Utility of Red Mud as an Embankment Material

December 2012 Volume 05 No 06 ISSN 0974-5904

INTERNATIONAL JOURNAL OF EARTH SCIENCES AND ENGINEERING

Indexed in: Scopus Compendex and Geobase (products hosted on Engineering Village)

Elsevier, Amsterdam, Netherlands, Chemical Abstract Services-USA, Geo-Ref Information

Services-USA

EARTH SCIENCE FOR EVERYONE

Published by

CAFET-INNOVA Technical Society Hyderabad, INDIA

www.cafetinnova.org

CAFET-INNOVA Technical Society

1-2-18/103, Mohini Mansion, Gagan Mahal Road, Domalguda

Hyderabad – 500 029, Andhra Pradesh, INDIA

Website: http://www.cafetinnova.org

Mobile: +91-9866587053, +91-7411311091

Registered by Government of Andhra Pradesh

Under the AP Societies Act. 2001 Regd. No.: 1575

The papers published in this journal have been peer reviewed by experts. The authors are solely

responsible for the content of the papers published in the journal.

Each volume, published in six bi-monthly issues, begins with February and ends with December

issue. Annual subscription is on the calendar year basis and begins with the February issue every

year.

Note: Limited copies of back issues are available.

Copyright © 2012 CAFET-INNOVA Technical Society

All rights reserved with CAFET-INNOVA Technical Society. No part of this journal should be

translated or reproduced in any form, Electronic, Mechanical, Photocopy, Recording or any

information storage and retrieval system without prior permission in writing, from CAFET-

INNOVA Technical Society.

INTERNATIONAL JOURNAL OF EARTH SCIENCES AND ENGINEERING

The International Journal of Earth Sciences and Engineering (IJEE) focus on Earth

sciences and Engineering with emphasis on earth sciences and engineering.

Applications of interdisciplinary topics such as engineering geology, geo-

instrumentation, geotechnical and geo-environmental engineering, mining engineering,

rock engineering, blasting engineering, petroleum engineering, off shore and marine

geo-technology, geothermal energy, resource engineering, water resources and

engineering, groundwater, geochemical engineering, environmental engineering,

atmospheric Sciences, Climate Change, and oceanography. Specific topics covered

include earth sciences and engineering applications, RS, GIS, GPS applications in earth

sciences and engineering, geo-hazards such as earthquakes, landslides, tsunami, debris

flows and subsidence, rock/soil improvements and development of models validations

using field, laboratory measurements.

Professors / Academicians / Engineers / Researchers / Students can send their papers

directly to: [email protected]

CONTACT:

For all editorial queries:

D. Venkat Reddy (Editor-in-Chief)

Professor, Department of Civil Engg.

NIT-Karnataka, Surathkal, INDIA

� +91-9739536078

� [email protected]

All other enquiries:

Hafeez Basha. R (Managing Editor)

� +91-9866587053


Raju. Aedla (Editor)

� +91-7411311091


EDITORIAL COMMITTEE

D. Venkat Reddy

NIT-Karnataka, Surathkal

EDITOR-IN-CHIEF

Trilok N. Singh

IIT-Bombay, Powai, Mumbai

EXECUTIVE EDITOR

R. Pavanaguru

Osmania University, Hyderabad

EXECUTIVE EDITOR

P. Ramachandra Reddy

Scientist G (Retd.), NGRI, Hyderabad

EXECUTIVE EDITOR

Hafeez Basha R

K L University, Vijayawada

MANAGING EDITOR

Raju Aedla


EDITOR

INTERNATIONAL EDITORIAL ADVISORY BOARD

Paul M. Santi Colorado School of Mines

USA

Choonam Sunwoo

Korea Inst. of Geo-Sci & Mineral

SOUTH KOREA

Hsin-Yu Shan National Chio Tung University

TAIWAN

Hyun Sik Yang

Chonnam National Univ Gwangu

SOUTH KOREA

Krishna R. Reddy University of Illinois, Chicago

USA

L G Gwalani NiPlats Australia Limited

AUSTRALIA

Abdullah MS Al-Amri King Saud University, Riyadh

SAUDI ARABIA

Suzana Gueiros Dra Engenharia de Produção

BRAZIL

Zhuping Sheng Texas A&M University System

USA

Luigia Binda DIS, Politecnico di Milano, Milan

ITALY

Gonzalo M. Aiassa Cordoba Universidad Nacional

ARGENTINA

Shuichi TORII Kumamoto University, Kumamoto

JAPAN

Ganesh R. Joshi University of the Rykyus, Okinawa

JAPAN

Kyriakos G. Stathopoulos DOMI S.A. Consulting Engineers

Athens, GREECE

Nguyen Tan Phong Ho Chi Minh City University of

Technology, VIETNAM

Robert Jankowski

Gdansk University of Technology

POLAND

Joanna Maria Dulinska Cracow University of Technology

POLAND

U Johnson Alengaram

University of Malaya, Kuala

Lumpur, MALAYSIA

Anil Cherian United Arab Emirates

DUBAI

Vahid Nourani Tabriz University

IRAN

Paloma Pineda University of de Sevilla, Seville

SPAIN

Nicola Tarque

Department of Engineering

Catholic University of Peru

Jaya naithani

Université catholique de Louvain

Louvain-la-Neuve, BELGIUM

P Hollis Watts

WASM School of Mines

Curtin University, AUSTRALIA

Abdullah Saand

Quaid-e-Awam University of Eng.

Sc. & Tech., Sindh, PAKISTAN

S Neelamani

Kuwait Institute for Scientific

Research, SAFAT, KUWAIT

Mani Ram Saharan

National Geotechnical Facility

DST, Dehradun, INDIA

S Viswanathan

IIT- Bombay, Powai, Mumbai

Maharashtra, INDIA

Ramana G V

IIT– Delhi, Hauz Khas

New Delhi, INDIA

Subhasish Das

IIT- Kharagpur, Kharagpur

West Bengal, INDIA

K U Maheshwar Rao

IIT- Kharagpur, Kharagpur

West Bengal, INDIA

Kalachand Sain

National Geophysical Research

Institute, Hyderabad, INDIA

Katta Venkataramana NITK- Surathkal

Karnataka, INDIA

M K Nagaraj

NITK- Surathkal

Karnataka, INDIA

R Sundaravadivelu

IIT- Madras

Tamil Nadu, INDIA

G S Dwarakish NITK- Surathkal

Karnataka, INDIA

M R Madhav JNTU- Kukatpally, Hyderabad

Andhra Pradesh, INDIA

Chachadi A G Goa University, Taleigao Plateau

Goa, INDIA

S M Ramasamy

Gandhigram Rural University

Tamil Nadu, INDIA

Gholamreza Ghodrati Amiri

Iran University of Sci. & Tech.

Narmak, Tehran, IRAN

C Natarajan NIT- Tiruchirapalli,

Tamil Nadu, INDIA

R Bhima Rao

IMMT, Bhubaneswar

Odissa, INDIA

Shamsher B. Singh BITS- Pilani, Rajasthan

Rajasthan, INDIA

Pradeep Kumar R IIIT- Gachibowli, Hyderabad


N Ganesan

NIT- Calicut, Kerala

Kerala, INDIA

D P Tripathy

National Institute of Technology

Rourkela, INDIA

E Saibaba Reddy JNTU- Kukatpally, Hyderabad


Chowdhury Quamruzzaman

Dhaka University

Dhaka, Bangladesh

Parekh Anant kumar B Indian Institute of Tropical

Meteorology, Pune, INDIA

Datta Shivane

Central Ground Water Board

Hyderabad, INDIA

Gopal Krishan

National Institute of Hydrology

Roorkee, INDIA

P Sreenivas Sarma

Chaitanya Bharathi Institute of

Technology, Hyderabad, INDIA

Prasoon Kumar Singh

Indian School of Mines, Dhanbad

Jharkhand, INDIA

A G S Reddy

Central Ground Water Board,

Pune, Maharashtra, INDIA

Rajendra Kumar Dubey

Indian School of Mines, Dhanbad

Jharkhand, INDIA

Subhasis Sen

Retired Scientist

CSIR-Nagpur, INDIA

M V Ramanamurthy

Geological Survey of India

Bangalore, INDIA

C Sivapragasam

Kalasalingam University,

Tamil Nadu, INDIA

Bijay Singh Ranchi University, Ranchi

Jharkhand, INDIA

B R Raghavan

Mangalore University, Mangalore

Karnataka, INDIA

CH Hanumantha Rao K. L. University, Guntur


Anand V. Shivapur SDM College of Engg. and Tech.

Karnataka, INDIA

S Suresh Babu Adhiyamaan college of Engineering

Tamil Nadu, INDIA

Kripamoy Sarkar

Assam University

Silchar, INDIA

Xiang Lian Zhou

ShangHai JiaoTong University

ShangHai, CHINA

Naveed Ahmad

University of Engg. & Technology,

Peshawar, PAKISTAN

Nandipati Subba Rao

Andhra University, Visakhapatnam


M Suresh Gandhi

University of Madras,

Tamil Nadu, INDIA

Debadatta Swain

National Remote Sensing Centre

Hyderabad, INDIA

H K Sahoo

Utkal University, Bhubaneswar

Odissa, INDIA

R N Tiwari Govt. P G Science College, Rewa

Madhya Pradesh, INDIA

B M Ravindra

Dept. of Mines & Geology, Govt. of

Karnataka, Mangalore, INDIA

M V Ramana

CSIR NIO

Goa, INDIA

N Rajeshwara Rao

University of Madras

Tamil Nadu, INDIA

R Baskaran

Tamil University, Thanjavur

Tamil Nadu, INDIA

Salih Muhammad Awadh

College of Science

University of Baghdad, IRAQ

Sonali Pati

Eastern Academy of Science and

Technology, Bhubaneswar, INDIA

K. Bheemalingeswara

Mekelle University

Mekelle, Ethiopia

Samir Kumar Bera

Birbal sahni institute of

palaeobotany, Lucknow, INDIA

Raj Reddy. Kallu

University of Nevada

1665 N Virginia St, RENO

Manish Kumar

Tezpur University

Sonitpur, Assam, INDIA

C N V Satyanarayana Reddy

Andhra University

Visakhapatnam, INDIA

S M Hussain

University of Madras

Tamil Nadu, INDIA

Safdar Ali Shirazi

University of the Punjab,

Quaid-i-Azam Campus, PAKISTAN

T J Renuka Prasad

Bangalore University

Karnataka, INDIA

S S S V Gopala Raju

GITAM University

Hyderabad, INDIA

Glenn T. Thong

Nagaland University

Meriema, Kohima, INDIA

Mohammed Sharif

Jamia University

New Delhi, INDIA

A M Vasumathi

K.L.N. College of Inf. Tech.

Pottapalayam, Tamil Nadu, INDIA

Ashutosh Das

PRIST, Tamilnadu

Tamil Nadu, INDIA

INDEX Earth Science for everyone

Volume 05 December 2012 No.06

EDITORIAL NOTE

Is the Prediction of Human Generation End by the Year 2012? A Geologic Perspective on

Mass Extension.

By BUSNUR RACHOTAPPA MANJUNATHA and D. VENKAT REDDY

RESEARCH PAPERS

Luminescence Techniques as a Low-Cost Geophysical Tool in Mineral Exploration: Some

Examples

By R. DHANA RAJU

1472-1480

Ensemble Empirical Mode Decomposition of the Lightning Return Stroke

By XUQUAN CHEN and WENGUANG ZHAO

1481-1491

Peak Ground Acceleration on Bedrock and Uniform Seismic Hazard Spectra for different

Regions of Behbahan, Iran

By SEYED ALI RAZAVIAN AMREI, GHOLAMREZA GHODRATI AMIRI, ARMAN SAED

and ALI SABZEVARI

1492-1499

Coal Mine Seal Design – Numerical Approach

By RAJ R. KALLU

1500-1509

A Note on Three Way Quality Control of Argo Temperature and Salinity Profiles - A Semi-

Automated Approach at INCOIS

By T. V. S. UDAYA BHASKAR, E. PATTABHI RAMA RAO, R. VENKAT SHESU

and R. DEVENDER

1510-1514

Peak Ground Acceleration on Bedrock and Uniform Seismic Hazard Spectra for Different

Regions of Zanjan, Iran

By GHOLAMREZA GHODRATI AMIRI, SEYED ALI RAZAVIAN AMREI, ALI SABZEVARI,

ARMAN SAED and YASER SHOKRANI

1515-1521

Heavy Minerals and Provenance of the Lower Gondwana Sandstones, Ong-River

Gondwana Basin, Odisha

By ARYA GOUTAMI NATH and RABINDRA NATH HOTA

1522-1529

Rating of Tunnel by Visual Field Inventory – A Case Study of Punasa Tunnel, District

Khandawa, M. P. India

By GUPTA. M. C, SINGH. B. K and SINGH. K. N

1530-1534

Geology and Characteristics of Metalimestone-Hosted Iron Deposit near Negash, Tigray

and Northern Ethiopia

By SOLOMON GEBRESILASSIE, BHEEMALINGESWARA KONKA and FISEHA ADHANOM

1535-1544

Ore Fluids Associated With the Metasediment Hosted Central Auriferous Zone of Gadag

Gold Field, Karnataka

By M. A. MALAPUR, S. MANJUNATHA, B. CHANDAN KUMAR and A. G. UGARKAR

1545-1551

A Case Study of Particle Size Distribution of Paleosols around Sargur Supracrustal

Terrain, Dharwar Craton, South India

By NARGES GOHARI RAD, PRAKASH NARASIMHA. K. N and MADESH. P

1552-1559

Effect of Ground Moisture on Spread of Contaminants in Sand Deposit -An Experimental

Study

By E. SAIBABA REDDY, SINA BORZOOEI and G. V. NARASIMHA REDDY

1560-1566

Identification of Rain/No-Rain Events using QuikSCAT Scatterometer Data

By PRITI SHARMA, RAJESH SIKHAKOLLI, B. S. GOHIL and ABHIJIT SARKAR

1567-1571

Genesis of Thermal Springs of Odisha, India

By S. C. MAHALA, P. SINGH, M. DAS and S. ACHARYA

1572-1577

Remote Sensing Studies in Delineating Hydrogeological Parameters in the Drought-Prone

Kuchinda-Bamra Area in Sambalpur District, Odisha

By NANDITA MAHANTA and H. K. SAHOO

1578-1583

Impact Analyses of Industrial and Mining Activities on Groundwater Regime -Case Studies

in Goa

By A. G. CHACHADI

1584-1589

Identification of Artificial Recharge Sites in a Hard Rock Terrain using Remote Sensing

and GIS

By S. SARAVANAN

1590-1598

Drinking and Irrigation Water Quality in Jalandhar and Kapurthala Districts, Punjab,

India: Using Hydrochemsitry

By P. PURUSHOTHAMAN, M. SOMESHWAR RAO, B. KUMAR, Y. S. RAWAT,

GOPAL KRISHAN, S. GUPTA, S. MARWAH, A. K. BHATIA, Y. B. KAUSHIK,

M. P. ANGURALA and G. P. SINGH

1599-1608

Groundwater Quality Assessment around Talabasta Area, Banki Sub-Division, Odisha,

India

By ROSALIN DAS, MADHUMITA DAS and SHREERUP GOSWAMI

1609-1618

Terrain Analysis and Hydrogeochemical Environment of Aquifers of the southern West

Coast of Karnataka, India

By S. S. HONNANAGOUDAR, D. VENKAT REDDY and MAHESHA. A

1619-1629

Water Quality Modeling and Management of Karanja River in India

By BASAPPA. B. KORI, SHASHIKANTH MISE and SHASHIDHAR

1630-1638

Prediction of Penetration Rate and Sound Level Produced during Percussive Drilling using

Regression and Artificial Neural Network

By S. B. KIVADE, Ch. S. N. MURTHY and HARSHA VARDHAN

1639-1644

Utility of Red Mud as an Embankment Material

By SUBRAT KUMAR ROUT, TAPASWINI SAHOO and SARAT KUMAR DAS

1645-1651

Rapid Chloride Penetration Test on Geopolymer Concrete

By SHANKAR H. SANNI and R. B. KHADIRANAIKAR

1652-1658

News and Notes

NIT Professor spoke at an International Conference in SanFransisco, USA. i

Dr. Subhash C. Yaragal, is conferred with Prof. Satish Dhawan Young Engineer State

Award -2011 ii

www.cafetinnova.org

Indexed in

Scopus Compendex and Geobase Elsevier, Chemical Abstract

Services-USA, Geo-Ref Information Services-USA

ISSN 0974-5904, Volume 05, No. 06

Dec 2012, Editorial Note

Volume 5, No. 6, December 2012

Is the Prediction of Human Generation End by the Year 2012?

A Geologic Perspective on Mass Extension.

BUSNUR RACHOTAPPA MANJUNATHA1 and D. VENKAT REDDY

2

1Department of Marine Geology, Mangalore University, Mangalagangothri-574 199, India 2Department of Civil Engineering-National Institute of Technology –Karnataka, India

Email: [email protected], [email protected]

There are many documentaries, movies and armature articles regarding the extinction of human generation by end of

the year 2012. The first prediction has been the Mayan astrological calendar that built out of stones indicates an

abrupt end of the world on 21st December 2012. Like-wise, some of religious books, including the Bible implicate

the end of human domain by 2012. This has caused curiosity and anxiety among people across the world. The main

purpose of this note is to address the public form geologist perspectives about the mass extension of life.

The age of the earth has been measured by radioisotopic method to about 4550 million years (Ma) before present.

Form the violent beginning of the earth, it took nearly 800 Ma to cool down (Hadean Eon) by allowing to from

mainly igneous and metamorphic rocks from 3800 Ma to 2500 Ma (Achaean Eon), principally sedimentary rocks

from 2500 Ma to 570 Ma (Proterozoic Eon) and mixed types of rocks on land as well as in the oceans from 570 Ma

to the present (Phanerozoic Eon). The origin of life and its commencement during the Achaean Eon is still infancy

because of the lack of preservation of fossils in igneous and metamorphic rocks as these rocks form at high

temperatures. The atmospheric conditions during Hadean and to some extent Archaean Eons were not conducive for

the advancement/evolution of life as the atmosphere was dominated by large quantities of water vapour, carbon

dioxide, sulphur and nitrogen gases and perhaps hydro chloric acid. However, there are some evidences of life

existed, for e.g., biogenic stromatolites dated older than 3500 Ma (belong to the Archaean Eon). The existence of

life during the Proterozoic Eon has been very well documented from the sedimentary rocks. During this Eon, the

free diatomic oxygen that necessary for evolution of life gradually built up in the atmosphere due increase in the rate

of photosynthesis by the cyanobacteria (blue green algae).

The life prevailed during the Proterozoic was principally of unicellular organisms, while at the beginning of

Phanerozoic Eon, a sudden advancement of life noticed. Over a little more than 500 Ma, most of species including

human beings originated as the last species. Based on the biotic evolution, the Phanerozoic Eon can be classified

into Palaeozoic Era (early life; 570-245 Ma), Mesozoic Era (middle life; 245-65 Ma) and Cenozoic Era (recent life;

65 Ma -today).

In contrast to the Proterozoic Eon, the Paleozoic Era began with the origin of multi-cellular organisms, for e.g.,

trilobites, jelly fish and worms (during the Cambrian period; 545-505 Ma), suggesting that an abrupt change in the

life on earth. This was followed by snail, sponges during the Ordovician (505-438 Ma), corals and star fish during

the Silurian (438- 408 Ma), fish, lungfish and fern during the Devonian (408 - 360 Ma), amphibianin, insect and

trees during Carboniferous (360-286 Ma), fin back reptiles and reptiles during the Permian periods (286-245 Ma).

The life existed during Mesozoic Era was quite different than that during the Paleozoic Era in terms of the

advancement of reptiles, primitive mammals and gymnosperms. The Mesozoic Era can be demarcated into three

periods Triassic (245-208 Ma) Jurassic 208-144 Ma) and Cretaceous (144-65 Ma). The life existed in the former

period comprises of (i) the first appearance of dinosaurs and ammonoite, followed by (ii) dominance of giant

dinosaurs both in the oceans and on land, first appearance mammals and conifers, and (iii) dominance of flying

dinosaurs, carnivorous dinosaurs and horned dinosaurs, flowering plants, and insects during the latter period

respectively. The Cenozoic Era dominated by the mammals began with dominance of flowering plants, ancestral

horse, insects, birds whale clawed mammals camel dating back to 65 to 1.6 million years. The first appearance of

humans noticed during the Quaternary period (1.6 Ma) along with mammoth saber toothed tiger.

However, the life existed earlier either became greatly reduced or extinct in the latter time span of the earth. Such

process can be termed as mass extinction of life. A mass extinction usually noticed during the end of particular

geologic period. Mass extinction refers to the species diminishing or almost end of biological world probably due to

BUSNUR RACHOTAPPA MANJUNATHA and D. VENKAT REDDY

Editorial Note

large-scale calamities or catastrophic processes such as earth quakes, tsunamis, volcanic eruptions and meteoritic

impact. However, there could have been climatic factors such as famine, drought, and biological factors for instance,

genetic factors causing infertility, and human factors leading to deforestation and large-scale hunting of animals.

There are as many as six major mass extinctions and a number of smaller ones occurred in the Phanerozoic Eon

(history of the earth):

(1) 52 % reduction in the families of trilobite, sponge and gasteropod occurred during the late Cambrian,

(2) 24 % reduction in the families of trilobite, brachiopod, crinoid and echinoid families during late Ordovician,

(3) 30 % reduction in the families of coral, stromatoporoid, trilobite, ammonoide bryozoan, brachiopod and fish

during late Devonian,

(4) 50 % reduction in the families of bryozoans and reptiles during late Permian,

(5) 35 % reduction in the families of brachiopod, ammonoite fish and reptile during late Triassic and

(6) 26% reduction in the families of beleminites, corals, echinoids, sponges and planktonic foraminifera during the

ate Cretaceous period.

Among the above, the major extinction of rugose corals, trilobites, blastoids, inadunate, flexibiliate, camerate,

crinoids, productid, brachiopods and fusulinid foraminifera occurred during the late Permian, conodonts during the

late Triassic and ammonoites, rudistid mollusks, dinosaurs and large marine reptiles at the end of Cretaceous period.

However, certain species, for e.g., sponges, brachiopods, bivalves and ostrocods continue to exist through out the

Phanerozoic Eon with variations in their populations. The dominance of sponges, bryozoans, brachiopods and

ostracods particularly during the transition particularly during the Palaeozoic Eon to Mesozoic Eon. This suggests

that the dominant species become extinct by the end of particular geological period by making a place for the

advanced life.

In most of major mass extinctions, the main causative factor was the trigger of large-scale volcanic eruption. The

Antrim volcano erupted in Ireland about 511 Ma that must have caused mass extinction noticed in between

Cambrian and Ordovician periods.

The boundary between Permian and Triassic has been the most conspicuous by a massive and large-scale Siberian

volcanic eruption that occurred at about 251-250 Ma ago. The volume of eruption estimated at 1.0 - 4 million cubic

kilometers covering an area of 2 million sq km. The dinosaurs extinction noticed at the end of Cretaceous period can

be linked to Deccan volcanic eruption (India) where the volume of lava erupted has been estimated to 1.5 million

cubic kilometers and cover an area more than 500,000 sq km. Nevertheless, meteorites bombardment on earth as

well as cold periods appeared to have been significant foe the mass extinction. Therefore, it appears that the rate of

extinction is controlled by the quantum of volcanic eruption, as it has been mentioned above that Siberian volcanic

eruption led to the extinction of about 50 families of biota while the Deccan volcanism accounts to about half of

that from Siberian volcanism.

