, Vasudevan AK. Strength and stiffness
response of coir fibre-reinforced tropical soil. Journal of
Materials in Civil Engineering, 2008; 20, 9: 571-577.

Page 173

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

BEHAVIOR OF CIRCULAR FOOTING RESTING ON REINFORCED SAND


Sangameshwar Patil1, P.G Rakaraddi2
1 Post -Graduate student, Department of Geotechnical engineering, Basaveshwar Engineering College,
Bagalkot 587102 Karnataka, India, sangameshwarpatil87@gmail.com
2 Professor in civil dept, Basaveshwar Engineering College, Bagalkot 587102 Karnataka, India,
pgraddi@yahoo.co.in

Abstract
In this investigation series of laboratory model test has been carried out to evaluate the response of circular footing
resting on plastic bottle reinforcement. These days disposal of plastic waste is a major issue and which leads to harmful
effect on environment. To avoid disposal problem and to improve the geotechnical properties of the soil these waste
products can be utilized. In this project plastic bottles are used as reinforcing materials to improve the load carrying
capacity of soil. Series of model test is carried out on circular footing with waste bottles as reinforcement. Bottle
reinforcement is used with different L/D ratio, number of layers and change in spacing of reinforcement layers. Test is
carried out for three different densities of sand (=16.48kN/m3, 17.07kN/m3 and 17.26kN/m3). The test result has shown
that providing reinforcement below the footing increase in the bearing capacity of soil. The effective utilization of bottle
reinforcement, the optimum depth should be (U/B=0.35) which is found to be good from the results and the foundation
soil should be in higher density.
Keywords: Sand, Ultimate load, Reinforcement, U/B ratio, L/D ratio

1. Introduction
Super Structure which is built on soil should resist the load
coming on the soil. In India some of the areas have low
bearing capacity which cannot carry load coming from the
super structure. Nowadays availability of land is less and
which leads to increase in the cost of the land as well as cost
of the structure. To overcome this problem either soil should
be replaced or foundation should be well designed. It is
difficult to replace the existing soil, so engineers should
design the foundation in such a way that it should resist the
load coming on it. Different types foundation are used
square footing, circular footing, strip footing. These
foundations undergo large unequal settlement in low bearing
areas, for this raft foundation is preferred which will avoid
the unequal settlement of the structure. In raft foundation
disadvantages are there as huge area has to excavate and
should provide bracing all over the area which will avoid the
caving of the soil. This method leads to increase in the cost
of construction. So to reduce the cost of the construction
reinforcement is provided below the footing within the
influencing depth which will reduce the settlement of the
structure as well as reduce the cost the construction.
Reinforcement which can be provided below the footing is,
geogrid, geocell, confinement etc., which are available in
market which are used as reinforcing materials. These

materials are costly which are available in market. So in this


investigation to avoid that waste plastic bottles are used as
reinforcing materials. These days plastic bags and plastic
materials causing environmental problem. Use of these
materials as a reinforcing material environmental problems
can be avoided and as well as cost also reduces.
In this investigation behavior of circular footing resting on
reinforced sand is evaluated through model plate load test.
More than 40 series of tests are carried out on varying the
parameters. The parameters varied in these tests are density
of soil, U/B ratio of first layer reinforcement, L/D ratio of
reinforcement, number of layers and spacing between the
layers. From the test results shown that the use of plastic
bottle as reinforcement increases in the load carrying
capacity of soil.
2. Literature Review
Research work is carried out on using reinforcement below
the footing. Some of the researchers carried out are reviewed
in this investigation they are as follows, Sujit Kumar et.al
(2001) carried out test on strip footing geocell as
reinforcement. In this some of the parameters varied are
pocket size, height and width of geocell mattress. From the
results concluded that increase in the bearing capacity of soil
by 8 times than the unreinforced case, Al- Aghbari (2007)
carried out test on circular footing with structural skirts
below the footing. The parameter varied in the test is
Page 174

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

structural depth ratio which results shown that increases in


the bearing capacity, Gupta et al. (2014), Chandrawanshi et
al. (2014), Gupta et al. (2014) they carried the series of tests
on confinement below the footing which leads to increase in
the bearing capacity of soil.
3. Methodology
For the present study the sand is used from the Malaprabha
river, Bagalkot district, Karnataka, is used as soil medium.
Properties of sand is determined as per IS 2720 and same is
presented in table 1. As per IS 2720 part 4 soil classified as
poorly graded sand which lies in the range of Cu<6 and
Cc1 hence soil is classified as poorly graded sand. The
model footing used in this investigation is circular of having
diameter of 11.30 cm and thickness 10 mm. Footing is
placed on the above sand bed for different densities which
is filled in the wooden box of size 600mm x 600mm x
600mm by raining technique method. Densities are
predetermined by varying height of fall in the laboratory and
which is presented in table 2. Table 2 shows the value for
different densities corresponding to angle of internal
friction. Reinforcement is used in this investigation is waste
plastic bottles which were collected from local areas and
which were cut for different L/D ratios which is presented in
table 3. These cut plastic bottles were joined together using
the tie wire and placed below the footing at different depth.
First reinforcement layer is determined and which is placed
at 0.35 U/B ratios which is given in table 4, while placing
the second layer and third layer the spacing between the
layers is varied are (i) zero (ii) 4cm (iii) 6cm and (iv) 8cm.
The test is carried out on the experimental setup as shown in
fig.2. Vertical loading on the footing is applied with the help
of screw jack of 5 tonne capacity is load is measured using
proving ring of capacity 1000 kg and displacement is
measured by LVDT which is placed diagonally on the
footing. LVDT capable of measuring 0.01mm.

Fig 2.Line diagram of experimental setup


Table 1.Properties of sand used in the study
Characteristics

Value

IS classification

SP

D10

0.39

D30

0.64

D60

1.05

Uniformity of coefficient (Cu)

2.69

Coefficient of curvature (Cc)

Minimum Void ratio

0.46

Maximum void ratio

0.73

Specific gravity

2.59

Minimum dry density

14.61 kN/m3

Maximum dry density

17.36 kN/m3

Table 2. Density and the associated heights of fall in the raining


technique with angle of internal friction for the sand used in the
present study.
Height of

Density (kN/m3)

Angle of internal
friction()

fall(cm)
15

16.48

330

30

17.07

360

45

17.26

370

Fig 1. Particle size distribution characteristics of sand used in the study

Page 175

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

4.2 Reinforced
Static tests are carried for circular footing on reinforced
case. Initially tests were carried to determine the optimum
depth of first reinforcement layer (U) which is as follows,

Fig 3. Layout and configuration three dimensional reinforcement


layers in the test.

For circular footing reinforced with waste plastic bottles the


load settlement curve is plotted for the density 16.48kN/m 3
is shown in fig 4 for different U/B ratio 0.30, 0.35, 0.40 and
0.45 respectively. As per IS (1888/1982 [15]) ultimate load
carrying capacity is determined by double tangent method
and also for densities 17.07kN/m3 and 17.26kN/m3 is
exclusively shown in table 4.

Table 3. Different L/D ratio of reinforcement


Diameter of

Length of

L/D

bottles

bottles

ratio

(D)

(L)

75mm

15mm

0.20

75mm

20mm

0.26

75mm

25mm

0.33

75mm

30mm

0.40

Sl.no

The U/B ratio corresponding to depth first reinforcement


was given in table 4.
Table 4. U/B ratio corresponding first reinforcement depth
U/B ratio

First reinforcement Depth


(U)

0.30
0.35

30 mm
35 mm

0.40

40 mm

0.45

45 mm

4. Results and Discussion

Fig.4. Load - Settlement curves for L/D ratio 0.2 for different U/B ratio
Table 4. Ultimate bearing capacity of reinforced sand for different
densities and U/B ratio.
Ultimate Load (kN) for the density of
U/B
ratio
0.30
0.35
0.40

16.48
kN/m3
0.42
0.68
0.50

17.07
kN/m3
0.57
0.80
0.55

17.26
kN/m3
0.84
1.00
0.80

0.45

0.47

0.50

0.77

Tests were carried out on reinforced and unreinforced soil.


Tests results are discussed as follows,

4.1 Unreinforced
Tests are carried out on circular footing without
reinforcement and the results obtained for the densities
16.48kN/m3, 17.07kN/m3 and 17.26 kN/m3 are 0.20 kN,
0.38 kN and 0.70 kN.

Increase in the density of soil increase in the ultimate load


carrying capacity of soil with the provision of reinforcement
below the footing. From the fig 4 and table 4 the ultimate
load carrying capacity values for U/B 0.35 for the densities
16.48 kN/m3, 17.07 kN/m3 and 17.26 kN/m3 are 0.68 kN,
0.80 and 1.00 kN respectively. The values obtained are
higher than the U/B ratio 0.30, 0.40 and 0.45, hence U/B
ratio 0.35 is optimum in this investigation.

Page 176

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

(a)

(c)

(b)

(d)

Fig 5. Load - Settlement curves for the density of 16.48 kN/m3 and L/D ratio (a) 0.20 (b) 0.26 (c) 0.33 and (d) 0.40

Figure 5 shows load-settlement curve for the density 16.48 kN/m3 for different reinforcement layers and for L/D ratios
0.20, 0.26, 0.33 and 0.40 shown in figure (a), (b), (c) and (d) respectively.

Page 177

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

(a)

(c)

(b)

(d)

Fig 6.Load - Settlement curves for density of 17.07 kN/m3 and L/D ratio (a) 0.20 (b) 0.26 (c) 0.33 and (d) 0.40

Figure 6 shows load-settlement curve for the density 17.07 kN/m3 for different reinforcement layers and for L/D ratios 0.20,
0.26, 0.33 and 0.40 shown in figure (a), (b), (c) and (d) respectively.

Page 178

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

(b)

(a)

(c)

(d)
3

Fig-7: Load - Settlement curves for the density of 17.26kN/m and L/D ratio (a) 0.20 (b) 0.26 (c) 0.33 and (d) 0.40

Figure 7 shows load-settlement curve for the density 17.26


kN/m3 for different reinforcement layers and for L/D ratios
0.20, 0.26, 0.33 and 0.40 shown in figure (a), (b), (c) and (d)
respectively.
From the figure 5, 6 and 7 the ultimate load carrying
capacity values are given in the table 5. Fig 5, 6 and 7
represents for the densities 16.48 kN/m3, 17.07 kN/m3 and
17.26 kN/m3 for L/D ratios 0.20, 0.26, 0.33 and 0.40 for one,
two and three layers respectively. From the figures 5, 6 and
7 and table 5 the values obtained for the density 16.48 for
L/D ratios 0.20,0.26,0.33 and 0.40 with one, two and three
layer are (i) 0.68,0.90, and 1.00 kN (ii) 0.80, 1.03 and 1.12
kN (iii)1.04, 1.32 and 1.59 kN and (iv) 0.72, 0.84 and 0.98
kN, for the density 17.07 kN/m3 (i) 0.80, 1.03 and 1.26 kN

(ii) 1.09, 1.24 and 1.42 kN (iii) 1.38, 1.51 and 1.80 kN and
(iv) 0.90, 1.12 and 1.27 kN, for the density 17.26 kN/m3 (i)
1.00, 1.19 and 1.48 kN (ii) 1.30, 1.60 and 1.80 kN (iii) 1.58,
1.79 and 2.21 kN and (iv) 1.00, 1.38 and 1.54 kN
respectively. From the values obtained it can be observed
that increase in the load carrying capacity for the L/D ratio
0.33 and which is for three layers and for L/D ratios 0.20,
0.26 and 0.40 decrease in the load carrying capacity. The
optimum value obtained for the density 17.26 kN/m3 for L/D
ratio 0.33 and for three layers is maximum i.e 2.21 kN and
the remaining values are given in table 5. For this series of
test second layer is placed at a depth of 5.5 cm from the
topmost layer and similarly third layer is placed at a depth of
5 cm from the second layers respectively.

Page 179

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

Table 5. Load carrying capacity values for various cases


Density (kN/m3)

L/D ratio
Unreinforced

Number of layers
1

Ultimate load (kN)


0.20
0.68

0.20

0.90

1.00

0.80

2
3
1
2
3

1.03
1.12
1.04
1.32
1.59

1
2
3

0.72
0.84
0.98

1
2
3
1
2
3
1
2
3

0.38
0.80
1.03
1.26
1.09
1.24
1.42
1.38
1.51
1.80

1
2
3
1

0.90
1.12
1.27
0.70
1.00

2
3
1
2
3
1
2
3
1
2
3

1.19
1.48
1.30
1.60
1.80
1.58
1.79
2.21
1.00
1.38
1.54

0.26
16.48

0.33

0.40
Unreinforced
0.20

17.07

0.26

0.33

0.40
Unreinforced
0.20

17.26

0.26

0.33

0.40

4.3 Variation in Spacing of Reinforcement


From the fig 5, 6 and 7 and table 5 the values obtained for
the reinforcement layer placed at the constant depth
throughout the test which is at 5.5 cm, hence the placing of
reinforcement layer which will effect on the load carrying
capacity, so for this case some tests are carried out by
varying the depth of reinforcement layers for the densities

16.48 kN/m3, 17.07 kN/m3 and 17.26 kN/m3 for the


optimum L/D ratio 0.33 respectively. Topmost layer kept
constant depth at 3.5cm from the surface, the depth of
second and third layer varied with (i) Zero (ii) 4cm (iii) 6 cm
and (iv) 8cm respectively.

Page 180

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

(a)

(b)

Fig 8. Load - Settlement curves for the density 16.48 kN/m for (a) Two and (b) Three layers with different spacings (i) Zero (ii) 4 cm (iii) 6 cm and
(iv) 8 cm
3

Fig 9. Load - Settlement curves for the density 17.07 kN/m3 for (a) Two and (b) Three layers with different spacings (i) Zero (ii) 4 cm (iii) 6 cm and
(iv) 8 cm

Page 181

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

(b)

(a)

Fig 10. Load - Settlement curves for the density 17.07 kN/m3 for (a) Two and (b) Three layers with different spacings (i) Zero (ii) 4
cm (iii) 6 cm and (iv) 8 cm

Table 6. Load carrying capacity values for different spacing, L/D ratio 0.33 and for two and three layers
Density (kN/m3)

L/D ratio

Number of layers (N)

2
16.48

0.33

17.07

0.33
3

17.26

0.33
3

Spacing (S)

Ultimate Load
(kN)

0
4
6
8

1.10
1.20
1.40
1.08

1.28

1.40

1.62

1.10

0
4
6

1.28
1.46
1.60

1.18

0
4
6

1.39
1.54
1.88

8
0

1.20
1.41

1.60

6
8
0

1.90
1.30
1.51

1.80

6
8

2.24
1.38

Page 182

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15


Fig 8, 9 and 10 shows the load settlement curves for the densities 16.48 kN/m3, 17.07 kN/m3 and 17.26 kN/m3 for two
and three layers with varying the spacing of the reinforcement i.e. (i) Zero, (ii) 4 cm, (iii) 6cm and (iv) 8 cm and the
ultimate values obtained from the fig 8, 9 and 10 is represented in Table 6 respectively. The load carrying capacity of
soil increases with increase in the spacing of reinforcement from zero, 4cm, 6cm and 8cm. the optimum values obtained
for the spacing 6 cm for the densities 16.48 kN/m3, 17.07 kN/m3 and 17.26 kN/m3 for two and three layers are 1.40 and
1.62 kN, 1.60 and 1.88 kN and 1.90 and 2.24 kN respectively.
5. Conclusion
The behavior of circular footing resting on reinforced
sand has been investigated by using model circular
footing. The results shown that the load carrying
capacity of circular footing increases with provision of
reinforcement below the footing. increase in the load
carrying capacity of footing is also depend on placing of
reinforcement. The optimum results obtained for the
density 17.26 kN/m3 for L/D ratio 0.33 for three layers
and depth of reinforcement is constant. for optimum L/D
ratio the varying the spacing of reinforcement given
better results for the depth 6 cm. the optimum usage of
reinforcement depth is 6cm and the density 17.26 kN/m3
respectively.

[6] Moghaddas Tafreshi, S.N and Dawson, A.R. (2010)


Comparison of bearing capacity of a strip footing on
sand with geocell and with planar forms of geotextile
reinforcement", Geotextile and Geomembranes, Vol.28,
pp.72-84.
[7] Kumar.K.V.S.P, Venkata Koteswara Rao Pasupuleti
and Satish Kumar Kolluru. (2012). Bearing Capacity Of
Square Footing on Geocell Sand Mattress Overlying
Clay Bed. International Journal of Emerging trends in
Engineering and Development, Issue 2, Vol.5, pp, 563-

References

[1] Dash, S. K., Krishnaswamy N.R and Rajagopal K.


(2001).Bearing capacity of strip footings supported on
geocell

reinforced

sand,

Geotextiles

and

Geomembranes, Vol.19, pp.235-236.


[2] Dash, S. K., Krishnaswamy N.R and Rajagopal K.
(2003). Model studies on circular footing supported on
geocell reinforced sand underlain by soft clay,
Geotextiles and Geomembranes, Vol.21, pp.197-219.
[3] Al-Aghbari, M.Y. (2007). Settlement of Shallow
Circular Foundations with Structural Skirts Resting on
Sand, The Journal of Engineering Research Vol. 4,
No.1 (2007) 11-16.
[4] Sireesh, S, Sitharam, T.G and Dash,S.K., (2009).
Bearing capacity of circular footing on geocell sand
mattress overlying clay bed with void, Geotextiles and
Geomembranes, Vol.27, pp.89-98.
[5] Gupta, R and Trivedi, A. (2009). Bearing capacity
and settlement of Footing Resting on Confined Loose
Silty Sands. Electronic Journal of Geotechnical
Engineering, Vol.14, pp.1-14.

573.
[8] Krishna. A, Viswanath B and Keshav Nikita (2014).
Performance of Square Footing Resting On Laterally
Confined Sand International Journal of Research in
Engineering and Technology, Vol.03, pp, 110-114.
[9] Ravi Gupta, Rakesh Kumar and P.K. Jain (2014).
Behavior of Circular Footing Resting On Two
Dimensionally

And

Three

Dimensionally

Skirted

Foundations In Medium Dense Sand International


Journal of Advanced Engineering Research and Studies,
Vol.3 pp, 45-47.
[10] Ravi Gupta, Rakesh Kumar and P.K. Jain (2014).
Behavior of Circular Footing Resting On Three
Dimensional Confined Sand, International Journal of
Advanced Engineering Technology, Vol.5, Issue 2, pp,
11-13.
[11] Sareesh Chandrawanshi, Rakesh Kumar, Dr. Suneet
Kaur and Dr. P.K. Jain, (2014), Effect Of Skirt on
Pressure Settlement Behavior of Model Circular Footing
In Medium Dense Sand,

International Journal of

Page 183

Sangameshwar Patil and P.G.Rakaraddi / INDECS-15

Advanced

Engineering

Technology,

Vol.

5,Issue

2,pp,01-05.

[14] Indian Standard - IS: 2720 (Part 13) 1986,


Method of test for soils, direct shear test, New Delhi,
India.

[12] Indian Standard - IS: 2720 (Part 3/See Indian


Standard - IS: 2720 (Part 3/See 2) 1980, Methods of

[15] Indian Standard - IS: 1888 1982, Method of load

test for soils, determination of specific gravity, fine,

test on soils, New Delhi, India

medium and coarse grained soils, New Delhi, India.


[13] Indian Standard - IS: 2720 (Part 4) 1985, Method
of test for soils, grain size analysis-mechanical method,
New Delhi, India.

Page 184

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SEISMIC EVALUATION AND RETROFITTING OF REINFORCED


CONCRETE BUILDINGS
Dr.A.Vijayakumar1* Dr.D.L.Venkatesh Babu2 Dr.A.R.Santhakumar3
1

Professor and HOD of Civil Engineering, Coimbatore Institute of Engineering and


Technology, Coimbatore,641109, Tamilnadu, India,vijayakumarciet@gmail.com
2
Professor and HOD of Civil Engineering, JSS Academy of Technical Education,
JSS Campus, Uttarahalli, Bangalore, 560060, India, drdlvbabu@gmail.com
3
Former, Dean of Civil Engineering, Anna University, Chennai and Former Emeritus
Professor, I.I.T.Madras,Tamilnadu,600025, santhaar@gmail.com,India.

Abstract
It is known a fact, that the globe is facing a serious threat of natural disasters from time to time. With particular
record to earthquake reoccurrence, the consequences are loss of human lives and destruction of properties, which
ultimately affects the natural economy. As the occurrence of an earthquake cannot be predicted and prevented,
preparedness of the structures to resist earthquake forces becomes more important. Keeping the view of constant
revision of the seismic zones in India, lack of proper design and detailing of structures against earthquake, this paper
aims to evaluate the performance of a typical selected R.C. building with respect to seismic vulnerability. For this
seismic evaluation, a pushover analysis has been performed by software SAP 2000. The analysis results showed the
performance levels, behaviour of the components and failure mechanism of the building. It also showed the sequence of
hinge formation. Based on the analysis the elements which needed retrofitting were identified. The deficiency of the
member was strengthened with varying configuration and layers of externally wrapped E- Glass Fiber Reinforced
Polymer sheets. More particularly, the effect of the number of GFRP layers and its orientation were investigated in
detail.

Keywords: RC building, Seismic Evaluation, Pushover Analysis, Performance level, GFRP.

1. Introduction
Coimbatore is a fast growing city in India
which is located in seismic Zone-III. Many reinforced
concrete frame buildings in Coimbatore were designed
and built prior to 2002. The seismic code IS 1893 was
revised in 2002. Hence, buildings built prior to 2002 do
not comply with the codal requirement. Further, some of
the Coimbatore buildings built OVER the past few years
even after 2002 are seismically deficient because of lack
of awareness of the seismic behavior of structures. Most
of the existing buildings in this city are designed for
gravity loads only. Most of the buildings which have
infilled walls have not considered infills in their design.
In Coimbatore many of the existing buildings
are with masonry infills as non-structural element. The
analyses as well as design of the frames are carried out
by considering the mass but neglecting the strength and
stiffness contribution of infill. Therefore, the entire
lateral load is assumed to be resisted by the frame only.
A large number of existing buildings in Coimbatore need
seismic evaluation due to various reasons such as

noncompliance with the codal requirements, updating of


code, poor design practice and change the use of the
building. However, the existing deficient structures in
Coimbatore, which is in Zone III, can be provided with
some rehabilitation to sustain the expected performance
level.
The reports ATC 40 [1] and FEMA 273 [2]
highlight the non-linear static pushover analysis. It is an
efficient method for the performance evaluation of a
structure subjected to seismic loads. The step by step
procedure of the pushover analysis was to determine the
capacity curve, demand curve and performance point.
These reports deal with modeling aspects of the hinge
behavior, acceptance criteria and procedures to locate the
performance point. Mehmet Inel and Hayri Baytan
Ozmen [3] point out that the effect of plastic hinges in
nonlinear analysis of reinforced concrete buildings.
Analysis was done for four and seven storey reinforced
concrete buildings to represent low and medium rise
buildings. The frames were modeled with default and
user defined hinge properties to study possible
differences in the results of pushover analysis.
Comparison of response was also made in terms of base
Page 185

A.Vijayakumar et. al. / INDECS-15

shear capacity, displacement capacity and deformation of


hinges. User defined plastic model was found to be
effective than the default hinge model. From the above
discussion it was concluded that comparison studies are
made between pushover analysis and inelastic time
history analysis in evaluating the performance of
existing building.
FEMA 172 [4] this handbook described
techniques that engineers can use to solve a variety of
seismic rehabilitation problems in existing buildings, a
broad spectrum of building types and building
components (both structural and nonstructural).
Techniques are illustrated with sketches, and the relative
merits of the techniques are also discussed.
Fig. 1. Plan of the Existing Hostel Building

Shailesh Agrawal and Ajay Chourasia [5] the


nonlinear static analysis of RC building was performed
using pushover approach before and after retrofitting.
The comparison of strength parameters and pushover
curve indicates that there was increase in ductility. As
regards to stiffness of the building, it is seen that it
remains more or less same up to linear stage, while in
nonlinear stage every point increase after retrofitting.
The strength of the building was correlated with base
shear, the net enhancement in strength after retrofitting.
Amar Prakash and Thakkar [6] the present
paper deals with the seismic retrofitting of an existing
fourteen storied RC building frame located in seismic
zone IV. The study included seismic evaluation and
retrofitting of RC framed building, by using steel bracing
and infill masonry walls. The seismic performance of
two retrofitting techniques such as steel bracing (V,
diamond and cross pattern) and infill walls were
relatively compared. Among three patterns of steel
bracing, cross pattern shows better performance than V
and diamond bracing patterns.
2. Description of the Frame Structures
The present study was to evaluate the behavior of
G+2 reinforced concrete frame building subjected to
zone III level earthquake forces. The three dimensional
reinforced concrete structures were analyzed by
nonlinear static analysis (Pushover Analysis) using
SAP2000 software. The analysis results showed the
performance levels, behavior of the components and
failure mechanism of the building. It also showed the
sequence of hinge formation. Based on the analysis, the
elements which need retrofitting were identified.
Figure 1 shows the plan of the building representing
the X and Y direction used for analysis. The building
exists in Coimbatore and is a typical example of many
such buildings in this region. It was built with M15
concrete and Fe415 steel.