It is clear from the recent studies that volcanic eruption can reduce the global temperature by 1- 2 oC because of the

absorption incoming solar radiation by volcanic materials. This gradually reduces photosynthesis leading to a

considerable reduction in the species as well as their population as a consequence of shortage of food. Another thing

that volcano emits lot of oxides of carbon, sulphur and nitrogen leading to the acidification of the ecosphere.

Volcanoes may often trigger ice age, for instance, the boundary between Permian and Triassic, and Cretaceous and

Tertiary periods marked by cold (glacial) conditions. Other processes responsible for the mass extinction of life on

earth are climate, physical, chemical and biological warfare. Climate change has serious consequence on sustenance

of life on earth. For instance, the severe drought that occurred over 4000 years ago collapsed the old world’s

civilization, for e.g., Akkadian, Mayan, Indus Valley and Chinese civilizations. Similarly long-term climate change

such as solar activity due to human impacts and prevalence of glacial (cold) conditions can disrupts life on the planet.

Now the puzzle is that when do the human’s generation going to extinct? Our answer to this question is not very

soon and certainly not during the end of 2012, because the catastrophic processes may not occur simultaneously on

the earth. More than that the highly civilized human beings today can bear the risk of extinction. However, gradual

processes, such as evolution can cause a considerable amount of species reduction and population explosion because

of scarcity natural resources. Still longer processes are the consequence of solar flares and Sun’s evolution can lead

to substantial extinction of life on earth.

In a nutshell, we conclude that although mass extinction processes are violent and catastrophic, however, it is

evident that they mark origin of new species. For example, the extinction of the dinosaurs led the rise of the

mammals. As William Shakespeare said life of a human being can be comparable to a “drama stage”, similarly the

whole earth is a stage for origin of new species, while extinction for living ones. In this context, extinction of

humans is inevitable, but a long way to go.

www.cafetinnova.org

Indexed in

Scopus Compendex and Geobase Elsevier, Chemical

Abstract Services-USA, Geo-Ref Information Services-USA

ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1472-1480

#02050601 Copyright ©2012 CAFET-INNOVA TECHNICAL SOCIETY. All rights reserved.

Luminescence Techniques as a Low-Cost Geophysical Tool in

Mineral Exploration: Some Examples

R. DHANA RAJU A.M.D., Dept. of Atomic Energy

Dept. of Applied Geochemistry, Osmania University

ASR Mining Company, Kondapur, Hyderabad – 500 084

Email: [email protected]

Abstract: Luminescence is the phenomenon of emission of light or photons, mainly in the visible domain, from a substance when it is stimulated by any means other than heating to incandescence. It is believed to result from motion of electrons. Based on the nature of motion, it is of numerous types and constitutes a low-cost geophysical tool. Of these types, three, viz., fluorescence (-phosphorescence), cathodoluminescence (CL) and thermoluminescence (TL) have many applications in mineral exploration, which are documented in this paper through the following examples: (i) fluorescence (-phosphorescence) of some uranyl- and non-radioactive-minerals; (ii) CL in discriminating Uraniferous sandstone/conglomerate from barren ones, and in probing hydrothermal

alterations, fluid migration pathways and fractures, all of which have critical bearing on mineralization; and (iii) Natural TL on whole-rock samples for: (a) broad classification of the schistose rocks

and (b) discrimination of the Uranium and Rare Element mineralized horizons fromethe adjacent barren zones, as demonstrated in the Jublatola, Domiasiat and Bast r-Malkangiri areas.

The above applications point to the utility of these low-cost, direct and sensitive geophysical techniques as a powerful exploration tool. These techniques can be profitably used during the reconnoitory stage of exploration, prior to costly drilling, in the virgin areas or in deciphering the concealed mineralized zones as well as for predicting the blind-extensions of the already established mineralized horizons.

Keywords: Fluorescence (-Phosphorescence), Cathodoluminescence, Thermoluminescence, Mineral Exploration.

Introduction:

Luminescence is the phenomenon of emission of light or photons, mainly in the visible domain, from a substance when it is stimulated by any means other than heating to incandescence. It is believed to result from motion of electrons and is classified as per the nature of motion. It is of several following types: radioluminescence is by excited X-ray photons and γ-rays, and bombardment of α- and β-nuclear particles; chemiluminescence produced by chemical reactions, e.g., oxidation of P; electroluminescence by electrical discharges; triboluminescence due to mechanical deformation by rubbing or crushing crystals like diamond and sphalerite; ionoluminescence generated under an energetic beam as in an ion microprobe; bioluminescence by living organisms like some bacteria, glowworms, fireflies and many deep-sea fish; fluorescence; phosphorescence; crystalloluminescence during crystallization from a solution like arsenic oxide, As2O3; cathodoluminescence produced by energetic electrons; thermoluminescence due to an activator in a mineral when it is heated; and photoluminescence

involving selective energy of photons to excite electronic levels of luminescent centers. Of these, fluorescence (-phosphorescence), cathodoluminescence (CL) and thermoluminescence (TL) are used as tools in (a) identification of minerals, including ore minerals, which emit these phenomena; (b) examination of certain features like zoning, overgrowth and ultra-fine inclusions present in such minerals as they are much easy (due to better clarity and contrast) to identify than by optical microscopy alone; and (c) mineral exploration. Fluorescence (-phosphore-scence), CL and TL, which constitute the low-cost, direct and sensitive geophysical tools, have notable applications in mineral exploration, especially during its reconnoitory and semi-detailed stages. These applications, together with some Indian examples, are presented in this paper to demonstrate their relevance in mineral exploration, particularly before undertaking costly drilling.

Fluorescence (-Phosphorescence):

The emission of light from a substance like mineral when it is exposed to direct radiation such as ultraviolet

1473 R. DHANA RAJU

International Journal of Earth Sciences and Engineering ISSN 0974-5904, Vol. 05, No. 06, December 2012, pp. 1472-1480

(UV) light, or in certain instances to an electrical discharge in a vacuum tube is called fluorescence that ceases when radiation stops. This phenomenon is best exhibited by the mineral, fluorite and, hence, the name. Fluorescence is the process of emission of electromagnetic radiation produced by energy transitions, not just in the visible range. A number of others minerals, like scheelite, malayaite and uranyl (U6+ or secondary U) minerals such as autunite, uranophane, gummite, torbernite and schroekingerite fluoresce due to the presence of activators or spikes (Mn, U, Rare Earths) as impurities in them. If a beam of white light were passed through a cube of colourless fluorite, a delicate violet colour can be seen in its path due to change of refrangibility in the transmitted light. The continued emission of light by a substance (phosphor) for a finite time even after exposure to light, or an electrical discharge and especially after heating is called phosphorescence. Fluorescence and phosphorescence differ only in the amount of time it takes for electrons to return to their ground states. With fluorescence, vacant lower energy positions are filled within small fractions of a second. Phosphorescent materials, however, continue emitting light significantly after the exciting radiation has been turned off, sometimes for hours. Since the process of displacing electrons into higher energy configurations absorbs electromagnetic radiation, the electrons as they return to the ground state emit radiation. The emitted radiation is always of lower energy than the radiation used to displace the electron, and of a definite wavelength, corresponding to the difference between the excited state and the ground state. This process is most transparent for X-rays. Many transitions contribute to fluorescence in the visible range, and the spectra are not as sharp as those in the range of X-rays. Furthermore, similar to colour, visible fluorescence depends critically on trance elements and defects, but is nevertheless a diagnostic property used in mineral identification and mineral prospecting. Fluorite is highly phosphorescent and gives off different colours of emerald green, purple, blue and reddish tints, after heating to above 1500C. Scheelite (CaWO4), an important ore of tungsten, is another mineral with strong fluorescence. To discriminate scheelite from carbonates is rather difficult in hand specimens, but it is immediately recognized by its bright white fluorescence, when irradiated with ultraviolet light. The phenomenon of fluorescence and phosphorescence under UV radiation is used as a guide in identification of some minerals as well as to get information on their zoning, inclusions and crystal growth. UV light is of two wavelengths, viz., short wave of 185-300 nm and long wave of 300-400 nm, with the former cannot and the latter can penetrate cover glass of a thin section. Since the short-wave UV light excites most fluorescence in minerals, it is better to

observe under it in darkness the fluorescence and phosphorescence on uncovered polished slabs, polished thin-sections and surface of hand-specimens. A useful model for laboratory work is MINERALIGHT model MPR2 (or later advanced ones of M/s. Ultraviolet Products Inc., San Gabriel, California 91778, USA). This has two externally clamped tubular bulbs that can be adjusted in position above the polished surface on either side of microscope-objectives and, with a short UV filter on one or both bulbs, the fluorescence colours can be studied through the microscope. Furthermore, the fluorescent minerals on the surface of a hand-specimen can be scrapped for their correct identification later by XRD. When once the identity of fluorescent mineral in a terrain or an area is established, then the UV fluorescent technique can be used as a rapid mineral exploration technique like in the Sn-W mineralized region of Southeast Asia. For a comprehensive list of minerals, rocks and gemstones, which exhibit UV reactions, the reader is referred to Gleason (1960).

Warning: Since UV light, especially of the short wavelength, can cause temporary or permanent blindness, it should not be looked either directly from the lamp or its reflection from surface. Hence, it is necessary to observe UV fluorescence through spectacles or microscope system.

Fluorescence of Uranium Minerals and Non-

Radioactive Minerals:

It may be noted that fluorescence (and phosphorescence) exhibited by certain non-radioactive minerals is of impurity-activated type, such as zinc orthosilicate activated by manganese as impurity, whereas that by uranium (and thorium) minerals is an intrinsic fluorescence due to uranyl [(UO2)

2+] ion. About the fluorescence of uranium minerals, the following generalizations can be made (Frondel, 1958, p. 357):

a) All the fluorescent uranium minerals contain uranyl ion and are secondary in origin, whereas the uranium minerals containing quadrivalent uranium, (U4+), such as uraninite, coffinite and uranoan varieties of niobate-tantalates, which are generally black in colour and primary in origin, do not show fluorescence.

b) The characteristic emission colour of the strongly and moderately fluorescent uranium minerals is greenish yellow to yellowish green (except andersonite and liebigite, which are green). These minerals comprise chiefly the hydrated uranyl phosphates and arsenates containing alkaline earths, alkalies or hydrogen (members of these groups containing copper or iron fluoresce weakly, if at all), together with certain uranyl sulphates and carbonates.

1474 Luminescence Techniques as a Low-Cost Geophysical

Tool in Mineral Exploration: Some Examples


c) The uranyl silicates/vanadates and hydrated oxides fluoresce weakly or not at all.

d) The uranyl minerals containing lead fluoresce either weakly in yellow-brown or brown rather than yellow-green colours, or do not fluoresce at all.

e) Dehydration, in general, reduces the intensity of fluorescence.

Fluorescence in Exploration for U:

As many uranium salts fluoresce under UV light of suitable wavelengths (2537 Å and 3660 Å), the technique of fluorescence is an appropriate tool for detection of some uranium minerals that might otherwise escape detection by routinely used GM counter or scintillometer because of their youth and fine subdivision. In fact this is one of the methods of radiation survey (the others are radiometric survey using gamma-ray detectors of GM counter and scintillometer, radio-activation analysis, autoradiography, radiation track analysis and radioactive tracers) (Boyle, 1982, pp. 166-174). Since certain substances (e.g., bitumen, petroleum) and non-radioactive minerals interfere with, inhibit or mimic uranium fluorescence, corroborating analyses should always be carried out to definitely establish the presence or absence of uranium. After establishing the presence of U in the fluorescent minerals, recorded either in the field or in laboratory examination of samples, their exact identification by XRD is a must. It helps in understanding the geochemical system of U operating in the sampled area, like establishing the exact nature of transport of U (after its release from a source rock like fertile granite) as a uranyl carbonate or phosphate complex, and their respective Eh-pH conditions. Such a probing of the geochemical system helps in better planning of exploration for U, like in selecting the potential target areas and their specific zones having necessary reducing environment for precipitation and concentration of U. For more details on the use of fluorescence as a tool in exploration for U, the reader is referred to Walenta (1959) and Gleason (1972).

Fluorescence in Exploration for Non-Radioactive

Minerals:

As many non-radioactive minerals like fluorite, scheelite, sphalerite, calcite, aragonite and willemite exhibit the phenomenon of fluorescence (with some even phosphorescence), the fluorescence technique may be used as a tool in exploration for these minerals. For example, in exploration for tin and tungsten, which are usually associated together in the S-type granitic rocks and their related pegmatites, reconnaissance survey using UV light can be carried out to locate zones of fluorescence in which the main tungsten mineral, namely scheelite occurs. Tin mineralisation, mainly in the form of cassiterite, occurs usually in the altered and

greisenised zones of S-type granitic rocks and their related/associated pegmatites, and more concentrated in the eluvial-colluvial-alluvial placers derived from these rocks. In all these, fluorite and occasionally scheelite may also occur. Hence, fluorescence survey may be employed during reconnoitory stage of exploration. This may help in delineating potential zones of Sn (and W)-mineralization. Like in exploration for U, this should be, of course, followed by semi-detailed and detailed exploration to establish the Sn-W mineralization by other more reliable tools. It is, thus, feasible to use fluorescence survey with UV light as a tool during reconnoitory stage of exploration in the areas like Bastar Sn-bearing pegmatite belt in Chhattisgarh, India as well as in other provinces such as Sn-W province in Malaysia and Myanmar. Furthermore, fluorescence survey may also help in differentiating S-type granitic rocks that are potential for Sn-W mineralisation from the other types, viz., I-, A- and M-types, which are usually devoid of such mineralisation. In such a scenario, the fluorescence survey becomes a potential tool even in selection of target areas in reconnoitory exploration for Sn and W.

Cathodoluminescence:

When minerals, with impurities of activators like U, Th, Rare Earths and Mn, are bombarded in vacuum by an electron-beam, they emit ectron excited luminescence that is termed as ‘Cathodoluminescence’ (CL). Since last few decades, CL is becoming a very useful technique having many applications in geosciences. Thus, CL can supplement petrographic observations and geological considerations, made by polarizing microscopy and other conventional analytical methods. CL, in combination with XRD, SEM and EPMA, contributes to the phase-characterization of technical products and waste materials. CL microscopy alone provides at least a clear differentiation of several phases even in samples with a high content of non-crystalline components or extremely heterogeneous material. In combination with spectral measurements of the CL emission, conclusions concerning structural states of solids and trace element incorporation are possible. This enables detection of differences in the crystal structure and monitoring of diffusion process or can reveal processes of alteration and formation of new phases. A combination of CL microscopy and image analysis allows quantification of different mineral phases. Only when the phases in rocks or technical products with a complex mineralogy have no luminescence, the use of CL to determine phase proportions is limited. All these advantages of CL offer promising perspectives in the specific investigations of minerals and technical products in research and industrial applications (Gotze, 2000). Furthermore with CL technique, new subjects like allochthonous vs.

1475 R. DHANA RAJU


autochthonous soil, alterations during transportation, palaeopermeability and palaeoporosity reconstruction, and paleoenvironmental studies could be investigated (Pagel et al. 2000).

Advantages of CL:

Compared to the UV light and XRF methods, CL has the advantages of: (a) high sensitivity, e.g., 6 x 10-3 ppm Dy; (b) no serious health hazard unlike XRF; (c) no excitation radiation to filter unlike in UV light excitation; and (d) excitation not variable with REE; however, it requires vacuum. Compared to the electron microprobe, CL has advantages like (a) features readily comparable with those seen in normal petrographic study; (b) use of uncoated and unpolished thin-sections; and (c) illumination of much larger areas, up to 10 x 15 mm.

Preparation of Samples for CL Study:

In preparation of samples for CL study, thinning by emery powder creates a dead or de-structured layer that is removed by polishing to observe the maximum luminescence. A well-polished slab or thin section is required to obtain a good CL image. The CL emission is linked to the temperature for both the position and the width of the peaks. Close to 0oK, the transitions are phonon-free, pure and noiseless and, therefore, the peaks appear as narrow lines, with a much higher intensity. The electronic beam increases strongly the temperature of the sample. This emphasizes the problem for the beam conditions in the scanning mode: focused or not focused spot, slow scan or TV scan. The observed colours are not stable over time with an optical microscope. In the first few seconds, there is a fugacious emission and, therefore, the sample is difficult to photograph. Afterwards, there is a new colour set-up that is often stable. Instead of optical observation of the colours, it is possible to record these changes in the spectra. With CCD detectors, the entire spectrum can be recorded in a very short time and repeatedly. The intensity could increase with time, when new centers are created under the electron beam. In the case of self-activated peaks corresponding to bound defects or oxygen vacancy, the number of defects increases under the beam with time. One of the new and very interesting approaches is valence changing during electron bombardment. For example, Eu3+ ions can capture electrons and be reduced to Eu2+ ions (Pagel et al. 2000).

Analytical Systems of CL:

The CL-analytical system is of two types: (1) CL generated by an electron gun, coupled to an optical microscope (cold cathode optical microscope CL system) and (2) CL as attachment to EMP, SEM and TEM. Other combinations include as the attachment of

a hot cathode to an optical microscope (Ramseyer et al. 1989, cited in Pagel et al. 2000). In cold cathode optical microscope CL-system, the electrons are generated by an electric discharge between two electrodes under a low gas pressure. The commercially available CL-stage consists of an electron gun, a vacuum chamber with windows and an X-Y stage movement. The conditions are usually 17 keV (16 ± 2 keV) and 450 mA. The advantages of this system are: low price of the stage that is adaptable to an optical microscope; easy to use due to simple low vacuum system; large field of observation, up to 2 cm; no coating, since the positive ions generated in the gas phase are sufficient to neutralize the charge effects; and easy check of chemical composition of minerals, when coupled to an EDS (Energy Dispersive Spectrometer). However, it has a few inconveniences like low spatial resolution, instability due to variation of gas-pressure, damage of surface due to electron bombardment and recording system that gives mainly qualitative results. A new cold CL equipment was introduced by the OPEA (Laboratoire de Optique Electronique Appliquee), which allows better observations and spectral analysis, as the stability is very good due to the use of argon as residual gas In the hot cathode CL-system, the electrons are generated by heating a filament (2000-3000oC) and are focused on the sample by magnetic optics. Compared to the cold CL-system, the advantages of hot cathode CL-system are: good spatial resolution due to low size of the electron beam; high magnification; possibility of imaging at a given wavelength; coupling with BSE X-ray mapping; and local high current density. Some disadvantages of this system are: necessity of coating (C, Al, Ni, Pa or Au) and high vacuum, and phosphorescence phenomenon in SEM-CL inducing difficulties in obtaining images in some materials like calcite (Pagel et al. 2000).

Geological Applications of CL:

The CL is an effective method with wide ranging applications in geology. Of these, the petromineralogical applications include: (a) identification of ultra-fine, submicroscopic radioactive minerals present in other minerals as inclusions that are not identifiable by normal transmission optical microscopy, e.g., zircon crystals revealed by reddish luminescence, surrounded by bluish luminescence of host mineral, quartz; (b) delineation of different parts like core and rim of a mineral, which were formed under different environments, e.g., quartz core formed at elevated temperature in an igneous environment shows bright bluish luminescence, whereas its overgrowth formed in a sedimentary environment gives dull red colour; (c) probing minerals formed in diverse rock types, e.g., apatite from (i) basalt yellow luminescence, alkaline plutonic rocks lavender-coloured




luminescence, and (ii) pegmatite many colours, including green; (d) easy identification of individual phases of different mineral groups like carbonates and feldspars, e.g., calcite – orange red and dolomite – darker colour luminescence, and in perthites, K-phase giving dark blue and Na-rich phase pink and red luminescence; (e) distinction between carbonate rocks of magmatic origin (carbonatites, containing high contents of lanthanide-activators, particularly Eu2+, e.g., apatite giving blue luminescence) from those of sedimentary origin (limestone, dolostone, e.g., yellow luminescence due to Mn in apatite); and (f) distinguishing cements precipitated from marine and fresh water, e.g., calcite cement - orange red and Fe-rich dolomite - dark bands. In exploration for radioactive minerals like of U, brilliant luminescence of coronas due to radiation damage, e.g., brilliant orange-red rim of a quartz grain, against brick red or bluish colour of non-damaged interior part of quartz, can be made use of to locate U-bearing sandstone/conglomerate. Other applications of CL in mineral exploration are identification of hydrothermal alteration of crystalline rocks like post-intrusion alteration sequence down to late stage hydrous alterations, fluid migration pathways and fractures recording complex history of mineral growth, and replacement and radiation damage of surrounding minerals like quartz and feldspars (Ramseyer and Mullis, 2000).

Thermoluminescence:

Thermoluminescence (TL) is the phenomenon of emission of light from a crystal previously irradiated, either by exposure to naturally occurring radioactive minerals in the field (Natural TL, NTL) or by exposure to artificial radioactive sources in the laboratory, like 60Co gamma rays (Artificial TL, ATL). When an ionizing radiation like gamma ray enters a crystal, it dislodges electrons from their atomic positions resulting in formation of free electrons and electronic holes or sites that have lost an electron. Although most electrons and holes recombine immediately, a small percentage will, however, be trapped on substitutional and structural defects. Thus in quartz, the most widely used mineral in TL investigations, these holes may be trapped on Al3+ sites and electrons on vacant oxygen sites. These charges, once trapped, can be released by heating the crystal. Once released, the holes and electrons will recombine, which may produce a pulse of light when recombination occurs at a colour centre. Such emission of light is measured with a photomultiplier (PM) tube and recorded as a glow peak. As release of trapped charges occur over a range of temperatures, a number of glow peaks results and these constitute a glow curve. The intensity and shape of the TL glow curve depend upon a few factors like the

number and type of defect centres capable of acting as traps and their occupancy rate, which is largely a function of ionizing radiation. As charge occupancy rate affects the strength of the TL signal, TL has been used as a dosimeter to gain meaningful information.

Important Parameters of TL of Geological

Materials:

TL of geological materials like minerals, rocks and soils involves four important parameters, viz., TL-Sensitivity, -Saturation, -Emission spectral peaks and Life-time of the TL trap. The physical significance and measurement of each of these are as follows (Sankaran et al. 1983): TL sensitivity: This depends upon the centres of available trapping and luminescence. In specific cases of rocks and soils, this can be correlated with a particular mineral and its relative presence in a suite of samples. It is measured by the slope of the linear region of TL-intensity (in coloumbs or photons, or amperes per rad) growth curve with increasing gamma dose over and above the natural dose present in the sample.

TL saturation: This depends on either the saturation of the trapping centres or luminescence centres. In many cases, the relative TL saturation level of a family of samples will indicate relative levels of available trapping centres, when the levels have been weighted for their respective TL sensitivity. TL saturation is measured by the light intensity level at which TL reaches saturation for large irradiations.

TL emission spectral peaks: The spectral peaks in a majority of cases are indicative of impurity-ions (transition elements and REE) that are responsible for emission. They are measured by scanning the emitted TL light, using a grating, prism or band pass filters.

Life-time of the trap: This directly gives the upper limit for the TL dating method for a particular sample. When the TL sensitivity and the self-dose rate of the sample are known, the dynamic equilibrium reached, if any, could be calculated. In samples with large trap life-times but of younger ages, any thermal exposure of the sample in antiquity can be worked out. The life-time of the trap is measured by the evaluation of the activation energy of trap and frequency factor. The value obtained should be checked independently by recording isothermal decay curves at two or more temperatures less than the plateau temperature.

Instrumental Set-up of TL:

The instrument set-up for TL study includes an arrangement for heating the sample-powder (–100 to +140 mesh, ASTM) in a strip of 15 x 10 x 1 mm central depression of a sample-heater, made up of a non-corrosive material, a thermocouple spot-welded to the heater strip to determine the temperature profile, a

1477 R. DHANA RAJU


temperature programmer for linear heating the sample-strip, a PM-tube and a recorder, with four selective chart speeds and five sensitivity ranges for monitoring the PM-output and the temperature (Dhana Raju et al. 1984). A representative portion of 30 mg of sample is to be hated on the heating strip from room temperature to 400OC, at a uniform rate of 5O s-1, and TL intensity is to be recorded in arbitrary units. Necessary precautions are to be taken to avoid the effects of light, UV radiation and other sources, during sample preparation and thermal read-out. First, background (BG) curves are to be taken, and usually the level of BG is negligible compared to the TL signal. Each sample is to be repeated for three to four times and the average temperature and intensity of glow peak are to be measured.

Geological Applications of TL:

TL of geological materials has found wide applications (Sankaran et al. 1983) during the last five decades in different branches of geology like stratigraphy (Saunders, 1953; Nambi and Mitra, 1978), mineralogy (Zeller, 1954; Kaul et al. 1972; Ramesh Babu and Dhana Raju, 1998), geothermometry (Johnson, 1968), geochronology (Ganguli and Kaul, 1968; Nambi et al. 1978), and ore prospecting (Zeschke, 1963; McDougal, 1966; Nambi et al. 1978). Most TL studies are carried out on TL-sensitive minerals like quartz, calcite, dolomite, fluorite, zircon and diamond, whereas TL study on whole-rock has received less attention (Sankaran et al. 1980; Dhana Raju et al. 1984).

TL of rocks, compared to that of minerals, is more complex and depends upon the combined effect of quantity and sensitivity of TL-sensitive minerals present in a rock. Like TL of minerals, TL data on whole-rock samples, containing TL-sensitive minerals, are useful in mineral exploration as well as occasionally in classification of ore-bearing rocks, as shown in a later section.

TL of Ore Bodies:

In the zones adjacent to ore bodes, the TL signatures are either enhanced or depressed due to the following causes (McDougal, 1966, 1968): Increased TL is due to (a) introduction of some amounts of either radioactive or other activating elements into pre-existing minerals; (b) recrystallization of pre-existing minerals at elevated temperatures; (c) formation of new TL minerals by either metamorphism or metasomatism; and (d) state of physical stress in wall-rocks, related to emplacement of ores.

Decreased TL is due to (a) introduction of suppressing elements into pre-existing minerals; (b) formation of non-TL minerals by metamorphism or metasomatism; and (c) natural discharge of trapped electrons from

normal TL minerals by elevated temperature (low TL due to natural heating of carbonate rocks by igneous intrusive or ore mineralization would be eliminated in about 104-106 years through refilling of traps by natural radioactivity).

Many factors govern the development or alteration of TL patterns in rocks hosting ore bodies, and these are as follows: (a) size of the ore body; (b) difference in the temperature between ore-bearing fluids and wall-rocks; (c) thermal conductivity and permeability of wall-rocks; (d) efficiency of diffusion of trace elements into wall-rocks during ore deposition; (e) composition of ore-fluids; (f) possibility of changes in the background radioactivity; (g) geological age of ore deposition; and (h) reactivity of wall-rocks.

Generally, the amplitude of glow curves shows an increase away from the non-radioactive ore deposits like Pb, Zn and Ag, with a perceptible difference in altered and unaltered rocks. The TL profile in the environs of ‘natural fossil reactor’ at Oklo, Gabon Republic in western Africa is of the following pattern (Durrani et al. 1975): near the core, samples showed absolute radiation damage and no TL, either natural or artificial, whereas the TL sensitivity gradually increased away from the zone; however, with depletion of TL sensitive minerals in host rocks, the TL also correspondingly changed.

TL as a Tool in Exploration for U and Rare Element

Pegmatites:

Some examples of how NTL has been used as a tool in exploration for two types of U-mineralization, viz., Hydrothermal type U-mineralization at Jublatola in the Singhbhum shear zone (SSZ), Jharkhand (Dhana Raju et al. 1984) and Sandstone-type U-deposit at Domiasiat and U-prospects at Gomaghat and Pdengshakap of Meghalya (Dhana Raju et al. 1989), as well as for Rare Element Pegmatites in the Bastar-Malkangiri pegmatite belt in central India (Ramesh Bahu and Dhana Raju, 1998) are described in the following.

Hydrothermal Type U-Mineralisation at Jublatola,

Singhbhum Shear Zone, Jharkhand:

Taking first the example of hydrothermal type U-mineralization at Jublatola (nearby to the Jaduguda U-deposit) in SSZ, many drill core samples from both the U-mineralized horizons and their adjacent barren zones were selected from complete core of a single bore hole of this area. Some of these samples were petromineragraphically studied, prior to TL study, for both assigning proper rock nomenclature and understanding the nature of U-mineralization. Then, powders (-60 to +140#) of these and a few other samples without any prior petromineragraphic study (subsequent petrographic study confirmed that both the




methods point to the same rock nomenclature) were subjected to NTL study by heating them from room temperature to 400oC, at a uniform rate of 5oC/second, and TL intensity (in arbitrary units) was recorded; each sample was repeated four times to get the average temperature and intensity of glow peak (Dhana Raju et al. 1984). The data are examined from the point of relating NTL to (i) rock type and (ii) radioactivity.

(i) Relating NTL (Temperature of Glow peak) to Rock Type: Different rock types yield glow peaks at different temperatures (Table 1). Quartzite and schistose rocks, containing TL-sensitive minerals, like quartz and biotite, have given significant glow peaks, whereas tremolite-actinolite rock has not recorded any significant glow peak, as it consists mostly of these two minerals, the TL sensitivity of which is very low. It is, thus, demonstrated that the NTL study on whole-rock samples can be used for broad rock-classification (metamorphic rocks in the present case) and even within the same group like schistose rocks, different specific types can be identified; this technique appears to be particularly suitable when diverse rock types are to be studied in the field itself, as within a bore hole. Table1.

(ii) Relation between NTL (intensity of glow-peak) with Radioactivity: Theoretically, in the same rock type, the intensity of TL glow peak from the sample near to a radioactive ore body should be higher than the one that is far away. In case of the schistose rocks, samples remote from the radioactive horizon show less TL intensity as compared to those close to it, by an order of 2 to 5. This is again well brought out by the abrupt increase in TL intensity of the sample in the radioactive horizon and the subsequent fall in the intensity for samples further down the bore hole. For evaluating the relative contribution of U, Th, and K to the observed NTL intensities, schistose rock samples with glow peaks around 174oC were compared with their radiometric data (Table 2). Samples, 1 to 4, 12 and 13, with more or less equal amounts of K and Th but different uranium contents, show markedly different TL intensities, indicating that such variation in TL intensities are mostly due to uranium. This study, thus, demonstrates that the TL study on whole-rock samples can be used as an aid in U-exploration, and this appears particularly promising for indicating hidden ore bodies that are not otherwise encountered in the bore hole. Table 2.

Sandstone-type U-mineralisation in the Domiasiat –

Gomaghat – Pdengshakap Area, Meghalaya:

NTL study of whole-rock and its corresponding quartz-predominant bromoform-light mineral fraction of the Upper Cretaceous, Lower Mahadek sandstone from the sandstone-type U-deposit at Domiasiat and U-prospects

at Gomaghat and Pdengshakap in Meghalaya in northeastern India has shown that NTL patterns on whole-rock sandstone and its quartz-rich mineral fraction are very much similar, except for a shift in TL glow peak temperature by about 30oC toward higher side in case of the former, as compared to that of the latter. Furthermore, NTL glow curve of uraniferous (with > 0.01% U3O8) samples is characterized by two glow peaks – one of low temperature (LT) at 210o ± 10oC for whole-rock and at 180o ± 14oC for quartz-rich bromoform-light mineral fraction, and another of high temperature (HT) at 260o ± 10oC and 230o ± 10oC, respectively --, whereas that of U-poor (ppm level) samples is marked by HT peak only (Dhana Raju et al., 1989). These observations, together with rapid and easy way of taking NTL pattern on whole-rock, point to the NTL technique on whole-rock as a potential tool in large scale exploration for sandstone-type U-deposits/-mineralization.

Advantages of NTL Study on Whole-Rock Samples:

The above account documents the advantages of NTL study on whole-rock samples, which is simple, direct, rapid and does not require any laborious and time-consuming separation of TL-sensitive minerals in a rock. It is profitable especially for (i) deciphering the concealed mineralized zones of even low-level radioactivity, as TL being the net effect of long time radiation exposure, and (ii) predicting the extensions of the already known uraniferous zones (Dhana Raju et al. 1989), before costly drilling. Thus, this study may be attempted to predict the blind-extensions on either side of the 6 km-long Tummalapalle – Giddankipalle U-deposit (~ 65,000 t, with average grade of 0.045% U3O8) in the Kadapa district of Andhra Pradesh, which is presently under exploitation by the Uranium Corporation of India Ltd.

Rare Metal Pegmatites in the Bastar-Malkangiri

Pegmatite Belt, Central India:

NTL on pegmatitic feldspars (perthitic microcline and albite) and quartz from the Bastar-Malkangiri Pegmatite belt (BMPB) of central India shows that NTL of feldspars is marked by low glow peak temperature (187o - 242oC; rarely 279oC) as compared to that of quartz (220o – 313oC). With increasing internal evolution of pegmatitic melt that resulted in the progressive formation barren (type I) to rare element mineralized pegmatites (types II to V with columbite-tantalite, beryl, lepidolite, amblygonite, and cassiterite mineralization; Ramesh Babu, 1993), there is an increase in the intensity of NTL glow peak of feldspars and to a lesser extent of quartz. Thus, perthite with glow peak of 3.9 to 9.0 mV intensity at 187o –229oC, albite with 0.3 to 1.9 mV at 210o-279oC, and quartz with up to 27 mV at 220o-313oC indicate mineralized nature (Nb-Ta, Sn, Be

1479 R. DHANA RAJU


and Li) of pegmatites emplaced in amphibolites. NTL of pegmatitc feldspars and quartz can, thus, be used as an aid to discriminate the mineralized from the barren pegmatites in other areas with similar geological set up (Ramesh Babu and Dhana Raju, 1998).

Conclusions:

1. Luminescence techniques like fluoresceince - phosphorescence, cathodo-luminescence (CL) and thermoluminescence (TL) constitute low-cost, sensitive and direct geophysical tools, having wide applications in geological sciences, in general, and mineral exploration, in particular. 2. Fluorescence – phosphorescence under ultraviolet light of both short and long wavelength are used for identification of fluorescent minerals, viz., radioactive (mainly uranyl-phosphates and –arsenates) and non-radioactive (fluorite, scheelite, etc.,) as well as their exploration. 3. CL, in combination with optical microscopy, XRD, SEM, EMP and Image Analysis, has many applications like phase-characterization and –differentiation, their quantification and in studies of palaeo-permeability, -porosity and –environment, and in U-exploration like identifying sandstone-type and quartz-pebble conglomerate type U-mineralization. 4. TL has diverse applications in stratigraphy, geothermometry, geochronology and ore prospecting. Natural TL (NTL) study of TL-sensitive minerals like quartz and whole-rock samples containing such minerals is useful in mineral exploration, as demonstrated by three case-studies of hydrothermal-type U-mineralization in the Jublatola area in Jharkhand, sandston-type U-mineralization in the Domiasiat-Gomaghat-Pedengshakap area in Meghalaya and rare element pegmatites in the Bastar-Malkangiri pegmatite belt in central India. NTL on whole-rock samples is a potential tool in exploration for radioactive and rare element minerals, especially for (a) deciphering concealed mineralization and (b) predicting blind-extensions of known mineralization, before undertaking costly drilling.

Acknowledgements:

Sincere thanks are due to my former colleagues in the Atomic Minerals Directorate for Exploration and Research (AMD), Dept. of Atomic Energy for many discussions and support during the course of the work.

References:

[1] Boyle, R.W. (1982). Geochemical prospecting for Thorium and Uranium. Elsevier Sci. Publ. Co., Amsterdam, 498 p.

[2] Dhana Raju, R., Bhargava, R.C., Selvam, A.P. and Virnave, S.N. (1989). Natural thermoluminescence of whole-rock as a potential tool in exploration

for sandstone-type uranium deposits: Application to Lower Mahadek sandstones of Meghalaya, India. Proc. Uranium Tech., v. 1, pp. 74-89, BARC, Mumbai (issued in Oct., 1991).

[3] Dhana Rraju, R., Venkataraman, B. and Ananthara-man, K.B.(1984). Natural thermo-luminescence of whole-rock samples as an aid in uranium exploration: A case study from Singhbhum shear zone, Bihar, India. Uranium, v. 1, pp. 279-287.

[4] Duranni, S.A., Kazal, K.A.R., Malik, S.R., Fremlin, T.H. and Hendry, G.L. (1975). Thermoluminescence and fission track studies of the Oklo fossil reactor materials. Proc. Symp. on ‘Oklo Phenomenon’, Libreville, June 23-27, 1975, IAEA, Vienna, pp. 207-222.

[5] Frondel, C. (1958). Systematic mineralogy of uranium and thorium. U.S. Geol. Survey Bulle. 1064, 400 p.

[6] Ganguli, D.K. and Kaul, I.K. (1968).The age of radioactive mineralisation of placer deposits of Kerala, India. Econ. Geol., p. 63, pp. 838-839.

[7] Gleason, S. (1960). Ultraviolet Guide to Minerals. Van Nostrand, Princeton, New Jersey, 244 p.

[8] Gotze, J. (2000). Cathodoluminescence in applied geosciences. Chapter 18, In: Pagel et al. (Eds.), Cathodoluminescence in Geosciences. Springer Verlag, Berlin, pp. 457-475.

[9] Johnson, N.M. (1968). Determination of magma temperatures from natural thermolumi-nescence. In: D.J. McDougal (Ed.), Thermoluminescence of geological materials. Academic Press, London, pp. 545-546.

[10] Kaul, I.K., Ganguli, D.H. and Hess, B.F.H. (1972). Influencing parameters in thermo -luminescence. In: D.J. McDougal (Ed.), Thermoluminescence of quartz. Mod. Geol., v. 3, pp. 201-207.

[11] Mcdougal, D.J. (1966). A study of the distribution of thermoluminescence around ore deposits. Econ. Geol., v. 61, pp. 1090-1103.

[12] Mcdougal, D.J. (1968). Natural thermoluminescence of igneous rocks and associated ore deposits. In: D.J. McDougal (Ed.), Thermoluminescence of geological materials. Academic Press, London, pp. 527-544.

[13] Nambi, K.S.V., Bapat, V.N. and David, M. (1978). Geochronology and prospecting of radioactive ores by their thermoluminescence. Indian Jour. Earth Sci., v. 5, pp. 154-160.

[14] Nambi, K.S.V. and Mitra, S. (1978). Thermoluminescence investigations of old carbonate sedimentary rocks. Neub. Jahrb. Miner. Abh., v. 133, pp. 210-226.

[15] Pagel, M., Barbin, V., Blanc, P. and Ohnenstetter, D. (2000). Introduction, Chapter 1, In: Pagel et al.




(Eds.), Cathodoluminescence in Geosciences. Springer Verlag, Berlin, pp. 1- 21.

[16] Ramesh Babu, P.V. (1993). Tin and rare metal pegmatites of the Bastar – Koraput pegmatite belt, Madhya Pradesh and Orissa, India: Characterisation and Classification. Jour. Geol. Soc. India, v. 42 (2), pp. 180-190.

[17] Ramesh Babu, P.V. and Dhana Raju, R. (1998). Natural thermoluminescence of pegmatitic feldspars and quartz as an aid in exploration for rare element pegmatites: A case study of Bastar-Koraput Pegmatite belt, central India. Indian Jour. Geochem., v. 13, pp. 51-60.

[18] Ramseyer, K. and Mullis, J. (2000). Geologic appli-cation of cathodoluminescence of silicates. Chapter 7, In: Pagel et al. (Eds.), Cathodoluminescence in Geosciences. Springer Verlag, Berlin, pp. 177-191

[19] Sankaran, A.V., Nambi, K.S.V. and Sunta, C.M. (1983). Progress of thermoluminescence research

on geological materials. Proc. Indian Nat. Sci. Acad., v. 49 A, pp. 18-112.

[20] Sankaran, A.V., Sunta, C.M., Nambi, K.S.V. and Bapat, V.M. (1980).Thermoluminescence studies in Geology. BARC-1060, Bhabha Atomic Research Centre (BARC), Bombay, 96 p.

[21] Saunders, D.F. (1953). Thermoluminescence and su-rface correlation of limestone. Bulle. Amer. Assoc. Pet. Geol., v. 37, pp. 114-124.

[22] Walenta, K. (1959). Uranprospektion mit der UV-Lampe. Zeit, Erzbergbau u. Metallhuttenw, v. 12, pp. 51-55.

[23] Zeller, E.J. (1954). Thermoluminescence of carbon-ate sediments. In: H. Faul (Ed.), Nuclear Geology. Wiley, New York, pp. 180-188.

[24] Zeschke, G. (1963). Thermal glow tests as a guide to ore deposit. Econ. Geol.., v. 58, pp. 800-803.

Table 1: NTL Glow Peak Temperatures of different Rock Types from a Borehole in the Jublatola Area, Singhbhum

Shear Zone, Jharkhand.

Rock type Glow peak (oC)

1. Quartzite 232 ± 2 2. Tremolite-actinolite rock No peak 3. Schistose rocks 87 and 170-212

a. Chlorite-hornblende schist 176 ± 2 b. Hornblende schist 190 ± 2 c. Biotite-quartz schist 87 ± 1 and 175 ± 1 d. Hornblende-biotite-quartz schist 202 ± 7 e. Tourmaline-biotite-chlorite schist 170 f. Tourmaline-biotite-quartz schist 212

Table 2: Comparison of Gamma-Ray Spectrometric Data with NTL Glow-Peak Intensity of the Schistose Rocks from

Jublatola, Singhbhum Shear Zone, Jharkhand

Sample no. eU3O8

(ppm) Ra.eq. (ppm)

ThO2

(ppm) K

(%) Glow peak (oC ± 2oC)

Peak intensity (arbitrary units)

1. 5.6 1.6 1.0 2.0 176 31 2. 3.5 1.0 1.0 1.1 175 30 3. 1.7 <1.0 <1.0 0.5 177 26 4. 7.0 1.7 1.3 2.6 178 18

12. 128 125 5.0 2.1 170 91 13. 10 4.7 4.3 2.5 174 140

www.cafetinnova.org

Indexed in Scopus Compendex and Geobase Elsevier, Chemical


ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1481-1491


Ensemble Empirical Mode Decomposition of the Lightning Return

Stroke

XUQUAN CHEN and WENGUANG ZHAO

School of civil engineering & mechanics, Huazhong University of Science and Technology, Wuhan, P. R. China


Abstract: In this investigation, the electric field waveforms of 16 negative lightning return strokes were analyzed.

Ensemble Empirical Mode Decomposition (EEMD) and Empirical Mode Decomposition (EMD) of the return

strokes were conducted. Contrary to the wavelet analysis, both the EEMD and EMD provide a better representation

of the multi scale characteristics of lightning discharge process for the advantage of better representation of the non-

stationary data. For the close return stroke, the ramp due to the electrostatic components can be represented by the

residual item, which was often treated as the data trend. We inferred that the mode mixing phenomenon may easily

occur on the EMD-based Hilbert Huang spectrum. After processing with EEMD, the EEMD-based Hilbert Huang

spectrum can significantly improve on the mode mixing for 11 return strokes, which demonstrated that the

intermittence of these return strokes was significant. However, the improvement for the other 5 return strokes was

not evident, which indicated that the intermittence of the 5 return strokes was not significant. The multi scale

characteristics of the lightning electric field waveforms can be clearly represented in the EEMD-based Hilbert

Huang Spectrum showing less mode mixing problem. The corresponding marginal spectrum of these electric field

waveforms showed that the energy of the close return strokes were mainly concentrated in the low-frequency ranges.

Taken together, this study demonstrates the utility of EEMD in analyzing the pattern of lightning return strokes and

provides new insights into the understanding of the lightning physical discharge structure.

Keywords: Eemd, Electric Field Waveform, Hilbert Huang Spectrum, Lightning Return Stroke, Wavelet Analysis

1 Introduction:

Cloud-to-ground lightning strikes often produce strong

electromagnetic pulse which has the potential to damage

the power systems, information and telecommunication

facilities. The destructive power of these strikes may

also cause lethal harm as well as cause significant

property damage. Therefore, it is necessary to develop

methods to locate and track lightning and design strong

protection systems against lightning [1]. In addition to

locating lightning strikes, studying the frequency

spectrum of the electric fields is also important not only

for the physical investigation of lightning phenomenon

but also for the engineering assessments of designing

lightning protection schemes [2].

The high-frequency content of the electric field

waveforms amplitude is attenuated because of

propagation effects, and the parameter characterizing

the lightning return stroke waveform such as the peak

value, rise time, and zero crossing time, will also vary

with the propagation distance [3] [4] [5]. The frequency

spectrum of lightning electric field changes produced by

the return strokes have been extensively studied in the

past [6] [7] [8] [9].Generally, two techniques have been

used to study the lightning electric fields spectrum; one,

is the narrowband recording device, which is often used

to measure the average spectrum in particular

frequency; and the other is wideband recording device,

which is able to record the shape of the waveform

corresponding to a wide frequency range [10]. In

comparison to the narrowband recording device, the

wideband recording device has the advantage of making

it possible to probe electric and magnetic field

waveforms in the time domain with minimal distortion,

which often facilitates the understanding of a particular

lightning process when the spectrum is obtained using

the Fast Fourier Transform. The spectrum of the

different lightning events such as the first and

subsequent return strokes, stepped and darted stepped

leaders, were analyzed by Willett et al who reported that

both of the spectra were in the range between 0.2 and 20

MHz [11].

Although the Fourier Transform has been successfully

applied to many different scientific fields, it still has

some limitations when applied to non-stationary data

analysis. A time-frequency analysis method named

wavelet analysis has become increasingly popular in

recent years, and is suitable for processing non-

stationary signals [12]. For the advantage of observing

the data in both time and frequency domain, it has been

1482 XUQUAN CHEN and WENGUANG ZHAO

International Journal of Earth Sciences and Engineering

ISSN 0974-5904, Vol. 05, No. 06, December 2012, pp. 1481-1491

widely applied to numerous scientific studies, such as

speech enhancement, image processing, fault diagnosis,

structure health monitoring and so on. Recently, the

wavelet analysis was applied to analyze the electric field

waveforms data of lightning return stroke, and the

results showed that the characteristics of the frequency

distributions can be clearly revealed in the wavelet

power spectrum [13]. The different lightning events

were also analyzed using the wavelet transform, and

they often correspond to different frequency distribution

ranges [14]. The wavelet multiresolution based

multifractal analysis of the return stroke was performed,

which often facilitates the understanding of lightning

scaling property [15]. Moreover, because of the capacity

of noise reduction, the wavelet transform can help

locate the lightning path accurately when used in

conjunction with a broadband interferometer [16].