IS 456:2000 demands M20 to be the concrete


grade for structural applications. However, many
buildings in Coimbatore region were built with only
M15 grade concrete. In order to check the vulnerability
of such buildings, the strength of concrete and grade of
steel based on test results was used in the pushover
analysis.
For the selected building the strength of
concrete was assessed as 16.5 N/mm2 using rebound
hammer test and strength of steel was assessed as Fe420
N/mm2 based on tensile strength test .These values were
used in the pushover analysis. The chosen G+ 2 existing
hostel building and total height of the building is 8.79 m.
The building is 25.41 m in length and 11.95 m in width.
Typical floor to floor height is 2.93 m and slab thickness
is 140 mm.
With reference to, Mohamed Nour El-Din AbdAlla [7] beam and column members had been defined as
frame elements with the appropriate dimensions and
reinforcements. Slabs were defined as an area element
having the properties of shell elements with required
thickness. Slabs had been modeled as rigid diaphragms
based on Sundar and Elangovan [8]. Soil structure
interaction was not considered and columns have been
restrained in all six degrees of freedom at the base. The
building was constructed in the medium soil site
conditions.
3. Pushover Analysis
The non-linear static procedure or simply push
over analysis is a simple option for estimating the
strength capacity in the post-elastic range. This
procedure involves applying a predefined lateral load
pattern which is distributed along the building height.
The lateral forces are then monotonically increased in
constant proportion with a displacement control node of
the building until a certain level of deformation is
reached. The applied base shear and the associated
lateral displacement at each load increment are plotted.
Page 186

A.Vijayakumar et. al. / INDECS-15

Based on the capacity curve, a target displacement which


is an estimate of the displacement that the design
earthquake will produce on the building is determined.
The extent of damage experienced by the building at this
target displacement is considered as representative of the
damage experienced by the building when subjected to
design level ground shaking.

analysis was 2924 kN at a displacement of 0.030 m in Y


direction. In Y direction the base shear was more
compared to X direction because of configuration of the
columns in the frames.

4. Nonlinear Plastic Hinges Properties


This requires the development of the force deformation curve for the critical sections of beams and
columns by using the guidelines FEMA 356 [9]. The
force deformation curves in flexure were obtained from
the reinforcement details and were assigned for all the
beams and columns. The Nonlinear properties of beams
and columns have been evaluated using the section
designer and have been assigned to the computer model
in SAP2000. The flexural default hinges (M3) and shear
hinges (V2) were assigned to the beams at two ends. The
interacting (P-M2-M3) hinges assigned for all the
columns at upper and lower ends.
It is the plot of the lateral force (V) on a
structure, against the lateral deflection (d), of the roof of
the structure. This is often referred to as the push over
curve. Performance point and location of hinges in
various stages can be obtained from pushover curves as
shown in the Figure 2. Five points labeled A, B, C, D
and E are used to define the force deflection behavior of
the hinge. The points labeled A to B Elastic state, B to
IO- below immediate occupancy, IO to LS between
immediate occupancy and life safety, LS to CP- between
life safety to collapse prevention, CP to C between
collapse prevention and ultimate capacity, C to Dbetween ultimate capacity and residual strength, D to Ebetween residual strength and collapse and greater than
E collapse.

Fig. 3. Pushover Curve in X- Direction

Fig. 4. Pushover Curve in Y- Direction

5.2 Capacity Spectrum and Building Performance


Level
The bare frame performance point was obtained
at a base shear level of 1016.436 kN and displacement of
0.015 m in the X direction as shown in Figure 5. In this
performance point the structure reached immediate
occupancy level as shown in Figure 6. The performance
point was obtained at a base shear level of 1,330.743 kN
and displacement of 0.0044 m in the Y- direction as
shown in Figure 7. In this performance point the
structure reached collapse prevention level as shown in
Figure 8. Therefore, it was seen that the structure has
more critical in the Y direction compared to the X
direction as expected.

Fig. 2. Different Stages of Plastic Hinge

5. Analysis Results and Discussions


5.1 Pushover Curve
Pushover curves for the building in X and Y
direction are shown in Figure 3 and Figure 4. These
curves depict the global behavior of the frame in terms
of its stiffness and ductility. For bare frame base shear
from pushover analysis was 1288 kN at a displacement
of 0.180 m in X direction. The base shear from pushover
Page 187

A.Vijayakumar et. al. / INDECS-15

Fig. 5. Capacity Spectrum in X-Direction

Fig. 8. Hinges Formation in Y- Direction at


Performance Level

5.3 Plastic Hinge formation


Plastic hinge formation of the building was
observed in X and Y direction obtained at different
displacements levels. Plastic hinge formation starts with
beam ends and later proceeds to base of columns of
lower stories then propagates to upper stories and
continues with the yielding of interior intermediate
beams. However, yielding occurs at designated events
such as B (yielding), IO (Immediate occupancy), LS
(Life safety) and last hinge CP (collapse prevention)
respectively, the amount of damage in this direction was
expected be limited based on the predicted displacements
as below.
Fig. 6. Hinges Formation in X- Direction at
Performance Level

Fig. 7. Capacity Spectrum in Y-Direction

In X direction the first hinge formation occurs


at a load of 664.45 kN and at a displacement of 0.007 m.
The moment and rotation value for point B (Yielding)
were 1.5767 kNm and 0.0017 radians respectively. The
moment and rotation value for point IO (Immediate
occupancy) were 1.5849 kNm and 0.00254 radians
respectively. The moment and rotation value for point
CP (collapse prevention) were 1.9344 kNm and 0.0356
radians respectively.
Plastic hinge formation of the building was
observed in Y direction obtained at different
displacements levels. Plastic hinge formation starts with
beam ends and later proceeds to base of columns of
lower stories then propagates to upper stories and
continues with the yielding of interior intermediate
beams. However, yielding occurs at designated events
such as B (yielding), IO (Immediate occupancy), LS
(Life safety) and last hinge CP (collapse prevention)
respectively, the amount of damage in this direction was
Page 188

A.Vijayakumar et. al. / INDECS-15

expected be limited based on the predicted displacements


as below.

be strengthened by using locally practiced strengthening


techniques, before an earthquake.

The first hinge formation occurs at a load of


664.45 kN and at a displacement of 0.007 m. The
moment and rotation value for point B (Yielding) were
1.5767 kNm and 0.0017 radians respectively. The
moment and rotation value for point IO (Immediate
occupancy) were 1.5849 kNm and 0.00254 radians
respectively. The moment and rotation value for point
CP (collapse prevention) were 1.9344 kNm and 0.0356
radians respectively.

The member-level retrofit approach can be used


for the seismic strengthening of the existing RC
structure. The member level retrofit approaches include
the Fiber Reinforced Polymer (FRP) wrapping
techniques which are used for strengthening of the
existing building.

Based on the analysis result, it is concluded


that, earthquake occurs in Coimbatore region which is
located in Zone-III. This existing hostel building will
first yield in the Y direction because of weak beams. The
base shear capacity was low in the Y direction as
compared to the X direction for first hinge formation
levels and also at the performance point level this
building reached immediate occupancy level in the X
direction and collapse prevention level in the Y
direction. It is seen that this building is more critical in
the Y direction compared to the X direction as expected.
Therefore weak beams in the Y-direction must be
strengthened by using any one of the strengthening
techniques, before an earthquake.
6. Seismic Retrofitting on Existing Rcc
Building
Many existing structures located in Coimbatore
are inadequate based on current seismic design codes. In
addition, a number of major earthquakes during recent
years underlined the importance of mitigation to reduce
seismic risk. Seismic retrofitting of existing structures is
one of the most effective methods of reducing this risk.
In recent years, a significant amount of research has been
devoted to the study of various strengthening techniques
to enhance the seismic performance of RC structures.
However, the seismic performance of the structure may
not be improved by retrofitting or rehabilitation unless
the engineer selects an appropriate intervention
technique based on seismic evaluation of the structure.
Therefore, the basic requirements of rehabilitation and
investigations of various retrofit techniques should be
considered before selecting an appropriate seismic
retrofit scheme up gradation of structures.
If an earthquake occurs in Zone - III
Coimbatore region, this building will first yield in Y
direction because of weak beams. The base shear
capacity is low in Y direction when compared to X
direction for first hinge formation levels and also at the
performance point level, this building has reached
immediate occupancy level in X direction and Collapse
prevention level in Y direction. It is found that this
building is more critical in the Y direction compared to
X direction. Therefore weak beams in Y-direction must

This experimental study was made to discuss


about the flexural behavior of reinforced concrete beams
which were strengthened using various configurations
and layers of GFRP composites, which were bonded to
the surface of the beams. The test observations and
results provided information about the effectiveness of
strengthening configuration, percentage increase in load
carrying capacity, moment rotation and ductility of
composite beams
Based on the scale factor 1:2, the RCC beams
of cross section 220 mm x 250 mm and length 2000 mm
was chosen for experimental studies. The following
strengthening configurations were adopted using GFRP
mats for flexure beams. Externally they were wrapped by
a (i) Single layer at two vertical sides and at the tension
bottom face (GB1) and (ii) Double layers at two vertical
sides and at the tension bottom face (GB2).
6.1 Experimental Programme
A two point loading system was adopted for
testing beams for flexure. During testing of beams,
deflection corresponding to each increment of load, yield
load, ultimate load and mode of failure were observed
and recorded. After the curing period of 28 days was
over, the beam was washed and its surface was cleaned
for clear visibility of cracks. The support points were
provided with a hinge support at both ends by steel rod
welded to the base plate. The load was applied by means
of 500kN capacity hydraulic jack powered by hand
operated hydraulic pump.
6.2 Experimental Results
The RCC beams were tested on the flexural test
as per specifications. The behavior throughout the static
test up to failure was recorded using acquisition of data
on deflection behavior and the ultimate load carrying
capacity. The crack pattern and the mode of failure of
each beam were also discussed. All beams were tested
up to their ultimate strengths. The beam (B1a) was taken
as a control beam; beams (GB1 and GB2) were
strengthened with single layer and double layer GFRP
wrapping.

Page 189

A.Vijayakumar et. al. / INDECS-15

6.2.1. Load versus Deflection and Failure Modes

dial gauges.

The load deflection histories of all the beams


were recorded. The mid-span deflection of each beam
was compared with that of their respective control beams
B1a. Also the load deflection behavior was compared
between wrapping beams having the same
reinforcement. It was noted that the behavior of the
flexure deficient beams when bonded with GFRP sheets
were better than their corresponding control beams. The
graphs comparing the mid-span deflection of wrapping
beams and their corresponding control beams are
discussed. The use of GFRP sheet had effect in delaying
the growth of crack formation.

In beam GB1, the strengthening was done by


wrapping of single layer of GFRP sheet for two vertical
sides and at tension bottom face. The ultimate load
carrying capacity of the beam was 67 kN and the
corresponding deflection was 29 mm. The yield load was
57 kN and at a deflection of 9.5mm as shown in Figure
11.

The control beam B1a was designed to fail by


flexure. The beam B1a was tested up to 42 kN and
corresponding deflection was 20 mm. The yielding of
steel was found at 37 kN and corresponding deflection
was 8 mm. Using this load and deflection of data, load
versus deflection curve was plotted as shown in Figure 9.
Fig. 11 . Load versus Deflection Curve for Beam
GB1

Fig. 12. Failure Pattern of Beam GB1


Fig. 9 . Load versus Deflection Curve for Beam
B1a

Fig.10. Failure Pattern of Beam B1a

During progress of the experiment, it was


observed that flexural cracks initiated near mid span and
further increase of loading caused widening of flexural
cracks in the portion between the loading points. In the
later stages of loading, shear cracks were also propagated
from the supports and further propagated diagonally as a
diagonal crack up to the top of the beam as shown in
Figure 10. Ultimately the beam was failed by flexure.
From the Figure 9 it may be noticed that the centre dial
gauge deflection was more compared to right and left

Initial cracks were not visible on the beams.


Further with increase of loading, propagation of the
cracks took place but it had poor visibility of cracks due
to the covering of the GFRP sheet. The failure was
initiated by the stretching of fiber wrap at the bottom in
the flexure zone followed by the crushing of concrete in
compression zone as shown in Figure 12.
From Figure 11, it was observed that GB1 had a
linear load of 54% higher when compared to B1a. By the
use of GFRP wrapping in the beam GB1, the initial
cracks were formed at higher loads than in their
respective control beam B1a. This showed that the use of
GFRP wrapping was more efficient in the case of
strengthening of flexure. Maximum percentage of
increase in ultimate strength was 60% for GB1 when
compared to B1a.

Page 190

A.Vijayakumar et. al. / INDECS-15

in the amount of strengthening materials. The deflection


at yield level increases as the stiffness of the beam
increases. This is due to the delay in the steel yielding
that occurs when the quantum of strengthening material
is increased.
For the beams strengthened with GFRP
laminates they experienced deflections of 9.5 mm and
11mm for beams GB1 and GB2 at yield level
respectively. These values indicated the higher deflection
when compared to control beam B1a of 8mm. Similarly
deflections of 29 mm and 38 mm for beams GB1 and
GB2 respectively at ultimate level were also observed.
These values indicated higher deflection when compared
to control beam of B1a 20 mm.
Fig. 13 . Load versus Deflection Curve for Beam
GB2

Therefore, the increase in the strength of the


beam depends upon the increase in the number of
laminates provided to the beam. The presence of GFRP
laminate beam inhibits the development of the diagonal
cracks. There was a significant difference in the load
which causes this initial crack. The load versus
deflection behavior was better for beams retrofitted with
GFRP wrapped beams.
6.2.2

Moment versus Rotation Curves

Fig. 14. Failure Pattern of Beam GB2

In beam GB2, strengthening was done by


wrapping of double layer of GFRP sheet at two vertical
sides and at tension bottom face. The ultimate load
carrying capacity of the beam was 92 kN and
corresponding deflection was 38mm. The yield load was
78 kN and deflection of 11mm as shown in Figure 13.

The control beam B1a had a linear moment


rotation curve in the initial stage. The moment found was
3.50 kNm and rotation 0.04 radians at the ultimate level.
The moment found was 3.08 kNm and rotation 0.016
radians at yield level. When the steel starts yielding, the
beam gradually loses its stiffness. Finally, it fails when it
reaches the maximum strain in concrete.

Initial cracks are not visible on the beams.


Further with the increase in loading propagation, the
cracks took place but it had poor visibility due to the
covering of the GFRP sheet. The failure was initiated by
the stretching of fiber wrap at the bottom in the flexure
zone followed by the crushing of concrete in
compression zone as shown in Figure 14.

The beam GB1 had a linear moment and


rotation value of
4.75 kNm and 0.019 radians
respectively. It showed 54% higher moment at the linear
level when compared to B1a. As a result, concrete
undergoes higher strains before failure. Due to improved
ductility of concrete, steel reaches its strain limit only at
the time of failure.

From the Figure 13, it was observed that GB2


had a linear load of 111% higher when compared to B1a
and 36% higher when compared to GB1. By the use of
GFRP wrapping in the beam GB2, the initial cracks were
formed at higher loads than in their respective beams
B1a and GB1. This showed that the use of GFRP
wrapping was more efficient in the case of strengthening
of flexure. Maximum percentage of increase in ultimate
strength was 119% and 37% when compared to B1a and
GB1.

The ultimate moment and rotation value of GB1


were 5.58 kNm and 0.058 radians respectively. The
value of moment carrying capacity of GB1 was 60%
higher when compared to B1a.

All the strengthened beams and that


experienced mid span deflections were higher than those
of the control beam at their failure loads as shown in
Figure 11 and Figure 13. The values of mid span
deflection were increased, because ductility of the beams
increases
due
to
the
increase

The beam GB2 had a linear moment and


rotation value of 6.5 kNm and 0.022 radians
respectively. It showed 111% higher moment at the
linear level when compared to B1a and 36% higher when
compared to GB1. As a result, concrete undergoes higher
strains before failure. Due to improved ductility of
concrete, steel reaches its strain limit only at the time of
failure. The ultimate moment and rotation value of GB2
were 7.66 kNm and 0.075 radians. The value of moment
carrying capacity was 119% higher when compared to
B1a and 37% higher when compared to GB1.
Page 191

A.Vijayakumar et. al. / INDECS-15

All the strengthened beams that experienced


mid span rotations were higher than those of the control
beams at their yield moments and ultimate moments.
Since high ductility of the strengthened beams the mid
span rotations were increased.
For beams strengthened with GFRP laminates
experienced rotation of 0.019 radians and 0.022 radians
for beams GB1 and GB2 at yield moment respectively.
These values indicated that higher rotation when
compared to control beams B1a of 0.016 radians.
Similarly, rotation of 0.058 radians and 0.075 radians for
beams GB1 and GB2 respectively at ultimate levels were
observed. These values indicated higher rotations when
compared to control beam B1a of 0.04 radians.
Therefore, the rotation capacity is also
increased dramatically when used with GFRP
strengthening techniques. This is advantageous in
seismic retrofitting of structures according to the study
of Shailesh Kr Agrawal and Ajay Chourasia (2003).

4.50% respectively. These beams exhibited higher values


as compared with the control beam B1a.
The
energy ductility index for GB2 was 27% higher when
compared to GB1.
The increase in ductility in strengthened beams
exhibited higher values in both the cases when compared
to control beams. Therefore all strengthened beams
showed adequate ductility, due to GFRP strengthening.
They can survive the Zone III earthquake forces.
Hence, it is concluded that GFRP wrapping improves
seismic resistance of the building.
7. Conclusions
Conclusion Based on Analytical Results:

6.2.3. Ductility of the Composite Beams


Ultimate
Yield
Types of deflection deflection
Beam
(u)
(y)
(mm)
(mm)
B1a
20
8

()

(Increase of
() over the
control
beam (%)

2.5

GB1

29

9.5

3.05

22

GB2

38

11

3.45

38

Conclusion based on Strengthening Results:

Maximum percentage of increased in ultimate


strength was 60% and 119% for single layer of
GFRP (GB1) and double layer of GFRP (GB2)
when compared to unstrengthen beam (B1a).
By using double layer of GFRP (GB2) ultimate
strength was increased 37% when compared to
single layer of GFRP (GB1). Therefore, the
increase in the strength of the beam depends
upon the increasing number of laminates
provided to the beam.

By using single layer of GFRP (GB1) observed


that had a linear load of 54% higher when
compared to unstrengthen beam (B1a). By
using double layer of GFRP (GB2) observed
that had a linear load of 111% higher when
compared to unstrengthen beam (B1a) and 36%
higher when compared to single layer of GFRP
(GB1).

For beams strengthened with GFRP laminates


experienced deflections of 9.5mm and 11mm
for beams GB1 and GB2 at yield level
respectively. These values indicated that higher
deflection when compared to control beam B1a
of 8mm. Similarly deflections of 29mm and
38mm for beams GB1 and GB2 at ultimate
level respectively were observed.

Table 1. Comparison of Displacement Ductility of Tested Beams

Increase of
(E) over the
( control
beam
(%)

Types of
Beam

Ultimate
load
(Eu)
(kN)

Yield load
(Ey)
(kN)

E)

B1

42

37

1.13

GB1

67

57

1.17

3.54

GB2

92

78

1.18

4.50

Table 2. Comparison of Energy Ductility of Tested Beams

As per table 1, the percentage of displacement


ductility index for strengthened beams GB1 and GB2
were 22% and 38% respectively. These beams exhibited
higher values as compared with the control beam B1a for
displacement ductility index. The displacement ductility
index for GB2 was 13% higher when compared to GB1.
Similar to the displacement ductility index, the
percentage of energy ductility index values for
strengthened beams GB1 and GB2 were 3.54% and

Based on analysis, the beam first reaches its


capacity at performance point level.
The base shear capacity was low in Y direction
as compared to X direction at first hinge
formation level.
If an earthquake occurs in Coimbatore region
which is in Zone-III, this building will first
yield in Y direction because of weak beams.
Therefore weak beams in Y-direction must be
strengthened by using any one of strengthening
techniques, before an earthquake.

Page 192

A.Vijayakumar et. al. / INDECS-15

8.
The moment rotation relation has discussed.
Maximum percentage of increased in moment
carrying capacity was 60% and 119% for single
layer of GFRP (GB1) and double layer of
GFRP (GB2) when compared to unstrengthen
beam (B1a). By using double layer of GFRP
(GB2) moment carrying capacity was increased
37% when compared to single layer of GFRP
(GB1). Therefore, the increase in the strength of
the beam depends upon the increasing number
of laminates provided to the beam.
The percentage of displacement ductility index
for strengthened beams GB1 and GB2 were
22% and 38% respectively. These beams
exhibited higher values as compared with the
control beam B1a for displacement ductility
index. The displacement ductility index for
GB2 was 13% higher when compared to GB1.
The percentage of energy ductility index values
for strengthened beams GB1 and GB2 were
3.54% and 4.50% respectively. These beams
exhibited higher values as compared with the
control beam B1a. The energy ductility index
for GB2 was 27% higher when compared to
GB1.
The increased in ductility for strengthened
beams exhibited higher values in both cases
when compared to control beams. Therefore all
strengthened beams were showed adequate
ductility, due to GFRP strengthening. They are
particularly important for strengthening of
existing concrete structures using composite
elements, to resist the seismic effects. This was
very helpful in seismic retrofitting of structures
to survive the Zone III earthquake forces.

References

1.

ATC 40. Seismic Evaluation and Retrofit of


Concrete Buildings. Applied Technology
Council, 1996.

2.

FEMA 273. NEHRP Guidelines for the


Seismic Rehabilitation of Buildings. Federal
Emergency Management Agency, Washington
D.C., 1997.

3.

Mehmet Inel, Hayri Baytan Ozmen. Effect of


Plastic Hinge Properties in Nonlinear Analysis
of Reinforced Concrete Buildings. Engineering
Structures, 2006, pp.1494-1502.

4.

FEMA 172. NEHRP Handbook of Techniques


for the Seismic Rehabilitation of Existing
Buildings. Building Seismic Safety Council,
Washington D.C., 1992.

5.

Shailesh KR, Agrawal, Ajay Chourasia.


Nonlinear Static Analysis For Seismic
Evaluation and Retrofit of RC Buildings.
Workshop on Retrofitting of Structures, Indian
Institute of Technology Roorkee, India, 2003,
pp.116-124.

6.

Amar Prakash, Thakkar SK. A Comparative


Study of Retrofitting of RC Building.
Workshop on Retrofitting of Structures, Indian
Institute of Technology Roorkee, India, 2003,
pp. 159-173.

7.

Mohamed Nour El-Din Abd-Alla. Application


of Recent Techniques of Pushover for
Evaluating Seismic Performance of Multistory
Buildings. Faculty of Engineering, Cairo
University Giza, Egypt, September 2007.

8.

Sundar N, Elangovan G. Evaluation of Inelastic


Response of Reinforced Concrete Frames using
Pushover Analysis. Thesis, Sastra University,
Kumbakonam, 2000.

9.

FEMA 356. Prestandard and Commentary for


the Seismic Rehabilitation of Buildings.
American Society of Civil Engineers, Reston
V.A., 2000.

Page 193

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15) ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

CONSTRUCTION ACCIDENTS INDIAN SCENARIO


D. Kamalnataraj 1, Dr. V. Ramasamy 2
1

Research Scholar, Anna University, Chennai, 600025, India,kamalnatarajsgi@gmail.com


Adhiparasakthi engineering college, Melmaruvathur, 603319, India, rams_apec@yahoo.co.in

Abstract
Construction industry is one of the most hazardous with an average of one fatal accident every ten minutes. Such huge
number of accidents and fatalities causes social and economic loss including the loss of productivity. The root causes
for such accidents were analyzed universally and majority of the cases indicates the constant negligence towards safety
by the project personnel, in recent years partly due to the structural collapse accidents. Both the human and the economic
costs of accidents worldwide are enormous, This paper identify the effect and their expensive defeat of investment and
human loss in construction accidents expected the knowledge shared Construction professional involved in construction
projects to plan an accident prevention strategy properly
Keywords: Construction Accidents, Construction Safety, Structural collapse, PPE, Construction Projects
1. Introduction

1.2 Structural Collapses in India:

The construction industry is significantly more hazardous


than most others economic sectors [3]. Trying to reduce the
incidence of construction failures is a continuous process and
organization such as Occupational Safety and Health
Administration [2] and others are dedicated to this goal [7].
The European Foundation found that 60% of the accidents it
surveyed could have been eliminated, reduced, or avoided
with more thought during the design stage (John A.
Gambatese [8]. A study on U.K. construction industry
revealed that there is a certain relationship between safety of
construction and decisions made in design process regarding
the accident causation.

Every year, India loses an average of 2,658 people to


different kinds of structural collapses; that is around 7 deaths
a day [3]. India has witnessed a significant number of
building collapses in the past 18 months. There have been
seven major building collapses in three Indian cities
(Mumbai, Chennai, and Vadodara) alone, resulting in a total
of 234 deaths, along with innumerable injuries and economic
losses [1]
.
The Indian professionals say, Safety standards always have
been an issue in India, whose building boom is helping the
country maintain an economic growth of almost 9% a year.
But as the number and scale of the airport, toll roads,
skyscraper and mall projects increase, so the risk of tragedy
rises. The number of accidents is also climbing in mining,
manufacturing and brick-making as those industries expand.
Big construction companies tend to train workers and
provide for safety. But the industry remains dominated by
small contractors who pay little attention.
A study of building failures in the US covering 11 years
from 1989 to 2000 recorded 225 failures (including both
partial and total collapses) with a total loss of lives of 97
persons. This implies about 9 lives lost per year on average
in the US, as compared to our record of 234 deaths in 18
months from just 7 selected collapses [3]. India has seen

1.1 Structural Collapses Worldwide:


There are annually at least 60,000 fatal accidents on
construction sites around the world, according to ILO
estimates. This means that one fatal accident occurs every
ten minutes in this sector, and that around 17% of all fatal
accidents at work (1 in every 6) happen on construction sites.
Young workers aged 15-24 are much more likely to suffer
non-fatal but serious accidents at work compared to their
older colleagues [3].

Page 194

Author et. al. / INDECS-15

frequent building collapses, many blamed on lax safety and


substandard materials [5].
The objective of this study is to find out factors leading to
accidents in construction projects using two approaches
which are event area breakdown structure and the nature of
fall accident analysis. In this paper, current profile of
accident records of India is presented. This paper also
presents the nature of accidents. The event area and their
potential active failures leading accidents are described. It is
expected the knowledge shared in this paper assist all people
involved in construction projects to plan an accident
prevention strategy properly.
2. BACKROUND PAPERS FOR CASE STUDIES
Apart from design failures which results in disasters, more
case studies prove that design criteria and limits adopted
does have strong impact over the accidents occurring at
construction sites. A 1996 paper showed that 50% of general
contractors interviewed identified poor design features are
affecting safety (Smallwood 1996).A European study
published that 60% of the accidents could have been
eliminated or reduced with more thought during the design
(European Foundation 1991).Researchers in the U.K. found
that, the design changes would have reduced the likelihood
of 47% of the 100 construction accidents (Gibb et al
2004).An American researcher found that construction
design was linked to accidents in approximately 22% of 226
injury incidents in Oregon, Washington, and California and
to 42% of 224 fatality incidents between 1990 and 2003
(Behm 2004).
3. Construction output in 2015 (billions in USD)
Construction industry spend investment worldwide ,the India
is the one of most leading investment for 620 billion USD in
the year (2015) This indicate the Indian construction
companies getting rapid development to improve the
infrastructure in this meanwhile the construction accidents
also it is crowning (India and china) in graph showing
worldwide In [3]

FIGURE 1 Construction output in 2015 (billions in USD)

Table 1 - Estimated numbers of fatal and non fatal


accidents world wide

Region

Estimated
fatal accidents
(ILO)

All accidents
report
to
(ILO)

15879
17416
40133
90295
76886
53292
39372
17977

Fatal
accidents
reported to
( ILO)
14316
7853
222
12736
3051
145
2009
1416

EME
FSE
IND
CHN
OAI
SSA
LAC
MEC
WORLD

351250

41748

9031431

7527083
343004
928
61329
141349
27015
776938
153785

4. Construction accidents an Indian scenario

In India, the construction industry is the second largest


employer next to agriculture. The annual turnover of the
construction industry is about Rs 4,000 billion, which is
more than 6 per cent of the National GDP even as it employs
a large work force. The average Fatal Accident Frequency
Rate (FAFR) in the Indian construction industries is 15.8 for
1000 employees as against 0.23 in the USA, avoid such
fatalities a culture of safety, including conduct of induction
training at the commencement at work place, mock drills
from time to time and periodic refreshing courses besides
appointment of safety officer, one each for every 1,000
construction workers is required .The number of fatality
caused by structural collapses in India in 10 years from 2003
to 2015 is given below graphically.

Page 195

Author et. al. / INDECS-15

Construction accidents like collapse structure make major


cost defeat from other industry ,in Tamilnadu various type of
structure collapse occurred in past three years ,In above fig
showing
Major defeat occurred in moulivakkam (Chennai) is
nearly 11 storied building is cost of construction is
approximately 250 cores and no of fatality is also very high

200
0

Collapse of
Structure (Bridge)
Collapse of
Structure (Others)

Fig 3 Accidental death and injured rate caused by collapse of structure


in the year 2013

Engineers generally concentrate only on productivity but it


became difficult when the execution is carried by way of
accidents in construction industry. This results losses in
investment; changes in schedule, affect in productivity, The
general attempts by the client, contractor and the designers
are always towards the attitude of cost savings where the
attempt to the prevention of accidents is negligible. This type
of attitude of construction professionals lack to concentrate
in safety aspects to protect this type major building accident.

2013
Collapse

400

2012
Suffocations

600

Collapse of
Structure
(Building)
Collapse of
Structure (Dam)

2011

Fall into

800

Collapse of
Structure (House)

Electrocutions

1000

Fire

1200

14000
12000
10000
8000
6000
4000
2000
0

Explosion

National Crime Records Bureau (NCRB) statistics show that


Tamil Nadu accounts for a significantly larger number of
deaths in building collapses than neighbors Andhra Pradesh
Karnataka and Kerala. NCRB data also reveals that more
people are likely to die in a single collapse in Tamil Nadu
than in other states (The Times of India, 2013).

Fall from height

Fig 2 Number of fatality caused by structural collapses in


India

As a predominant factor in controlling the accidents the


expected hazards at site can be controlled during the design
stage of the project itself. This study is an attempt to analyze
the role of the designers in controlling the hazards at site and
their attitude towards safety, the root causes for such
accidents were analyzed universally and majority of the
cases indicates the constant negligence towards safety by the
project personnel.

2014

Fig 4 various types of accidents in past four years


There are many accidents leading to fatal and serious injuries
in construction. Fig 5 shows the natures and expense of
accidents in under construction .This results make afraid to
Investment the capital was complete defeat to the
construction companies this make vanish the company
India faced major destructions within last two decades in
both under and post construction sites. Studies reveal that
major structural collapses are due to deficient structural
design, poor detailing, substandard materials, worst
workmanship, poor maintenance, combination of any of
these factors. It is expected that even if a building does fail,
it would give adequate warning so as to ensure safe egress
and prevent loss of life.
Nevertheless, recent building
collapses in India were instant without much warnings /
indications or there was not much inspections on the
structure. Lack of such regular inspections in both under and
post construction structures are also a reason for these type
of major loss. The subject study attempted to assess the

Page 196

Author et. al. / INDECS-15

potential damages of the structure using a detector and


communicate the same to the occupants / employees for safe
evacuation.
India has had a national law in place since 1996 that
aims to ensure the welfare of construction workers. But
Indian states have been slow to adopt the law and implement
it, which is required for its provisions to take effect. India

also faces a shortage of safety inspectors for construction


sites
Fig 6 show the rate of fatal occurred past three years this
graph show to excite construction industry is most hazards
industry to the construction labour this type of accidents
make demand to labour

Collapse of old
building(Triplicane,
Collapse of school
building(Viruthunagar)
Collapse of multi storey
building(Thenampet,
Collapse of star hotel
building(Mahabalipuram,
Collapse of multi storey
building(Madurai)
Collapse of shopping
complex(Trichy)
Collapse of hotel
building(Puzhuthivakkam,
Collapse of multi storey
building(Moulivakkam,
Collapse of godown
compound wall(Thiruvallur)
Collapse of formwork of
university

Collapse of multi storey


building(Madurai)

Collapse of multi storey


building(Madurai)

Scaffolding
falls(Puducherry)

Crane crash at a metro rail


site(Chennai)

Collapse of collage building


(Kancheepuram)

Collapse of
building(coimbatore)

Collapse of two storey


building(Thiruvallikani,

DEFEAT IN COST(APPROXIMATE) IN LAKHS

300
250
200
150
100
50
0

1436910- 23- 24- 3-Oct- 14- 30- 19- 23- 19- 4-Apr- 28- 6-Jul- 30Jun- Aug- Aug- Aug- Sep- Sep- Sep- 12 May- Jun- Jul-13 Jul-13 Sep- 14 Jun- 14 Mar12
12
12
12
12
12
12
13
13
13
14
15

Fig 5 OCCRANCE OF CONSTRUCTION ACCIDENTS IN 2010-2015 IN TAMIL NADU

Page 197

Author et. al. / INDECS-15

70
60
50
40
30
20
10
0

Collapse of
building(coimbatore)
Collapse of two storey
building(Thiruvallikani,
Collapse of collage
building (Kancheepuram)
Crane crash at a metro rail
site(Chennai)
Scaffolding
falls(Puducherry)
Collapse of multi storey
building(Madurai)
Collapse of multi storey
building(Madurai)
Collapse of old
building(Triplicane,
Collapse of school
building(Viruthunagar)
Collapse of multi storey
building(Thenampet,
Collapse of star hotel
building(Mahabalipuram
Collapse of multi storey
building(Madurai)
Collapse of shopping
complex(Trichy)
Collapse of hotel
building(Puzhuthivakka
Collapse of multi storey
building(Moulivakkam,
Collapse of godown
compound
Collapse of formwork of
university

NUMBER OF FATALITY

1436910- 23- 24- 3-Oct- 14- 30- 19- 23- 19- 4-Apr- 28- 6-Jul- 30Jun- Aug- Aug- Aug- Sep- Sep- Sep- 12 May- Jun- Jul-13 Jul-13 Sep- 14 Jun- 14 Mar12
12
12
12
12
12
12
13
13
13
14
15
Fig 6 Numbers of fatal in past three years in Tamilnadu constructions

3. Conclusion
Recent building crashes are instant collapse of heavy
weighed structures over the occupants due to the soil
settlement that is down ward movement of building. The
occupants are unaware about the dames and hesitant in
reacting towards such emergency situations resulting in
loss to life, property etc. Engineers generally concentrate
only on productivity but it becomes difficult when the
execution is carried by way of accidents in construction
industry. This approach further emphasize to loss in
investment; changes in schedule, affect productivity. The
engineers should concentrate on safety aspects through
design in initial stage, in order to eliminate the
construction accidents. Safety of human life is a serious
threat in these structural collapses. Both the human and
the economic costs of accidents worldwide are
drastically increasing frequently worldwide. Even
though there are enormous economic losses, the loss of
human lives can be avoided if the structural collapse is
predicted .this paper recognizes the responsiveness to
construction accidents challenge to civil Engineering
professional.
Reference
[1] Complete collapse of building standards, The
Hindu business line, October 5, 2014, article no.
6473308.

[2] Eric MARKS and Jochen TEIZER (2012),


Proximity sensing and warning technology for heavy
construction equipment operation, Construction
Research Congress 2012, ASCE 2012.
[3] I.L.O (2005), Prevention: A Global Strategy
Promoting Safety and Health at Work. The ILO Report
for World Day for Safety and Health at Work,
International Labour Office , Geneva, 2005.ISBN 92-9117107-8
[4] In South India, TamilNadu records most building
collapse deaths, The Times of India, July 4, 2012,
article no. 14661723.
[5] India building collapse: Scores Trapped in
Chennai, BBC News India, 29 june, 2014, no.
28076572.
[6] Ivan W.H. Fung, Y.Y. Lee, Vivian W.Y. Tam, H.W.
Fung (2014), A feasibility study of introducing chin
straps of safety helmet as a statutory requirement in
Hong Kong construction industry, Safety Science 65
(2014) 70 78.
[7] Janet K. Yates and Edward E. Lockely (2002),
Documenting and Analyzing Construction Failures,
Journal of Construction Engineering and Management,
January/February 2002. DOI: 10.1061/(ASCE) 07339364(2002) 128:1
[8] John A. Gambatese Michael Behm Jimmie

Page 198

Author et. al. / INDECS-15


W. Hinze, Viability of Designing for Construction
Worker Safety, Journal of Construction Engineering
and Management (ASCE)/September 2009.
DOI: 10.1061/(ASCE) 0733-9364(2005) 131:9(1029)
[9] Mr.N.Sambamurty, SK.Hasane Ahammad (2013),
Prevention of Train Accidents Using Wireless Sensor
Networks, International Journal of Engineering
Research and Applications, ISSN: 2248-9622, Vol.3,
issue 6, Nov-Dec 2013, PP.1592-1597
[10] Naresh Fernandes, Samir Patil, Jennifer OBrien,
Shivam Vij (2015), Across India 2600 people die every
year in building and other structural collapses, scroll.in,
article no. 668636.
[11] Narito KURATA, Billie F.SPENCER, Jr. Manual
RUIZ SANDOVAL (2004), Building Risk
Monitoring Using Wireless Sensor network, 13th world
conference on Earthquake Engineering, Vancouver,
B.C., Canada, August 1-6, 2004, Paper No.1406.
[12] Nie Jia Yau, Ming Kung Tsai, Hao Lin Wang,
Dong Mou Hung, Chih Shian Chen, Wen Ko Hsu
(2014), Improving bridge collapse detection and on-site
emergency alarms: A case study in Taiwan, Safety
Science 70 (2014) 133 142.
[13] P.H. Galanis, J.P. Moehle (2012), Development of
Collapse Indicators for Older-Type Reinforced Concrete
Buildings, 15th World Conference on Earthquake
Engineering, Lisbon, Portugal, 24 28 september, 2012.
[14] Szymberki.R (1997), Construction Project Safety
Planning, TAPPI journal, 80(11), 69-74.
[15] Ung Kyun Lee, Joo Heon Kim, Hunhee Cho,
Kyung In Kang (2009), Development of a mobile
safety monitoring system for construction sites,
Automation in Construction 18 (2009) 258 264.
[16] Vicki Kaskutas, Ann Marie Dale, Hester Lipscomb
et al (2010), Changes in fall prevention training for
apprentice carpenters based on a comprehensive needs
assessment, Journal of Safety Research 41 (2010) 221
227.

Page 199

International Conference on Infrastructure Development for Environmental Conservation and Sustenance


(INDECS-15) ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

ECOLOGICAL BEHAVIOUR OF WASHBOWL FROM DISCARDED PRINTED


CIRCUIT BOARD
A.G.GANESHKUMAR1, Dr.G.RANGANATH2, Dr CHANNANKAIAH3,
1 Adhiyamaan College of Engineering, Hosur, 635109, India, aggkdesign@gmail.com
2 Adhiyamaan College of Engineering, Hosur, 635109, India, principal@adhiyamaan.ac.in
3 Adhiyamaan College of Engineering Hosur, 635109, India, hod_mech@adhiyamaan.ac.in

Abstract
The environmental costs of e-waste disposal, from toxic incinerator emissions to groundwater landfill poisons, have been reported
with increasing frequency and alarm. The Printed circuit board PCB is one of the major e-waste, which causes toxic to our
environment. In India people started using sanitary wares is increasing from last 5 years only. The foreign countries are starting to
establish their companies in India. The cost of the product is also increasing because of demand in the market. Therefore PCB boards
are powdered and sieved it to 80 micron size. The sieved powder is mixed with the epoxy resin, epoxy hardener and White color
pigment for fabricating the specimen. The specimens are tested under different mechanical and ecological conditions. Finally, the
composition 50% of PCB powder is mixed with 50% of bonding material is suitable for fabricating the washbowl by compression
molding technique. The washbowl which is fabricated is used to achieve lower cost and weight when compared to ceramic washbowl.
Keywords: E-Waste PCB, Epoxy resin, Colour Pigment, Fabrication and Ecological.

1. Introduction
An e-waste consists of a highly toxic material like Arsenic,
Barium, Beryllium, Cadmium, Lead, Mercury, etc., which
cannot be decomposed easily and occupies more land area. The
dumping of e-wastes spoils the fertility of the soil and causes a
lot of hazards to the ecosystem. These wastes can be reduced by
properly recycling it and converting them into some useful
products. The recycling process involves the collection of
various types of E-waste like Printed circuit board (PCB). Then
soldering parts and waste materials are segregated from the
exceeding E-waste materials and shredded into small size
particles. The small sizes of Printed circuit board are then made
into powder. The technique that we use here to make the
washbowl is compression molding. The normal washbowl
production involves the raw materials which are of more cost
and requirement is high. But here we use the E-wastes that are
less economical and performance is far better than the normal
washbowl.

These procedures outline the requirements for the management


of PCB-contaminated waste materials generated in laboratories,
other academic areas and facilities and services operations. The
identification of possible sources of PCB-contaminated
materials, analysis and administrative requirements are outlined.
Decontamination procedures as they apply to emergencies are
also addressed.

Fig.1. Waste PCBs


2.1. PCB Recycling Processes

2. Experimental

Page 200

The waste PCBs recycling is the process of recovering land or


waste PCBs and reprocessing the material into useful products,
sometimes completely different in form from their original state.
For instance, this could mean convert from PCBs to powder by
routing machine.

material, giving it an advantage when working with expensive


compounds.

2.2. Materials: Waste PCBs Powder


The obtained powder contains some broken parts which cannot
be grinded and some abrasive particles which are very fine. So
the powder obtained is separated to some standard mesh sizes by
using sieves. The sieves used by us in the process of segregation
are 80 micron size and 150 micron size. These sizes are selected
because of the smaller particle size of the powder obtained.

Fig.2. Waste PCB Powder

Fig.5. Compression molding

Fig.3. 80 micron size


Fig.6. Molding Process
2.3.1. Parameters of Compression Molding

Fig.4. 150 micron size


As two different types of powders are obtained after mashing, in
order to select a better powder with good physical properties
such as Density, impact strength, tensile strength, compressive
strength and hardness, some standard tests are made of the
powder obtained by making sample specimens.

The quantity of charge (molding material) pouring into the mold.


Pressure of the molding process Range of pressure 13.8 - 20.7
MPa, Mold temperature ranges from 149C - 191C. Cure time
variables. The period required to harden thermosetting material
to partial and complete polymerization is called cure time.
2.4. Mold Preparation for Specimen
For preparation of specimens, initially mold is prepared by using
sheets. For making specimen for tensile test, and hallow
cylindrical mold is prepared with 20mm diameter and 100 mm
length.

2.3. Compression Molding


Compression molding is a method of molding in which the
molding material, generally preheated, is first placed in an open,
heated mold cavity. The mold is closed with a top force or plug
member, pressure is applied to force the material into contact
with all mold areas while heat and pressure are maintained until
the molding material has cured. The process employs
thermosetting resins in a partially cured stage, either in the form
of granules, putty-like masses, or performs. Compression
molding is a high-volume, high-pressure method suitable for
molding complex, high strength fiberglass reinforcements.
Advanced composite thermoplastics can also be compression
molded with unidirectional tapes, woven fabrics, randomly
oriented fiber mat or chopped strand. The advantage of
compression molding is its ability to mold large, fairly intricate
parts. Also, it is one of the lowest cost molding methods
compared with other methods such as transfer molding
and injection molding; moreover, it wastes relatively little

Fig.7. Mold for specimen


2.5. Specimen Preparation
The 80 micron size powder is chosen because it has more
density. The strength is also more when compared to the other
micron sizes. The powder is mixed with the epoxy resin in 50:50
ratios along with Gardner and white color pigment is added.
These are mixed by using hand stirring process. The mold is
coated with the wax for non-sticky purpose. Then the mixture is

Page 201

poured into the both cylindrical and square shape molds.


Afterwards is tightened with bolts and nuts. Finally, it is placed
in the oven for the time period of 1 hour at 60 O C.

Fig.8. Hand stirring

Fig.10. Mixture is pasted into mold

Fig.9. Wax coated molds

Fig.11. Mold is tightened

replaces transmission of load through levers and knife edges,


which are prone to wear and damage due to shock on rupture of
test pieces. Load is applied by a hydrostatically lubricated ram.

Fig.14. Before Tensile Test

Fig.15. After Tensile Test


(sample-I)

Fig.16. After Tensile Test (sample-II)


3.1.1. Calculations
Sample I - (50% PCB POWDER + 50% EPOXY)
Breaking load
= 850 N
Area
= 28.286 mm2
Tensile strength = Breaking load / Area
= 850 / 28.286
= 30. 050 N/mm2
Elongation
= 6. 000%

Fig.12. Cylindrical specimen Fig.13. Square shape specimen

Figure: Stress Vs Strain

3. Mechanical Tests
The fabrication of PCB (Printed Circuit Board) is tested by
Universal Testing machine, Hardness testing machine. It helps to
identify the strength and hardness of the washbowl.
The following test is carried out by using the above machines.
1. Tensile test
2. Compression test
3. Rockwell Hardness test
3.1. Tensile Test
Tension test is conducted by gripping the test specimen between
the upper and lower cross-head, Compression, transverse,
bending, shear and hardness tests are conducted between the
lower crosshead and the table. The lower cross-head can be
raised or lowered rapidly by operating the screwed columns thus
facilitating ease of fixing of the test specimen.
Universal Testing Machine is designed for testing metals and
other materials under tension, compression bending, transverse
and shear loads. Operation of the machines is by hydraulic
transmission of load from the test specimen to a separately
housed load indicator. The hydraulic system is ideal since it

Fig.17. Tensile Test (sample 1)

Sample 2 - (50% PCB POWDER + 50% EPOXY)


Breaking load
= 770 N
Area
= 28.286 mm2
Tensile strength = Breaking load / Area
= 770 / 28.286
= 27. 221 N/mm2

Page 202

Elongation

= 4. 000%

Figure: Stress Vs Strain

Breaking load

= 29,095 N

Area

= 316.490 mm2

Compressive strength

= Breaking load / Area


= 29,095 / 316.490
= 91. 930 N/mm2

Figure: Load Vs Displacement

Fig.18. Tensile Test (sample 2)


3.2. Compression Test
Compression tests are used to determine how a product or
material reacts when it is compressed, squashed, crushed or
flattened by measure fundamental parameters that determine the
specimen behavior under a compressive load. These include the
elastic limit, which for "Hookean" materials is approximately
equal to the proportional limit, and also known as yield point or
yield strength, Young's Modulus (these, although mostly
associated with tensile testing, may have compressive analogs)
and compressive strength. Compression tests can be undertaken
as part of the design process, in the production environment or in
the quality control laboratory, and can be used to: Assess the
strength of components, e.g. automotive and aeronautical control
switches, compression springs, bellows, keypads, package seals,
PET containers, PVC / ABS pipes, solenoids etc.