Despite its advantages, wavelet analysis retains some

shortcomings that the Fourier analysis always possesses.

Similar to the Fourier analysis, the wavelet analysis is

essentially an adjustable window Fourier Transform.

Before analyzing the data, the wavelet basis is to be

selected, thus the selection of different basis may give

rise to different decomposition results for one set data,

leading to difficulties in data interpretation. Moreover,

the pre-fixed wavelet basis often does not make the

decompositions physical sense. For the lightning multi-

scale irregular discharge phenomenon, the

decompositions produced by wavelet analysis does not

represent the physical multi-scale discharge

characteristics. In other words, the pre-fixed wavelet

basis may lead to decompositions which are

meaningless physically and thus the analysis cannot

process the data in an adaptive manner.

Another time-frequency analysis approach named

Hilbert-Huang Transform (HHT) based on Empirical

Mode Decomposition (EMD) was invented by Huang

and his colleagues [17]. The core of the method is

EMD, which works through decomposing the data into

a set of complete and almost orthogonal components

named intrinsic mode functions (IMFs) that often can

make sense for the IMFs. Subsequently applying the

Hilbert Transform to the IMFs, one can obtain the

Hilbert Huang spectrum. The variable amplitude and the

instantaneous frequency often enable the IMFs to give a

better representation of the non-stationary data.

Contrary to the previous time-frequency analysis tools,

this method has an adaptive basis and offers excellent

resolution both in time domain and frequency domains.

Due to the advantages in data analysis, this technique

has been quickly applied to many different scientific

fields, such as biomedical signals processing,

geophysics, image processing, fault diagnosis, structure

testing and so on. Even so, there were still some

shortcomings unresolved for the EMD, for example, the

mode mixing phenomenon has always been an obstacle

when applying this technique for the data processing. In

resolution of this pitfall, Huang and his colleagues

invented an improved approach named Ensemble

Empirical Mode Decomposition (EEMD) [18], which

can essentially eliminate the mode mixing phenomenon.

Thus, the IMFs without mode mixing problem often can

provide a better representation of the physical meaning

of processed data.

In this study, the EEMD based Hilbert-Huang spectrum

and the EMD based Hilbert-Huang spectrum of the 16

negative return strokes were obtained, respectively. The

natural multi-scale characteristics of the electric field

waveforms were revealed in the corresponding Hilbert-

Huang spectrum, which provides new insights into the

understanding of the lightning physical discharge

structure. This paper begins with a brief introduction of

EEMD method (Section 2). This is followed by an

explanation of the data acquisition system (Section 3)

along with the results and discussion (Section 4).

Section 5 is the conclusions and prospects.

2 EEMD Method:

Before introducing the EEMD technique, a brief

introduction of EMD and HHT approaches is provided.

2.1 EMD and HHT:

As it is known that the Hilbert Transform of signal x(t)

has the definition of the convolution of x(t) with 1/t, as

follows:

+

-

( )( )

p xy t d

t

ττ

π τ

∞

∞=

−∫ (1)

Where p indicates the Cauchy principle value. Then

x(t)’s analytical signal denoted as z(t) is as follows: ( )( ) ( ) ( ) ( )

i tz t x t iy t a t e

θ= + = (2)

Where, 2 2 1/2( ) [ ( ) ( )]a t x t y t= +

(3)

( )( ) arctan ( ) / ( )t y t x tθ =. (4)

a(t) is the instantaneous amplitude of x(t), andθ(t) is the

instantaneous phase of x(t). Thus, one can see the

Hilbert Transform has the function of emphasizing the

local properties of x(t). Then the instantaneous

frequency having the physical meaning can be obtained

by the time derivative of instantaneous phaseθ(t), as

follows: ( )

( )( )

d t

d tt

θω = (5)

However, the Hilbert Transform requires that the

instantaneous frequency of the x(t) have physical

meaning when the signal x(t) is monocomponent.

Consequently, in almost all of the practical applications,

the signal is hardly treated as monocomponent which

1483 Ensemble Empirical Mode Decomposition of the Lightning Return Stroke

International Journal of Earth Sciences and Engineerin


makes the concept of instantaneous frequency not

applied extensively. To address this issue, Huang and

his colleagues invented a method named EMD which

decomposes a signal into a set of almost complete and

orthogonal monocomponents named as IMFs to which

the instantaneous frequency can be applied. This form

of Hilbert Transform based on EMD is named as the

Hilbert-Huang Transform.

An IMF is a function that satisfies the following two

conditions: (1) in the whole data set, the number of

extreme value point and the number of zero crossing

must either equal or differ at most by one; and (2) at

every point, the mean value of the envelope defined by

the local maxima and the envelope defined by the

minima is zero. An IMF represents the oscillation mode

embedded in the data. With this definition, the IMF in

each cycle, defined by the zero crossings, involves only

one mode of oscillation and no complex riding waves

superimposed on one cycle. For this reason, IMF is not

restricted to a narrow band signal, and it can be both

amplitude and frequency modulated. In fact, it can be

non stationary, which is very significant in analyzing

the non stationary data, such as the lightning return

stroke data. The IMF does not always guarantee a

perfect instantaneous frequency under all conditions,

however, even under the worst conditions, the

instantaneous frequency defined as Eq.(5) is still

consistent with the physics of the system. Actually,

most of the data are not IMFs but instead may involve

more than one oscillatory mode. The EMD method can

then decompose a signal into a set of IMFs (cj, j=1,...,n)

through a sifting process, which is as follows: (1) For

the data rj-1, Extract all the extreme value point of the

data and connect the local maxima and the local minima

by a cubic spline as the upper envelope and lower

envelop, respectively; (2) Obtain the first component h

by taking the difference between the data and mean of

the upper and lower envelopes;(3) let h the data and

repeat step1 and step 2 as many times as is required

until h satisfies the IMF definition. Thus, the final h is

denoted as cj. When the sifting process stops, all the

IMFs (cj, j=1, ... , n) and the residue rn can be obtained.

Then a set of IMFs and a residue rn can be obtained

from above procedure. When summing up all the IMFs

and residue, we can have

1

( )n

i n

i

x t c r=

= +∑ (6)

Applying the Hilbert Transform to each of the IMFs,

and computing the instantaneous frequency and

instantaneous amplitude based on the Eqs. (3) and (4),

we can obtain

1

( ) exp( ( ) ( ))n

j j

j

x t a i t d tω=

=∑ ∫ (7)

Equation. (7) enables us to represent the amplitude and

instantaneous frequency as functions of time in a three-

dimensional plot, in which we can observe the data in

the frequency-time space. The frequency-time

distribution of the amplitude is designated as the

Hilbert-Huang spectrum, H(ω,t). With the Hilbert-

Huang spectrum defined, one can define the marginal

spectrum, h(ω), as

0( ) ( , )

T

h H t dtω ω= ∫ (8)

The marginal spectrum can offer a measure of total

amplitude (or energy) contribution from each frequency

value, and represents the cumulated amplitude over the

entire data span in a probabilistic sense. Contrary to the

Fourier spectrum, the marginal spectrum has meaning

physically when the data are non-stationary. More

details about the EMD and HHT can be found in Huang

et al [17].

2.2 EEMD:

Although the EMD method has been used widely in

many different fields, the mode mixing problem always

troubles the researchers. Mode mixing is defined as a

single IMF including oscillations of dramatically

disparate scales, often caused by intermittency of the

driving mechanisms. Inspired by the noise-added

analysis initiated by Flandrin and Gledhill [19], a new

method named EEMD was proposed by Huang and his

colleagues to alleviate the mode mixing problem. This

method defines the true IMFs as the mean of an

ensemble of trials, and each trial consists of the

decomposition results of the signal added with a

normally distributed white noise with a constant

standard deviation (STD). The principle of the EEMD is

simple: white noise added to the system would populate

the whole time-frequency space uniformly with the

constituting components of different scales. When a

signal is added to the uniformly distributed white noise

background, the components in different scales of the

signal are automatically projected onto proper scales of

reference established by the white noise in the

background. Since each trial produces the noisy results,

each of the noise-added decompositions consists of the

signal and the added noise. As a result, the noise can be

canceled out in the ensemble mean of enough trials, and

the ensemble mean is treated as the true answer. Then

the only persistent part is the signal as more and more

trials are added in the ensemble.

The EEMD method is developed as follows:

(a) Add a white noise series with a constant standard

deviation to the targeted data;

(b) Decompose the data with the added white noise into

IMFs;




(c) Repeat step 1 and step 2 again and again, but with

different white noise series each time; and

(d) Treat the ensemble means of corresponding IMFs of

the decompositions as the final results.

The relationship among the number of ensemble

number, the amplitude of the added noise and the final

standard deviation of error is as follows:

nN

εε =

(9)

Or

ln ln 02

n Nε

ε + = (10)

Where N is the ensemble number, ε is the amplitude of

the added noise, and εn is the final standard deviation of

error, which is defined as the difference between the

input signal and the corresponding IMFs. Generally, an

ensemble number of a few hundred will make a good

result, and the remaining noise would cause less than

one percent of error if the added noise has an amplitude

that is a fraction of the standard deviation of the input

signal. Huang and his colleagues recommended that in

most cases, that the amplitude of the added noise be

about 0.2 standard deviation of that of the input data,

whereas the data is dominated by high-frequency

signals, the noise amplitude needs to be smaller, and if

the data is dominated by low-frequency signals, the

noise needs to be increased. More details about EEMD

can be found in [18].

3. Data:

In the summer of 2008, the observations on natural

lightning discharges were conducted at Guangzhou

Field Experiment Site, China (113.6E°, 23.6N°). In the

experiment, a slow antenna, fast antenna and the

radiation antenna, respectively, were employed to

record the different lightning discharge characteristics

simultaneously.

The radiation antenna consists of one broadband holder

(Schwarzbeck HFBA9122) and a pair of biconical

elements (Schwarzbeck BBVU9135). The output of the

radiation antenna was filtered with a low-pass filter

(40MHz) and digitized by a National Instruments

PCI5122 (14 bit 512 MB) data acquisition card with a

sampling rate of 100 MHz. The zero-to-peak risetimes

for both the slow and fast antenna was less than 30 ns.

The decay time constant of the slow antenna and the fast

antenna were 6 s and 2 ms with a frequency bandwidth

of 10 Hz~3.5MHz and 1 kHz~2MHz, respectively.

Then the outputs of the antennas were digitalized by a

14-bit A/D converter and recorded by a computer at a

sampling rate of 15 MHz. The recording length was

800ms, and the pre-triggered time length was 160ms,

which can almost cover a lightning event.

The distances of the cloud-to-ground return strokes

from the station were obtained by measuring the time

difference between the arrivals of the lightning visual

observation and the hearing of thunder. Notice that the

measured distances were the approximate values. Two

thunderstorms occurred respectively on July 30 and

August 4 of 2008 within the observation station of about

5~20km, and the corresponding lightning data were

acquired. In this paper, 16 electric field waveforms of

negative cloud-to-ground return stroke recorded by the

fast antennas are chosen for analysis. More details about

the data acquisition system can be found in Lan et al

[20].

4. Results and Discussions:

Before applying the EEMD method to the return

strokes, the wavelet analysis and the EMD of the return

stroke would be first performed to make a comparison

among these methods. This next section is arranged as

follows: Section 4.1 gives a brief introduction of the

wavelet analysis of lightning return strokes. The EMD

and HHT of the return strokes were performed in Sect.

4.2. Section 4.3 is the EEMD-based HHT of the return

strokes, and the marginal spectrum of the return strokes

was also performed in this section. Here, the analyzed

data length is 7500 points corresponding to about 500µs.

4.1 The Wavelet Analysis of Lightning Return

Stroke Field:

Wavelet transform has become a well accepted time-

frequency analysis tool and has been extensively

employed in many scientific studies. Since it is very

well known, the introduction about it will be omitted.

Figure 4.1 depicts a negative return stoke waveform

with a range of about 15km. Different from what were

discussed by Miranda [13], for the close negative stroke,

the return stroke waveform can be divided as two

portions: one is called return stroke stage denoted as RS

stage, and the other is called ramp, which is due to the

electrostatic component domination when the return

stroke was close to the observer [3]. The continuous

wavelet spectrum of this negative return stroke is shown

in Fig.4.2, which was represented by color, whose

amplitudes are represented by the color bar at the right

side of the plot. The DOG wavelet function was chosen

for analysis because it is more appropriate to analyze

the transient signals, such as the return stroke [13].

From the wavelet spectrum, it could be observed that

the characteristics of the frequency distribution varying

with time can be clearly revealed. Moreover, the

lightning discharge process in different scales from the

wavelet spectrum of the corresponding electric field

waveforms could be observed, especially in the RS

stage. Contrary to the Fourier spectral analysis, the

wavelet spectral analysis has the potential to observe the

lightning electric field waveforms both in time and




frequency domain. Thus, from the time-frequency

spectrum, the characteristics of lightning multi-scale

discharge can be represented.

4.2 The EMD and HHT of Return Stroke:

Since the EMD has the adaptive basis when compared

to the wavelet analysis, the decompositions by the EMD

often have physical meaning. Figure 4.3 depicts the

IMFs produced by the EMD for that return stroke

depicted in Fig.4.1. Obviously, the first item c1 in

Fig.4.3 is the high-frequency item, which can be treated

as the system noise. The last item is the low-frequency

residual, which is often treated as the trend of the data.

As discussed above, for the close return stroke, the ramp

is due to the electrostatic, which is predominant in the

electric fields and mainly composed of the low-

frequency components. In comparison with the IMFs,

the residual has larger amplitude with an increasing

trend. Based on this, we can infer that the residual is

mainly the result of the electrostatic interactions. The

other IMFs can be treated as the multi-scale discharge

signals which often have physical meaning.

Applying the Hilbert Transform to each of the IMFs,

one can obtain the lightning return stoke EMD-based

Hilbert-Huang spectrum depicted in Fig.4.4, which was

represented by color, whose amplitudes are represented

by the color bar at the right side of the plot. Apparently,

the natural multi-scale characteristics of the return

stroke have already been revealed. From the spectrum

one can observe the time-frequency distribution

characteristics of the return stroke. The multi-scale

characteristic in RS stage is significant, which is due to

the multi-scale dramatic discharge process. However, in

the ramp stage, the multi-scale characteristic in the

Hilbert-Huang spectrum is not evident, which is due to

slow-change and low-frequency electrostatic

components. By the comparison of the return stroke

between the wavelet spectrum and the Hilbert-Huang

spectrum corresponding to Fig.4.2 and Fig.4.4,

respectively, one can see the spectral distribution in the

Hilbert-Huang spectrum is different from that of the

wavelet spectrum. The spectral distribution in the

wavelet spectrum is continuous while it is distributed in

discrete scales in the Hilbert-Huang spectrum, which

exactly represent the natural multi-scale characteristics

of the return stroke. From the Hilbert-Huang spectrum

in Fig.4.4, one can see the spectrum spread distribution

is mainly in the range less than about 400 kHz with the

adaptive scale levels, which shows more detailed

frequency distribution characteristics than that of the

wavelet spectrum.

As analyzed above, the differences of the frequency

spread distribution between wavelet spectrum and

EMD-based Hilbert-Huang spectrum are determined by

the different decomposition basis. For the wavelet

analysis, the pre-fixed basis cannot change adaptively in

the analyzing process, which often leads to decomposed

components which are physically meaningless. In the

contrary, using the adaptive basis of the EMD, one can

obtain the IMFs often having physical meaning. Aside

from that, the number of the decomposed scale in the

wavelet analysis is man-made while it is determined by

the data in the EMD. In comparison with the Hilbert-

Huang spectrum, because of the effect of the

Heisenberg uncertainty principle, the time and

frequency resolution of wavelet analysis cannot be

achieved at a best value simultaneously, and they were

of mutual influence. In the contrary, the EMD-based

Hilbert-Huang spectrum has better resolution in both

time and frequency domain, and they were independent

of each other.

From the above, so we can infer that the EMD-based

Hilbert-Huang spectrum can give a better description of

the natural multi-scale characteristics of the return

stroke in contrary with the wavelet spectrum. The

EMD-based HHT is more appropriate to analyze the

lightning return stroke’s multi-scale characteristics

rather the wavelet transform.

4.3 The EEMD of the Return Stroke:

Here give a detail description of the return stroke

waveforms in Fig.4.1. Notice that there were several

larger subsidiary peaks following the first peak, which

is due to the effects of branches [21]. For the return

stroke with a distance of about 15 km, in the RS stage,

the radiation is dominant, then the electrostatic and

induction will become larger at longer times,

corresponding to the ramp stage. Thus, the measured

electric field waveforms were the superposition of the

electric fields emitted from the tortuous and branching

channel [22]. Therefore, electric fields’ multi-scale

characteristics are determined by the tortuous and

branching channels’ multi-scale characteristics. The

large scale corresponds to the low-frequency signal such

as the subsidiary peaks’ signal due to the branches.

Then the small scale corresponds to the high-frequency

signal, for example, the radiation signal due to channel’s

tortuosity [23].

As discussed in section 2.2, one of the drawbacks of

EMD is the mode mixing problem, which often yields a

nonsensical solution to IMFs. As illustrated by Wu and

Huang [18], the mode mixing will occur when the data

was a mixture of intermittent high-frequency

oscillations riding on a continuous low-frequency

signal. For the electric field waveforms of the lightning

return stroke, the multi-scale discharge characteristics

determine that they were intermittent signal. The

reasons are as follows: As discussed above, electric

fields’ multi-scale characteristics were due to the

tortuous and branching channels’ multi-scale




characteristics. Once the intermittent high-frequency

signals corresponding to small scale ride on the low-

frequency signals corresponding to large scale, it may

cause the intermittence of the data. Thus, one can see

that the intermittence may easily occur on the lightning

return stroke signals, which is based on the multi-scale

characteristics of the discharge process. Thus, the IMFs

produced by EMD can cease to have physical meaning

due to mode mixing. In this section, the EEMD of the

return stroke will be performed, and then the EEMD-

based Hilbert-Huang spectrum and the corresponding

marginal spectrum based on EEMD will also be

obtained.

Figure 4.5 depicts the IMFs decomposed with the

EEMD of the return stroke depicted in Fig 4.1. Here, we

adjusted the noise standard deviation to 0.01 and the

ensemble number to 200.One can see c1 is still the noise

item, and the trend is mainly due to the electrostatic

components. Applying the Hilbert Transform to each of

the IMFs, one can generate an EEMD-based Hilbert-

Huang spectrum as shown in Fig.4.6. From that one can

see the multi-scale characteristics are very significant

especially in the RS stage as the same as the EMD-

based Hilbert Huang spectrum in Fig.4.4.

When observing the EMD-based Hilbert-Huang

spectrum depicted in Fig.4.4, one can see the alias at the

transition points from one scale to another is clearly

visible at about 50µs and 150µs, which is the result of

mode mixing. Then observing the EEMD-based Hilbert-

Huang spectrum as depicted in Fig.4.6 at about 50µs

and 150µs, one can see the transition is not evident at

about 50µs and 150µs. Moreover, notice that the scale

distribution in Fig.4.6 becomes smoother than that in

Fig.4.4. Obviously, the EEMD-based Hilbert-Huang

spectrum allows an improvement on the mode mixing

problem for the lightning return stroke. Therefore, the

multi-scale discharge characteristics of the lightning

return stroke can be precisely represented in the EEMD-

based Hilbert-Huang spectrum.

Attention should be paid to how the selection of the

STD may influence the results. The authors have tried

the STD level with a step size of 0.01 from 0.01 to 0.1

and 0.05 from 0.1 to 0.4. For this return stroke, setting

STD=0.01 gives a better result. However, the results

may be even worse if the selection of STD is not

appropriate. Figure 4.7 depicts the corresponding

EEMD-based Hilbert-Huang spectrum of this return

stroke with STD=0.05 and N=200. From that one can

see the mode mixing phenomenon becomes severe in

general. We can infer the reason may be due to the

result of the lightning return stroke’s serious

intermittency characteristics, which is determined by its

multi-scale property. In other words, the selection of

some determined STD can only eliminate the mode

mixing corresponding to that level, which can not be

eliminated by selecting the other STD.

Despite this issue, the mode mixing problem still can be

ameliorated by choosing the appropriate STD.

Observation of the EEMD-based Hilbert-Huang

spectrum and the EMD-based Hilbert-Huang spectrum

of 16 return stokes was done, respectively. Still, try the

STD level with a step size 0.01 from 0.01 to 0.1 and

0.05 from 0.1 to 0.4, and set N=200. In order to make a

best improvement for mode mixing, select the

appropriate STD for each of the return stoke for EEMD.

The results showed that 11 out of 16 return strokes have

a significant improvement on the mode mixing

phenomenon, and the 11 STD values correspond to

different levels, respectively. However, the

improvement was not significant for the other 5 return

strokes no matter what STD was selected. Then one can

infer the 11 return strokes have significant intermittent

characteristics, and the intermittent characteristics were

not evident for the other 5 return strokes. Therefore, the

multi-scale characteristics determine that the return

stroke signals are intermittent, which proves our

inference at the beginning of this section. Together, it is

concluded that the EEMD is more suitable for the

lightning return stroke spectra analysis than both the

wavelet analysis and EMD in investigating the natural

multi-scale characteristics of lightning discharge

process.

Then the marginal spectrum based on EEMD of this

return stroke can be obtained from Eq.(8) and is

illustrated in Fig.4.8, from which one can see the

frequency distributions were mainly concentrated in the

low-frequency range. As it is known that the marginal

spectrum has meaning physically when the data are non-

stationary, therefore, it has the better representation of

the characteristics in frequency domain. Then the

EEMD based marginal spectrum of 11 negative return

strokes and the EMD based marginal spectrum of 5

negative return strokes were also obtained, respectively.

The results show that the distributions of the peak value

of marginal spectrum were in the range from about 0.4

kHz to 9 kHz, which demonstrates that the energy of

close return stroke was mainly concentrated in the low

frequency range.

From the above, we can conclude that the EEMD-based

Hilbert-Huang Transform is more suitable for

processing the lightning return stroke signals than both

the EMD-based Hilbert-Huang Transform and wavelet

analysis. The multi-scale characteristics showing less

mode mixing problem of the return stroke in the

EEMD-based Hilbert-Huang spectrum can be clearly

revealed, which contributes to understanding the

physical discharge structure of the tortuous and

branching channels.




5. Conclusions:

This study focused on the utilization of EEMD and

EMD technique to 16 close negative return strokes, and

the natural multi-scale characteristics of lightning

electric field waveforms have been investigated. The

decompositions denoted as IMFs having physical

meaning were often amplitude and frequency

modulated, and they were more meaningful to represent

the non stationary discharge process. We found that the

trend or residual was mainly the result of ramp. Then

the corresponding Hilbert-Huang spectrum based on

EEMD and EMD was obtained, respectively. Results

show that the multi-scale characteristics were very

significant in the RS stage. However, they were not

evident in the ramp stage due to the slow-change and

low-frequency electrostatic components for the close

return stroke.

In contrary with the wavelet spectrum, both of the

Hilbert-Huang spectrum based EMD and EEMD

method can give a better representation of the return

stroke natural multi-scale characteristics. We also found

that the return stroke signals were intermittent, which

was determined by the tortuous and branching channels’

multi-scale characteristics. Thus, the mode mixing

phenomenon may easily occur on the EMD-based

Hilbert-Huang spectrum. After being performed with

the EEMD method, the corresponding Hilbert-Huang

spectrum can give a significant improvement on mode

mixing problem for 11 return strokes, demonstrating

that these return strokes have significant intermittence

characteristics. However, there was not evident

improvement on the mode mixing for the other 5 return

strokes, indicating that the intermittence characteristics

of these return strokes were not significant. Therefore, it

is more accurate to observe the lightning multi-scale

characteristics in the EEMD-based Hilbert-Huang

spectrum. Moreover, the corresponding marginal

spectrum of the return strokes was also obtained. The

peak value distributions show that the energy of return

strokes was concentrated in the low-frequency range.