Fig.22. Compression Test (sample 1)

Sample 2 - (50% PCB POWDER + 50% EPOXY)

Breaking load

= 31,195 N

Area

= 341.568 mm2

Compressive strength

= Breaking load / Area


= 31,195 / 341.568
= 91. 328 N/mm2

Figure: Load Vs Displacement

Fig.19. Before Compression Test Fig.20. After Compression


Test (sample-I)

Fig.23. Compression Test (sample 2)

3.3. Rockwell Hardness Test


Fig.21. After Compression Test (sample-II)

3.2.1. Calculations

Sample I - (50% PCB POWDER + 50% EPOXY)

The Rockwell Hardness test is a hardness measurement based on


the net increase in depth of impression as a load is applied.
Hardness covers several properties: resistance to deformation,
resistance to friction and abrasion. The term refers to stiffness or
temper or to resistance to scratching, abrasion, or cutting. The
greater the hardness of the metal, the greater resistance it has to
deformation. By measuring the depth of the indentation,

Page 203

progressive levels of forcing are measurable on the same piece.


The indenter may either be a steel ball of some specified
diameter or
a spherical diamond-tipped cone of 120 angle
and 0.2 mm tip radius, called Brale. The hardness number is
expressed by the symbol HR and the scale designation.
3.3.1. Tabulation
Rockwell Hardness L scale, Major Load: 60kg, Minor Load:
10kg, Ball Indenter
Table 1. Rockwell hardness
Hardness Value
(hr L scale)
78.5
80.7
87.3

3.Hot water test 4.Cold water test


4.3.1.

Test Using Acid as the Medium,

The weight of the slab is weighed by using the digital weighing


machine.3% HCL acid is prepared by using 200 ml of water and
6ml of concentrated hydrochloric acid. It is diluted by using the
distilled water. The pH value of the acid has been noted before
dipping the slab. The slab is submerged in the acid for the
duration of 120 hours. After that the slab is removed and heated
in the oven for 1 hour at 600C. The slab is removed and weighed
again to check whether any changes in weight. The pH of the
acid solution is also checked.

88.0
4. Ecological Tests

Fig.24. while dipping in acid

The fabrication of PCB (Printed Circuit Board) is tested by using


HCL acid and Water (hot and cold). It helps to identify the
corrosion and decolourisation of the washbowl.
The following test is carried out by using the above machines.
1. Corrosion test 2.Decolourisation test 3.Odour test.
4.1. Corrosion Test
Corrosion testing refers to the processes conducted by
laboratories in order to solve, prevent or mitigate problems
related to corrosion. These processes can be applied in industrial
materials and infrastructure products, and are often used in
failure analysis.
All corrosion laboratories are composed of expert failure
analysts, chemists and engineers that are all certified in
corrosion testing. Such tests can provide useful information in
order to make sound decisions regarding selection of materials,
processing and treatment.
4.2. Decolourisation Test
After the corrosion test there may be removal of material will
take place. That will lead to change in color is known as
decolourisation. Decolourisation is inspected by visual method.
4.3. Odour Test
An odor is caused by one or more volatilized chemical
compound, generally at a very low concentration, that humans or
other animals perceive by the sense of olfaction. Odours are also
commonly called scents, which can refer to both pleasant and
unpleasant. The terms fragrance and aroma are used primarily by
the food and cosmetic industry to describe a pleasant odor, and
are sometimes used to refer to perfumes. In contrast, malodor,
stench, reek, and stink are used specifically to describe
unpleasant odor.
Above three tests are carried under by
1.Acid test
2. Water test,

Fig.25. After dipped in acid

Table 2. Test Using Acid as the Medium


Time Period: 120 hours
Sl
No

Weight in gms
Before

After

1.

14.964

2.
3.

pH Value
Before(28C)

After( 29C)

14.964

2.28

2.38

14.811

14.811

2.28

2.30

14.833

14.831

2.28

2.40

4.3.1.1 Conclusion for Acid as the Medium


No appreciable change in weight and pH value. Because
negligible amount of ions only moved from the surface of the
material with acid.
4.3.2. Test Using Demineralized (DM) Water at 23oc 32oc
The testing procedure for DM water, The weight of the slab is
weighed by using the digital weighing machine. Water is poured
in a beaker up to 200ml. The pH value of the water has been
noted before dipping the slab. The slab is submerged in the
water for the duration of 120 hours. After that the slab is
removed and heated in the oven for 1 hour at 60 0C.The slab is
removed and weighed again to check whether any changes in
weight. The pH of the water is also checked.

Page 204

Fig.31. After dipped in soft water


Fig.26. DM water

Fig.27. While dipping in DM


Water

Table 4. Test using soft water at 23oC 32oC


Time Period: 120 hours
Weight in gms
Sl.
No

Fig.28. After dipped in DM water

Before

After

9.

14.540

10.
11.

pH Value
Before(30C)

After(28C)

14.538

7.76

9.37

14.477

14.478

7.76

8.94

14.391

14.398

7.76

8.78

4.3.3.1. Conclusion for Soft Water


Table 3. Test using DM water at 23oC 32oC
No appreciable change in weight. Increase in basic nature is
noted due to the slight dissolution of the temporary oxide film
from the surface of the sample.

Time Period: 120 hours


Weight in gms

pH Value

4.3.4. Test Using Hard Water At 23oc 32oc

Sl.
No

Before

After

Before (30C)

After (28C)

6.

14.713

14.712

7.12

9.01

7.

14.792

14.794

7.12

9.11

8.

14.637

14.636

7.12

9.03

4.3.2.1. Conclusion for DM Water

Fig.32. Hard water

Fig.33. While dipping in hard water

Demineralized water means no ions, Oxides from the surface of


sample will be free to dissolve into the bulk of the medium
therefore pH increases drastically.
4.3.3. Test Using Soft Water At 23oc 32oc

Fig.34. After dipped in Hard Water


Table 5. Test using hard water At 23oC 32oC
Fig.29. Soft water

Time Period: 120 hours

Fig.30. While dipping in soft water

Weight in gms
Sl.
No

Before

After

pH Value
Before(30C)

After(28C)

Page 205

12.

15.440

15.454

9.33

9.62

13.

15.220

15.230

9.33

9.66

14.

15.427

15.441

9.33

9.66

DM water means no ions. Oxides from the surface of sample


will be free to dissolve into the bulk of the medium therefore pH
increases drastically. Oxide is facilitated by an increase in
temperature. Increase in temperature leads to decrease in
solubility products, which in turn results in increased
precipitation of salt particles.

4.3.4.1.Conclusion for Hard Water

Table 7. Test using soft water at 50oC

Increase in weight in hard water medium is due to the protective


coating formed by the ions present in the medium, Increase in
basic pH is noted, due to the dissolution of temporary oxide
films formed on the surface of the sample.

Sl.
No

Weight in gms

pH Value

Before

After

Before
(30C)

After
(31C)

18.

15.749

15.804

7.18

7.43

19.

15.427

15.468

7.18

7.50

20.

14.966

15.013

7.18

7.58

4.4 Test Using Water at 50oc


The weight of the slab is weighed by using the digital weighing
machine. Water is poured into a beaker up to 200 ml. The pH
value of the water has been noted before dipping the slab. The
slab is submerged in the water. The beaker is placed in the oven
at 50OC for the duration of 168 hours. After that the slab is
removed and heated in the oven for 1 hour at 60 0C. The slab is
removed and weighed again to check whether any changes in
weight. The pH of the water is also checked.

Time Period: 168 hours

4.4.2.Conclusion for Soft Water


No appreciable change in weight. Increase in basic nature is
noted due to the slight dissolution of the temporary oxide film
from the surface of the sample. Oxide is facilitated by an
increase in temperature. Increase in temperature leads to
decrease in solubility products, which in turn results in increased
precipitation of salt particles.

Fig.35. Beaker placed in oven

Fig.36. After dipped in DM


water (hot)

Table 8. Test using hard water at 50oC


Time Period: 168 hours
Sl
No

Weight in gms
Before

Fig.37. After dipped in soft


water (hot)

Fig.38. After dipped in


hard water (hot)

Table 6. Test using DM water at 50oC


Time Period: 168 hours
Sl.
No

Weight in gms
Before

Before
(30C)

After
(31C)

21.

14.800

14.846

8.96

8.77

22.

15.772

14.822

8.96

9.03

23.

14.842

14.884

8.96

9.10

4.4.3. Conclusion for Hard Water

pH Value

After

Before(30C)

After( 31C)

15.

15.060

5.107

7.15

8.06

16

15.097

5.142

7.15

8.07

17.

14.993

5.042

7.15

7.99

4.4.1.Conclusion for DM Water

After

pH Value

Increase in weight in hard water medium is due to the protective


coating formed by the ions present in the medium. Increase in
basic pH is noted, due to the dissolution of temporary oxide
films formed on the surface of the sample. Oxide is facilitated by
an increase in temperature. Increase in temperature leads to
decrease in solubility products,, which in turn results in increased
precipitation of salt particles.
4.5. TEST USING WATER AT 10OC
The weight of the slab is weighed by using the digital weighing
machine. Three different types of water are selected.

Page 206

Demineralized water, Soft water and Hard water. Water is


poured into a beaker up to 200ml. The pH value of the water has
been noted before dipping the slab. The slab is submerged in the
water. It is placed in the refrigerator at 10 oC for the duration of
168 hours. After that the slab is removed and heated in the oven
for 1 hour at 600C. The slab is removed and weighed again to
check whether any changes in weight. The pH of the cold water
is also checked.

25.

15.079

15.086

6.28

7.25

26.

15.780

14.795

6.28

7.48

27.

15.806

15.816

6.28

7.23

4.5.2. Conclusion for Soft Water


No appreciable change in weight. Increase in basic nature is
noted due to the slight dissolution of the temporary oxide film
from the surface of the sample. Inappreciable difference in
weight. Decrease in solubility products, value of salts not
appreciably affected low ion deposition.
Table 11. Test using hard water at 10oC

Fig.39. Beaker placed in the


refrigerator

Fig.40. After dipped in DM


water (cold)

Time Period: 168 hours


Sl
No

Fig.41. After dipped in


soft water (cold)

Fig.42. After dipped in hard


water (cold)

Table 9. Test using demineralized water at 10oC


Sl
No
Before

pH Value

4.5.1.

After

Before
(30C)

After
( 31 C)

5.

14.990

15.090

6.90

8.65

17.

15.244

15.251

6.90

8.83

24.

15.132

15.137

6.90

8.82

pH Value

Before

After

Before
(30C)

After
(31C)

28.

14.747

14.758

8.09

8.26

29.

14.963

14.978

8.09

8.37

30.

14.815

14.826

8.09

8.35

4.5.3. Conclusion for Hard Water

Time Period: 168 hours

Weight in gms

Weight in gms

Increase in weight in hard water medium is due to the protective


coating formed by the ions present in the medium. Increase in
basic pH is noted, due to the dissolution of temporary oxide
films formed on the surface of the sample. Inappreciable
difference in weight. Decrease in solubility products, value of
salts not appreciably affected low ion deposition.
5.

3D Model of the Washbowl

Conclusion for Demineralized Water

DM water means no ions. Oxides from the surface of sample


will be free to dissolve into the bulk of the medium therefore pH
increases drastically. Decrease in solubility products, value of
salts not appreciably affected low ion deposition.
Table 10. Test using soft water at 10oC
Fig.43. 3D model of the washbowl

Time Period: 168 hours


Sl
No

Weight in gms
Before

After

pH Value
Before
(30C)

After
(31C)

5.1 Area Calculation of Washbowl


Area at top

= length breadth
= 22.522.5

Page 207

Area at bottom

= 506.25 mm2
= length breadth
= 14.514.5
= 210.25 mm2

6. Recycling the Specimen


The prepared specimen is also can be recycled by grinding
process. The powder which we got is also mixed with the
bonding material to fabricate the any product. The recycled
powder has some bonding materials it will increase the strength.
When every time of recycling the strength value increases.

suitable to attain the increase in property of strength (tensile and


compression) and hardness. The washbowl is fabricated by using
compression molding technique. Because of better surface finish
and increase in strength.
The fabrication of washbowl which is made from PCB powder is
used to reduce the increasing number of electronic waste. The
product made from E-waste is also again can be converted into
powder by grinding process.
Acknowledgement
We wish to thank all the faculties in Mechanical Department,
Adhiyamaan College of Engineering, Hosur, helped us giving
suggestions and motivations for this work. Also thanking all that
indirectly involved in this work.

REFERENCES

[1]
Fig.44. After powdering the specimen
7.

Results and Discussion

The E-waste cannot be decomposed easily and occupies more


land area. The dumping of e-wastes spoils the fertility of the soil
and causes a lot of hazards to the ecosystem. These wastes can
be reduced by properly recycling it and converting them into
some useful products. The PCB is powdered and added to the
bonding materials like epoxy resin, epoxy hardener and white
colour pigment to prepare the cylindrical and square shape
specimens and is used to analyze the mechanical and ecological
behavior of washbowl. The specimens are prepared by using
50% of PCB powder and 50% of bonding material
The mechanical property of the specimen is
Tensile strength
: (25 30) N/mm2
Compressive strength
: (91 92) N/mm2
Hardness
: (80 88) HR L
SCALE
7.1. The Ecological Property of the Specimen

Cheng-Hsien Wua, Yu-Rey Panb June 2008 The study


of injection/compression liquid composite molding.

[2]

K. Moria, M. Umedab, M. Murakamia April 2005


determination of material constants and prediction of
fracture in sintering of ceramic products.

[3]

. Aksoy, Z. Ertrk, K. ztrk, H. Celtic, J.S. Ward,


March 2007 A study of ergonomic factors in
washbasin design.

[4]

Peeranart Kiddee, Ravi Naidu, Ming H. Wong 2013


Electronic waste management approaches: An
overview, Waste Management 33 12371250.

No appreciable change in weight and pH value, There are only


negligible amount ions transformation only takes between the
specimen and the medium. Corrosion takes in the hard water
medium only that too negligible amount, There is no change in
the odour while before and after testing.
8.

Conclusion

The Washbowl is fabricated from the results of the test


specimens which is made from the Printed Circuit Board
powder, Epoxy Resin, Epoxy Hardner and white color pigment
it helps us to achieve low cost and weight when compared to the
ceramic washbowl. The washbowl is fabricated by using 50% of
PCB powder and 50% of bonding material. The selected ratio is

Page 208

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SUSTAINABLE TRANSPORATION PLANNING IS KEY TO EMISSIONS AND ENERGY


CONSERVATION
P.Ponnurangam*
Assistant Professor, S.A. Engineering College, Chennai 600 077, India, yabas123@gmail.com

Abstract
The Transport Sector of India is largely dependent on the Energy resources accounting to nearly 11% of its total primary
energy use which is predicted to reach 20% by 2030. Fuel usage in transportation sector is bound to increase by 87% in
20 years with a decrease in the available diesel stock and an alarming increase in the demand. Also the growing vehicle
population has increased the contribution of Vehicular Pollution to the Urban Air Pollution from 60% in the year 1990 to
90% in 2010. In order to eliminate these risks the interaction between these sectors should be studied in detail. Since
there was a lack of interdisciplinary works involving the sectors of Transport Energy and Emissions interaction based
on a Systems Approach this study is focused on carrying out an in depth study of the same based on Systems Dynamics
principles. The major objective of the work is to study and appreciate the existing energy and emissions scenarios in
Chennai city, to procure data through inventory on energy requirement and emission standards from transportation
sector and to build a System Dynamics (SD) model using STELLA simulation software to determine the Energy
requirement and Emissions levels from the transport sector in the year 2026.
Keywords: Transportation Energy; Transportation Emission; Demand and supply of Energy consumption, Emission
standards; Sustainable Planning; Simulation Modeling.

1. Introduction
India is the sixth largest energy consumer in the world. The
transport sector of the country accounts to nearly 19% of global
energy use which is projected to reach 50% by 2030. Studies
disclose the fact that at the present rate of economic growth, energy
needs may increase by 16% and more with respect to the present
scenario. This increase in energy use leads to an increase in Carbon
di-oxide emissions also. It has been projected that the greenhouse
gases emission will increase at a rate of 7% per annum. Hence
proper planning measures should be adopted to reduce the
increasing energy demand and concomitant reductions in
emissions.

1280Mtoe in 2030, which is about 3.5%. This demand is higher


than the World average of 1.6%. As a result, India will become
more dependent on imports and hence will be exposed to global
supply risks and external price shocks casting a shadow on the long
term sustainability, economy and security of its energy
requirements. Hence a collective effort and unified policy direction
is the need of the hour to face the challenges.
1.2. Objectives of the Study

1.1. Need for the Study


In 2005, Indias transport sector consumed 11% of its total primary
energy demand (16.9 % of commercial energy supply). 78 % of
this demand was consumed by Road Transport, 11% by Aviation,
10% by rail transport and 1 % by Inland water. The transport sector
is set to grow at over 6% per annum on the back of rising economic
activity and a rapid surge in the vehicle stock. By 2030, the share
of transport sector is likely to double to about 20% of the primary
energy demand. Globally, the share of the Indian transport sector is
likely to triple from its low of 2% in 2005 to about 6% in 2030.
The growth of transport sector, primarily driven by road transport
will remain heavily dependent on the availability and affordability
of oil. For the energy requirement of the country as a whole, coal
will however remain a dominant fuel with its share of 48% of total
energy needs in 2030, followed by oil at 25%. The total demand of
primary energy is likely to go up from 566Mtoe in 2006 to

To study and appreciate the existing energy scenarios in


Chennai city.
To collect secondary data on energy requirement in
transportation sector and level of emissions in
transportation sector.
To build a System Dynamics model using STELLA
software that would address transport, energy and
emissions interactions.
To determine the Energy demand levels from the
transport sector in the year 2026.
To suggest appropriate measures that would ensure
developments towards achieving sustainability in
transportation planning.

2. Review of Literature
Several researches have been conducted to study the trend of
energy scenarios in Chennai city with regards to Transportation and
energy sector. It is time to look back at the past to know what has
been achieved and what new techniques have been developed so
far. Thus related works in this field of study have been reviewed
and are presented below.

Corresponding author. Tel.: 09952608829


E-mail: yabas123@mail.com

Page 209

P.Ponnurangam / INDECS-15

2.1.1. Transportation and Energy Interaction


Beimborn (2005) in his paper studied about the Transportation
energy: its supply, demand and the future. An energy model was
built to determine the energy demand for the year 2000 to 2050.
Population, energy intensity and GDP were taken as input, the
values were projected for the 50 years and the results were found.
Population was predicted to increase by 50%. This leads to an
increase in GDP of about 300%. Demand for petroleum was found
to increase by 91%. A decline of 1.11 % in energy intensity led to a
total energy demand increase by 172% and 42% decline in
hydrocarbon production. Stock of petroleum is declining at a faster
rate and a marketed imbalance between supply and demand was
observed. The author has concluded that there is no easy long term
solution, but only a combination of thousands of actions like Price
increase, Conservation, use of alternative Fuels like Natural gas,
LPG, CNG, Methanol, Ethanol etc., Electricity, Hydrogen and Biodiesel can lead towards a sustainable transportation.
Kibira et al (2011) in their study System Dynamics Modeling of
Corn Ethanol as a Bio Transportation fuel in the United States
modelled using system dynamics the use of Corn Ethanol as a bio
transportation fuel in the United States. The authors say that
continued strong growth of the corn ethanol industry will depend
on profitability by both suppliers and producers. This in turn will
be influenced by several factors such as demand, government
incentives, feedstock availability and prices, processing plant
capacity, and efficient farm and ethanol processing technologies.
Therefore they use system dynamics modeling to construct the
causal-loop structure of the corn ethanol industry and a stock and
flow diagrams to explore how changes in projected factors will
affect the industry. This paper explores different possible growth
scenarios of the industry for the next twenty one years. The paper
has described a system dynamics model that can be used to study
the state of corn ethanol fuel industry in the U.S., and the effect of
changes to the forecast future trend on the operating parameters.
Kwon and Hyeong (2009) did A System Dynamics Model of
Alternative Fuel Vehicles Market under the Network Effect.
According to the system dynamics model of this study, if there is a
significant network effect on vehicle operating costs, it is difficult
to achieve the shift to AFV even in the long term without a policy
intervention because the car market is locked in to the current
structure. Network effect was caused by an increasing return to
scale in fuel supply sector as well as in maintenance service sector.
It was also related to the fact that the reliability and awareness of
consumers on new products increases with the growth of the
market share of the new products. The authors say that if there was
a significant network effect for vehicle operating costs, it would
have been difficult to achieve the shift to AFVs even in the long
term without a strong policy intervention because the car market is
locked in to its current structure. Network effects can be caused by
increasing returns to scale in the fuel supply sector as well as in the
maintenance service sector. They are also related to the fact that the
reliability and awareness of consumers of new products increases
with the growth of the market share of the new products. AFVs
which rely on the fuel supply infrastructure of conventional cars
such as hybrid cars should have less of a network effect for
operating costs. These vehicles would have less difficulty in
increasing their market share than AFVs which require a
fundamentally different fuel supply infrastructure from

conventional cars (e.g. hydrogen fuel cell cars). Therefore the


authors concluded that, it could be effective to put a policy focus
on the former type of AFVs (with less of a network effect: same
fuel supply infrastructure as conventional cars) even if the latter
type of AFVs (with a significant network effect: different fuel
supply infrastructure from conventional cars) have more
environmental benefit in the long term, by the time the cost gap
between conventional cars and AFVs decreases substantially.
Mi and Zhou (2009) in their paper studied Transportation, energy
consumption and carbon emission in china. The authors says that
road transport plays a dominant role while railway transport has
been in short supply in Chinese inter-city \transportation system.
Rail transportation is the most energy-efficient as for the unit
energy consumption of t transportation. Fossil fuels such as coal,
oil, and gas are the major source of world energy, accounting for
more than 80 % of the total energy produced. The distribution of
population is uneven in China because 94% of the population is
beneath the southeast of the Heihe-Tengchong line. Rail
transportation becomes the first choice of the citizens for its large
capacity, low energy consumption, safety and economic under such
situation. Because of the inadequacy of the railways transport
capacity, the new transportation demand mainly depends on the
roads, which leads to high energy consumption, carbon emission
and severe environment pollution. Provided the cargo turnovers
research to 40% of the total turnover, 210 million tons of coal will
be saved and 400 million tons of CO2 will be reduced. China has
invested huge amount of money in the construction of high speed
railways. However, both the operating mileage and the
passenger/freight turnovers are lag behind the roads. The low
efficiency of the sector monopoly is an important cause. The
government should learn from the development mode of the
highway and the aviation to achieve the diversification of
construction-operation and permits the local and private capital
to enter into the railway market to enhance the line coverage and
transport capacity.
2.1.2. Ttransport and Energy Sector in India
Kumar et al (2000) in his study National Energy Map for India:
Technology Vision 2030 examined the role that various
technological options could play under alternative scenarios of
economic growth and development, resource availabilities, and
technological progress. For this purpose, they have chosen an
integrated energy-modelling framework that would facilitate the
creation and analysis of various scenarios of energy demand and
supply at the national level, as well as provide a detailed
representation and analysis at the technological level for each
category of resource as well as sectoral endues demand. After an
extensive survey of existing models, the MARKAL (MARKet
ALlocation) model was selected to examine the pathways for
optimal energy supply to meet the endues services in the five
energy-consuming sectors (agriculture, commercial, residential,
industrial, and transport) under various scenarios. MARKAL is a
dynamic linear programming representation of a generalized
energy system. Exogenously estimated useful energy demands
were provided to the model over a modelling time frame of 2001
31 using population and GDP as the key drivers to growth. Seven
alternative scenarios were set up against the business-as-usual
(BAU) scenario to examine variations with regards to economic
growth and technological development.

Page 210

P.Ponnurangam / INDECS-15

The total energy consumption in the transport sector was found to


increase by about 14 times from 43 Mtoe in 2001 to 461 Mtoe in
2031. By targeting action on the demand as well as supply sides in
the transport sector by enhancing the use of public transportation a
reduction of about 190 Mtoe of energy can be achieved in 20201.
Introduction of cleaner fuels such a low sulphur diesel, ethanol
blending, and bio-diesel are recommended by the author to reduce
the consumption of petroleum products and thereby their import
dependency.
Ramachandra and Shwetmala (2009) in their paper synthesised
about The emissions from Indias transport sector: State-wise. In
this study, a decentralized emission inventories were prepared for
road transport sector of India in order to design and implement
suitable technologies and policies for appropriate mitigation
measures. The author have stated that Globalization and
liberalization policies of the government in 90s have increased the
number of road vehicles nearly 92.6% from 19801981 to 2003
2004. These vehicles mainly consume non-renewable fossil fuels,
and are a major contributor of green house gases, particularly CO2
emission. In this paper they have focussed on the state wise road
transport emissions (CO2, CH4, CO, NOx, N2O, SO2, PM and
HC), using region specific mass emission factors for each type of
vehicles. The country level emissions (CO2, CH4, CO, NOx, N2O,
SO2 and NMVOC) are calculated for railways, shipping and
airway, based on fuel types. In India, transport sector emits an
estimated 258.10 Tg of CO2, of which 94.5% was contributed by
road transport (20032004). Among all the states and Union
Territories, Maharashtras contribution is the largest, 28.85 Tg
(11.8%) of CO2, followed by Tamil Nadu 26.41 Tg (10.8%),
Gujarat 23.31 Tg (9.6%), UttarPradesh 17.42 Tg (7.1%), Rajasthan
15.17 Tg (6.22%) and, Karnataka 15.09 Tg (6.19%). These six
states account for 51.8% of the CO2 emissions from road transport.
The total CO2 emission for Indian transport was 258.10 Tg
in20032004. Among all type of transport, road and aviation were
first and second major contributor of air pollution. The road
transport sector has contributed 94.5% and 53.3% of total transport
emission of CO2 and CO. Among all the states and UT,
Maharashtras contribution is the largest, 28.85 Tg (11.8%) of
CO2, followed by Tamil Nadu 26.41 Tg (10.8%), Gujarat 23.31 Tg
(9.6%), Uttar Pradesh 17.42 Tg (7.1%), Rajasthan 15.17 Tg
(6.22%) and, Karnataka 15.09 Tg (6.19%). The total of these six
states accounts for 51.8% of the CO2emissions from road transport.

3. Study Area
Chennai is the fourth largest metropolitan city of India which
covers an area of 426 sq.km and recorded a population of 46.81
lakhs in 2011. The Chennai Metropolitan Area which extends over
an area of 1189 sq.km recorded the population of 86.96 lakhs in
2011 and the density is 11,000 per sq.km. The population of
Chennai in 1639 was 40,000 and today the city is estimated to have
a population of 7.5 million, which gives a population density of
about 6482 per sq.km. This rapid increase in population leads to
traffic congestion and imbalanced supply and demand of transport
facilities. Thus it is imperative to study a transportation and energy
interaction in Chennai city and to provide an early warning on the
robustness of both present and future situation. Fig.1 depicts the
index map of the study area.

Chennai City
CBD

Fig.1 Index Map of Study Area

3.1. Software Used


The model of transport and energy interaction using the System
Dynamics (SD) has been implemented in the STELLA
environment (STELLA 9.1 package). The modeling tool which is
an object-oriented simulation environment allows the development
of Transport Energy interaction models with significantly less
effort than using traditional programming languages. It has a userfriendly graphical interface and supports modular program
development. Using this tool, the modeler defines objects
representing physical or conceptual system components and
indicates the functional relationships among these objects. Building
on these strengths, the general architecture of a Transport and
Energy Interaction model will be described.

4. Methodology
The methodology for the model development and analysis of the
same has been provided in the form of a flow chart in Figure 2.
Carrying out a review of various literatures in the study area is the
initial work required. Based on the review, a methodology to
analyse the work to be carried out has been figured. The need for
study and the primary objectives to be fulfilled have been
established as given in the earlier chapters. The analysis starts with
the data collection required for the study. As far as this study is
concerned, the data required is only secondary data which has been
collected from various journals and reports pertaining to the data.
After the model conceptualisation has been carried out, model
building, analysis and testing the model for various scenario
options has been carried out. Based on the results obtained suitable
recommendations for policy options have been given.

Page 211

P.Ponnurangam / INDECS-15
Need for the Study

Formulation of Objectives

Selection of Study area

Collection of Data
Secondary data

CCTS, Wilbur Smith Associates, Aug 2010

Energy and Emission data from various


literatures

Projected/Expected Modal split form Second


master plan

Primary data

Relevant Information on Transport and


Energy and Emission Sector and Service
volume levels on urban roads

Existing Transportation Modal mix

Existing Levels of Service

SD Model Development using STELLA


Model Calibration
Model Behavior
Model Evaluation & Scenario Analysis
Results, Inferences & Recommendations
Fig.2. Study Methodology Chart

5. Data Collection
5.1. Population Sector
Total birth and death rates are increasing from decade to decade but
their rates decreasing. The registered birth rate in Chennai City
segment was 24.06 in 1991 and reduced to 22.62 in the year 2003.
The death rate also declined from 9.20 in 1981 to 8.01 in 2003.
However, net natural increase in population had been decreasing
from year to year from 22.00 in 1981 to 14.61 in 2003. According
to 2001 census, migrants to Chennai city from other parts of Tamil
Nadu State constitute 74.5 percent. The growth of in-migrant
population shows a declining trend from 36.80 percent in 1961 to
21.57 percent in 2001. Migrants from other parts of India constitute
23.8 percent and the remaining 1.71 percent of the migrants is from
other countries. Table.1depicts the migration details of the Chennai
city.
5.2 Transportation Sector
With respect to the transport sector, the data which are of
prime concern are vehicle population and the existing Modal Split
in the city which have been obtained from the Second Master Plan

for Chennai, CMDA report. Statistics of Transport Department,


Chennai which gives the data regarding the total number of
commercial and non-commercial vehicles in the city and the data
considered for this study have been given in Table.2. The MTC
buses with a fleet size of around 3500 buses cater to 36 lakhs
trips/day.
5.3 Energy Sector
The Energy sector the fuels which were of prime concern were
Petrol and Diesel. Hence data pertaining to the fuel consumption
per day by Petrol and Diesel driven vehicles have been collected
from earlier studies and presented in the Table.3 and the respective
Fuel Efficiency values have been used in the model analysis.
5.4 Emission Sector
The Emission sector the emission factors for trace gas emissions
from vehicles have been collected from earlier studies and
presented in the Table.4 and the respective emissions values have
been used in the model analysis.

Page 212

P.Ponnurangam / INDECS-15

Table 1. Migration to Chennai City (in Lakhs)

Total Migration in the City

1961

17.49

4.47

69.45

1.71

26.6

0.25

3.90

6.44

Percent of
Total
Migrants in
the Total
Population
36.80

1971

26.42

5.51

70.61

2.00

25.63

0.29

3.76

7.80

29.52

1981

32.84

7.19

71.28

2.55

25.31

0.34

3.41

10.08

30.70

1991

38.43

6.44

70.51

2.42

26.47

0.28

3.01

9.18

23.90

2001

43.43

6.98

74.49

2.23

23.80

0.16

1.71

9.37

21.57

Total
Population

Year

Other Parts of
Tamil Nadu
No.
Percent

Other Parts of
India
No.
Percent

Other Countries
No.

Percent

Total

Source: www.cmdachennai.gov.in
Table 2. Vehicle Population in Chennai City in the 2006 to 2011

Vehicles / Year

2006

2007

2008

2009

2010

2011

Buses

2803

3084

3260

3280

3421

3464

0.11

Auto Rickshaw

41316

39330

51113

44973

49062

63640

Taxi

283

284

1165

1252

1259

1268

Private Bus

883

926

2376

874

2702

2906

Mini Bus

902

961

1709

1129

2095

2217

Motor cycles

6.72

7.86

8.96

10.41

13.71

15.63

Scooters

2.86

2.98

3.12

3.20

3.33

4.03

Mopeds

4.69

4.76

4.82

4.90

4.97

6.15

Two Wheelers

14.27

15.60

16.90

18.51

22.01

25.81

Cars

3.35

3.66

4.00

4.41

4.82

5.80

Total (in lakhs)

18.08

19.71

21.5

23.43

27.41

32.34

2.00

0.16

97.73

100

Source: www.tn.gov.in
Table 3. Average Fuel Consumption by Different Classes of Vehicle

Average Distance Covered

Fuel Efficiency

Fuel Consumed

Two Wheelers

(km/day)
18

(km/l)
53

(Litres/veh/year)
124

Three Wheelers

96

Cars

22

Type of Vehicle

Mini Bus (Diesel)

22

Bus

151

21 (Petrol)
13.5(Petrol),

1669
593

14.0(Diesel),

571

13.0(LPG),

618

15.0(CNG)

535

8.7
4.1(Diesel)

897
13415

Source: Report of the Expert Group, Government of India Report, February, 2010

Page 213

P.Ponnurangam / INDECS-15

Table 4. Trace Gas Emissions Factors for Vehicles

Pollutant

Two

Auto

Wheeler

Rickshaw

515.20

26.6

3.60

3.60

NOX

12.00

CH4

Bus

Omni / Bus

Cars

Taxi

CO2

515.20

60.30

223.60

208.30

CO

2.20

5.10

1.98

0.90

12.00

0.19

1.28

0.20

0.50

0.09

0.09

0.18

0.18

0.17

0.01

SO2

1.42

1.42

0.013

0.029

0.053

10.30

PM

0.56

0.56

0.05

0.20

0.03

0.07

HC

0.87

0.87

1.42

0.14

0.25

0.13

(g/km)

Source: Ramachandra, T.V., Shwetmala, Emissions from Indias transport


Sector: Statewise synthesis, Atmospheric Environment (2009)

6. System Dynamics Model Development


The system Dynamics model development process is summarized
in the schematic diagram of a model life cycle in Figure 3. The
modeling process starts with defining the purpose/goal of the
system. Then boundaries of the system to be modeled are specified.
This is followed by identification of key variables in the system
that affects the system, the most. Then behavior of the key variable
is described, the stocks and flows are indentified, and their
structure is mapped in the modeling tool using basic building
blocks. Quantitative information, i.e., equations and data, is
included in the model structure. The model is run to test the
behavior. The model is then evaluated and adjustments are made.
Once the model is replicating system behavior, it is ready for
simulation modeling. Figure 4. Depicts the overall casual loop
diagram of Transport, Energy and Emission interaction.

Define purpose of
system
Specify System
boundaries
Identify key variables of system

Describe behavior of key variables

Identify stocks and flows in the system

Map system structure in modeling

Run model to test system

Apply model for simulation


Fig.3 System Dynamics Model development Life Cycle

6.1 Population Sector


After the identification of key variables, the model for the
population sector has been developed based on the causal loop
diagram. The System Dynamics model for Chennai City segment
population sector is shown in Figure 5. The various parameters
considered for building the population sector model are as follows:
Fig. 4 Overall Casual Loop Diagram

POP
BR
DR

Total Population
Birth rate
Death Rate

Page 214

P.Ponnurangam / INDECS-15

IMR
OMR
BN
DN
IMN
OMN

In Migration rate
Out Migration rate
Birth Normal
Death Normal
In Migration Normal
Out Migration Normal

In this sector, Population of the Base Year, Birth Rate, Death Rate,
In Migration and Outmigration are considered. The size of the
population is influenced by both the net birth rate and net migration
rate. The net birth rate equals the total number of births per year
minus the total number of deaths. Similarly, the net migration rate
equals the number of in-migrants minus the number of out
migrants. However, the number of births and deaths as well as net
in-migrants can be defined as a yearly percentage of the population.
Hence, the population model contains one level and four rates as
stated below:

ADT
FE
FC

-Average distance travelled by the vehicle per day


-Fuel efficiency of the vehicle
-Total fuel consumed by the vehicle

The following vehicles were considered under various sectors for


the study. The major sectors taken into account were Public
transport vehicles (Bus), IPT vehicles (Auto Rickshaw and Taxi),
Personalised vehicles (Two Wheelers and Cars) and Other Vehicles
Sector (Private Bus and Mini Bus). The vehicle growth is assumed
from the past trend. Using the values of Average Distance travelled
per day and Fuel Efficiency in km/litre and the projected values of
vehicle population, the final value of Fuel consumption has been
determined. In this study, the total trips made by Rail contributing
to Public Transport sector have been analysed. Sub-Urban rail
system within the city operates on electricity and hence is devoid of
emissions. The demand of electricity by rail sector has not been
accounted in this research.

Population (t) = Population (t-dt) + (Birth_Rate + migration_Rate


Death_Rate - Out_Migration_Rate) * dt

IPT VEHICLES

TOT Auto

Chennai City segment population is the level variable. The level is


influenced by the Birth Rate, Death Rate, In-Migration and OutPOPULATION MODEL
Migration rate.

Total FC Auto
Auto Inc

ADT per day Auto


FE per Auto

~
GR Auto

DN

~
Auto TripsAuto TR

IMN
IMR

CAR VEHICLES

POP

DR

BR

Tot FC PC
Tot PC
ADT per day PC

Table 1
BN

OMR

Tot Cars

OMN

FE per PC

Graph 1

Car Inc
Tot FC DC

Fig. 5 Population Sector Model for Chennai City

6.2 Transport, Energy and Emission Sector


Based on the causal loop diagram the System Dynamics model
addressing the interactions between transport energy, emission
sector has been developed which is given in Figure 6. The various
parameters considered for building of transport sector model are as
follows:
TOT
GR

Tot DC

~
GR Cars

~
Car Trips

ADT per day DC


FE per DC

Car TR

-Total Vehicle Population


-Vehicle growth rate

Page 215

P.Ponnurangam / INDECS-15

SHARE OF TRANSPORT TRIPS

Two Wheeler Vehicles

Table 4

Auto Trips

Tot TW
Tot FC TW
Tot Trips

Bus Trips

Tot Public Trips


Car Trips

TW Inc

Tot Priv ate Trips

ADT per day TW


FE per TW

GR TW

TW Trips

TW TR
TW Trips

PUBLIC VEHICLES
FUEL CONSUMPTION

Tot bus

Total FC Auto

Tot FC bus

Tot FC bus
Bus Inc

Tot Public FC

ADT per day Bus


Tot FC

FE per bus

~
~

GR bus

Tot FC PC
Tot Priv ate FC

Bus TR
Tot FC TW

Bus Trips

Tot FC DC

Fig.6 Transport, Energy and Emission Sector Model for Chennai City

Page 216

P.Ponnurangam / INDECS-15

8. Recommendations
7. Results
This system dynamics models are tested with various scenarios
analysis and policy options. They are following:

Scenario I - Do Minimum (Allowing the existing trend to


continue)

Scenario II Partial Condition Scenario (Augmentation


of Public Transport and Restricting growth of
personalized modes)

Scenario III - Desirable Scenario (Augmentation of


Public Transport and Restricting growth of personalized
modes)
7.1 Scenario I Do Minimum
At the present growth rate of vehicles in Chennai, Public transport
would share 51% of the total trips whereas personalized transport
would share 49% of the total trips in the year 2026. Fuel Consumed
by the personal modes of travel viz. Petrol Driven cars, Diesel
driven cars and Two Wheelers are 12, 13.3 and 20.8 lakhs liters per
day respectively in the year 2026. This shows an alarming increase
of about 300% to 400% in the fuel consumption by personalized
modes of travel which would correspondingly increase the demand
for Petrol and diesel. Personalised modes of transport contribute to
the major share of pollution in the city amounting to 40% roughly.
7.2 Scenario II Partial Condition Scenario
In the Partial Efforts scenario (Scenario II), simulation has been
carried out such that minimal efforts are undertaken by the
government to achieve a Modal Split of 64:36 between Public and
Private Mode. In order to facilitate the above condition, the Public
transport has been augmented in a phase wise manner to reach a
growth rate of 7% by 2026 and simultaneously the growth rate of
Two Wheelers and Cars have been restricted to almost half of the
existing value i.e. 4% and 5.5% respectively in 2026. On
comparison with Scenario I, a remarkable decrease to 13% and
23% could be observed in Scenario II with regard to the trips made
by cars and two wheelers. This decrease leads to a modal split
value of 64:36 between public and private modes of travel.
Decrease of about 60% is noted in the emissions from the
personalised vehicles and emissions per capita in public transport
are much lesser than that observed in personalised vehicle sector.
7.3 Scenario III Desirable Scenario
Based on this scenario, the Modal Split between Public and private
modes of travel have been achieved as 70:30. This increase in
Public transport trips can be attributed to the introduction of Metro
Rail, Mono Rail, BRTS and increasing the ridership in MRTS.
When compared with Do Minimum condition, Fuel consumption in
personalized mode of travel is found to reduce by about 65% in the
case of Petrol driven cars, 65% in diesel driven cars and by 50% in
Two Wheelers. Totally 40% reduction in Energy has been achieved
by incorporating CNG by 50% and 2 lakhs Fuel cell vehicles. This
reduces the load on Petrol and Diesel demand. Adoption of EURO
IV emission norms in addition to achieving a modal split of 70:30
reduces the emissions from vehicles to a sustainable 50 90%.

From the study carried out, the following recommendations are


given.

There is an immediate need to restrict the growth of


personalised vehicles and augment the public transport.
This augmentation can be attributed to increasing the
MTC fleet size, introduction of Metro Rail and Mono rail
in the city.

In order to achieve the Modal Split of 70:30, the MTC


fleet size increases to 9000 efficient services. Also the
Metro rail and Mono Rail services should be
implemented on time and operated with maximum
efficiency.

The growth rate of Two Wheelers and Cars should be


restricted to 2% and 1.5% respectively in the year 2026
to achieve the above mentioned policy of the
government.

EURO IV emission norms should be made mandatory by


2018 if not at least by 2026. This reduces the emission
load from the vehicles.

Since there is no specification for CO2 emissions from


vehicles it is found to be on the rising trend. Hence the
government should take needful steps to provide
emission standards for CO2 as well and hence reduce the
contribution of CO2 emissions in the Greenhouse gases.

This policy leads to a considerable reduction in pollution


load of the city making it a place with reduced
environmental hazards.

Since the results of Partial scenario prove to be much


better than Do Minimum condition, it is recommended
that even if the desired condition of achieving a Modal
Split value of 70:30 cannot be done at least the
government should strive towards achieving a modal split
of 64:36.
Reference
[1]

Armenia S. A System Dynamic Energy Model for a


Sustainable Transportation System Anna University,
Chennai, 1995.

[2]

Azhaginiyal. A Transport, Energy, Emissions and Economy


Interaction - A Systems Approach Anna University,
Chennai, 2012.

[3]

Beimborn. E Transportation Energy: Supply, Demand and


the Future, UWM-CUTS Energy, ITE District 4 Meeting,
2005.

[4]

Banks. J Discrete- Event system simulation, second


Edition, Eastern Economy Edition, Prentice Hall of India,
New Delhi, 1996.

[5]

Ghosh. A Service Sector in India & Role of Foreign Direct


Investment, the Management Accountant, Vol. 46 No. 11,
2011; pp. 984-987.

[6]

Armenia S., Fabrizio B., Diego F. and Emanuel T.A System


Dynamic Energy Model for a Sustainable Transportation
System,1995.

Page 217

P.Ponnurangam / INDECS-15

[7]

Beimborn E., Transportation Energy: Supply, Demand and


the Future, UWM-CUTS Energy, ITE District 4 meeting,
2005.

[8]

Banks J. and Carson J. Discrete- Event system simulation,


second edition, Eastern Economy Edition, Prentice Hall of
India, New Delhi, 1996.

[9]

Ghosh A. (2011) Service Sector in India & Role of Foreign


Direct Investment,

[10]

The Management Accountant, Vol. 46 No. 11, pp. 984-987

[11]

Kadiyali.L.R. Traffic and Transportation


Khanna publishers, New Delhi, 1985.

[12]

Karandikar V. and Rana A.Future of energy options for


India for an interdependent world, 2006.

[13]

Kavin. J, Naveen Kumar. S and Raja. P.K., Macro Level


Simulation Model Building for Sustainable Transportation
Planning, Research project, Anna University, Chennai,
2009.

[14]

Kumar A., Shirish S. G. and Mahesh V. National Energy


Map for India: Technology Vision 2030, TERI press, 2000.

[15]

Poudenxa P. and Merida W., Energy demand and


greenhouse gas emissions from urban passenger
transportation in Canada, 2005.

[16]

Paul. H. A System Dynamics Model of Transportation


Energy Demand, the 1987 International Conference of the
System Dynamic Society, China, 1987, pp. 383-397.

[17]

Ramachandra, T.V., Shwetmala, (2009) Emissions from


Indias transport sector: State wise syntheses, Atmospheric
Environment doi:10.1016/j.atmosenv.2009.07.015

Planning,

[18] Report of the Expert Group on a Viable and Sustainable


System of Pricing of Petroleum Products, Government of
India Report, New Delhi, February 2010.

Page 218

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SLOPE STABILITY ANALYSIS BY FINITE ELEMENT


APPROACH
1,

Photo of First
Author
JPEG Format

Barnali Ghosh S. K. Prasad,


E. P.C.E.T, Bangalore,560049, India, barnalighosh08@gmail.com
2
S. J. College of Engineering, Mysore, 570006, India, prasad_s_k@hotmail.com
1

econd Author (if


any)

Abstractin
Earthen
JPEGslopes
Format are beautiful and most important geotechnical structures. It is very important to maintain their stability.
Stability of earth slopes by Limit equilibrium approach is a familiar analysis. Presently, a number of geotechnical
software are being used for this purpose. This paper describes numerical analysis of slope stability problems by using
the software Slope/w (Geo-slope 2007) based on limit equilibrium principles. The software also uses finite element
analysis for the estimation of stress distribution in the sloping mass. The minimum factor of safety (FOS) and the
critical slip surface are determined using the MorgensternPrice method and the results are compared with those from
Bishops modified method, Janbus Simplified method and Mohr-Coulombs expression. A parametric study is made on
the influence of pore water pressure, cohesion, angle of internal friction, and unit weight on the factor of safety for slope
stability problems.
Keywords: Limit equilibrium, Factor of safety, Geo-slope, slope stability problems
2.
1.

Methodology

Introduction

The slope stability analyses are performed to find the


safe and economic design of slopes. In the analysis of
slopes, factor of safety is primarily used to determine
how near or far slopes are from failure. When this ratio
is more than one, resistive shear strength is greater than
driving shear stress and the slope is considered as stable.
When this ratio is nearer to one, shear strength is nearly
equal to shear stress and the slope is close to failure. If
Factor of Safety is less than one, the slope has already
suffered failure. Limit equilibrium types of analysis for
assessing the stability of earth slopes have been in use in
geotechnical engineering for many decades. The
software SLOPE/W (GEOSLOPE 2007) provides
solution to many methods such as: Bishops Modified
method, Janbus Simplified method, Spencer method,
Morgenstern-Price method (M-PM) and others.
SLOPE/W is a component of a complete suite of
geotechnical products called GeoStudio. SLOPE/W has
been designed and developed to be a general software
tool for the stability analysis of earth structures.
GEOSLOPE
is
developed
by
GEO SLOPE
International, Canada, based on limit equilibrium
principles. A finite element analysis is also introduced to
estimate the distribution of stresses in the sloping mass.

In this study, SLOPE/W has been used to compare the


accuracy and efficiency of most common Limit
Equilibrium based methods for the analysis of stability
of slopes..
General Limit Equilibrium Method
A general limit equilibrium (GLE) method encompasses
the key elements of all the methods available in
Slope/W. The GLE formulation is based on two factors
of safety equations. One equation gives the factor of
safety with respect to moment equilibrium, F m (equation
1) while the other equation gives the factor of safety with
respect to horizontal force equilibrium, Fl (equation 2).