Some attention should be paid to the selection of the

different STDs which may influence on the

decomposition. Different return strokes may correspond

to different STDs to give an improvement on mode

mixing phenomenon. The reason we inferred may be the

serious intermittency characteristics. Even so, the mode

mixing can still be ameliorated by selecting the

appropriate STD for different return strokes.

Taken together, this study demonstrates the utility of

EEMD and EMD in analyzing the pattern of lightning

return strokes and provides new insights into the

understanding of the lightning physical discharge

structure. Apart from that, the EMD method may help

obtain a more accurate time of arrival (TOA) of the

return stroke’s peak value, which often can improve the

lightning location precision. Further study will focus on

the utilization of EEMD or EMD to that and to

investigate other lightning events to further assess the

utility of the EEMD method in investigating lightning

discharge process.

Acknowledgments:

This work was supported by the national key technology

program under grant NO.2008BAC36B00. The authors

would like to express their appreciation to Professor

Meng Qing and Lan Yu for their help in providing the

lightning measuring data.

References:

[1] Cummins, K. L. and Murphy, M. J.2009. An

overview of lightning locating systems: history,

techniques, and data uses, with an in-depth look at

the U.S. NLDN, IEEE Transansactions on

Electromagnetic Compatibility. 51, 499–518.

[2] Levine, D. M.1987. Review of measurements of the

RF spectrum of radiation from lightning,

Meteorology and Atmospheric Physics. 37, 195-

204.

[3] Lin, Y.T., Uman, M.A., Tiller, J.A., Brantley, R.D.,

Beasley, H.W., Krider, E.P. and Weidman,

C.D.1979. Characterization of lightning return

stroke electric and magnetic fields from

simultaneous two station measurements. Journal of

Geophysical Research. 84, 6307–6314.

[4] Weidman, C.D., Krider, E.P. and Uman,

M.A.1981.Lightning amplitude spectra in the

interval from 100 kHz to 20 MHz, Geophysical

Research Letters. 8, 931–934.

[5] Cooray, V. and Lundquist, S.1983. Effects of

propagation on the rise times and the initial peaks

of radiation fields from return strokes, Radio

Science. 18, 409–415.

[6] Krider, E. P. and Radda, G. J.1975. Radiation field

wave forms produced by lightning stepped leaders,

Journal of Geophysical Research.80, 2653-2657.

[7] Uman, M.A., Swanberg, C.E., Tiller, J.A., Lin,

Y.T. and Krider, E.P.1976. Effects of 200 km

propagation on Florida lightning return stroke

electric fields, Radio Science .11, 985–990.

[8] Serhan, G.I., Uman, M.A., Childers, D.G. and Lin,

Y.T.1980. The RF spectra of first and subsequent

lightning return strokes in the 1-to 200-km range,

Radio Science. 15, 1089-1094.

[9] Weidman, C.D. and Krider, E.P.1986. The

amplitude spectra of lightning radiation fields in the

interval from 1 to 20MHz, Radio Science. 21, 964–

970.

[10] Willet, J.C., Bailey, J.C. and Krider, E.P.1989. A

class of unusual lightning electric field waveforms




with very strong high-frequency radiation, Journal

of Geophysical Research. 94., 16,255–16,267.

[11] Willet, J.C., Bailey, J.C., Leteinturier, C. and

Krider, E.P.1990. Lightning electromagnetic

radiation field spectra in the interval from 0.2 to 20

MHz, Journal of Geophysical Research. 95,

20,367–20,387.

[12] Mallat, S.2003.A wavelet tour of signal processing,

China Machine Press, Beijing.

[13] Miranda, F.J.2008. Wavelet analysis of lightning

return stroke, Journal of Atmospheric and Solar-

Terrestrial Physics. 70, 1401-1407.

[14] Sharma, S.R., Cooray, V., Fernando, M. and

Miranda, F.J.2011. Temporal features of different

lightning events revealed from wavelet transform,

Journal of Atmospheric and Solar-Terrestrial

Physics. 73. 507-515.

[15] Gou, X.Q., Chen, M.L, Zhang, Y.J., Dong, W.S.

and Qie, X.S.2009. Wavelet multi-resolution based

multifractal analysis of electric fields by lightning

return strokes, Atmospheric Research. 91, 410–415.

[16] Qiu, S., Zhou, B. H., Shi, L. H., Dong, W. S.,

Zhang, Y. J. and Gao, T. C.2009. An improved

method for broadband interferometric lightning

location using wavelet transforms, Journal of

Geophysical Research. 114, D18211.

[17] Huang, N. E., Shen, Z., Long, S. R., Wu, M. C.,

Shih, H. H., Zheng, Q. N., Yen, N. C., Tung, C. C.

and Liu, H. H.1998. The empirical mode

decomposition and the Hilbert spectrum for

nonlinear and non-stationary time series analysis,

Proceedings of the Royal Society of London Series

a-Mathematical Physical and Engineering Sciences.

454, 903–995.

[18] Wu, Z .H. and Huang, N. E.2009. Ensemble

empirical mode decomposition: a noise assisted

data analysis method, Advances in Adaptive Data

Analysis. 1, 1–41.

[19] Wu, Z .H. and Huang, N. E.2004. A study of the

characteristics of white noise using the empirical

mode decomposition method, Proceedings of the

Royal Society of London Series a-Mathematical

Physical and Engineering Sciences. 460, 1597–

1611.

[20] Lan, Y., Zhang, Y. J., Dong, W.S., Lu, W. T., Liu,

H.Y. and Zheng, D.2011. Broadband analysis of

chaotic pulse trains generated by negative cloud-to-

ground lightning discharge, Journal of Geophysical

Research. 116, D17109.

[21] Lee, B.H., Eom, J.H., Kang, S.M., Peak, S.K. and

Kawamura, T.2004. Characteristics of radiation

field waveforms by the lightning return strokes,

Japanese Journal of Applied Physics. 43, 4379-

4385.

[22] Levine, D.M. and Willett, J. C.1995. The influence

of channel geometry on the fine scale structure of

radiation from lightning return strokes, Journal of

Geophysical Research. 100, 18629-18638.

[23] Meredith, S.L, Earles, S. K. and Kostanic,

I.N.2010. How lightning tortuosity affects the

electromagnetic fields by augmenting their

effective distance, Progress in Electromagnetics

Reaserch B, 25, 155-169.

Figure 4.1: An Example of Electric Field Waveforms of

a Negative Return Stroke with a Range of About 15 Km.

The Close Return Stroke is Divided as Two Portions:

One is Called Return Stroke Stage Denoted as Rs Stage,

and the other is called Ramp

Figure 4.2: The Wavelet Amplitude Spectrum of the

Negative Return Stroke Depicted in fig.4.1. Here, the

Dog Wavelet Function was Chosen for Analysis




Figure 4.3: The IMFs Produced with Emd Method of the Return Stroke Depicted in Fig. 4.1.

Figure 4.4: The EMD-Based Hilbert Huang Spectrum of the Return Stroke Depicted in Fig.4.1




Figure 4.5: The IMFs Produced with Eemd Method of the Return Stroke Depicted in Fig.4.1

with STD=0.01 and N=200.

Figure 4.6: The EEMD-Based Hilbert Huang Spectrum of the Return Stroke Depicted in Fig.4.1

with STD=0.01and N=200




Figure 4.7: The EEMD-Based Hilbert Huang Spectrum of the Return Stroke Depicted in Fig.4.1

with STD=0.05 and N=200

Figure 4.8: The EEMD Based Marginal Spectrum of the Return Stroke Depicted in Fig.4.1

with STD=0.01and N=200

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1492-1499


Peak Ground Acceleration on Bedrock and Uniform Seismic Hazard

Spectra for different Regions of Behbahan, Iran

SEYED ALI RAZAVIAN AMREI1, GHOLAMREZA GHODRATI AMIRI

2, ARMAN SAED

3

and ALI SABZEVARI4

1Department of Civil Engineering, Payame Noor University, Tehran, Iran

2Center of Excellence for Fundamental Studies in Structural Engineering, School of Civil Engineering Iran,

University of Science & Technology, Tehran, Iran 3Department of Civil, Behbahan Branch, Islamic Azad University, Behbahan, Iran 4Department of Civil Engineering, Islamic Azad University, Shahrekord Branch

Email: [email protected], [email protected], [email protected], [email protected]

Abstract: The present paper was done under the title of peak ground acceleration (PGA) on bedrock and uniform

seismic hazard spectra (UHS) for different regions of Behbahan city. A set of seismic sources, historical, and

instrumental seismic were used by means of data regarding the time since 8th century until now with a radius of 150

km. Kijko program (2000) was applied for calculation of seismic parameters considering lack of suitable seismic

data and uncertainty of magnitude in different periods. The calculations were performed by using the logic tree

method. Three weighted attenuation relations were used; including, Ramazi (1999), 0.35, Campbell (1997), 0.25 &

Ghodrati Amiri et al (2007), 0.4. In order to determine the seismic spectra based on weighted attenuation spectral

relations, and also for the reason of being spectral and more suitable with the conditions of the zone, Ambraseys et

al (1996), 0.35, Ghodrati Amiri et al (2010), 0.4 & Campbell (1997), 0.25 were used. The SEISRISK III software

was used to calculate the earthquake hazard. The results of this analysis were submitted including the spectra and

maps for 2% and 10% PE in 50 years.

Keywords: Seismic hazard analysis, Peak Ground Acceleration, Uniform seismic hazard spectra, Behbahan, Iran

1. Introduction:

Behbahan with the area 3743 sq. km where dates back

to the second millennium BC is located in the extreme

southern part of the Khuzestan province. It is one of the

most important cities of the province and country.

There are numerous important historical, religious and

industrial places in Behbahan, including two gas and

two large dams; Kosar and Maroon. Given the past

recorded history of earthquakes and faults in the range

of the city, the possibility of another earthquake event is

inevitable. Due to the lack of detailed engineering

design of structures, especially in older structures, if a

large earthquake in this city happens, a terrible tragedy

will be experienced. Probabilistic analysis is one of

modern methods in analyzing seismic risk. In this

analysis, the uncertainty within different parameters is

taken in consideration and results are presented

logically. In this study, it was sought to accurately

identify both local area faults and numerous attenuation

relations suitable for the region to ensure more reliable

results [1].

2. Seismotectonics:

Due to active faults in the surrounding areas, Behbahan

is among active seismic places. Throughout the present

study, in order to evaluate the seismic hazard in the

region, all sources of possible earthquakes and their

ability to generate strong ground movement have been

collected. A list of critical faults in the range of 150 km

is given in Table 1 and in Figure 1 some of the faults in

the studied areas are shown.

Table 1: Main Faults within 150 km Radius of

Behbahan

Length of the faults Km Faults No.

150 Aghajari 1

80 Rag Sefid 2

150 Lahbari 3

76 Mishan 4

100 Dena 5

In Figure 1 some of the faults which were studied in the

area were shown.

1493 SEYED ALI RAZAVIAN AMREI, GHOLAMREZA GHODRATI AMIRI,

ARMAN SAED and ALI SABZEVARI



3. The Peak Earthquake Magnitude and Fault

Rupture Length Appendix:

To estimate the relationship between the peak expected

magnitude and fault length, seismotectonic and

geotectonic behaviors of the concerned area were taken

into consideration and the following relation was used

[2].

LogLMs 244.1259.1 += (1)

Where, In Eq. (1), Ms is the surface magnitude and L is

the rupture length in meter.

Figure 1: Main Faults within 150 km Radius of the

Behbahan City [3]

4. Seismicity:

The history of past earthquakes in each width is an

indication of the Seismicity of that area. Thus, in order

to conceive the Seismicity features, we should have a

comprehensive list of occurred earthquakes in the area.

In the present work, a number of earthquakes in a radius

of 150 km of Behbahan city were collected and

considered.

4.1 Historical Earthquakes:

The historical documents indicate that the historical

earthquakes occurred before 1900. As far as the

collected information from old and historical books was

concerned, their validity may be under question because

they may have exaggerated the extent of the damage

and destruction in excess negligence. However, the

existence of such places could be important in the

process of gathering information. Among the most

important historical earthquakes, the following ones are

to be mentioned [4].

The first Earthquake occurred in the year 452 (AD)

Arjan - successive earthquakes in the fifth century AD -

as it is stated by Ebne-Kasir in his book, two severe

earthquakes occurred in 445 AD and 478 AD in Arjan

(the old name of the present Behbahan) written by one

of two severe earthquakes in 445 years.

4.2 Instrumental Earthquakes:

In spite of the uncertainties in estimating the epicenter,

focal depth, and magnitude of earthquakes in seismic

data in twentieth century, these earthquakes are crucial

with regard to the instrumental registration. From 1963,

with the installation of seismography network, the

uncertainties in their estimations were prominently

decreased. A list of instrumental earthquakes in

Behbahan from 1900 to the present has been collected,

the most important of which is the website of

international seismological center [3].

4.3 Earthquake Magnitude:

In the present study, the surface-wave magnitude, Ms,

was used in order to analyze the seismic hazard

magnitude. As far as, the collected magnitudes were not

of Ms type, they were converted to Ms. Thus, in order to

convert the wave magnitude and Local magnitude to

Ms, Table 2 [1] was employed. Moreover, in order to

convert mb to Ms, equation 2 was used [5].

MS = 1.21*Mb - 1.29 (2)

Where, Ms and Mb stand for the surface-wave

magnitude and the body-wave magnitude respectively.

Table 2: Magnitude Convert [1]

Ms Mw ML

3.6 4.5 4.8

4.6 5.2 5.3

5.6 5.8 5.8

6.6 6.6 6.3

7.3 7.3 6.8

5. Seismicity Parameters of Behbahan:

Seismicity parameters or the peak expected magnitude,

Mmax, λ, and β are among the basics of the seismicity

of a place. They are used to indicate the seismicity of a

place. Collecting earthquake data for Behbahan

according to the fundamental assumption in estimating

Seismicity parameters, Filtered data was evaluated in

Poisson distribution. The method that was used for

elimination of foreshocks and aftershocks is the variable

windowing method in time and space domains Gardner

and Knopoff [6].

5.1 Determination of Seismicity Parameters:

In this paper, in order to estimate the seismic parameters

due to the shortage of appropriate seismic data and the

uncertainty of earthquake magnitude, the Kijko method

[7] was used based on the probabilistic method of peak

likelihood estimation. In this method, according to the

faults of Seismic data and the low accuracy at different

times, their occurrence in determination of seismicity

parameters Mmax, λ, and β are used. The results of

1494 Peak Ground Acceleration on Bedrock and Uniform Seismic

Hazard Spectra for different Regions of Behbahan, Iran



applying this method includes determination of the

seismic parameters (Table 3), the return period,

probability of event and the magnitude of seismic events

at different times. In Figure (2) estimation of the return

period of earthquakes in Behbahan (Kijko method) is

presented.

Table 3: values of Seismic Parameters in the range of

150 km Behbahan (Kijko method) [6].

Results

Beta 1.61 + - 0.11 (b=1.05+-0.03)

Lambda 3.5 For Mmin=4

Figure 2: Estimation of the Return Period of

Earthquakes in Behbahan (Kijko method)

6. Seismic Hazard Analysis:

The present paper aims at providing Peak Ground

Acceleration (PGA) on bedrock and uniform seismic

hazard spectra (UHS) for different regions of Behbahan

using Probabilistic Seismic Hazard Analysis.

In this method, the procedure starts with identifying and

modeling the distribution of seismic sources, evaluation

of recurrence relationship, evaluation of local site

effects such as soil types, estimation of activity rate for

probable earthquakes, evaluation of attenuation

relationships for peak ground acceleration, geotechnical

characteristics of sediments, topographic effects

resources and the probabilistic analysis of the risk of

earthquakes the likely location of earthquakes causing

them the determination of the acceleration spectra on

bedrock has been used.

6.1 Attenuation Relationship:

The selection of an appropriate attenuation relationship

is of high importance in the reliability of the results

taken from seismic hazard assessment. Throughout this

process, the following points are to be taken into

consideration. The needed points are as follow, source

specifications, direction of the wave propagation,

geology and topography effects of the site, magnitude,

refraction and energy absorption due to the properties of

the material through which the waves pass, fault

mechanism, reflection, and distance from seismic

source. Knowing about just above mentioned points,

here are three different attenuation relationship,

Ghodrati Amiri et al [8], Campbell [9], Ramazi [10]

using the logic-tree method with the weighs of 0.4, 0.25,

0.35 respectively, used in the process of providing Peak

Ground Acceleration (PGA) on bedrock and uniform

seismic hazard spectra (UHS). Moreover, in order to

provide spectra aceleration map and uniform seismic

hazard spectra, Ghodrati Amiri et al [11], Ambraseys et

al [12] and Campbell [9] using logic-tree method with

the weighs of 0.4, 0.25, and 0.35 respectively.

The reason for using the Logic-tree method is that using

a single attenuation relationship is not an appropriate

choice because the uncertainty of given data is not as

reliable as desired. Moreover, the local and global

relationships which enjoy a higher accuracy in

comparison with those of Iran, the other countries’ data

are used in the provision of their model. Therefore, as a

logical conclusion, the best method is the use of both

different attenuation relationships together with the

Logic-tree. Performing in this way, each one

compensate for the other one’s shortage. There are two

parameters in assigning the weigh to the branches of

each Logic-tree, including conditions in the given site

and considering higher effect of local relationship. In

Figures 3 and 4 the used Logic-trees with the weight of

each branch are indicated.

Figure 3: the Used Logic-Tree Together With Weight of

Each Category for Determination of PGA

Figure 4: The Used Logic-Tree Together With Weight

of Each Category for Determination of UHS





Figure 5: Zoning Maps of PGA with 10% and 2% PE in 50 years (Left to Right), and the Border of Behbahan

(Thick Line)

Figure 6: Zoning Maps of 0.2 s Spectral Acceleration with 10% and 2% PE in 50 Years (Left to Right) in Soil Type

2, and the Border of Behbahan (Thick Line)







Figure 8: Zoning Maps of 0.5 s Spectral Acceleration with 10% and 2% PE in 50 years (Left to Right) in Soil Type


Figure 9: Zoning Maps Of 0.5 S Spectral Acceleration with 10% and 2% PE in 50 years (Left to Right) in Soil Type


Figure 10: Zoning Maps of 1.0 S Spectral Acceleration with 10% and 2% PE in 50 years (Left to Right) in soil type

2, and the border of Behbahan (Thick Line)





Figure 11: Zoning Maps of 1.0 S Spectral Acceleration With 10% and 2% Pe in 50 Years (Left To Right) in Soil

Type 3, and The Border of Behbahan (Thick Line)









6.2 Probabilistic Seismic Hazard Analysis:

In this part, based on the modeled seismic sources,

seismic parameters, and SEISRISK III software [13],

the peak horizontal acceleration on bedrock (PGA) and

horizontal spectral acceleration, each with 10% and 2%

PE in 50 years (equivalent to a return period of 475 and

2475 years) in accordance with the levels of 1 and 2 of

Seismic Rehabilitation of Existing Building [14], for a

4x6 network where surround Behbahan appropriately,

were estimated. As far as the type of soil is unknown, in

each place of the network, exact calculations for soil

type 2 and 3 (the most probable soil type in the area)

based on Iranian Code of Practice for Seismic Resistant

Design of Buildings [15] were done. The maps for peak

ground acceleration on bedrock (PGA) and horizontal

spectral acceleration based upon the two soil types of

Behbahan city for the 0.2, 0.5, 1, 2 periods are presented

in Figures 5 to 13.

7. Uniform Hazard Spectra (UHS):

The uniform hazard spectra (UHS) are formed in the

form of a response in which any time, there is the same

probability range for its occurrence. Throughout these

spectra, within all periods in the life of the structure, the

probability of occurrence is considered as the same. In

other words, in designing a structure, the return period

of spectral acceleration for different periods is

considered the same. For this purpose, in any part of the

network with given seismic hazard, this range is

obtained for different periods.

In Figures 14 and 15, the uniform hazard spectra for the

two soil types and risk levels 1 and 2 based on Seismic

Rehabilitation of Existing Building for Behbahan city

are presented. The spectra in these figures are presented

as the peak, minimum and average values of spectral

acceleration at different points in the range of the

network. Also to compare the risk levels 1 based on

Iranian Code of Practice for Seismic Resistant Design of

Buildings [15] and 70% and the standard range for the

risk level 2, the standard spectra range of Iranian Code

of Practice for Seismic Resistant Design of Buildings

[15] and 150% of it is presented.

Figure 14: UHS for Soil Type 2, with 10% and 2% PE in 50 years (Up to Down)

Figure 15: UHS for Soil Type 3, with 10% and 2% PE in 50 Years (Up to Down)

8. Conclusion:

1- PGA with the probability event, 10% PE in 50 years

(the period to 475 years or risk level 1) in the range of

Behbahan varies from 0.17 to g 0.24, also, this amount

in the Iranian Code of Practice for Seismic Resistant

Design of Buildings [15] is presented as 0. 3 g.





2- PGA with the probability event, 2% PE in 50 years

(the period to 2475 years or risk level 2) in the range of

Behbahan varies from 0.17 to g 0.24, also, this amount

in the Iranian Code of Practice for Seismic Resistant

Design of Buildings [15] is presented as 0. 3 g.

3- Acceleration and spectral acceleration values in the

northeast region have the highest value in comparison to

other regions.

4- The peak spectral acceleration at the surface of the

soil type 2 is obtained 0.87 g in the risk level 1.

5- The peak spectral acceleration at the surface of soil

type 3 is obtained 1.04 g in the risk level 1.

References

[1] A. Saed, “Peak Ground Acceleration (PGA) on

bedrock and uniform seismic hazard spectra for

different regions of Behbahan city”, M.Sc. thesis,

Azad University of Shahrekord, Supervised by

Prof. Ghodrati Amiri, and Dr. Razavian Amrei,

2011.

[2] Nowroozi, “Empirical relations between magnitude

and fault parameters for earthquakes in Iran”,

Bulletin of the Seismological Society of America

,Vol. 75, No. 5, pp. 1327-1338, 1985.

[3] International Institute of Earthquake Engineering

and Seismology website: http://www.iiees.ac.ir

[4] N.N. Ambraseys, and C.P. Melville, A History of

Persian Earthquakes, Cambridge University Press,

Cambridge, Britain, 1982.

[5] “IRCOLD, Iranian Committee of Large Dams

“Relationship between Ms and mb,” Internal

Report, 1994. (in Persian)

[6] J.K. Gardaner, L. Knopoff, “Is the sequence of

earthquake in southern California, with aftershocks

removed, poissonian?”, Bulletin of the

Seismological Society of America ,Vol. 64, No. 5,

pp. 1363-1367, 1974.

[7] A. Kijko, “Statical estimation of peak regional

earthquake magnitude Mmax”, Workshop of

Seismicity Modeling in Seismic Hazard Mapping,

poljce, Slovenia, May, 22-24, 2000.

[8] G. Ghodrati Amiri, A. Mahdavian, F. Manouchehri

Dana, “Attenuation Relationship for Iran”, Journal

of Earthquake Engineering, Vol. 11, Issue 4, pp.