The terms in the equations are c' for effective cohesion,


' for effective angle of friction, u for pore water
pressure, N for slice base normal force, W for slice
weight, D for Concentrated point load, for inclination
of slice base and , R, x, f, d are the geometric
parameters. F is Fm when N is substituted into the

Barnali Ghosh Tel.: 9845108358


E-mail: barnalighosh08@gmail.com

Page 219

Barnali Ghosh, S.K.Prasad. / INDECS-15


moment factor of safety equation and F is Fl when N is
substituted into the force factor of safety equation.
Ordinary method
The simplest form of factor of safety by Ordinary
Method in the absence of any pore-pressures for a
circular slip surface is given in the equation

MorgensternPrice method and the results are compared


with those of Bishops modified method and Janbus
Simplified method. The situations studied included,
(i) Stability analysis changing the slopes (H: V).
(ii) Stability analysis with changed values of cohesion
(C) and angle of internal friction ().
(iii) Stability analysis of dry slope.
(iv) Stability analysis with different piezometric level.
2.1 Stability Analysis with changing slopes

Here, c is the cohesion, is the slice base length, N is the


base normal (W cos ()), is the friction angle, W is the
slice weight, and is the slice base inclination
Bishop's simplified method
In 1950's Bishops simplified method was developed
which included inter-slice normal forces, but ignored the
inter-slice shear forces. A simple form of equation for
factor of safety in the absence of any pore water pressure
is by this approach is given by:

FS is on both sides of the equation as noted above. The


equation is not unlike the Ordinary factor of safety
equation except for the ma term, which is defined as

Janbus Method
Janbus simplified method is identical to Bishops
method, except that it satisfies only horizontal force
equilibrium. Like Bishops method, Janbus method
includes the interslice normal forces but not shear forces.
Morgenstern-Price Method
This method considers not only the normal and
tangential equilibrium but also the moment equilibrium
for each slice in circular and non-circular slip surfaces. It
is solved for the factor of safety using the summation of
forces tangential and normal to the base of a slice and the
summation of moments about the center of the base of
each slice. The equations were written for a slice of
infinitesimal thickness. The force and moment
equilibrium equations were combined and a modified
Newton-Raphson numerical technique was used to solve
for the factor of safety satisfying force and moment
equilibrium. The solution required an arbitrary
assumption regarding the direction of the resultant of the
interslice shear and normal forces.

In this paper, the minimum factor of safety (FS) and the


critical slip surface have been determined using the

The objective of the analysis was to compute the


minimum factor of safety and locate the critical slip
surface. Fig.1 presents the schematic diagram of the
slope with piezometric line considered 1m below the toe
level for the present study. The slope (2-horizontal: 1vertical) is having total height of 14 m over rock at 4m
below the base of the cut.

Fig. 1 Schematic diagram of a slope

The problem was first modeled in SLOPE/W using a


MohrCoulomb soil model without tension cracks and
solved using M-PM with half-sine inter-slice force
function. The problem was solved with same cohesion,
angle of internal friction and unit weight but with
different slopes. The corresponding Factors of Safety
(FS) were computed (Table 1). The Critical Slip Surface
corresponding to lowest FS obtained is shown in Fig.2.
Table 1: Factor of Safety with changing slopes

GEOSLOPE (SLOPE/W)
Slope stability
method used
M-PM

H:V
1.5:1.0
1.672
1.7:1.0
2.0:1.0

FS
1.232
1.806
1.320
1.484

Fig. 2 Critical Slip Surface for SLOPE/W Analysis using M-PM


with a slope of 1.5H: 1V

Page 220

Barnali Ghosh, S.K.Prasad. / INDECS-15


Table 3 : Factor of Safety with changing unit weight

GEOSLOPE (SLOPE/W)
3

Slope stability
method used

Unit weight (kN/m )

FS
Upper layer

M-PM

Lower layer

15.0
1.672
18 1.734
.0
20 .0

1.806
18.0
1.983
20.0
22.0

1.484
1.425
1.401

Fig. 3 Free body diagram and force polygon of a slice

2.2 Stability Analysis with varying cohesion (C) and


angle of internal friction ()
Analysis is carried out with a slope of 2:1(H:V), keeping
the unit weight same and piezometric line constant. The
values of cohesion and angle of internal friction are
varied. Unit weight for the upper layer is taken as 15
kN/m3 and that for the lower layer is taken as 18 kN/m3.
The entry and exit values of the slope are kept the same.
The corresponding FS was computed (Table 2). The
Critical Slip Surface corresponding to lowest FS
obtained is shown in Fig. 3.
Fig. 5 Critical Slip Surface for SLOPE/W Analysis using M-PM
with varying unit weight

Table 2: Factor of Safety with changing C and

GEOSLOPE (SLOPE/W)
Slope stability
method used

M-PM

C (in kN/m2) and (in


degrees)
Upper
Lower
layer
layer

5 & 10
1.672
5 & 10
1.734
5 & 20

5&
1.806
20
10 1.983
& 20
10 & 25

2.4 Stability Analysis with a dry slope


FS
1.023
1.228
1.484

Analysis is done with a slope 2:1(H:V) and keeping the


slope, unit weight as same and piezometric line constant
we have changed the values of cohesion and angle of
internal friction. Unit weight for the upper layer is taken
as 15kN/m3 and for the lower layer is taken as 18kN/m3.
The entry and exit values of the slope were kept as same.
The corresponding FS was computed (Table 4). The
Critical Slip Surface corresponding to lowest FS
obtained is shown in Fig.6.
Table 4: Factor of Safety with changing C and for a dry slope

GEOSLOPE (SLOPE/W)
Slope stability
method used

M-PM
Fig. 4 Critical Slip Surface for SLOPE/W Analysis using M-PM
with a varying C and

C and
Upper
layer

Lower
layer

5 & 10
1.672 5 &1.806
20
5 & 10
1.734 10 &1.983
20
5 & 20
10 & 25

FS
1.139
1.395
1.685

2.3 Stability Analysis with varying unit weight ()


Analysis is done with a slope 2:1(H:V) and keeping the
slope, cohesion and angle of internal friction as same and
piezometric line constant we have changed the values of
unit weight. Entry and exit values of the slope were kept
as same. The corresponding FS was computed (Table 3).
The Critical Slip Surface corresponding to lowest FS
obtained is shown in Fig. 5.
Fig. 6 Critical Slip Surface for SLOPE/W Analysis using M-PM
with changing C and for a dry slope

Page 221

Barnali Ghosh, S.K.Prasad. / INDECS-15

2.4 Stability Analysis with different piezometric level


Analysis is done with a slope 2:1(H:V) and keeping the
slope, unit weight cohesion and angle of internal friction
as constant every time we have changed the water level.
Unit weight for the upper layer is taken as 15KN/m3 and
for the lower layer is taken as 18 kN/m3. Cohesion for
the upper layer and lower layer is taken as 5 and 10 and
angle of internal friction for upper and lower layer was
taken 20 and 25 respectively. The corresponding FOSs
were computed (Table 5). The Critical Slip Surface
corresponding to highest FS obtained is shown in Fig.7.
Table 5: Factor of Safety with changing piezometric level

GEOSLOPE (SLOPE/W)
Water level
Slope stability
above the toe of
method used
the slope
M-PM
1.672
1m
1.806
1.734
2m
1.983
3m
4m

FS

1.450
1.462
1.485
1.524

Fig. 7 Critical Slip Surface for SLOPE/W Analysis using M-PM


with Water level 4m above toe

3. Comparison with other methods and M-PM


Comparison is made with Ordinary method, Bishops
method and M-PM for a slope 2:1(H:V), Unit weight for
the upper layer is taken as 15 kN/m3 and that for the
lower layer is taken as 18 kN/m3. Cohesion for the upper
layer and lower layer are taken as 5 and 10 kN/m2 and
angle of internal friction for upper and lower layer ares
taken 20 and 25 degrees respectively The entry and exit
values of the slope were kept as same. The
corresponding FS was computed (Table 6). The Critical
Slip Surface corresponding to lowest FS (Janbus
method) obtained is shown in Fig. 8.
Table 6 : Factor of Safety with different methods

Slope stability method


used

FS

Janbu
Ordinary
Bishops
M-PM

1.297
1.386
1.483
1.484

Fig. 8 Critical Slip Surface for SLOPE/W Analysis using Janbus


method

4. Concluding Remarks
The scope of this study was to compare various stability
evaluation methods using GEOSLOPE. Firstly, the factor
of safety with different slopes was calculated and found
that the maximum stability was achieved with higher H:
V ratio (flatter slope), increased cohesion and angle of
internal friction. Further, it was observed that the slope
was stable under dry condition than that piezometric
line. Besides, it was inferred that stability of slope
gradually decreased when the water level increased. The
Factor of safety obtained by different methods was
compared and it was found that Janbus method gave the
least factor of safety, whereas M-PM produced the
highest factor of safety.
The analysis of slope stability problems by using the
geotechnical engineering software SLOPE/W provides
global understanding of the problem with a possibility to
view the detailed forces on each slice and to understand
failure mechanisms. From the present analysis, it is
found that the four parameters studied, namely shear
parameters, height of slope, inclination of slope and unit
weight of sliding mass have significant influence on the
slope stability.

References
[1] Bishop, A. W. (1955). The use of slip circles in stability
analysis of slopes. Geotechnique, 5(1), 7 17.
[2]

GeoStudio (2004). Tutorial Manual,


International Ltd, www.geo-slope.com.

GEO-SLOPE

[3]

Janbu, N. (1968). Slope Stability Computations.


(Geoteknikk, NTH). Soil Mechanics and Foundation
Engineering, Technical University of Norway.

[4] Krahn, J. (2004). Stability Modeling with SLOPE/W. An


Engineering Methodology, Published by GeoSlope
International.
[5] Morgenstern, N. R. and Price, V. E. (1965). The Analysis of
the Stability of General Slip Surfaces. Geotechnique,
15(1), 77 93.
6] D.G Fredlund, and J. Krahn, Comparison of slope stability
methods of analysis, Canadian Geotechnical Journal. Vol.
14, No. 3, pp. 429-439. 1977

Page 222

Barnali Ghosh, S.K.Prasad. / INDECS-15


[7] D.G. Fredlund, J. Krahn, and D.E. Pufahl, The

Relationship between Limit Equilibrium Slope Stability


Methods, Proceedings of the International Conference on
Soil Mechanics and Foundation Engineering, 1981 .
[8] J.M. Duncan, State of the art: Limit equilibrium and finite
element analysis of slopes, Journal of Geotechnical
Engineering, Vol. 122, pp. 577-596, 1996.

[10] SLOPE/W., An Engineering Methodology", Geo-Slope


International, Calgary, Canada, 2007.
[11] K.E. Petterson, The early history of circular sliding
surfaces". Gotechnique, 5: 275296, 1955.
[12] W. Fellenius , Calculation of the stability of earth dams,
In Transactions, 2nd Congress on Large Dams,
Washington, Vol. 4, pp. 445, 1936.

[9] E. Spencer, A method of analysis of embankments


assuming parallel inter-slice forces, Geotechnique, Vol.
17, No. 1, pp. 11-26, 1967.

Page 223

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

SIMULATION MODELING OF SUSTAINABLE INFRASTRUCTURES


DEVELOPMENT IN PERI URBAN AREAS
- A CASE STUDY OF CHENNAI
Abirami Krishna. A1 , Devi Priyadarisini. K2, Dr.Umadevi. G3
JPEG Format

M.E Student,Anna University, Chennai, 600025, India, abiramikrishna92@gmail.com


Research Scholar, Anna University, Chennai, 600025, India, devpriya910@gmail.com
3
Professor, Anna University, Chennai, 600025, India, gudevi11@gmail.com

Third Author (if


any)
in
Abstract
JPEG Format

Peri urban areas refer to the settlements beyond or about cities. The population growth rate of peri urban areas of Chennai is
3.5 compared to the city growth rate of 1.25. The current population of Chennai Metropolitan Area is about 8.7 million and
estimated to reach about 12.5 million by 2026. Majority of this increase is expected to settle in suburban and peri urban areas
as the city is already reaching saturation. Fundamental root cause for traffic congestion is improper neighbourhood planning
which does not account the holding capacity of basic infrastructures and its impact on transportation infrastructure. So
planning should be done to estimate holding capacity based on all basic infrastructures.
The main objective is to develop a system dynamics simulation model for peri urban areas. The developed model will be used
for assessing its possible and preferable growth directions through various scenario analyses based on economic considerations
and suggest the best scenario to ensure a growth towards sustainable development in the long run. The outcome of the study is
evolving optimum density norms by considering basic infrastructures such as transportation, water supply, sanitation,
electricity and telecommunication. Analysis would be done towards achieving dynamic balancing between holding capacity of
land use development and carrying capacity of transportation network of study zones. In do minimum condition, the existing
trend of growth rates has been allowed to continue till 2026. For basic infrastructures sector, projects under construction are
accounted. In desirable scenario simulation has been carried out such that supply augmentation and demand management is
incorporated. For basic infrastructures sector, all new proposals are introduced to make Demand/Supply ratio less than 1 with
quality of life as criteria. In land use sector a density of 259 persons/hectare is achieved by accelerating the land use
intensification trend on par with the population increase.
Keywords: Peri Urban, Sustainable Infrastructure , System Dynamics
growth in the worlds urban population is taking place in the
1. INTRODUCTION

cities of the developing countries and most of this urban


growth is occurring on the metropolitan fringe. Nearly

Over the last 20 years, the urbanization rates in the


advanced industrialized countries have stabilized, and in
many countries declined but contrary to that the urban
population growth in the low-income and developing
countries has risen dramatically. Nearly, all of the future

percentage of the worlds total urban population lives in


informal settlements, many located on the urban fringe.
While the overall increase in urbanization have been
striking, considerable variations exist between the different
regions of the world. Chennai city is literally bursting and
there is inadequacy of proper housing and other basic

Page 224

Abirami Krishna et. al. / INDECS-15

infrastructures for thousands of people. The urban area

3.OBJECTIVES OF THE STUDY

problems of traffic congestion, pollution, lack of basic


infrastructure facilities and environmental degradation also

haunts the life of city people. Life of people in Chennai is

infrastructures of peri urban areas to build appropriate

filled with unpleasant experiences on a daily basis. High

simulation model

living costs, extended travel time of daily trips, overloaded

public transportation

and

systems, lack of proper basic

To study the transportation network and other basic

To test and validate the model for various scenario


policy

infrastructure, water scarcity, pollution, health problems,

development.

cramped spaces, absence of open spaces are some common

problems faced by the people of Chennai city.

sustainable

options

for

ensuring

sustainable

To recommend the best scenario for achieving


transportation

and

infrastructure

development to enhance quality of life in the study area


Unplanned and unrealistic growth of built up areas
inclusive of roads, water supply, electricity, drainage,

4. SCOPE OF THE STUDY

telecommunication, population density limits, insufficient


mass transport facility are the major causes for the problems

In this study, a System Dynamics model relating


Transport, land use and infrastructure sectors have been

faced by the urban dwellers.

built. Data relating to vehicular growth and population


in the present scenario has been collected and simulated

2. NEED FOR STUDY

to find the future demand and supply. The holding


The growth in peri urban areas is very rapid, land use
changes occur in a large scale from agricultural to
residential and commercial. The population growth rate of
peri urban areas of Chennai is around 3.5 compared to the
city growth rate of around 1.25. Unregulated growth,
haphazard

developments,

low

quality

housing

and

inadequate infrastructure facilities are synonymous with peri


urban areas of Chennai. The current population of Chennai
Metropolitan Area (CMA) is about 8.7 million and is
estimated to reach about 12.5 million by 2026. Majority of
this increase is expected to settle in suburban and peri urban
areas as the city is already reaching saturation. Lack of
proper transportation and other various infrastructure
facilities in the peri urban areas leads to people migrating to
places with proper transportation facilities and other
infrastructure facilities. This can be prevented by providing
the proper needs of the people but at the same time proper
and planned growth have to be carried out with a check on
population density norms for the peri urban areas for a
sustainable growth.

capacity of the land use for present and future demand


is simulated. The level of infrastructure facility such as
water and electricity from the present and future
simulated levels of population has also been identified.
The model has been subjected to various scenario
analysis aimed at studying the interaction between
Transport, landuse and infrastructure of the study area
under various conditions. The impacts of the policies
framed by the government to achieve sustainability in
transport and infrastructure with respect to economic
development have been analyzed at a macro level. A
comparison of the results under various scenario
options has been carried out to determine the
advantages of the policy measures to be adopted. Based
on the results, suitable recommendations of policies
have been provided. The impact of Transport, landuse
and infrastructure on the Economy of the nation can
also be studied. The impact of these scenarios quantify
the quality of life has been analyzed in this study
5
SYSTEM
DEVELOPMENT

DYNAMICS

MODEL

Page 225

Abirami Krishna et. al. / INDECS-15

algebraic relationships, graphs and Tables. For ease of


Generally the model building process can be
divided into two phases namely the Conceptual phase

presentation, the symbols used for flow diagramming of


System Dynamics are presented in the Figure 2.

and the Technical phase. The various phases in model


building process are as shown in Figure 1 and each phase
with its characteristics is presented here. The first phase
of problem definition involves recognizing and defining
a problem. System Conceptualization is to commit to
paper the important influences believed to be operating
within the system mostly in the form of causal-loop
diagram. In the next phase, models are represented in the
form of computer code that can be fed into the computer
in the form of computer programming languages. In the
Model Behaviour phase, computer simulation is used to
determine how all the variables within the system behave
over a period of time. Numerous tests must be performed

Figure 2. Flow Diagramming Symbols

on the model in the Model Evaluation phase to evaluate

The concept of the stocks and flows in System

its quality and validity. In the final phase, the model is

Dynamics is very appropriate to deal with a transport energy

used to test alternative policies that

economy interaction modeling. The System Dynamics

might be

implemented in the system under study.

approach is based on theory of feedback processes. A


feedback system is influenced by its own past behaviour.
This system has a closed loop structure that brings result
from past interactions of the system. The System Dynamics
model development process is summarised in the schematic
diagram of a model life cycle in Figure 3. The modeling
process starts with defining the purpose/goal of the system.
Then boundaries of the system to be modelled are specified.
This is followed by identification of key variables in the
system that affects the system, the most. Then behaviour of
the key variable is described, the stocks and flows are

Figure 1. Phases in Model Building Process

identified, and their structure is mapped in the modeling tool


using basic building blocks. The model is run to test the

The System Dynamics tool used in this study has four


basic building blocks namely Stock, Flow, Connector and
Converter. Stocks or levels are used to represent anything
that accumulates. Flows or rates represent activities that

behaviour. The model is then evaluated and adjustments are


made. Once the model is replicating system behaviour, it is
ready for simulation modeling.
5.1 Model Verification

increase and decrease stocks. Connectors are used to


establish the relationship among variables in the model

The purpose of model verification is to assure that the

represented graphically as arrows. Converters transform

conceptual model is reflected accurately in the computerized

input to output which can accept input in the form of

representation. The conceptual model quite often involves


some degree of abstraction about system operations, or some
Page 226

Abirami Krishna et. al. / INDECS-15

amount of simplifications of actual operations. It provides

5.3 Model Validation

answers to the questions like Is the conceptual model


(assumptions on system components and system structure

Validation is the overall process of comparing the

parameter values, abstractions and simplification) accurately

model and its behaviour to the real system and its behaviour.

represented by the operational model (i.e., by the

After the model has been calibrated using the original

computerized representation).

system data set, a final validation is conducted using the


second system of data set. If unacceptable discrepancies
between the model and the real system are discovered in the
final validation effort, the modeler must return to the
calibration phase and modify the model until it becomes
acceptable.

6. METHODOLOGY

The methodology of this work is depicted in the figure


5. Essentially the steps involved are data collection,
analysis, model building and to suggest appropriate
mitigation measures.
Figure 3. System Dynamics Model Development Life Cycle
5.2 Model Calibration
Calibration is the iterative process of comparing the
model to the real system, making adjustments (or)
manipulations (or even major) changes to the model,
comparing the revised model to reality, making additional
adjustments comparing again and so on. The model
calibration process can be pictorially represented as in
Figure 4.

Figure 4. Iterative Process of Calibration of a Mode

Page 227

Abirami Krishna et. al. / INDECS-15

7.

STUD
Y
ARE
A

Peri
Urban
settle
ments along OMR are selected as study area. It includes
Perungudi,

okkiyam

Thoraipakkam,

karapakam

and

sholinganallur. Perungudi and Sholinganallur are Town


Panchayats and Okkiyam Thoraipakkam and karapakkam
comes under Village Panchayats of St.Thomas Mount
Panchayat union as of old Chennai City limit concerned as
depicted in Fig.1. The reason behind the selection of study
area is that the southern group consisting of developments
from Perungudi to Sholinganallur exhibits a very high
growth rate.

Page 228

Abirami Krishna et. al. / INDECS-15

Fig 8. Peak hour volume

Fig 7.Study Area


Fig 9. Vehicle Composition
8.

DATA COLLECTION AND ANALYSIS


Figure 9 illustrates the vehicle composition of

8.1

General

Chennai IT corridor. Two wheeler and car proportion in the

Holding capacity can be determined by considering

total volumes are observed around 53 and 35 percent and

Traffic volume as primary data and demography, land use

Buses (governments and private & institutional) is around 4

and norms on basic infrastructures as secondary data. The

percent only. This seems to be an unhealthy proportion with

collected data are briefly discussed.

regard to environment consideration and Level of Service

8.2 Primary data collection


Primary data collection involves Local trips bounded in

(LOS).
8.3

the study area. It is achieved through traffic volume surveys


at 3 different locations of Old Mahabalipuram Road
between 3.30 p.m and 8.30 p.m. The locations include
Perungudi, Thoraipakkam, Sholinganallur.

Secondary data collection


Collection of secondary data includes Demographic

data, Land use data and norms on basic infrastructures such


as Water supply, and Electricity.
8.3.1

Demographic data

Fig 8 shows the peak traffic volume observed at 3


survey locations and it depicts that the peak traffic volume is
the highest at Perungudi location.

Fig 10 shows the demographic profile of study area.


From the figure it is observed that Thoraipakkam census
town has highest population of 56642 with 14524
households.

Page 229

Abirami Krishna et. al. / INDECS-15


of the study area is around 500 units/month out of which
TNEB is able to meet only 80% of the demand. Details of
existing substations in study area is obtained from Tamil
Nadu Transmission Corporation Limited (TANTRANSCO)
and tabulated below.
Table 2 Details of Substations in Study area
Fig 10. Demographic profile
Source: Compiled from census

2011

8.3.2 WATER SUPPLY DATA


Water supply requirement and supply data have been
collected from CMDA Second master plan as listed in Table

Name of the Sub

Voltage ratio in

Capacity in

Station

KV

MVA

Perungudi

110/33/11

132

Sholinganallur

110/11

32

Source: TANTRANSCO

2 and 3.
Table 1 Estimated water requirement 2026

8.3.4 Land Use data

Category in
CMA

Population
in lakhs

Rate of water
supply in lpcd

Total
requirement
in MLD

Chennai City

58.56

150

1230

Municipalities
and Town
panchayats in
CMA

47.90

125 for
Municipalities
100 for Town
panchayats

796

Rest of CMA

19.88

80

223

Total

126.34

Land area corresponding to each and every land use is


extracted from Geo referencing Land use map obtained from
CMDA by using AutoCAD 2015. The extracted results are
given in table 3.

2249

Table .3 states the Total domestic supply and industrial supply


by CMWSS board.

8.3.3 ELECTRICITY DATA


Electricity demand and rate of supply is obtained by Tamil
Nadu Electricity Board (TNEB). Average electricity demand

Table 3.Extent of Land use

Land Use

Perungudi

Thoraipakkam

Karapakkam

Sholinganallur

Ha

Ha

Ha

Ha

Primary Residential

152.6

39

289

48

22.4

611.4

40

Mixed Residential

51.8

13

79.7

13

34

14

116.4

Institutional

84.2

21

17.5

18.1

32.36

Industrial

49.2

12

135.9

23

93.5

39

423.9

28

Water body

17.7

48.6

140.29

Agricultural

8.6

Commercial

21

8.7

33.24

Page 230

Abirami Krishna et. al. / INDECS-15

Non Urban

1.3

14.11

CRZ

140.74

Road Network

30.9

30.9

8.5

23.36

Total

395

100

601.6

100

242

100

1535.8

100

Source: Compiled from CMDA Land use map

9. MODEL DEVELOPMENT
9.1 General
Simulation model has been developed for population sector
and infrastructures sector using the STELLA simulation
software

With

the

System dynamics (SD)

model,

relationship between the population growth and its level of


population and its supply of all basic infrastructures are
established and density norm for all basic infrastructures are
determined. Optimum density norm is said to be the

Fig 10. SD Model for Population Sector

minimum among all. Holding capacity is calculated based


on the optimum density norm.
The existing land use data is used to know the existing area

9.2 Model Development Land Use Sector

for allocating the projected population and the future land


use proportion can be formulated for corresponding

Stella model for land use sector is built by considering

population growth. Intuitive scenarios can be analysed for

existing residential land area and future developable floor

the developed model. The different patterns of possible

area. By accounting residential growth rate extent of land

present and future growth are generated and holding

area in projected year is calculated.

capacity is determined towards achieving sustainable

Existing residential floor area for every year can be

development for the study area.

developed with respect to the growth rate of residential


intensity and available residential area of each zone.