469-492, 2007.

[9] K.W. Campbell, “Empirical near-source attenuation

relationships for horizontal and vertical components

of peak ground acceleration, peak ground velocity,

and pseudo-absolute acceleration response spectra”,

Seismological Research Letters, Vol. 68, No. 1, pp.

154–179, 1997.

[10] H.R. Ramazi, “Attenuation laws of Iranian

earthquakes”, proceedings of th 3rd International

Conference on Seismology and Earthquake

Engineering, Tehran, Iran, 1999.

[11] G. Ghodrati Amiri, M. Khorasani, R. Mirza Hesabi,

and S.A .Razavian Amrei, “Ground-Motion

Prediction Equations of Spectral ordinates and

Arias Intensity for Iran”, Journal of Earthquake

Engineering, Vol. 14, Issue 1, pp. 1-29, 2010.

[12] N.N. Ambraseys, K.A. Simpson and J.J. Bommer.

“Prediction of horizontal response spectra in

Europe”. Earthquake Eng. Struct. Dynam. Vol. 25,

pp. 371-400, 1996.

[13] A. Bender, D.M. Perkins, “SEISRISK-ІІІ: A

computer program for seismic hazard estimation”,

US Geological Survey, Bulletin 1772, 1987.

[14] IIEES. Seismic Rehabilitation Code for Existing

Buildings in Iran, International Institute of

Earthquake Engineering and Seismology, Tehran,

Iran, 2002.

[15] BHRC. Iranian Code of Practice for Seismic

Resistant Design of Building, Standard No. 2800,

Third Revision, Building and Housing Research

Center, Tehran, Iran, 2005.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1500-1509


Coal Mine Seal Design – Numerical Approach

RAJ R. KALLU Macky School of Earth Sciences- Mining and Metallurgical Engineering , University of Nevada, Reno, NV, 89557,

United States of America


Abstract: This paper addresses the design of reinforced concrete (RC) seals for coal mines for the 120 psi design

standard, in the USA, using numerical simulation techniques. The 3D explicit finite element code, ABAQUS, is

used in the simulation to understand the basic seal mechanical response to the design explosion loading. Structural

testing data available from full-scale explosion tests on concrete structures conducted in National Institute of

Occupational Safety and Health (NIOSH)’s Lake Lynn Experimental Mine (LLEM) is used for verifying the validity

of the modeling results. Based on the results obtained from the simulation of verification cases and the modeling

experience, design guidelines are proposed for reinforced concrete seals that meet the 120 psi design standard.

Finally, based on the comprehensive modeling work, RC seal design charts are proposed for typical coal mine entry

dimensions in the USA.Various other issues related to RC seal designs, such as, the influence of the roof-to-floor

convergence on the ability of the seal, the maximum allowable roof-to-floor convergence on the seal for a particular

entry dimension, etc., are discussed in detail. The effect of duration of explosion loading on the RC seal response is

also discussed in this paper.

Keywords: Coal Mine Seal, Roof-To-Floor Convergence, Explosion Loading, Numerical Modeling

1. Introduction:

The new 120 psi (827 kPa)mine seal design and

evaluation practices in the USA are mostly based on

simple analytical solutions.Considering the simplistic

assumptions and limitations involved in simple

analytical solutions and practical difficulties involved in

conducting conventional full-scale explosion tests for

evaluating 120 psi (827 kPa) seal designs, there is a

need for developing new scientific methods and

engineering practices for design, analysis and evaluation

of mine seals.

Among all other available alternatives, this paper

focuses on the use of numerical simulation techniques

for design of reinforced concrete (RC) seals. Three

dimensional finite-element based numerical simulation

program, ABAQUS, is used as a tool for design of RC

seals for 120 psi (827 kPa)design standard. This paper

provides insight into how the reinforced concrete

structures respond to the dynamic explosion pressure

loading, and also discusses the failure mechanism of the

seal structures under these loading conditions.

Appropriate design guidelines are proposed for design

of RC seals that meet the 120 psi (827 kPa)design

standard. Also, based on the comprehensive modeling

work RC seal design chart isdeveloped for use in typical

coal mine entries. Various other issues related to RC

seal designs, such as, influence of roof-to-floor

convergence, nature of explosion loading (different p-t

curves), multiple explosions, and duration of explosion

loading are discussed in detail.

2. Factors Considered in RC Seal Design:

A number of factors that affect the performance of the

reinforced concrete seals are considered in this paper.

Numerical models are used for evaluating these factors

and help identifying the key influential parameters in

the mine seal design. Factors considered in the current

study include:

• Nature of explosion loading – The pressure-time (p-

t) curves proposed in the 30 CFR is used for design

of reinforced concrete seals to comply with the

regulatory requirements.

• Concrete constitutive behavior – Complete stress-

strain curve in compression as well as in tension.

• Surrounding rock constitutive behavior - including

roof, floor and coal ribs

• Interaction of the roof, floor and coal ribs with the

seal structure

• Roof-to-floor convergence

3. ABAQUS Model: General:

Figure 1 shows the general layout of the ABAQUS

model. Two sets of vertical and horizontal rebars (grade

60 steel) are laid about 2.5 in (0.063 m). (cover

distance) from both faces of the seal, in accordance with

the guidelines suggested by the ACI (American

1501 RAJ R. KALLU



Concrete Institute) [1] for design of reinforced concrete

structures.

After applying the appropriate initial and boundary

conditions, interface properties, and constitutive

behavior of materials the model is solved dynamically

in two steps. In the first step, a very small amount of

velocity is applied on the top face of the model to allow

initialization of self- weight of the structure. In the

second step, explosion pressure is applied on the inby

face of the seal using the MSHA specified 120 psi (827

kPa) pressure-time curve and the model is solved for the

prescribed amount of time. A number of predefined

monitoring points are selected in the model to record the

history of seal’s lateral displacements, seal response

pressures and axial forces in the rebars.

Figure 1: Shows the General Layout of the ABAQUS.

4. Material Models:

The immediate roof and floor rocks and coal are

modeled using the extended Drucker-Prager model.

Deformation prior to yielding is assumed to be linear

elastic governed by the elastic parameters E and υ. The

modified Drucker-Prager plasticity model in ABAQUS

is intended for geological materials that exhibit

pressure-dependent yield. The model uses non-

associated flow in the shear failure region.

Geo-mechanical properties of the coal, immediate roof

and floor rock, and reinforced steel rebars used in the

models are given in Table 1.

The reinforced steel rebar is assumed to behave as a bi-

linear elastic-perfectly plastic material. A classical Von-

Mises yield criterion with associated plastic flow is used

in the modeling. The reinforced rebars are modeled as

beam elements embedded in the host concrete/rock

elements. These rebars are embedded into the roof and

floor to a distance of 2 ft (0.61 m) in the simulation

models. The concrete and the internal steel rebars

interaction is approximately considered in the modeling

by tension stiffening of the concrete material model.

The interface between the reinforced concrete seal and

the surrounding rock is modeled using the bi-linear

Coulomb friction model with zero cohesion. According

to this model, sliding will occur if the magnitude of the

shear stress along the interface reaches a critical value,

τmax, regardless of the magnitude of the contact pressure.

Table 1: Geo-Mechanical Properties of Rock, Coal and Steel

Property Rock Coal Steel

Young’s Modulus (E), psi (GPa) 3.00E+06 3.00E+05

2.9e+07 (199.948) -20.684 -2.068

Poisson’s ratio (υ) 0.19 0.35 0.28

Density, lbs/ft3 (kg/m

3)

162.3 86.5 486 (7785)

-2600 -1386

Friction angle 35 29 -

Cohesion, psi (MPa) 370 psi 163 psi

- (2.55 MPa) (1.12 MPa)

Yield strength, psi (MPa) - - 60,000 psi (413.68 MPa)

5. Ceb & Barth et al., Concrete Model:

Non-linear behavior of concrete is simulated with a

uniaxial stress-strain curve with strain softening and

tension stiffening (Figure 2). The compressive part of

concrete material model was initially developed by

Comite European Du Beton (CEB) and has been

verified by Barth and Wu [2] by comparing the

complete load-deflection relationships and ultimate

capacities resulting from the finite element analysis of

two simply supported composite girders and a 4-span

continuous composite steel bridge with experimental

data. Simply supported composite girder experimental

tests were conducted by Mans [3] at the University of

Nebraska and the 4-span continuous composite steel

bridge experimental test was reported by Burdette et al.

[4]. Barth et al.[3] used the finite element based

software ABAQUS in their analysis.

The CEB compressive concrete model relates the stress-

strain by the following equation which is a function of

only one parameter (f’c), yet it can accurately reflect the

1502 Coal Mine Seal Design – Numerical Approach



variation in response to different strengths that are

observed in practice, Hognestad et al. [5]. However, it is

noted that the compressive concrete constitutive law

appears to have only minor effects on the overall results,

ASCE [6] and Cope and Rao [7].

Figure 2: Stress-Strain Behavior of Concrete in

Compression and Tension

c

cccc

b

aff

ε

εε

+

−=

1

)000,206(85.0 '

Where, a = 6193.6(0.85 fc + 1.015)-0.953

b = 8074.1(0.85 fc + 1.450)-1.085

- 850

fc= 28-day concrete compressive strength, ksi

εc = compressive strain in concrete

Based on available data in the literature,Wittry [8], a

value of 0.0038 is assumed for the maximum

compressive strain (εcmax). This value is typical

American Concrete Institute (ACI) maximum

compressive strain value and is thought to be

conservative since higher values of crushing strain have

been reported in physical tests by Wittry [8].

Barth and Wu [3] used a non-linear concrete tension

model in the analysis, according to which the tensile

stress increases linearly up to concrete cracking stress

(ft) and then unloads gradually. A parabolic curve

passing through the concrete cracking point (ft,εt) and

concrete maximum strain point (εtu) is suggested for the

gradual unloading portion by Barth and Wu (2006).

Beyond the cracking stress concrete exhibits a

significant tension stiffening behavior and the tensile

load carrying capacity of the concrete decreases

exponentially. The concrete and the internal steel rebars

interaction is approximately considered by the tension

stiffening behavior of the concrete.

Concrete Damaged Plasticity:

The Concrete Damage Plasticity option in ABAQUS is

used to define the yield function and flow potential. An

isotropic damaged elasticity in combination with

isotropic tensile and compressive plasticity is used in

the Concrete Damage Plasticity model to better

represent the inelastic behavior of concrete.

Figure 3 shows the relationship between concrete

damage and plastic strain for compression and tension,

Barth and Wu [3]. A damage variable close to 1.0

indicates complete failure of the material and, close to

0.0 indicates no-damage in the material.

6. Verification Case:

Although NIOSH engineers conducted extensive full-

scale explosion tests in LLEM on mine seals

constructed from various types of construction

materials, there were only two tests that were conducted

on seals constructed from concrete like materials with

internal steel reinforcement that meet the reinforced

concrete seal criteria, Zipf et al. [9]. These seals were

tested for the old 20 psi (137 kPa) design standard.

Explosion overpressure was created by igniting a

methane-air mixture in a confined area and monitored

using pressure transducers. Lateral displacements on the

outbycenter of the seal were monitored using LVDT’s.

The results from these full-scale explosion tests are used

for validation of the predictions from the ABAQUS

models. Kallu et al. [10] reported complete description

of the verification cases and the modeling results in

their work.

Figure 3: Relationship between Concrete Damage and

Plastic Strain

7. RC seal mechanical response:

A number of finite element seal models were

constructed,for 6 ft(1.83 m) entry height and 20 ft (6.1

m) entry width with #9 vertical rebars spaced at 10 in.

(0.25 m) apart and #6 horizontal rebars spaced at 18 in.

(0.46 m) apart, to understand RC seal mechanical

response to the dynamic explosion loading and the seal

failure mechanism under these loading conditions.The

seal thickness varied from 12 to 30 in. (0.30 m to 0.76

m) and is subjected to 120 psi (827kPa) instantaneous

explosion loading pressure lasting for 4 sec and

1503 RAJ R. KALLU



removed instantaneously. Figure 4 shows typical seal

response curves. Although the seal is subjected to 120

psi (827 kPa) instantaneous pressure, the lateral

pressures in the seal reached a maximum of about 160

psi (1100kPa) at various locations in the seal.

Figure 4: Typical Seal Response Curves: Lateral

Pressure Monitored at different Locations on the Seal

At 12 in. (0.30 m)thickness, seal exhibits extensive

damage of concrete in tension and offers no

bending/shear resistance to absorb the explosion

pressure. The seal is under continuous deformation for

the entire duration of loading. A very high level of

tensile forces is observed in the rebars on the outby as

well as on the inby side of the seal. With the increase in

seal thickness to 24 in. (0.61 m) and beyond the extent

of the damage (tensile cracks)in the seal reduces

significantly. The structural integrity of the seal

improvesquite significantly at the same time, and offers

high internal bending and shear resistance to absorb and

transfer the explosion pressure effectively to the

surrounding rock through the internal steel rebars. At

higher seal thickness, high axial forces concentrated in

the sections of the rebars near the roof and floor line.

Figure 5 shows the plot of the maximum lateral

dipslacements for different seal thickness measured at

various locations in the seal. The displacements in the

seal near the roof and floor are very much the same. The

difference in the displacements at Outby_M and

Outby_T/B decreases sharply with increasing in seal

thickness but once the seal reaches sufficient thickness

the difference approaches near zero and remains

constant with further increase in seal

thickness,indicating that at higher thickness seal

behaves like a rigid body (plug) and tries to shear along

the seal-rock/coal interfacesunder the applied explosion

loading. The seal eventually transfers the applied

explosion loading to the surrounding rock through the

steel rebars and seal-rock interfaces.

Figure5: Maximum Lateral Displacements Vs. Seal Thickness

Under the applied explosion loadingseal, when it is

sufficiently thin, bends over its heightand causes the

horizontal tensile cracks develop on the surface of the

seal and when the seal is sufficiently thick, 24 in. (0.61

m) and beyond in this particular case, it bends over its

length and develops vertical tensile cracks on the outby

surface of the seal. Figure 6 shows the tensile damage

contour (red) and bending of the 24 in. (0.61 m) thick

seal.

In either case, bending of the seal under the explosion

loading causes the inby section of the seal to be subject

to compressive stresses and the outby section of the seal

to be subject to tensile stresses. Concrete being weaker

in tension develops tensile cracks on the outby surface

of the seal, whereas the inby surface of the seal shows

no signs of damage. These tensile cracks extend only

half way deep into the seal (Figure 7). Also notice in

Figure 7 the development of shear failure surfaces

originating from the top outby corner of the seal and

extending deep into the seal. The extent of the shear

failure is reduced significantly with increase in seal

thickness to 21 in. (0.53) and nonexistent in seals with

thickness more than 24 in. (0.61 m)

It is important to recognize the fact that development of

the tensile cracks in normal concrete material cannot be

completely avoided unless the seal is highly




overdesigned or constructed using the fiber reinforced

concrete material. Simple cracking in the concrete does

not necessarily indicate the complete failure of the seal.

The primary purpose of installing the reinforced rebars

close to the seal surface is to take care of the high

tensile stresses that may develop near the surface of the

seal in the event of an explosion.

Figure 6: Tensile Cracks (red) in 24 in. (0.61 m) Thick

Seal

Figure 7: Different Failure Modes in 18 in. Thick Seal

(Mid-Vertical Section)

Figure 8 shows the plot of weighted average damage

factor (WADF) for different seal thicknesses. The

WADF decreases sharply with increase in seal thickness

to 21 in. (0.53 m) and reaches a value of about 0.057.

Further increase in seal thickness beyond 21 in. (0.53

m) showed very little change in WADF. The trend of

WADF plot can be effectively used as a tool in

determining the optimal thickness of a reinforced

concrete seal for a particular entry situation.

7.1 RC Seal Design Guidelines:

Based on the results obtained from modeling of the

verification cases, in-depth analysis of the seal response

to the explosion loading, and the extensive modeling

work the

Figure 8: Weighted Average Damage Factor for

different Seal Thickness

Following guidelines are proposed for the design of the

reinforced concrete seals.

• Tensile cracking: The seal design can be limited to a

single or no tensile crack in the seal

• WADF: The weighted average damage factor

decreases exponentially with the increase in seal

thickness and reaches a ‘near-steady’ value once the

seal reaches sufficient thickness. The ‘Near-steady’

value can be used to identify the optimal seal design

• Lateral displacements: Exhibits very much similar

behavior compared to that of WADF

• No shear failure in the seal is allowed in the seal

• No yielding in the steel rebars is allowed

8. RC Seal Design Chart:

A large number of numerical models have been built to

design the RC seals for typical coal mine entry sizes

based on the guidelines discussed in the previous

section. Entries with width-to-height (W/H) ratio

greater than or equal to 2.0 are modeled based on one-

way slab loading concept i.e., only the vertical rebars

are assumed to transfer the explosion loading and

therefore keyed to the roof and floor. Entries with

width-to-height ratio less than 2.0 are modeled based on

two-way slab loading concept i.e., both vertical and

horizontal rebars are keyed to the surrounding rock.

Figure 9 shows the RC seal design chart for typical coal

mine entry dimensions. Entry height showed greater

influence on the seal design than the entry width. At

lower entry heights, width of the entry showed a very

little or no influence on the seal design. On the other

1505 RAJ R. KALLU



hand, at higher entry heights, width of the entry has

some influence on the seal design.

Figure 9: Reinforced Concrete Seal Design Chart for

Typical Entry Dimensions

9. Roof-to-Floor Convergence:

After the initial development of coal mine entries they

are subjected to different levels of roof-to-floor

convergence. The rate of convergence depends on many

factors including but not limited to: the stress around the

entry, the geological conditions, the age of mine entry,

the mining method, the proximity of the entry to the

mined out areas, etc. In general, it is observed in

practice that the convergence rate is high during the

initial phases of life of the mine entry and in due course

of time the rate of convergence decreases exponentially

and very much stabilizes after a certain period of time.

Figure 10 shows the measured rate of convergence at

acoal mine in Western Kentucky byKallu et al. [11].

Figure 10: Measured Rate of Convergence for different

Entry Ages by Kallu et al. [11].

While designing a particular type of seal it is important

to understand how the seal responds to the roof-to-floor

convergence. This is particularly true with the

reinforced concrete seals because they are stiffer than

any other type of seals and cannot take a large amount

of roof-to-floor convergence.

This particular section tries to address how the

reinforced concrete seal respond to convergence loading

and its ability to sustain higher explosion loading

pressures when subjected to limited amount of roof-to-

floor convergence. Previous research by Kallu et al. [11]

on ‘pre-stressed’ seals showed that the capacity of the

seal to sustain explosion pressure loading increased,

within in certain limits, with increase in roof-to-floor

convergence.

Table 2 shows the summary of the stability condition of

the seals at different levels of roof-to-floor convergence

and explosion pressure. The convergence values shown

in the table are calculated by averaging the relative

displacements of the nodes at the top and bottom of the

seal at mid vertical section. Actual convergence values

measured in coal mine entries may be a lot higher than

these values since the convergence stations are usually

installed at least a few feet away from the mine seals. In

order to use the convergence values shown in the table

below in a meaningful way, there is a need to develop

appropriate relationship between the actual convergence

measured in the mine entries and the convergence on

the seals for a particular mining situation.

The letter ‘S’ shown in the table indicates that the seal is

stable and the letter ‘F’ indicates that the seal is ‘not

suitable’ under the given loading conditions. From the

table it can be observed that the explosion resistant

capacity of the seal increases with increase in

convergence on the seal. At zero convergence, the seal

is only able to withstand 120 psi (827 kPa) explosion

pressure but the same seal can withstand about 280 psi

(1930kPa) explosion pressure at 0.073 in. (1.854 mm)

of convergence. This positive influence of roof-to-floor

convergence is only valid as long as the stresses induced

in the seal due to the convergence loading are well

within the strength limits of the concrete material. Once

these induced stresses exceed the strength of the

concrete the seal fails under the convergence loading

alone. At about 0.085 in. (2.16 mm) of convergence the

seal fails completely under the convergence loading

alone.

Figure 11 shows the contours of plastic strain in the seal

and rebars at 0.1 in. (2.54mm) of convergence on the

seal. Observing the plastic strain and damage contours

in the seal at this stage indicate the onset of shear failure

in the seal due to the excessive convergence loading on

the seal.




Table 2: Convergence Table (6’x20’x24’’ seal)

Convergence

on seal, in.

Explosion pressure, psi

12

0

14

0

16

0

18

0

20

0

22

0

24

0

26

0

28

0

30

0

0 S F F F

0.0023 S S F F

0.0047 S S S F

0.0093 S S S S F F

0.014 S S S S S F

0.0235 S S S S S F F F

0.0345 S S S S S S F F

0.044 S S S S S S F F

0.0528 S S S S S S S F

0.0629 S S S S S S S S F

0.0728 S S S S S S S S S F

0.0827 S S S F F F F F F F

0.0929 F F F F F F F F F F

0.104 F F F F F F F F F F

0.1825 F F F F F F F F F F

S – Stable, F – Not suitable

Figure 11: Contours of Plastic Strain in The Seal (half

section) and in Rebars at 0.1 in. (2.54 mm)

Convergence

9.1 Maximum Allowable Convergence:

This section tries to determine the maximum amount of

roof-to-floor convergence that a particular RC seal,

designed for various entry dimensions, can sustain

without losing its ability to withstand the explosion

pressure loading.

Based on the comprehensive modeling work, the

maximum allowable convergence for different seal

heights are plotted in figure 12. The slope of the line

shown in figure 12 represents the maximum amount of

strain (0.0012) that can be allowed on the RC seal.

Figure 12: Maximum Allowable Convergence on the

RC Seal for Various Seal Heights

10. Explosion Loading Nature:

Four different pressure-time (p-t) curves with a

maximum of 120 psi (827 kPa) explosion pressure as

shown in figure 13 are considered for investigating the

seal response. In the first two p-t curves, the explosion

pressure is applied for 4 seconds and then released

instantaneously. On the other hand, in the later two p-t

curves the explosion pressure is gradually reduced from

120 psi (827 kPa) to zero over a period of 0.1 sec.

1507 RAJ R. KALLU



Figure 13: 120 psi (827 kPa) Pressure-Time Curves

with different Rise and Down Times

Figure 14: WADF for Seals Subjected to different P-T

Curves Shown in Figure 13

It is observed from the results that the seal subjected to

instantaneous pressure loading lasting for 4 seconds

sustained higher damage compared to all other cases.

The seal subjected to explosion pressure with 0.1 sec

rise time and lasting for 4 sec showed similar but lower

damage compared to the previous case. On the other

hand, seals subjected to explosion pressures with

different rise times and 0.1 sec down time (Figures 13

(c) & (d)) showed no damage (tensile cracks) on the

surface of the seal. Figure 14 shows the weighted

average damage factor for seals subjected to different p-

t curves.

Explosion pressures maintained for longer durations at

peak values significantly influence the seal stability

compared to any other parameter. Explosion pressures

applied using p-t curve having instantaneous rise time

produces comparatively more damage in the seal

compared to the p-t curve with 0.1 sec rise time; this

influence is more significant when the peak explosion

pressure is maintained for longer durations.

10.1 Explosion Loading Duration:

It is evident from the discussion in the previous section

that explosion pressures maintained for longer durations

at peak values significantly influences the seal stability.

At the same time,p-t curves with longer duration of

loading require significant amount of computational

time for solving the numerical models. There is a need

to determine appropriate loading duration that can

sufficiently represent the longer duration explosion

loadings.