9.2

Model Development Population Sector

FDFA1= Existing residential area (Remaining % from


already developed) * Plot coverage for constructing house*

Population model has been built to estimate the


projected population as given in Fig 7 by accounting birth
rate, death rate, immigration and out migration rate.
Population (t) = Population (t-dt) + (Birth_Rate +
Inmigration_Rate Death_Rate Out_Migration_Rate) *
dt

Undeveloped remaining FSI


FDFA2= Existing residential area (% yet to be developed) *
Plot coverage for constructing house* Undeveloped FSI
FDFA = FDFA1 + FDFA2
Residential area= Existing residential area (% from already
developed)

Plot

coverage

for

constructing

house*Developed FSI

Page 231

Abirami Krishna et. al. / INDECS-15

If existing growth trend continues, V/C of horizon year


is 2.53 which is 2 times higher than base year V/C.

SD model for density estimates an average density of


233,176,186 persons/hectare in do minimum scenario
for Transportation, Water supply, and Electricity sector
respectively.

It infers that the density corresponds to water supply


sector is very minimum when compared to other
infrastructures which explicit the stress on water
infrastructure and emergent need of funds for its

Fig 11. SD Model for Land Use Sector

improvement.

9.3 Model Development Transport Sector

Scenario II Partial (partial implementation of new


proposals)

Model for transportation sector is built by accounting


present vehicular volume and growth rate of vehicles in

In partial scenario few new proposals are introduced in

order to determine the volume/ Capacity ratio and given in

all basic infrastructures and corresponding density is

Fig 12.

estimated based on available land area in Partial


scenario.

In this scenario D/S comes around 0.97 which is 22%


reduction when compared to do minimum scenario.

Introducing an elevated corridor reduces the V/C ratio


to 1.52 which is 40% reduction than do minimum
scenario.
Scenario III Desirable (Effective implementation
of all new proposals)

In desirable scenario implementation of desalination


plant, Power plants in phase wise manner to enhance
Quality of Life is considered.

In this scenario average D/S ratio of 0.8 is achieved


through effective implementation of various proposals

Fig 12. SD Model for Transport Sector

which is 36% reduction than do minimum.

10.1 FINDINGS
Scenario I Do Minimum (Allowing the existing trend
to continue)

Results of Demand/ Supply (D/S) ratio model shows


that in do minimum scenario the D/S is found to be
1.245.

In desirable scenario, combination of growth restriction


and introduction of elevated corridor results in a V/C
ratio of 0.96.

In desirable scenario by incorporating elevated


corridor, desalination plants, Power plants in phase
wise manner the dynamic balancing of holding

Page 232

Abirami Krishna et. al. / INDECS-15

capacity of land use sector with carrying capacity of

For electricity sector, a power plant has to come up

basic infrastructures is achieved.

which may be implemented in 4 phases to augment

The resultant maximum density is in a range of 94 to

the

259 persons/hectare for Sholinganallur and Perungudi

units/capita/month to 111.11 units/capita/month.

respectively which is actually the minimum density of

all infrastructures put together.

supply

rate

of

electricity

from

88.88

An approximate cost of 70.45 billion will be


needed to implement all the proposals of desirable
scenario whereas 35.22 billion will be needed for

10.2

INFERENCES

Desirable scenario is achieved by incorporating growth


restriction eliminating certain percentage of car and two
wheelers and introducing an elevated corridor of
capacity 3600 PCUs/hour and bypassing HCV and
allowing only 30% of LCV.
For Water supply sector, the D/S ratio of horizon year
comes around 1.32, 1.20 and 1 in do minimum, partial
and desirable scenario respectively

REFERENCES

For Transportation Sector, a reasonable reduction in V/C


ratio of 2.53 in Do minimum scenario to 0.96 in

the partial implementation of proposals.

For Electricity sector Since there is no power plant


proposal in the study area, present D/S ratio of 1.25 gets
carried over up-to horizon year in do minimum scenario
whereas in Partial scenario, if 2 phases of a power plant
proposal gets completed D/S ratio comes around 1.10 in
horizon year and in desirable scenario, if all 4 phases of a
power plant proposal gets completed, a D/S ratio of 1.00
is achieved in horizon year.
10.3

RECOMMENDATIONS

[1] Douglas Webster (May 2002),On the Edge:


Shaping the Future of Peri urban East Asia, a
discussion paper, pp 5-8, Asia/Pacific Research Center
(A/PARC).
[2] Haroldo Torres, HumbertoAlves and Maria
Aparecida De Oliveira (2007), Sao Paulo peri urban
dynamics:

Some

social

environmentalconsequences,

causes
Environment

and
and

Urbanization, Vol. 19, No.1 , pp 207-223.


[3] Agricultural and ResourceEconomics Review,
Vol.32/1. pp. 83-102.
[4] John O. Browder, James R. Bohland and Joseph L.
Scarpaci (Sep 1995),Patterns of development on the
metropolitan

fringe:

Urban

fringe

expansionin

Bangkok, Santiago and Jakarta, Journal of American


Planning Association, Vol. 61, Issue 3, pp 310-327.
[5]

Karthik.N

and

et

al.

(2011),Infrastructure

Development of Medavakkam, B.E thesis, Division of


Transportation Engineering, Anna university, Chennai.
[6] Second Master Plan for Chennai Metropolitan Area

For land use sector, a desirable density is achieved

2026, Vol.1 pp 68-80, and Vol.3 pp 157-181, Chennai

by restricting the population growth trend as well as

Metropolitan Development Authority.

augmenting the land use intensification rate on par

[7] Sekar S P and kanchanamala S (2011), Analysis of

with population increase.

growth dynamics of Chennai metropolitan Area, ITPI

For Transportation sector, the growth trend of

journal December 2011.

personalized mode has to be decreased by


augmenting the public transportation facilities such
as BRT.

For water supply sector a total of 4 desalination


plants are to be implemented before 2026 in order
to achieve a D/S ratio of 1.

Page 233

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

RELIABILITY ANALYSIS OF POST TENSIONED PSC GIRDERS DESIGNED


AS PER IS 1343:2012
Bharath B.M. 1, Dr. K. Manjunath 2 Puja M. Singh 3
1

M.Tech Student, M.C.E., Hassan, 573201, India, bharath0491@gmail.com


Professor and H.O.D., M.C.E., Hassan, 573201, India, kmnrpur@gmail.com
3
Scientist, ISRO Satellite Centre, Bangalore, 560017, India, pujam@isac.gov.in
2

Abstract
An attempt is made to evaluate the safety of a post tensioned girder in different limit states. The explicit level of safety
can be assessed only by employing probabilistic methods. Reliability Analysis of post tensioned PSC girders is carried
out in limit states of flexure, shear and deflection. Fifteen typical girders are designed as per IS 1343:2012 for different
spans by developing a program in MATLAB software. Monte Carlo simulation technique and Advanced First Order
Second Moment (AFOSM) methods using Fisselers algorithm are employed in the reliability analysis. The geometric
properties, material properties and loading are considered as random variables.
The study investigates reliability index and probability of failure of post tensioned PSC girders. Programs are developed
in MATLAB for each of limit states and the histograms and probability density functions are plotted which would assist
in assessment of probability of failure. The deterministic design of post tensioned PSC girder is carried out at ISRO
satellite centre Bangalore, where actual construction of a roofing system with PSC girders for Satellite Integration Hall
is going on at the time of present work. STAAD Pro V8i software is used for modeling and analysis of the structural
system. The reliability analysis is carried out by developing programs in MATLAB software.
Keywords: PSC girders, IS 1343:2012, STAAD Pro V8i, MATLAB
1. Introduction

2.1. Available Data

The prestressed concrete is considered entirely as a concrete


structure with tendons supplying the pre-compression which
transforms brittle material like concrete into an elastic one. It can
be visualized as a beam subjected to two types of forces- initial
prestress and external load.
In recent years, PSC girder design is moving towards a more
rational method commonly known as limit states design. This type
of design procedure takes into account more information than the
elastic methods of design of structural components. This semiprobabilistic design includes uncertainties in the strength of various
materials, in loads and load combinations, and modeling errors in
analysis procedures. Further the limit state design is being replaced
by reliability based design which considers the probability,
however low it may be, of the resistance becoming lower than the
action. The resistance and action being random due to inherent
randomness of material strength, geometry of members and loads
acting upon, the reliability based design attempts to ensure explicit
level of safety in the presence of randomness.

The effective span of the simply supported PSC girder is 34.3m.


The slab thickness is of 100mm and the girder is designed for live
load and dead load.
Live load = 1.5 x 5 = 7.5 KN/m
Super imposed dead load = 4.013 x 5 = 20.07 KN/m
Dead Load is works out to be = 28.98 KN/m
2.1.1. STAAD Pro Model
The roofing system is modeled and analyzed in STAAD Pro Vi8
software. The results of bending and shear at different sections are
obtained and the section is designed based on this results.

2. Deterministic Design
The deterministic design of the PSC girder is carried out at ISRO
satellite center, Bangalore. The roofing system consists of 12 Post
tensioned PSC girders with 5m centre to centre spacing. One of the
PSC girders is selected for the present study.
Fig. 1. STAAD Pro model of Roofing System
*

Bharath B.M. Tel.: +91 8970099510


E-mail: bharath0491@gmail.com

Page 234

Bharath et. al. / INDECS-15

The section obtained from preliminary design is checked for


requirements of minimum section modulus, eccentricity,
prestressing force and permissible tendon zone.

3.1. Design of PSC Girders using MATLAB.


A computer program is developed in MATLAB software and hence
many number of typical PSC girders for different span range shall
be designed for the reliability analysis.
Fifteen typical girders are designed from span 20m to 34m, the
thickness of the top flange is fixed as 100mm. Lins method is
adopted in the computer program and the sections are checked for
its safety in limit states of Flexure, Shear and Deflection.
It should be noted that the cross sections of these fifteen girders are
uniform along the length of the span, but the section which
designed and constructed in the field (at ISRO Satellite centre) is as
shown in the fig 2 and fig 3.
3.2. Reliability Index and Probability of Failure

Fig. 2. Cross section at centre span

The adequacy of a design can be measured by means of a non


dimensional number called reliability index. It is a traditional
notation of the safety limit associated with the ultimate limit states.
A limit state function is given by,
g=R-S
(1)
Where, R is the resistance and S is the action.
If g > 0, the structure is safe and the probability of failure is given
by, Pf = Prob (g < 0)
3.3 Methods of Assessment of Reliability

Fig. 3. Cross section at support

2.2. Prestress Losses and Check for Stresses


The stresses in PSC girder are checked at 14th day, 28th day before
and after the application of super imposed dead load and at infinity
days (at service) after accounting for live load.
2.2.1. Prestress Losses
14th day Loss due to friction and Loss due to elastic shortening of
the girder
28th day Loss due to creep and shrinkage from 14th to 28thday and
Relaxation loss of prestressing steel from 14th to 28th day
Infinity Loss due to creep and shrinkage from 28th to infinity day
Relaxation loss of prestressing steel from 28th to infinity
days

Monte carlo simulation technique and Advanced first order second


moment methods using Fisselers algorithm are used for reliability
evaluation of PSC girders in the present study.
The statistics of basic variables are extracted from the available
literature and tabulated in table 1.
Variable

Bias

Geometric Properties
Dimensions (H,b,d,e)
1.03
Area (Tendon, Concrete, Steel)
1.03
Material Properties
Characteristic strength of tendon
1.04
Characteristic strength of concrete
1.1
Characteristic strength of steel
1.04
Loading
Dead load
1.05
Live load
1.2

COV
(%)

Distribution
Type

3
1.5

Normal
Normal

2
18
2

Normal
Normal
Normal

15
30

Normal
Normal

Table 1. Statistics of basic variables

3.4 Generation of Resistance Statistics


2.3. Check for Limit States of Flexure, Shear and Deflection.
The designed section is checked for its moment carrying capacity,
shear carrying capacity and its resistance against deflection.
Suitable amount of anchorage reinforcement is provided to take
care of the bursting tension.
The results are discussed in section 4.

3. Reliability Analysis
Reliability is the probability of an item performing its intended
function over a given period of time under the operating
conditions encountered.
Reliability analysis is carried out with the help of MATLAB
software.

Resistance models corresponding to each limit states are developed.


During the reliability analysis of the current design, resistance
model is calculated by performing Monte-Carlo simulations and
AFOSM methods using Fisselers Algorithm. 30000 data sets were
randomly generated for each cross section, and each data set
randomly as a function of statistical models for the variables
involved bias [mean/nominal], coefficient of variation [ COV =
standard deviation/mean], distribution type etc.
3.4.1. Resistance Model for Limit State of Flexure
Two resistance models are used depending upon the position of
neutral axis. If the neutral axis lies in the flange, the resistance
model is given by,
Mu = fpu Aps (d 0.42 Xu)
(2)

Page 235

Bharath et. al. / INDECS-15

If the neutral axis lies in the web, the resistance model is given
by,
Mu = fpu Apw (d 0.42Xu) + 0.675 fck (b-bw) Df (d 0.5 Df)
(3)
Where,
fpu
= Ultimate stress of prestressing strands
Aps
= Area ofc total prestressing steel
Apw
= Area of prestressing steel for web

Nominal longitudinal reinforcement of 2050 mm2 (0.2% of the cross


section) is provided in addition to tensioned reinforcement. Two
legged 10mm and 12mm diameter stirrups are provided at 200mm
c/c near center span and support respectively.
The deflection of the girder is works out to be 69.81mm and the
permissible deflection being higher i.e., 137.2mm, the section is safe
against deflection. 16mm diameter Fe 500 spiral eight turns at
50mm pitch is provided as anchorage reinforcement for bursting
tension of 284.93 KN.

3.4.2. Resistance Model for Limit State of Shear


The ultimate shear resistance of the section is estimated by
considering flexure shear and web shear cracking modes.
For web shear cracking mode, equation 3 is used as resistance
model and equation 4 is used for flexure shear cracking mode in
addition to shear carried by the stirrups and it is given by equation
5.
Vc = 0.67bwD(ft2 + 0.8fcpft)
(3)
Vc = [1-0.55(fpe /fpu)]cbwd + M0 (V/M)
(4)
Vcs = (0.87fydtAsv / Sv)
(5)
Where, ft = 0.24fck
M0 = 0.8 ft(I / Y)

Resistance statistics of the fifteen typical girders in terms of


reliability index and probability of failure in three limit states is
tabulated in table 4 and table 5.
Limit States
Flexure
Uncracked in
flexure
Shear

Cracked in
flexure
Deflection

Table 4.

3.4.3. Resistance Model for Limit State of Deflection


The resistance model for deflection of flanged section is obtained
by combining equation 6 with maximum permissible limit of
deflection l/250.
a = - (5(g + q)L4 / (384EI))
(6)
Equation 6 is the deflection of a beam with parabolic tendons
having an eccentricity e at the centre and zero at the support.

Limit States
Flexure
Uncracked in
Shear

The loss of prestress and check for prestress losses are tabulated in
table 2 and table 3.
Time
Period
(Days)

Frictio
n Loss
(KN)

ES
Loss
(KN)

Creep
Loss
(KN)

14

400.99

135.07

14 - 28

55.124

57.74

147.66

28 infinity

223.506

182.842

268.316

Table 2.
Time
Period

2.312
(compression)

28th
day

6.257
(compression)

At
infinity
days

7.917
(compression)
Table 3.

Relaxati
on Loss
(KN)
-

Prestress Losses

Actual Stress
Top
Bottom
( N/mm2)
(N/mm2)

14th
day

Shrinkage
Loss (KN)

Permissible Stress
Top
Bottom
(N/mm2)
(N/mm2)
17.39
15.744
0
(compressi
(compression)
(tension)
on)
15.6
5.762
(compress
0 (tension)
(compression)
ion)
15.6
0.4
(compress
0 (tension)
(compression)
ion)
Check for Stresses

Cracked in
flexure

Fifty strands are provided in the design with three cables having
twelve strands and two cables having seven strands. The strands
are of 12.7mm diameter with an effective area of 98.7 mm2. The
prestressing force provided is 6983.62 KN.

4.71 to 5.73

10-5 to 10-8

8.35 to 6.18

10-16 to 10-9

10.95 to 8.07

10-21 to 10-15

6.27 to 6.21

10-9

Resistance Statistics ( Mote Carlo Simulation


Technique)

flexure

4. Results

Monte Carlo simulation technique

Pf

Deflection
Table 5.

FOSM method using Fisselers algorithm

Pf
3.57 to 3.94

10-4

5.30 to 4.68

10-7 to 10-5

8.82 to 5.56

10-18 to 10-7

7.12 to 6.80

10-12 to 10-11

Resistance Statistics ( FOSM Method)

Histograms and probability distribution curves are plotted for all the
girders in limit states of flexure, shear and deflection. A typical
histogram and probability curve of moment of resistance for girder
of 20m span are shown in figure 4 and figure 5. Similarly
histograms and probability distribution curves are plotted for all
fifteen typical girders in other two limit states also.

F
R
E
Q
U
E
N
C
Y

MOMENT OF RESISTANCE (Nmm)


Fig. 4. Histogram of Moment of Resistance

Page 236

Bharath et. al. / INDECS-15

[4]

[5]

[6]

[7]
Fig. 5. Probability Distribution Curve

[8]

5. Conclusion
An attempt is made to evaluate the safety of a post tensioned
girder in different limit states. The explicit level of safety can be
assessed only by employing probabilistic methods.

Scientific Engineering and Research (IJSER), ISSN: 23473878, Volume 2, Issue 8, August 2014.
Ayman M. Okeil, Abdeldjelil and Daniel A. kuchma.
Reliability assessment of FRP strengthened concrete bridge
girders in shear. Journal of composites for construction @
ASCE / January/February 2013.
Ayman M. Okeil, Sherif El-Tawil, Mohsen Shahawy. Flexural
reliability of RC bridge girders strengthened with CFRP
laminates. M.S. thesis Work,
Dept. of Civil and
Environmental Engineering, University of Central Florida,
Orlando, FL 32816-2450.
Gunjan Agrawal, Baidurya Bhattacharya. Partial safety factor
design of rectangular partially prestressed concrete beams in
ultimate flexural limit state. Journal of Structural Engineering,
No. 37-31, Vol. 37, No. 4, October-November 2010 pp. 257
267.
Khaliq Mohamed Burhan. Generation of resistance statistics
of PSC bridge girder in limit state of flexure. M.Tech thesis
work, Dept of Civil Engineering, M.C.E. Hassan
Hayder A. Rasheed; Kyle H. Larson; and Robert J. Peterman.
Analysis and design procedure for FRP-strengthened
prestressed concrete T-girders considering strength and
fatigue. Journal of composites for construction asce /
september/october 2006.

Reliability index denotes the probability of failure in a given limit


state. The reliability index can be evaluated by digital simulation
like Monte Carlo simulation method or by closed form solutions
like first order second moment method using Fizzlers algorithm.
The deterministic design is obtained by limit state method as per
IS 1343:2012. A program is prepared for the deterministic design
using MATLAB software. Digital simulation is done by using
MATLAB platform.
Probabilistic variations of moment of resistance and that of
external bending moment are obtained by digital simulation.
Normal probability distribution is used to characterize randomness
in resistance and applied action. The goodness of fit of normal
distribution is established by K.S. test.
Similar studies are made in limit states of shear and deflection.
The reliability index is evaluated in these limit states.
It is concluded that the reliability index is not uniform in different
limit states. Hence the explicit safety factor varies for different
limit states.
Reliability based design can be used to achieve a target level of
reliability. The target reliability index can be specified depending
upon the importance of the structure.

References
[1]

[2]

[3]

Amit Saxena, Dr. Savita Maru. Comparative study of the


analysis and design of T-beam girder and box girder
superstructure. International Journal of Research in
Engineering & Advanced Technology (IJREAT), ISSN:
2320-8791, Volume 1, Issue 2, April-May, 2013.
Vishal U., N. G. Gore, P. J. Salunke. Analysis and design of
prestressed concrete girder. International Journal of
Inventive Engineering and sciences (IJIES), ISSN: 23199598, Volume-2, Issue-2, January 2014.
Rajamoori Arun Kumar, B. Vamsi Krishna. Design of
prestressed concrete T-beams. International Journal of

Page 237

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

EFFECT OF NUMBER OF LOADING CYCLES ON DYNAMIC PROPERTIES OF


SOLANI RIVER SAND
J. Chavda1, B. K. Maheshwari2, G. R. Dodagoudar3
1

Research Scholar, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, India,
jiteshchavda03@yahoo.in
2
Professor, Department of Earthquake Engineering, Indian Institute of Technology, Roorkee, India,
bkmahfeq@iitr.ac.in
3
Professor, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, India, goudar@iitm.ac.in
Abstract
During earthquakes, soil plays a great role for the safety of supported structures. Dynamic soil properties are very useful
in Earthquake Geotechnical Engineering problems like ground response analysis, seismic soil-structure interaction,
problems due to liquefaction, etc. The two important parameters required for analysing the above problems are the shear
modulus and damping ratio of the soil. The recent developments in numerical analysis for the nonlinear dynamic
response of ground due to strong ground motions have increased the demand for proper evaluation of dynamic soil
properties corresponding to medium and large strains. In the present work, the dynamic soil properties of Solani river
sand from Uttarakhand, India were evaluated using cyclic triaxial and bender element tests. The number of loading
cycles were varied and their effects on the degradation of modulus values and increase in damping have been studied.
The modulus reduction curves at different strains were obtained. From the test results, for the low to high strains range,
the shear modulus of the sand is changing with change in number of cycles of loading. At low shear strain, the shear
modulus increases with increase in number of cycles. At high strain of 1%, the shear modulus is reducing with increase
in number of cycles. This understanding will be helpful in better design of foundations subjected to earthquake loads.
Keywords: Shear modulus, Damping ratio, Cyclic triaxial test, Bender element test, Modulus reduction curve.
1. Introduction
The recent developments in numerical analysis of nonlinear
dynamic response of ground due to strong ground motions
have increased the demand for realistic evaluation of dynamic
soil properties corresponding to large strains. The parameters
affecting the dynamic properties of soils have to be considered
and evaluated properly. In numerical analysis, the modulus
reduction and damping curves are needed for wide range of
strains so that the proper nonlinear analysis can be attempted.
The present study provides the results of modulus values at
four different strains levels, i.e., 0.06%, 0.1%, 0.3% and 1%.
As part of the studies, the sample preparation, saturation and
consolidation have also been explained in the paper.

Sr.
No.
1

Particular

Notation

Value

Soil type

SP

2
3

Specific gravity
Coefficient
uniformity
Coefficient
curvature

of

Gs
Cu

Poorly graded
sand
2.68
1.96

of

Cc

1.15
0.120 mm
0.180 mm
0.210 mm
0.235 mm
0.86
0.54

Grain size

Maximum void ratio

D10
D30
D50
D60
emax

Minimum void ratio

emin

2. Properties of Sand
A grain size distribution curve obtained for the Solani river
sand is depicted in Fig. 1. and the grain size parameters are
given in Table 1 along with the other index properties.

Table 1: Index properties of Solani river sand

3. Sample Preparation
The modulus reduction and damping curves are less sensitive
to the specimen preparation method, degree of saturation and
drainage conditions [21] [10]. The remolded specimens are
prepared using moist tamping compaction method putting the
sand in three layers. Membrane correction is not considered,
since the membrane penetration per unit area is negligible.
Area correction is also not considered to the cyclic triaxial
loading data.
Fig. 1. Grain size distribution curve

The void ratio (e) corresponding to the relative density of sand


(Dr) is expressed as

J. Chavda Tel.: +919773134790


E-mail: jiteshchavda03@yahoo.in

Page 238

J. Chavda, et al. / INDECS-15

e emax Dr (emax emin )

(1 )

where Dr is the relative density, emax is maximum void ratio,


emin is the minimum void ratio, and e is the desired void ratio at
a particular relative density of sand.

d
max

E
2(1 )

After evaluating the value of void ratio e, dry unit weight of


sand () was determined by the following equation

G
d s w
1 e

where d is dry unit weight of sand, Gs is specific gravity of


sand, and w is unit weight of water.