Figure 15 shows the plot of WADF of the seal over the

complete duration of the explosion loading (MSHA p-t

curve). The WADF curve showed a sharp initial rise and

reached a near steady value in about 1.5 sec and

remained almost unchanged thereafter. A similar

behavior is also observed with the lateral displacements

in the seal. Based on these results, the author believed

that the explosion pressure maintained for about 1.5 to 2

seconds can sufficiently represent the longer duration

explosion loadings.

Figure 15: Change of WADF over the Duration of

Explosion Loading (p-t Curve Also Shown)




11. Multiple Explosions:

In some instances in underground coal mines there is a

possibility that mine seals may be subjected to multiple

explosions. In order to understand the effect of multiple

explosion loadings on the seal behavior different

pressure-time (p-t) curves shown in figure 13 are used

with two and three explosion pulses.

Figure 16: Multiple Explosions: Maximum Axial Forces

Developed in Steel Rebars

Figure 17: Multiple Explosions: Max. Lateral

Displacements Developed in Steel Rebars

The extent of the tensile damage in the concrete

material increased with increase in number of

explosions (to three) in all cases, but is very much

limited to a couple of tensile cracks on the outby surface

of the seal. Similarly,maximum axial forces in the steel

rebars and lateral displacements on the outby surface of

the seal also increased with each explosion loading

pulse. It is interesting to note that the maximum axial

forces developed in the steel rebars are well within yield

strength of the steel, and the lateral displacements

remained steady over time after each explosion pulse.

Refer to figures 16 to 18 for comparison of results.

Despite of the above mentioned factors, the seal designs

suggested in Table 2 can withstand multiple explosions

without any catastrophic failure.

Figure 18: Multiple Explosions: Weighted Average

Damage Factor.

12. Discussion and Conclusions:

There are different methodologies available, though

some are simple and some are complex, for design of

mine seals but no single method could provide accurate

solutions to various issues associated with the mine seal

design for the new MSHA 120 psi design standard

because of their unique strengths and weakness.

Although numerical simulation methods demand

significant amount of quality input data and user

expertise in their application to design problems, the

author believe that those are currently the best available

tools for a design engineer for addressing the seal

design issues with certain degree of acceptability.

In the absence of any specific guidelines for the design

of reinforced concrete seals, the recommendations

provided in this paper can serve as a reference for

design of a RC seal. Based on the design guidelines

proposed in this paper and comprehensive modeling

work, RC seal designs are summarized into a simple

design chart that can help readily identify appropriate

design for a particular entry dimension.

Various topics discussed in this paper, such as, seal

response to dynamic explosion loading and failure

mechanism, effect of roof-to-floor convergence, nature

of explosion loading and multiple explosions, and

duration of explosion loading on the seal behavior,

could help answer many unanswered questions about

the reinforced concrete seal design.

Acknowledgements:

The authors would like to thankSr. Syd S. Peng of

WVU, Morgantown, USA, Dr. Zipf and Dr. Morsy of

1509 RAJ R. KALLU



the National Institute for Occupational Safety and

Health (NIOSH), Pittsburgh, USA, and Dr. Yassien for

their help and adviceduring this work. Also, would like

to recognize CDC (NIOSH) for sponsoring this work.

Disclaimer:

The opinions expressed in this paper are belong to the

authors and do not necessarily reflect those of the

institutions they belong to.

References:

[1] ACI 318-95, Building code requirements for

structural concrete and commentary, American

Concrete Institute.

[2] Barth, K. E., and H. Wu, (2006). Efficient nonlinear

finite element modeling of slab on steel stinger

bridges, Finite elements in analysis and design,

Volume 42, No. 14, pp. 1304-1313.

[3] Mans, P.H., (2001). Full scale testing of composite

plate girder constructed using 70-ksi high

performance steel, M.S. Thesis, 2001, University of

Nebraska-Lincholn, USA.

[4] Burdette, E.G., and Goodpasture, D. W., (1971).

Final report in full scale bridge testing an

evaluation of bridge design criteria, Department of

Civil Engineering, the University of Tennessee,

USA.

[5] Hognestad, E., Hanson, N. W., and McHenry, D.

(1955). “Concrete stress distribution in ultimate

strength design.” J. American Concrete Institute,

27(4), pp. 454-479.

[6] ASCE (1993). State-of-the-Art Report on Finite

Element Analysis of Reinforced Concrete, ASCE,

New York, NY.

[7] Cope, R. J., and Rao, P. V. (1981). “Non-Linear

finite element strategies for bridge slabs.” IABSE

Colloquium, Delft University, Netherlands, pp.

273-288.

[8] Wittry, D. M. (1993). “An analytical study of the

ductility of steel-concrete composite sections.”M.S

Thesis, the University of Texas at Austin, USA.

[9] Zipf, R. K., E. S. Weiss, S. P. Harties, and M. J.

Sapko, (2009). Compendium of Structural Testing

Data of 20 psi Coal Mine Seals. Pittsburgh, PA:

U.S. Department of Health and Human Services,

National Institute for Occupational Safety and

Health, IC in press.

[10] Kallu, R. R., Yassien, A., Morsy, K. M., and Peng,

S. S., (2009), 3D Dynamic Simulation of

Reinforced Concrete Seals for 120 psi Design

Standard, 28th

International Conference on Ground

Control in Mining, July, pp 162-168.

[11] Kallu, R. R., Peng, S. S., Turner, D., Morsy, K. M.,

Zingano, A., (2007), Effect of Roof Convergence

on Stability of Mine Seal Subjected to Explosion

Loading – Numerical Approach, 26th

International

Conference on Ground Control in Mining, Aug, pp

370-378.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1510-1514


A Note on Three Way Quality Control of Argo Temperature and

Salinity Profiles - A Semi-Automated Approach at INCOIS

T. V. S. UDAYA BHASKAR1, E. PATTABHI RAMA RAO

1, R. VENKAT SHESU

1 and R. DEVENDER

1

1Indian National Centre for Ocean Information Services (INCOIS), Pragathi nagar (BO), Nizampet (SO),

Hyderabad – 500090

Email: [email protected], [email protected], [email protected], [email protected]

Abstract: A three way semi-automated quality control system established at Indian National Centre for Ocean

Information Services (INCOIS) for quality control of near real time temperature and salinity profiles obtained from

Argo deployed by India as well as other countries in the Indian Ocean is presented. At the outset, all the temperature

and salinity profiles are passed through 18 automated quality checks as suggested by the International Argo Data

Management Team (ADMT). Further, all the profiles are utilized in generating objectively analyzed product. Bad

profiles appearing as bulls eye are automatically rejected based on preset statistics. These bad profiles are then

visually checked (which requires manual intervention) using a visual quality control tool developed in house, for

their correctness. At the end of the quality control procedures, 10days and monthly objectively analyzed gridded

product of 1˚ X 1˚ resolution for the period 2002 - 2012 is generated and is made freely available on INCOIS Live

Access Server (las.incois.gov.in) in NetCDF format.

Keywords: Argo, Objective Analysis, Visual Quality Control, Real Time Quality Control

1. Introduction:

Argo is an internationally coordinated program directed

at deploying the global ocean with 3000 profiling floats

that measure temperature and salinity (T/S). The

profiling float sinks after launch to a prescribed pressure

level, typically 1000 dbars and after a preset time

(typically 10 days) the float dives to 2000 dbars and

returns to the surface, collecting T/S measurements

(Argo Science Team, 2001; Ravichandran et al., 2004).

On the surface the float transmits the data to ARGOS

satellite which are received at the ground station and are

analyzed.

Since the year 2000, the number of Argo floats in the

world oceans has been increasing year by year and the

projects goal of 3000 operating floats in the world ocean

was reached in November 2007, At the time of writing

this manuscript 3573 floats are in operation in the world

oceans. The number of profiles obtained annually by

Argo in the world oceans was more than 30,000 in 2003

and this number tripled to about 90,000 in 2006. Thus,

the annual total of Argo profiles obtained every year is

now equivalent to three quarters of the total number of

historical Conductivity-Temperature-Depth (CTD)

profiles deeper than 2000 m archived in the world ocean

data base 2001 (Ohno et al., 2009).

Quality control (QC) is an important part of ocean data

assimilation system. If erroneous values are assimilated

they can cause immediate spurious overturning and also

error propagation due to the sparse distribution of

oceanographic data (Bruce et al., 2007). In preparing his

ocean atlas Levitus (1982) noted that “By far the biggest

problem faced in this project concerned quality control

of the data”. Various data centers have developed

different QC procedures and also documented the same.

The semi-automated system described here processes

T/S profiles obtained from Argo profiling floats

deployed by India and other countries in the Indian

Ocean. With minor variations this system is suitable for

QC of near real-time as well as old archived T/S data

pertaining to global and/or regional domain. The output

of this QC process is specifically targeted for generation

of objectively analyzed gridded product which can be

used for studying the variability of the ocean over a

period for which the gridded product is available.

Further, this gridded product can also be used for

assimilation into ocean models.

Taking in to view the vast number of T/S profiles

produced by Argo floats, manual quality control of all

these profiles is time consuming and not affordable. In

this work, we present a semi-automated three way

quality control system developed at Indian National

Centre for Ocean Information Services (INCOIS) for

near-real time quality control of the Argo T/S profiles.

The system use a combination of automated QC

procedures (real time checks and Objective Analysis)

and visual QC system (which require manual

intervention). These QC procedures increase the quality

1511 T. V. S. UDAYA BHASKAR, E. PATTABHI RAMA RAO,

R. VENKAT SHESU and R. DEVENDER



and reliability of the data there by increasing the value

of these important data sets for climate and ocean

studies and reanalysis efforts.

The rest of the paper is organized as follows. Section 2

briefly reviews the data and methods and describes the

QC processing and checks which are relevant to these

observations. Section 3 provides a discussion and

summary.

2. Data and Methods:

INCOIS being the Argo Data Assembly Centre (DAC)

for India and Argo Regional Centre (ARC) for the

Indian Ocean, archives Argo T/S data from Indian

Ocean (20° – 140° E and 70° S – 30° N). Profiles

obtained from various floats deployed by countries

participating in International Argo program are archived

at INCOIS on a day to day basis. Each DAC at various

countries are responsible for the real time quality

control (RTQC) of floats deployed by them, as

prescribed by the Argo Data Management Team (Wong

et al.,2012). In the light of this, INCOIS DAC is

performing automatic RTQC of profiles obtained from

274 floats deployed by India over the years, starting

from October 2002. Similarly profiles from floats

deployed by other countries will undergo RTQC at their

respective centres and are made available at Global Data

Assembley Centres (GDAC). All T/S profiles obtained

from Indian as well as non-Indian floats, are archived at

INCOIS for additional QC processes. These additional

QC procedures along with the automatic RTQC

procedures forms the three way QC system which is

described in detail below.

Figure 1 shows the flow of the three way quality control

system in place at INCOIS for near real time quality

control of Argo T/S profiles. The T/S profiles obtained

from Argo floats are subjected to three levels of QC as

described below:

• Argo Real-Time QC: The first level is the real-

time system that performs a set of agreed checks (as

prescribed by Argo Data Management Team (Wong et

al., 2012)) on all profiles obtained from Argo floats.

Real-time data with quality flags assigned are available

to users within the 24 hrs timeframe at GDACs and

Global Telecommunication Systems (GTS).

• Objective Analysis: The second level of QC

involves usage of objective analysis method for

identification of outlier which appear as bulls eyes.

• Visual Quality Check: The third level of quality

control involves adjustment of suspicious quality flags

set by the Argo RTQC procedures, by visual inspection.

2.1 Argo Real-Time Quality Control (RTQC) Tests:

Because of the requirement for delivering data to users

within 24 hours of the float reaching the surface, the QC

procedures on the real-time data are limited and

automatic. The tests followed in RTQC are listed in

Table 1. There are total 18 RTQC tests conducted on the

Argo profile data. More detail on the tests can also be

found in IOC Manuals and Guides #22 or at

http://www.meds-sdmm.dfo-mpo.gc.ca/ ALPHAPRO/

gtspp/ qcmans/ MG22/ guide22_e.htm.

Even though majority of the profiles pass the RTQC

tests, there can be cases where in the profiles are

incorrectly assigned flags. Also as most of the automatic

QC tests for T/S profiles rely on individual records, a

bad value in one record might cause others records to be

assigned a bad flag.

2.2 Quality Control using Objective Analysis:

To over come some of the problems with the automatic

QC procedures, data is further subjected to additional

quality checks. Objective analysis is used for identifying

profiles with bad T/S values. All the available T/S

profiles are objectively gridded. Kessler and McCreary

(1993) method is used for objective analysis of the Argo

T/S profiles on to regular grids (1˚X1˚). Details of this

method is given in (Udaya Bhaskar et al., 2007). Bad

T/S profiles appear as bull’s eye like structure in the

objectively analysed output. Figure 2 shows a typical

objective analysis output where the bad temperature

profile is clearly shown as a bull’s eye (within the circle

pointed by arrow). Once the bad profiles are pinpointed

by the objective analysis, they can be visually checked

for assigning correct quality flags using visual quality

control (VQC) tool.

2.3 Visual Quality Control:

All the profiles which appear as bull eyes in the

objective analysis product (described above) are isolated

and subjected to further visual treatment. In this method,

all the Argo T/S profiles that are identified to be

appearing as bull's eye are segregated and visually

checked for their quality using a VQC tool.

Geographical position and month of the observation

under scanner (appearing as bulls eye) is used to obtain

climatological mean T/S profile and associated standard

deviations from world ocean atlas 2001 (WOA01)

climatology (Conkright et al., 2002). The profile under

examination is then overlaid on this climatological

profile and the quality flags are corrected.

This VQC system was developed in house and is being

used for quality control of CTD data. Udaya Bhaskar et

al., (2012) describes the details of the VQC system.

Argo profiles which are deviating beyond 2 standard

deviations from the mean profile obtained from the

WOA01 are assigned a bad flag by the visual operator.

Figure 3 shows a typical case of T/S profiles which are

checked visually for the correctness of the quality flags

assigned. For instance one can observe from figure 3

1512 A Note on Three Way Quality Control of Argo Temperature and Salinity Profiles

- A Semi-Automated Approach at INCOIS



that, T/S records with in the ellipses are incorrectly

assigned bad flags. Using the VQC system one can

thoroughly check the profiles for their quality, visually

and flags can be corrected. At the end of this exhaustive

quality process, the good quality T/S profiles are

utilized in generation of objectively analysed gridded

products with a spatial resolution of 1° x 1° and

temporal resolution of 10 days and month. Various

value added products viz., Heat Content, Mixed Layer

Depth, D20, Geostrophic Currents etc. are derived using

this gridded product and are also made available on

INCOS web site. Details about gridding procedure and

derived value added products can be obtained in Udaya

Bhaskar et al., (2007). The objectively analysed gridded

product for the period 2002 to 2012 is made available

on INCOIS Live Access Server (www.las.incois.gov.in)

in Network Common Data Format (NetCDF) format.

3. Discussion and Summary:

Quality control is a important part of ocean data

assimilation system. As the automated QC of Argo T/S

profiles are sometimes susceptible to errors, a system

consisting of additional quality control procedures was

developed and is put to use at INCOIS. In this system

all the profiles obtained from Argo profiling floats are

passed through three mode of QC procedures viz.,

RTQC, objective analysis and VQC. Compared to the

purely automated system, this systems requires a

manual intervention (particularly for visual inspection)

preferably by experts in the field of oceanography hence

the system is deemed as semi-automatic. Overall, it

appears to be performing well although there is scope

for improvement in any QC system. At the end of the

exhaustive QC process objectively analysed gridded

product on 10 day and monthly scale is generated and

made available on INCOIS Live Access Server in

NetCDF format.

Acknowledgments:

Autors are grateful to Dr. SSC Shenoi, Director,

INCOIS for his encouragement and providing the

facilities to carry out the work. Authors are also grateful

to Dr Christine Coatanoan and Thierry Carval of

Coriolis/IFREMER for their useful discussion on the

VQC system. Argo data is made freely available by the

Argo community. Authors thank the anonymous

reviewers for their constructive comments which helped

in improving the quality of the manuscript. This is

INCOIS contribution number 132.

References:

[1] Argo Science Team, 2001. The Global Array of

Profiling Floats, in observing Oceans in 21st

century. C.Z. Koblinsky and N.R. Smith, eds.,

Godae Proj. Off., Bur. Meteorol., Melbourne,

Australia, (2001) 248 – 258.

[2] Ravichandran M., P. N. Vinaychandran, S. Joseph

and K. Radhakrishnan, 2004. Result from the first

Argo float deployed by India, Current. Science, Vol

86, 651 – 659.

[3] Conkright, M.E., R. A. Locarnini, H.E. Garcia,

T.D. O’Brien, T.P. Boyer, C. Stephens, J.I.

Antonov, 2002. World Ocean Atlas 2001:

Objective Analyses, Data Statistics, and Figures,

CD-ROM Documentation. National Oceanographic

Data Center, Silver Spring, MD, 17 pp.

[4] Ohno, Y., N. Iwasaka, F. Kobashi, Y. Sato, 2009.

Mixed Layer depth climatology of the North Pacific

based on the Argo observations, Journal of

Oceanogaphy, Vol 65, 1 - 16.

[5] Bruce, I., M. Hudleston, 2007. Quality control of

ocean temperature and salinity profiles - Historical

and real-time data, Journal of Marine systems, Vol

65, 158 - 175.

[6] Levitus, S., 1982. Climatological Atlas of the

World Ocean. NOAA Professional Paper, Vol. 13.

U.S. Government Printing Office, Washington,

D.C.

[7] Wong, A., R. Keeley, T. Carval and the Argo Data

Management Team, 2012. Argo quality control

manual, Ver. 2.7, Report, 47 pp.

[8] Kessler, W.S., J.P. McCreary, 1993. The Annual

Wind-driven Rossby Wave in the Subthermocline

Equatorial Pacific, Journal of Physical

Oceanography, Vol 23, 1192 -1207.

[9] Udaya Bhaskar, T.V.S., M Ravichandran, R.

Devender, 2007. An operational objective analysis

system at INCOIS for generation of Argo value

added products, Tech. Rept, INCOIS-MOG-

ARGO-TR-04-2007.

[10] Udaya Bhaskar, T.V.S., E. Pattabhi Rama Rao, R.

Venkat Shesu, R. Devender, 2012. GUI based

interactive system for quality control of Argo data,

Tech. Rept, INCOIS-DMG-TR-2012-02.

1511 T. V. S. UDAYA BHASKAR, E. PATTABHI RAMA RAO,

R. VENKAT SHESU and R. DEVENDER



Table 1: Real Time Quality Control (RTQC) Tests as Prescribed by the Argo Data Management Teal

(Wong et al., 2012)

1. Platform identification 2. Impossible date test

3. Impossible location test 4. Position on land test

5. Impossible speed test 6. Global range test

7. Regional range test 8. Pressure increasing test

9. Spike test 10. Top and bottom spike test

11. Gradient test 12. Digit rollover test

13. Stuck value test 14. Density inversion

15. Grey list 16. Gross salinity or temperature sensor drift

17. Frozen profile test 18. Deepest pressure test

Figure 1: Flow Diagram Showing the Modules of Three Way Quality Control System Implemented

at INCOIS for Quality Control of Temperature and Salinity Data Obtained from Argo Floats

1512 A Note on Three Way Quality Control of Argo Temperature and Salinity Profiles

- A Semi-Automated Approach at INCOIS



Figure 2: Error Profiles Appearing as Bulls Eye in the Objectively Analysed Product.

The Error Profiles are Shown with in the Circle Pointed by Arrow

Figure 3: Typical Example of Temperature and Salinity Profiles with Flags Assigned by Real Time Quality Control.

Blue Dots are Argo Observations and Green Thick Lines are Climatological means obtained from Woa01. Red Dots

are Records under Observation for their Quality.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1515-1521


Peak Ground Acceleration on Bedrock and Uniform Seismic Hazard

Spectra for Different Regions of Zanjan, Iran

GHOLAMREZA GHODRATI AMIRI1, SEYED ALI RAZAVIAN AMREI

2, ALI SABZEVARI

3,

ARMAN SAED4 and YASER SHOKRANI

5

1Center of Excellence for Fundamental Studies in Structural Engineering, School of Civil Engineering Iran, University of Science

& Technology, Tehran, Iran 2Department of Civil Engineering, Payame Noor University, Tehran, Iran

3Department of Civil Engineering, Islamic Azad University, Shahrekord Branch, Islamic Azad University 4Department of Civil, Behbahan Branch, Islamic Azad University, Behbahan, Iran

5School of Civil Engineering Iran, University of Science & Technology, Tehran, Iran

Email: [email protected], [email protected], [email protected], [email protected],

[email protected]

Abstract: The present paper was done under the title of peak ground acceleration (PGA) on bedrock and uniform

seismic hazard spectra (UHS) for different regions of Zanjan. A set of seismic sources, historical, and instrumental

seismic were used by means of data regarding the time since 1678 until 2011 with a radius of 150 km. Kijko

program (2000) was applied for calculation of seismic parameters considering lack of suitable seismic data and

uncertainty of magnitude in different periods. The calculations were performed by using the logic tree method. Four

weighted attenuation relationship were used; including, Ramazi (1999), 0.15, Campbell (1997), 0.30, Ambraseys et

al (1996) ,0.15& Ghodrati Amiri et al (2007), 0.40. Furthermore, Ambraseys et al (1996), 0.25, Ghodrati Amiri et al

(2010), 0.45 & Campbell (1997), 0.30 were used in order to determine the acceleration spectra based on weighted

attenuation spectral relations, and also to be spectral and more suitable with conditions of the zone in question. The

SEISRISK III software (1987) was used to calculate the earthquake hazard. The results of this analysis were

submitted including the spectra and maps for 2% and 10% probability in 50 years.

Keywords: Seismic Hazard Analysis, PGA, Uniform Seismic Hazard Spectra, Zanjan, Iran.

1. Introduction:

The city of Zanjan is one of densely populated cities

located in the north west of Iran. With a population of

more than 349,800, Zanjan is a main connecting road

between many important cities. In addition, it contains

many important industrial centers. Given the past

history of earthquakes and faults in the range of likely

event, another earthquake is inevitable. Due to the lack

of detailed engineering design of structures especially in

older structures if a large earthquake happens in this

city, a terrible tragedy will be seen. Probabilistic

analysis is one of modern methods in analyzing seismic

hazard in which the uncertainty in different parameters

is considered and the results are presented logically. In

this study, it was sought to accurately identify both local

area faults and numerous attenuation relations suitable

for the region to ensure more reliable results [1].

2. Seismotectonics:

Due to active faults in the surrounding areas, Zanjan is

among active seismic places. Throughout the present

study, in order to evaluate the seismic hazard in the

region, all sources of possible earthquakes and their

ability to generate strong ground movement have been

collected. A list of critical faults in the range of 150 km

is given in Table 1 [1].

Table 1: Main Faults within 150 km Radius of Zanjan

No. Faults

1 Zanjan

2 Soltaneyeh

3 Lahbari

4 Roudbar

5 Gazvin North

6 Manjil

7 Takhte Solayman

8 Fouman

9 Tarom

In Figure 1 some of the faults which were studied in the

area were shown.

3. The Peak Earthquake Magnitude and Fault

Rupture Length Appendix:

To estimate the relationship between the peak expected

magnitude and fault length, seismotectonic and

geotectonic behaviors of the concerned area were taken

1516 GHOLAMREZA GHODRATI AMIRI, SEYED ALI RAZAVIAN AMREI,

ALI SABZEVARI, ARMAN SAED and YASER SHOKRANI



into consideration and the following relation was used

[2].