4. Saturation and Consolidation


All the specimens are saturated by passing deaired water so as
to achieve higher saturation at lower back pressure and in less
time. The incremental back pressure saturation is adopted for
saturation and all the specimens are saturated using a back
pressure of 313 5 kPa to achieve Skempton's pore pressure
parameter B in excess of 0.98. The back pressure is kept
constant for all the tests to eliminate the eect of back pressure
on modulus and damping. The specimens have been tested at a
relative density of 50% with void ratio (e) of 0.7 and are
isotropically consolidated to an eective conning pressure of
100 kPa.
For the estimation of G, one of the most reliable methods to
characterize small strain shear modulus (Gmax) is the
measurement of shear wave velocity (V s) through sample by
the bender element test:

Gmax vs2

max

where G is the shear modulus, is the shear strain and is the


Poisson's ratio that may be taken as 0.5 for saturated undrained
specimen [20]. A damping ratio D, is a measure of the
dissipated energy versus the elastic strain energy and is
computed by

1 AL
4 AT

where AL = Area enclosed by the hysteresis loop and AT =


Area of the shaded triangle as shown in Fig. 2.
Figure 3 depicts variation of axial strain with number of cycles
of loading. As a consequence of loading, a decrease in
deviatoric stress is noted and is shown in Fig. 4. The variation
of pore pressure build up with number of cycles of loading is
shown in Fig. 5. The effective stress path is drawn for the
above strain loading at a axial strain of 0.67% with confining
pressure of 100 kPa for the sand sample having void ratio of
0.7 (Fig. 7).

where Gmax is maximum shear modulus of sand at low strain,


Vs is shear wave velocity obtained from the bender element
test, and is density of the soil sample.

5. Formulation Used
The shear modulus is evaluated as the slope of a secant line
that connects the extreme points on a hysteresis loop at a given
shear strain, as shown in Fig. 2. As the cyclic strain amplitude
increases, the shear modulus decreases.

Fig. 3. Applied axial strain

Fig. 2. Hysteretic stress strain relationship


From the cyclic triaxial test results, a hysteresis loop similar to
Fig. 2 is obtained by plotting the axial strain () versus deviator
stress (d). The slope of the secant line connecting the extreme
points on the hysteresis loop is the young modulus (E) [20]:

Fig. 4. Induced deviator stress

Page 239

J. Chavda, et al. / INDECS-15

Fig. 5. Pore pressure variation with no. of cycles


Fig. 8. Hysteretic loop at axial strain of 0.1%

Fig. 6. Effective stress path for e = 0.7, f = 1 Hz, effective


confining pressure = 100 kPa, = 0.67%
Fig. 9. Hysteretic loop at axial strain of 0.3%
With increase in shear strain, the effect of number of cycles
should be considered for the evaluation of dynamic shear
modulus.

Fig. 7. Hysteretic loop at axial strain of 0.06%


Figures 7-10 depict the hysterisis loops at the axial strains of
0.06%, 0.1%, 0.3% and 1% respectively. It is noted from Fig. 7
that at low shear strain the effect of number of cycles of
loading on the dynamic shear modulus is less as compared to
the higher shear strain.

Fig. 10. Hysteretic loop at axial strain of 1%

Page 240

J. Chavda, et al. / INDECS-15

Fbb ( )

Gmax
G
1 ( max )

max

Loop 10

Loop 1

where Gmax is the maximum shear modulus at low strain, is


the shear strain amplitude and max is the maximum shear
stress.
The modulus reduction curve obtained for Loops 1 and 10 with
considering all the strain levels is shown in Fig. 13. At low
strain, the modulus related to Loop 10 is higher and at higher
strain the modulus is lower as compared to the results
corresponding to Loop 1. Figure 14 shows the backbone curves
along with the hyperbolic soil model. The hyperbolic soil
model captures the backbone curve well in the compression
quadrant. The modulus reduction curve developed using the
hyperbolic soil model is shown in Fig. 15.

Fig. 11. Hysteretic loop for e = 0.7, f = 1 Hz, Effective


confining pressure = 100 kPa, = 1%
The sand was compacted to achieve relative density of 50%.
The test has been performed with sample height of 88 mm and
diameter of 50 mm. The test was conducted at an effective cell
pressure of 100 kPa. From Fig. 11, it can be observed that at
high strain of 1%, the shear modulus reduces with increasing
number of cycles. It is noted that around 10th loop modulus
gets stabilized, however a slight variation is observed around
30th loop. Figure 12 depicts the backbone curve developed as
part of the testing at different strain levels for Loops 1 and 10.

Fig. 13. Modulus reduction curve considering Loops 1 and 10

Fig. 12. Backbone curve for Loops 1 and 10


The maximum value of shear modulus, Gmax, (i.e., small strain
shear modullus) is obtained from the bender element tests. The
value of Gmax = 90.123 MPa.
The hyperbolic soil model has been used to check the data
obtained from the cyclic tri tests. The performance of cyclic
nonlinear models can be illustrated by a very simple example
in which the shape of the backbone curve is described by =
Fbb(). The shape of any backbone curve is tied to two
parameters the initial (low strain) stiffness and the high strain
stiffness of the soil. For the simple example, the backbone
function, Fbb(), can be described by a hyperbola [9]:
Fig. 14. Backbone curve with hyperbolic soil model

Page 241

J. Chavda, et al. / INDECS-15

Symposium on Earthquake Engineering, Indian Institute


of Technology, Roorkee, December, 2010; 17-19, 2010.
[9] Kramer SL. Geotechnical Earthquake Engineering. New
Jersey: Prentice Hall, 1996.
[10] Kokusho T. Cyclic triaxial test of dynamic soil properties
for wide strain range. Soils and Foundations, 1980; 20(2),
45-60.
[11] Kirar B, Maheshwari BK, Jakka RS. Dynamic properties
of Solani sand reinforced with coir fibers. Proc. of 15th
World Conference on Earthquake Engineering,
September, Lisbon, Portugal, 2012.

Fig. 15. Modulus reduction curve with hyperbolic soil model

6. Conclusions
Based on the test results, it is concluded that at low to high
strain, the shear modulus of sand is changing with change in
the number of cycles of loading. With specific to low shear
strain, the shear modulus increases with increase in number of
loading cycles. At high strain of 1%, the shear modulus
decreases with increase in number of loading cycles. The
hyperbolic soil model has been used to check the data obtained
from the tests. It is noted that the basic hyperbolic soil model
can predict the behavior of sand realistically under cyclic and
dynamic loading.

7. References
[1] ASTM D3999-91 (2003). Test method for the
determination of the modulus and damping properties of
soils using the cyclic triaxial apparatus, Annual book of
ASTM
standards,
ASTM
International,
West
Conshohocken, PA.
[2] Govindaraju L. Liquefaction and dynamic properties of
sandy soils, Ph. D. Thesis, Department of Civil
Engineering, Indian Institute of Science, Bangalore, 2005.
[3] Hardin BO, Drnevich VP. Shear modulus and damping in
soils: measurement and parameters effects. J. of Soil
Mechanics and Foundations Division, ASCE, 1972;
98(SM6), 603-624.

[12] Kale SS, Maheshwari BK, Kaynia AM. Dynamic


properties of Solani sand under cyclic loads. Proc. of the
14th Symposium on Earthquake Engineering, IIT Roorkee,
Roorkee, December 2010.
[13] Maheshwari B.K, Kale SS, Kaynia AM. Dynamic
properties of Solani sand at large strains: A parametric
study. International J. of Geotechnical Engineering,
2012;6 (3): 353-358.
[14] Prakash S. Soil Dynamics. Singapore: McGraw Hill Book
Co., 1981; 274 -336.
[15] Ravishankar BV, Sitharam TG, Govindaraju L. Dynamic
properties of Ahmedabad sands at large strains. IGC2005, Ahmedabad, 369-372.
[16] Seed HB, Idriss IM. Soil moduli and damping factors for
dynamic response analyses. Rep. No. EERC-70-10,
Earthquake Engg. Research Center, Univ. of California at
Berkeley, Berkeley, California, 1970.
[17] Seed HB, Wong RB, Idriss IM, Tokimatsu K. Moduli and
damping factors for dynamic response analyses of
cohesionless soils. J. of Geotechnical Eng., ASCE, 1986;
112(11), 1016-1032.
[18] Sitharam TG, Govindaraju L, Shridharan A. Dynamic
properties and liquefaction potential of soils. Current
Science, 2004; 87(10), 1354-1362.
[19] Towhata I, Haga K, Nakamura S. Effects of cyclic
drained shear or rigidity of sand. Proc. 20th Nat. Conf.
Soil Mech. Found. Eng., 1985; 1, 591592.

[4] Hanumanthrao C, Ramana GV. Dynamic soil properties


of Yamuna sand. Proc. 13th Symposium on Earthquake
Eng., IIT Roorkee, India, 2006; 430-438.

[20] Towhata I. Geotechnical Earthquake Engineering, Berlin:


Springer-Verlag, 2008.

[5] Ishibashi I. Discussion to Effect of soil plasticity on


cyclic response. by M Vucetic, R Dobry, J. of
Geotechnical Eng., ASCE, 1992; 118 (5), 830-832.

[21] Tatsuoka F, Iwasaki T, Fukushima SH. Stress conditions


and stress histories aecting shear modulus and damping
of sand under cyclic loading. Soils and Foundations,
1979; 19, 2943.

[6] Ishihara K. Liquefaction and flow failure during


earthquakes. Geotechnique, 1993; 43(3), 351-415.
[7] Kale SS. Dynamic properties of soils under cyclic loads,
M. Tech. Dissertation, Department of Earthquake
Engineering, IIT Roorkee, India, 2010.

[22] Wykeham Farrance Engg. Manual of Automatic Dynamic


Triaxial System,
Controls Testing Equipment,
Wykeham Farrance Engineering Ltd., Slough, England,
2008.

[8] Kale SS, Maheshwari BK, Kaynia AM. Dynamic


properties of Solani sand under cyclic loads. Proc. 14th

Page 242

International Conference on Infrastructure Development for Environmental Conservation and Sustenance (INDECS-15)
ACE, Hosur, Tamilnadu-635 109, INDIA, 28-30 October 2015

CONCRETE PERFORMANCE WITH ADMIXTURES OF FLY ASH AND


COPPER SLAG CONCERNING MECHANICAL AND
DURABILITY ASPECTS FOR NORMAL STRENGTH
VELUMANI.M 1, NIRMALKUMAR.K2 SATHISH.S 3
1

Assistant Professor, K.S.Rangasamy College of Technology, Tiruchengode, Tamil Nadu, India,


mvelumani84@gmail.com
2
Professor, Kongu Engineering College, Perundurai, Erode,
nirmalkumar@kongu.ac.in
3
UG Student, K.S.Rangasamy College of Technology, Tiruchengode, Tamil Nadu, India,
sathish.ce00@gmail.com

Abstract
Fine Aggregate is one the most important material in the concrete. River sand is extensively used for fine
aggregate. On overuse of the aggregate, the source is declined and also the cost of aggregate is increased. On the other
side, the copper slag and fly ash is disposed in massive amount and these waste is ensue many environmental
complications. Copper slag is an industrial by-product produced in large quantities by the process of manufacturing
copper in Sterlite industries. Fly ash is produced from the thermal power plant for the period of burning the coal. The
objective of this work is to study the strength and corrosion resistive properties of concrete containing copper slag as
fine aggregate (0% to 100%) along with fly ash (0% to 30%) by weight of cement. The property of resistance to
corrosion is evaluated based on the performance of the concrete for the penetration of chloride ions by means of Rapid
Chloride Penetration Test (RCPT) and Gravimetric weight loss method. From the results obtained through the above
experiments, it is found that replacement of sand by fly ash along increases the strength of concrete; increases corrosion
resistance in addition to overall properties of concrete. Assessment of slump aspects of concrete including copper slag as
a fine aggregate and fly ash as a cement replacement is as important as the mechanical properties. In this study of
compressive strength and durability aspects of concrete such as ultrasonic pulse velocity, rapid chloride penetration, acid
resistance test and sulphate resistance tests these test results indicate that copper slag and fly ash in concrete give the
good results. When compared to normal concrete this conventional concrete gives good resistances. After 28, 60, 90
days of immersion in 5% magnesium Chloride solution, Copper slag concrete showed better resistance to chloride ion
penetration. The conventional concrete gives the good resistance for water absorption with compare than the normal
concrete. After done the all the primary experiments of concrete, optimum percentage addition of the fly ash by weight
of cement in concrete containing copper slag as fine aggregate was also determined.
Keywords: copper slag, fly ash ultrasonic pulse velocity, rapid chloride penetration, acid resistance test, sulphate
resistance tests.

1. Introduction

In the separation of copper, slag is a by-product


obtained during the matte smelting and refining of
copper has been reported by Biswas and Davenport
(2002). The major constituent of a smelting charge are
sulphides and oxides of iron and copper. The charge also
contains oxides such as SiO2, Al2O3 CaO and MgO,
which are either present in original concentrate or added
as flux. It is Iron, Copper, Sulphur, Oxygen and their
oxides which largely control the chemistry and physical
constitution of smelting system. The main reason for the

premature failure of reinforced concrete structures is


corrosion of the reinforcements. The use of new mortars
based on ternary mixtures, an alternative to ordinary
Portland cement (OPC), requires extensive research in
order to check its passivizing properties for
reinforcements and the instability or permanence of the
passive state achieved. Pozzolonas and slag extends the
market for concrete by improving specific properties of
concrete products, allowing them to be constructed with
other materials or placed in environments that would
have precluded the use of Portland cement alone. In
properly formulated concrete mixtures, Pozzolonas and
slag have been shown to enhance long-term strength,
decreased permeability, increased durability, and
reduction in thermal cracking of bulk concrete. The

Page 243

Author et. al. / INDECS-15

workability of concrete reduces due to the free flow of


water left in the concrete due to less absorption of water
by copper slag. On 50% replacement of copper slag, it
also gives more strength than conventional concrete
Arivalagan.s, (2013). An experimental investigation was
conducted on the properties of concrete to study the
effect of using copper slag as a fine aggregate. There
was more than 70% improvement in the compressive
strength of mortar with 50% copper slag substitution in
comparison with the control mixture. The volume of
permeable voids decreased with the replacement of up to
50% copper slag Khalifa S. Al-Jabri (2011). This work
reports an experimental procedure to investigate the
effect of using copper slag as partial replacement of
sand. The result indicated that workability increases with
increase in copper slag percentage. The highest
compressive strength obtained was 46Mpa (for 100%
replacement) and the corresponding strength for control
mix was 30Mpa Meenakshi Sundarvizhi. S (2011). High
performance concrete should be designed to have the
advanced workability and better durability than those of
conventional concretes. Thus this research was
performed to evaluate the potential use of copper slag as
sand replacement in the production of high performance
concrete D. Brindha, S. Nagan, (2010). This research
study was conducted to investigate the performance of
high strength concrete (HSC) made with copper slag as a
fine aggregate. The result shows that the water demand
was reduced by almost 22% at 100% copper slag
replacement compared to the control mixture .Khalifa
S.Al-Jabri et al (2009). Investigated the effect of using
copper slag as a replacement of sand on the properties of
high performance concrete (HPC). Concrete mixtures
were prepared with different proportions of copper slag
ranging from 0% (for the control mix) to 100%.
Addition of up to 50% of copper slag as sand
replacement yielded comparable strength with that of the
control mix. However, further additions of copper slag
caused reduction in the strength due to an increase of the
free water content in the mix Al-Jabri (2009). Copper
slag is also used as fine aggregate in the design of
bituminous concrete and semi dense bituminous
concrete, which increases the property of the bituminous
mixes Gorai, B., Jana, R.K. and Premchand, M (2003).
Thus this study focused on the effect of copper slag by
surrogating fine aggregate in M35 grade Portland
Pozzolona cement concrete and they were cured for 7
days, 28 days, 60 days and 90 days. Then the obtained
results were compared with the conventional concrete
made using Portland Pozzolona cement. Reinforced
concrete structures are exposed to harsh environments
yet is often expected to last with little or no repair or
maintenance for long periods of time (often 100 years or
more). To do this, a durable structure needs to be
produced. For reinforced concrete bridges, one of the
major forms of environmental attack is chloride ingress,
which leads to corrosion of the reinforcing steel and a
subsequent reduction in the strength, serviceability, and
aesthetics of the structure. The different mechanisms of
chloride penetration are presented, followed by a further
elaboration of the chloride diffusion theory. The

influence of basic properties of concrete on its chloride


penetrability is also discussed. Chloride penetration of
concrete pavement structures is determined through the
Rapid Chloride Permeability test (RCPT), which
typically measures the number of coulombs passing
through a concrete sample over a period of six hours at a
concrete age of 7, 28, and 56 days. In a composite
material, such as concrete, the parameters of the mixture
design and interaction between them determine the
behaviour of the material.
The rate of ingress of chlorides into concrete depends
on the pore structure of the concrete, which is affected
by factors including materials, construction practices,
and age. The penetrability of concrete is obviously
related to the pore structure of the cement paste matrix.
This will be influenced by the water-cement ratio of the
concrete, the inclusion of supplementary cementing
materials which serve to subdivide the pore structure
McGrath, 1996, and the degree of hydration of the
concrete. The older the concrete, the greater amount of
hydration that has occurred and thus the more highly
developed will be the pore structure. This is especially
true for concrete containing slower reacting
supplementary cementing materials such as fly ash that
require a longer time to hydrate Tang and Nilsson, 1992;
Bam forth, 1995. Because these conductors influence the
results so that a higher coulomb value than would
otherwise be recorded is determined, the method still
could serve as a quality control test. It can qualify a mix,
but not necessarily disqualify it Ozyildirim, 1994.
Chloride permeability is an inherent property of concrete
and needs to be assessed independently for long term
durability, especially in the design and construction of
structures to be built in a salt-laden environment.
Wherever there is a potential risk of chloride-induced
corrosion, the concrete should be evaluated for chloride
permeability Joshi and Chan, (2002).
2. MATERIALS AND PROPERTIES
2.1 CEMENT
The cement used in this project is Ordinary Portland
Cement of 43 Grade from Ultratech Cement Company.
This cement is most widely used in the construction
industry in India.
2.2 COARSE AND FINE AGGREGATE
Coarse aggregate of 20mm size and fine aggregate
of zone III from Karur area of TamilNadu
2.3 COPPER SLAG
The slag is a black glassy granular material, by
product of Sterlite Industries Limited (SIL), Tuticorin,
Tamil Nadu, India
2.4 FLY ASH

Page 244

Author et. al. / INDECS-15

Fly ash of class C is obtained from thermal


power plant, Mettur, TamilNadu, India was used.
3. PHYSICAL AND CHEMICAL PROPERTIES
3.1 Physical properties of OPC and Fly ash
Ordinary Portland cement from Ultratech Cement
Company is used for super grade. The cement is
produced as per the IS (Indian standard) specification
given in IS: 1489 Part-I-1991.Fly ash of class C is
obtained from thermal power plant. IS: 3812 Part-I 2003 The Physical properties of Ordinary Portland
Cement and Fly ash are given in Table 3.1
Table 3.1 Physical properties of OPC and Fly ash
S.No

4.

Physical
properties
Fineness
modulus
Initial setting
time
Final setting
time
Soundness

5.

Specific gravity

1.
2.
3.

OPC

Fly ash

335.7
m2/kg
28 min

397
m2/kg
130 min

595 min

290 min

0.8%

0.20%

3.15

2.14

Component
Silica (sio2)
Alumina ( Al2so3)
Iron oxide ( Fe2o3)
Calcium oxide (Cao)
Magnesium oxide (MgO)
Sulfuric trioxide
( so3)

3.4 VARIOUS REPLACEMENTS OF COPPER


SLAG AND FLY ASH IN CONCRETE
The various replacements of Copper slag and
Fly ash are given in Table 3.4
Table 3.4 Replacement of Copper slag and Fly ash
S.NO
1.
2.
3.
4.
5.
6.
7.
8.
9.

Cement
(%)
100
70
70
70
70
70
70
70
70

Fly ash
(%)
0
30
30
30
30
30
30
30
30

3.2 Chemical properties of OPC and Fly ash

4. TESTING OF MATERIALS

The Chemical properties of Ordinary Portland


Cement and Fly ash are given in Table 3.2

4.1 SIEVE ANALYSIS

Table 3.2 Chemical properties of OPC and Fly ash


Component

OPC (%)

Fly ash (%)

Silica (sio2)
Alumina
(Al2SO3)
Iron oxide
(Fe2O3)
Calcium oxide
(CaO)
Magnesium oxide
(MgO)
Sulfuric trioxide
(SO3)

20.85

58.65

4.78

15.65

3.51

6.08

63.06

3.50

2.32

0.28

2.48

0.16

CS (%)
33.05
2.79
53.45
6.06
1.56
1.89

FA
(%)
100
90
80
70
60
50
40
20
0

CS
(%)
0
10
20
30
40
50
60
80
100

The sample of aggregate was split into various


fractions, each consisting of particles of the same size.
The sieve analysis is conducted to determine the particle
size distribution in a sample of aggregate. The aggregate
used for making concrete are 4.75mm, 2.36mm,
1.18mm, 600 micron, 300 micron, and 150 micron. The
aggregate passes through 40mm and retained at 4.75mm
as coarse aggregate and the aggregate passes through
4.75mm and retained at 150 micron as fine aggregate.
Sieve can be done manually or mechanically. From the
below Fig.4.1, it is unstated that the fineness modulus of
copper slag (3.76) is more than the fineness modulus of
fine aggregate (2.73).Thus, the copper slag can be used
as fine aggregate in concrete.

3.3 CHEMICAL COMPONENTS OF COPPER


SLAG
The Chemical components of Copper Slag are given
in Table 3.3

Table 3.3 Chemical components of Copper Slag

Page 245

Author et. al. / INDECS-15

Fig.4.1 Sieve Analysis


4.2 SPECIFIC GRAVITY
The Specific gravity of aggregate is made use of in
design calculation of concrete mixes. The specific
gravity is calculated as the ratio between the weight of a
given volume of the material and weight of an equal
volume of standard material. Specific gravity of
aggregate is required in calculating the factor in
connection with the workability measurements. The
Specific gravity of materials are given in Table 4.1
Table 4.1 Specific gravity of materials
Specific
S.No
Material
gravity
1.
Cement
3.15
2.
Fly ash
2.14
3.
Fine aggregate
2.65
4.
Copper Slag
3.91
5.
Coarse aggregate
2.64

Fig.5.1 Workability of concrete


6. TESTS ON HARDENED CONCRETE
6.1 COMPRESSION TEST
In order to determine the compressive strength cube
mould of size 150150150 mm were casted. The cubes
were casted for different percentage of copper slag from
0% to 100%. The mould is cleaned and oiled properly
along its faces. Then the concrete is compacted properly
using tamping rod. Then the cubes are kept curing for
7day,28 day,60 day and 90 day. The compression test is
done according to the specification IS 516:1959. The
compressive strength is calculated using the formula
Compressive strength =P/A

5. TESTS ON FRESH CONCRETE

Slump test is the most commonly used method of


measuring the consistency of concrete. A concrete is
said to be workable if it can be easily mixed and easily
placed, compacted and finished. This result in large
voids, less durability and less strength the increase in
water cement ratio increases the slump and workability
but decreases the strength of concrete. From the below
Fig.5.1 shows the workability of concrete as Slump
value

COMPRESSIVE STRENGTH
(N/mm2)

5.1 Slump test

COMPRESSIVE STRENGTH
150
100
50
0
0 10 20 30 40 50 60 80 100
% OF REPLACEMENT OF COPPER SLAG

7th day

28th day

60th day

90th day

Fig.6.1 Replacement % Vs Compressive Strength


From the above fig 6.1 it is known that in the 7 th day
testing the compressive strength is maximum at 40% on
surrogating fine aggregate by copper slag which is
about 46.98 N/mm2 where the conventional concrete
having compressive strength of 21.62N/mm2 . On 28th
day testing the compressive strength 63.54N/mm2 by
surrogating 40% of fine aggregate were