LogLMs 244.1259.1 += (1)

Where, In Eq. (1), Ms is the surface magnitude and L is

the rupture length in meter.

Figure 1: Main Faults within 150 km Radius of the Zanjan City [3]

4. Seismicity The history of past earthquakes in each width is an

indication of the seismicity of that area. Thus, in order

to conceive the seismicity features, we should have a

comprehensive list of occurred earthquakes in the area.

In the present work, a number of earthquakes in a radius

of 150 km of Zanjan were collected and considered.

4.1 Historical earthquakes

The historical documents indicate that the historical

earthquakes occurred before 1900. As far as the

collected information from historical books was

concerned, their validity may be under question because

they may have exaggerated the extent of the damage

and destruction is in excess negligence. However, the

existence of such places could be important in the

process of gathering information.

4.2 Instrumental Earthquakes

In spite of the uncertainties in estimating the epicenter,

focal depth, and magnitude of earthquakes in seismic

data in twentieth century, these earthquakes are crucial

with regard to the instrumental registration. From 1963,

with the installation of seismography network, the

uncertainties in their estimations were prominently

decreased. A list of instrumental earthquakes in Zanjan

from 1900 to the present has been collected, the most

important of which is the website of international

seismological center [3].

4.3 Earthquake magnitude

In the present study, the surface-wave magnitude, Ms,

was used in order to analyze the seismic hazard

magnitude. As far as, the collected magnitudes were not

of Ms type, they were converted to Ms. Thus, in order to

convert the wave magnitude and Local magnitude to

Ms, Table 2 [1] was employed. Moreover, in order to

convert Mb to Ms, equation 2 was used [15].

MS = 1.21*Mb - 1.29 )2(

Where, Ms and Mb stand for the surface-wave

magnitude and the body-wave magnitude respectively.

Table 2: Magnitude Convert [1]

Ms Mw ML

3.6 4.5 4.8

4.6 5.2 5.3

5.6 5.8 5.8

6.6 6.6 6.3

7.3 7.3 6.8

5. Seismicity Parameters of Zanjan:

Seismicity parameters or the peak expected magnitude,

Mmax, λ, and β are among the basics of the seismicity

of a place. They are used to indicate the seismicity of a

place. Collecting earthquake data for Zanjan according

to the fundamental assumption in estimating Seismicity

parameters, Filtered data was evaluated in Poisson

distribution. The method recommended used for the

elimination of foreshocks and aftershocks are the

variable windowing method in time and space domains

Gardner and Knopoff [5].

5.1 Determination of Seismicity Parameters:

In this paper, in order to estimate the seismic parameters

due to the shortage of appropriate seismic data and the

uncertainty of earthquake magnitude, Kijko method [6]

was used based on the probabilistic method of peak

likelihood estimation. In this method, according to the

faults of Seismic data and the low accuracy at different

times, their occurrence in determination of seismicity

parameters Mmax, λ, and β are used. The results of

applying this method includes determination of the

seismic parameters (Table 3), the return period,

probability of event and the magnitude of seismic events

at different times. In Figure 2 estimation of the return

period of earthquakes in Zanjan (Kijko method) is

presented.

Table 3: values of Seismic Parameters in the range of

150 km Zanjan (Kijko method) [6].

Results

Beta 1.61 + - 0.11 (b=0.70 -+0.05)

Lambda 3.5 For Mmin=3

1517 Peak Ground Acceleration on Bedrock and Uniform Seismic Hazard




Figure 2: Estimation of the Return Period of

Earthquakes in Zanjan (Kijko method)

6. Seismic Hazard Analysis:

The present paper aims at providing Peak Ground

Acceleration (PGA) on bedrock and uniform seismic

hazard spectra (UHS) for different regions of Zanjan

using Probabilistic Seismic Hazard Analysis.

In this method, the procedure starts with identifying and

modeling the distribution of seismic sources, evaluation

of recurrence relationship, evaluation of local site

effects such as soil types, estimation of activity rate for

probable earthquakes, evaluation of attenuation

relationships for peak ground acceleration, geotechnical

characteristics of sediments, topographic effects

resources and the probabilistic analysis of the risk of

earthquakes the likely location of earthquakes causing

them the determination of the acceleration spectra on

bedrock has been used.

6.1 Attenuation Relationship:

Of an appropriate attenuation relationship is of high

importance in the reliability of the results taken from

seismic hazard assessment. Throughout this process, the

following points are to be taken into consideration. The

needed points are as follow, source specifications,

direction of the wave propagation, geology and

topography effects of the site, magnitude, refraction and

energy absorption due to the properties of the material

through which the waves pass, fault mechanism,

reflection, and distance from seismic source. Knowing

about just above mentioned points, here are three

different attenuation relationships, Ghodrati Amiri et al.

[7], Ambraseys et al. [8], Campbell [9] and Ramazi

[10], using the logic-tree method with the weighs of 0.4,

0.15, 0.30 & 0.15 respectively, used in the process of

providing Peak Ground Acceleration (PGA) on bedrock

and Uniform Seismic Hazard Spectra (UHS). Moreover,

in order to provide spectra Acceleration map and

uniform seismic hazard spectra, Ghodrati Amiri et al.

[11], Ambraseys et al. [8] and Campbell [9] using logic-

tree method with the weighs of 0.45, 0.25, and 0.30

respectively.

The reason for using the Logic-tree method is that using

a single attenuation relationship is not an appropriate

choice because the uncertainty of given data is not as

reliable as desired. Moreover, the local and global

relationships which enjoy a higher accuracy in

comparison with those of Iran, the other countries’ data

are used in the provision of their model. Therefore, as a

logical conclusion, the best method is the use of both

different attenuation relationships together with the

Logic-tree. Performing in this way, each one

compensate for the other one’s shortage. There are two

parameters in assigning the weigh to the branches of

each Logic-tree, including soil conditions in the given

site and considering higher effect of local relationship.

In Figures 3 and 4 the used Logic-trees with the weight

of each branch are indicated.


of Each Category for Determination of PGA


of Each Category for Determination of UHS





Figure 5: Zoning Maps of PGA with 10% and 2% PE in 50 years (Left to Right), and the Border of Zanjan (Thick

Line)


2, and the Border of Zanjan (Thick Line)









Figure 9: Zoning Maps of 0.5 S Spectral Acceleration with 10% and 2% PE in 50 years (Left to Right) in Soil Type


Figure 10: Zoning Maps of 1.0 S Spectral Acceleration with 10% and 2% PE in 50 years (Left to Right) in soil type

2, and the border of Zanjan (Thick Line)





Figure 11: Zoning Maps of 1.0 S Spectral Acceleration With 10% and 2% PE in 50 Years (Left To Right) in Soil

Type 3, and the Border of Zanjan (Thick Line)



6.2 Probabilistic Seismic Hazard Analysis:

In this part, based on the modeled seismic sources,

seismic parameters, and SEISRISK III software [12],

the Peak Horizontal Acceleration on bedrock (PGA) and

horizontal spectral acceleration, each with 10% and 2%

probability event in 50 years (equivalent to a return

period of 475 and 2475 years) in accordance with the

levels of 1 and 2 of Seismic Rehabilitation of Existing

Building [13], for a 7x11 network where surround





Zanjan appropriately, were estimated. As far as the type

of soil is unknown, in each place of the network, exact

calculations for soil type 2 and 3 (the most probable soil

type in the area) based on Iranian Code of Practice for

Seismic Resistant Design of Buildings [14] were done.

The maps for peak ground acceleration on bedrock

(PGA) in Figure 5 and horizontal spectral acceleration

based upon the two soil types of Zanjan city for the 0.3,

0.5, 1 periods are presented in Figures 5 to 11.

7. Uniform Hazard Spectra (UHS):

The uniform hazard spectra (UHS) are formed in the

form of a response in which any time, there is the same

probability range for its occurrence. Throughout these

spectra, within all periods in the life of the structure, the

probability of occurrence is considered as the same. In

other words, in designing a structure, the return period

of spectral acceleration for different periods is

considered the same. For this purpose, in any part of the

network with given seismic hazard, this range is

obtained for different periods.

In Figures 12 and 13, the uniform hazard spectra for the

two soil types and risk levels 1 and 2 based on Seismic

Rehabilitation of Existing Building for Zanjan are

presented. The spectra in these figures are presented as

the peak, minimum and average values of spectral

acceleration at different points in the range of the

network

8. Conclusion:

1- PGA with the probability event, 10% in 50 years (the

period to 475 years or risk level 1) in the range of

Zanjan varies from 0.23g to g 0.295g, also, this amount

in Iranian Code of Practice for Seismic Resistant Design

of Buildings is presented as 0.3g.

2- PGA with the probability event, 2% in 50 years (the

period to 2475 years or risk level 2) in the range of

Zanjan varies from 0.47g to 0.68g.

3- Acceleration and spectral acceleration values in the

northeast region have the highest value in comparison to

other regions.


soil type 2 in the range of Zanjan varies from 0.75g to

1.8g in the risk level 1.


soil type 3 in the range of Zanjan varies from 0.78g to

2.2 g in the risk level 1

9. References:

[1] A. Sabzevari, “Peak Ground Acceleration (PGA)

on bedrock and uniform seismic hazard spectra for

different regions of Zanjan city”, MA thesis, Azad

University of Shahrekord, Supervised by Professor

Gh. Ghodrati Amiri, and Dr. S.A. Razavian Amrei,

2011.

[2] A. Nowroozi, “Empirical relations between

magnitude and fault parameters for earthquakes in

Iran”, Bulletin of the Seismological Society of

America ,Vol. 75, No. 5, pp. 1327-1338, 1985.

[3] International Institute of Earthquake Engineering

and Seismology website: http://www.iiees.ac.ir

[4] N.N. Ambraseys, and C.P. Melville, A History of

Persian Earthquakes, Cambridge University Press,

Cambridge, Britain, 1982.

[5] J.K. Gardaner, L. Knopoff, “Is the sequence of

earthquake in southern California, with aftershocks

removed, poissonian?”, Bulletin of the

Seismological Society of America ,Vol. 64, No. 5,

pp. 1363-1367, 1974.

[6] A. Kijko, “Statical estimation of maximum

regional earthquake magnitude Mmax”, Workshop

of Seismicity Modeling in Seismic Hazard

Mapping, poljce, Slovenia, May, 22-24, 2000

[7] G. Ghodrati Amiri, A. Mahdavian, F. Manouchehri

Dana, “Attenuation Relationship for Iran”, Journal

of Earthquake Engineering, Vol. 11, Issue 4, pp.

469-492, 2007.

[8] N.N. Ambraseys, K.A. Simpson, J.J. Bommer,

“Prediction of Horizontal Response Spectra in

Europe”, Earthquake Engineering and Structural

Daynamics, Vol. 25, pp. 371-400, 1996.

[9] K.W. Campbell, “Empirical near-source

attenuation relationships for horizontal and vertical

components of peak ground acceleration, peak

ground velocity, and pseudo-absolute acceleration

response spectra”, Seismological Research Letters,

Vol. 68, No. 1, pp. 154–179, 1997.

[10] H.R. Ramazi, “Attenuation laws of Iranian

earthquakes”, proceedings of the 3rd International

Conference on Seismology and Earthquake

Engineering, Tehran, Iran, 1999.

[11] G. Ghodrati Amiri, M. Khorasani, R. Mirza Hesabi,

and S.A .Razavian Amrei, “Ground-Motion

Prediction Equations of Spectral ordinates and

Arias Intensity for Iran”, Journal of Earthquake

Engineering, Vol. 14, Issue 1, pp. 1-29, 2010.

[12] B. Bender, D.M. Perkins, “SEISRISK-ІІІ: A

computer program for seismic hazard estimation”,

US Geological Survey, Bulletin 1772, 1987.

[13] IIEES. Seismic Rehabilitation Code for Existing

Buildings in Iran, International Institute of

Earthquake Engineering and Seismology, Tehran,

Iran, 2002.

[14] BHRC. Iranian Code of Practice for Seismic

Resistant Design of Building, Standard No. 2800,

Third Revision, Building and Housing Research

Center, Tehran, Iran, 2005.

[15] “Relationship between fault length and maximum

expected magnitude”, Iranian Committee of Large

Dams (IRCOLD), Internal Report, 1994 (In

Persian).

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1522-1529


Heavy Minerals and Provenance of the Lower Gondwana

Sandstones, Ong-River Gondwana Basin, Odisha

ARYA GOUTAMI NATH and RABINDRA NATH HOTA

Department of Geology, Utkal University, Vani Vihar, Bhubaneswar-751004, Odisha


Abstract: Heavy mineral analysis of the Talchir and Karharbari sandstones of the Ong-river Gondwana basin,

Odisha, has been carried out in the present study. The characteristic transparent heavy minerals of the Talchir

Formation are garnet, biotite, brown tourmaline, sillimanite, apatite, zircon, enstatite, hypersthene, epidote, rutile,

chlorite, staurolite and sphene. The Karharbari sandstones, in addition to the above mentioned heavy minerals,

contain brookite, hornblende, andalusite, monazite, spinel and pink tourmaline. These heavy minerals are

comparable with the heavy accessories of the Eastern Ghats Supergroup rocks and suggest their derivation from

these rocks. Variation of heavy mineral frequencies can be ascribed to prolonged and deep weathering in the source

area, climate change and/or prevalence of geochemical conditions that prohibited dissolution.

Keywords: Heavy mineral, Talchir Formation, Karharbari Formation, Ong-river Gondwana basin.

Introduction:

Determination of provenance and its different aspects

like its position with respect to the depositional basin,

lithology, prevailing climate and tectonic setting are

some of the significant parameters of basin analysis.

Location of the source area can be inferred from the

scalar and vector attributes of the sediments while the

source rock lithology, climate and tectonic setting can

be determined from the composition of the sediments.

Common minerals like quartz and feldspars, which

constitute bulk of the sediment, are contributed by an

array of rocks whereas the uncommon heavy

accessories provide valuable clue to ascertain the source

rock lithology precisely. The study of heavy minerals

has been used to decipher the provenance lithology both

in unconsolidated and consolidated sediments of

different ages and countries (Tewari and Trivedi, 2001;

Mishra and Tiwari, 2005 etc).

Out of the eight Gondwana basins of Odisha (Fig. 1),

the Talchir and Ib-river basins have been studied in

detail for their economic potential of coal. Little

attention has been paid to the remaining basins. In the

past few years, Hota et al. (2008a, b and 2010) have

studied the texture, palaeocurrent, palaeohydrology and

cyclic aspect of the Talchir and Karharbari formations

of the Ong-river basin. The present study deals with the

heavy mineral analysis of the Talchir and Karharbari

sandstones of the Ong-river Gondwana basin with a

view to deduce the source rock lithology and prevailing

climatic condition.

Geological Setting:

The Lower Gondwana rocks of the Ong-river basin

occur in the Bargarh, Balangir and Sonapur districts of

Orissa extending over an area of about 2 – 6 km wide

and 50 km long (Fig. 1). The rocks of the Eastern Ghats

Supergroup composed of khondalite, charnockite,

garnetiferous granite gneiss, biotite schist, quartzite etc.

intruded by pegmatite, quartz vein and anorthosite

forms the basement (Pandya, 2006). The foliations of

metamorphic rocks strike N80ºE – S80ºW to E – W and

dip towards south at high angle ranging from 40º to 70º.

The strike of the Gondwana sedimentary rocks, on the

other hand, vary from E – W to NE – SW and they show

variable dip ranging from 5º to 25º due north to

northwest. The stratigraphic sequence based on surface

and subsurface studies is given in Table 1.

The Talchir Formation is dominated by coarse clastics.

It is composed of diamictite, matrix-supported

conglomerate, graded conglomerate, massive sandstone,

very coarse- to medium-grained flat- and cross-bedded

sandstones and interbedded sandstone-shale formed in

glacial, glacio-fluvial and fluvial environments (Hota et

al. 2010). Sandstones are the dominant constituents of

the Karharbari Formation. Various lithic units are thin

conglomerate, trough cross-bedded very coarse- to fine-

grained sandstones, interbedded sandstone-shale, gray

shale and interbedded carbonaceous shale-shaly coal-

coal. These lithofacies are characteristics of channel,

point bar, flood plain and swamp deposits of the fluvial

environment.

1523 ARYA GOUTAMI NATH and RABINDRA NATH HOTA



Methodology:

Forty three sandstone samples, 13 from the Talchir

Formation and 30 from the Karharbari Formation were

collected from exposures. These were disaggregated by

a porcelain mortar and pestle and sieved. The heavy

minerals were separated from 0.125 – 0.088 mm (fine-

to very fine-grained) sand fraction by standard

technique using bromoform as the heavy liquid.

Separated heavy minerals were washed with acetone

and stannus chloride and mounted on glass slide with

Canada balsam. The minerals were identified under

microscope by their diagnostic optical characters and

their number frequencies were determined by field

count method following the procedure outlined by Hota

(2011). Parameters like ZTR, shape and density indices

were calculated following Hubert (1962) and Flores and

Schideler (1978). The ZTR index is the percentage of

zircon, tourmaline and rutile among the transparent

nonmicaceous detrital heavy minerals. It is a measure of

the mineralogical maturity of the sediment. The shape

index is the ratio of bladed and elongated minerals like

pyroxene, hornblende, tourmaline, rutile, zircon,

sillimanite vs. equant minerals like garnet and epidote.

Density index is the ratio of opaque and transparent

heavies.

Heavy Minerals:

Both opaque and transparent heavy minerals constitute

the heavy mineral crop of the Gondwana sandstones.

The opaque minerals, which could not be identified

precisely in the present study, are mostly oxides and

hydroxides of iron like magnetite, hematite, limonite,

goethite, ilmenite etc. In the present study, they have

been categorized as rounded (Fig. 2a) and elongated

(Fig. 2b). The elongated minerals are more likely to be

ilmenite. Many of the ilmenite grains have been altered

to leucoxene, which are identified by their dead white

colour. In some instances they are associated with

unaltered ilmenite (Fig. 2c). Since it is an altered

product of ilmenite, it has been treated at par with

opaque minerals in the present study. The opaque

minerals (51%) are nearly equal to the transparent heavy

minerals (49%) in case of the Talchir Formation while

in case of the Karharbari Formation the opaque minerals

(63%) dominate over transparent heavies (37%).

The transparent heavy minerals of the Talchir

Formation are garnet, biotite, brown tourmaline,

sillimanite, apatite, zircon, pyroxene, epidote, rutile,

chlorite, staurolite and sphene. The Karharbari

Formation, in addition to these minerals, contains

brookite, hornblende, andalusite, monazite, spinel and

pink tourmaline. The characteristic optical properties of

these heavy minerals are as follows:

Garnet: Garnet is the dominant heavy mineral of the

sandstones. On the basis of colour, two types of garnets

have been recognised. These are colourless (Fig. 2d)

and pink (Fig. 2e). The grains are angular to subrounded

and show conchoidal fractures, inclusions are frequently

present and in a few instances the surfaces are pitted.

These are isotropic.

Biotite: Detrital biotite grains occur as brown flakes

(Fig. 2f) and are strongly pleochroic. The basal sections

are isotropic and alteration is a common feature.

Zircon: Zircon occurs as elongate, oval to rounded

grains, identified by high relief, dark border and parallel

extinction. Colourless (Fig. 2g) and pink (Fig. 2h)

varieties have been reported. Inclusions of other

minerals, liquid and gas are frequently present.

Sillimanite: Sillimanite occurs as non-pleochroic,

colourless, needle to long prism shaped grain marked by

striations parallel to length, high order interference

colour, straight extinction and negative sign of

elongation (Fig. 2i).

Apatite: Grains are oval and elongate prismatic,

colourless, low relief, negative sign of elongation,

parallel extinction and show first order interference

colour (Fig. 2j). Some grains contain abundant

inclusions arranged in rows.

Tourmaline: On the basis of colour, two types of

tourmalines have been identified. These are brown and

pink. The brown tourmaline (Fig.2k) is pleochroic from

yellowish brown to dark brown and pink variety (Fig.

2l) is pleochroic from pink to brownish pink. They

occur as short prismatic grains and a few of them

contain inclusions. The grains show straight extinction

and negative sign of elongation. The brown tourmaline

is more abundant in the Talchir Formation whereas the

pink variety has been reported from the Karharbari

Formation only.

Pyroxene: Two varieties of pyroxenes, viz. enstatite

and hypersthene have been recorded. The enstatite is

identified by its colourless, nonpleochroic nature, first

order interference colour, positive sign of elongation

and straight extinction. Some grains appear slightly

brown due to iron incrustation (Fig. 2m). Hypersthene

occurs as elongate to irregular grains, greenish in

colour, pleochroic from green to pink, presence of

cleavage, straight extinction and positive sign of

elongation (Fig. 2n).

Epidote: The grains are subrounded, weakly pleochroic

from pale green to lemon yellow (Fig. 2o) and show

straight extinction.

Rutile: Rutile is characterised by deep red colour, broad

and thick boarder around the grain. A few are sub-

rounded but most of them are elongated prismatic (Fig.

1524 Heavy Minerals and Provenance of the Lower Gondwana Sandstones,

Ong-River Gondwana Basin, Odisha



2p) and show the same colour under crossed nicols as in

ordinary light.

Chlorite: Chlorite is marked by pale green colour and

flat irregular grain of micaceous habit (Fig. 2q). Grains

show blotchy ultrablue abnormal interference colour.

Staurolite: Occurs as prismatic grain of golden yellow

colour (Fig.2r), marked by hackly to subconchoidal

fracture, pleochroic from straw yellow to russet brown,

bright yellow interference colour and straight extinction.

Many grains contain profuse inclusions.

Sphene: It occurs as subrounded grain of high refractive

index, colourless to light yellow, show same colour

under crossed nicols as in polarised light, incomplete

extinction and turn bluish near the extinction position

(Fig. 2s).

Brookite: The grains are squarish with truncated

corners, striated, orange to brown in colour and show

incomplete extinction (Fig. 2t). It occurs only in the

Karharbari sandstones.

Hornblende: It occurs as subrounded grain, bluish in

colour, pleochroic from greenish blue to bluish green

(Fig. 2u) having two sets of cleavage and extinction

angle of 15° - 25°.

Andalusite: The grains are nonpleochroic or pleochroic

from colourless to pink, show first order interference

colour, negative sign of elongation and straight

extinction (Fig.2v).

Monazite: The grains are rounded or oval in habit,

yellow to greenish yellow in colour, show the same

interference colour as under plane polarized light and

low angle of extinction varying from 2°-10° (Fig.2w).

Spinel: Grains are grass green in colour, irregular to

subrounded, marked by conchoidal fracture and

isotropic (Fig. 2x). It occurs in the Karharbari

sandstones only.

Provenance:

The frequency percentages of the transparent heavy

minerals of the Talchir and Karharbari formations along

with probable source rocks is presented in Table 2 and

graphically shown in Fig. 3. Garnet and biotite are the

most abundant heavy minerals of both the formations.

The transparent heavy crop of the Talchir Formation

consists of garnet (50.19%), biotite (20.85%), brown

tourmaline (12.02%), sillimanite (6.59%), apatite

(3.81%), zircon (2.04%), pyroxene (1.67%), epidote

(1.30%), rutile (0.80%), chlorite (0.60%), staurolite

(0.09%) and sphene (0.04%). The transparent heavy

mineral assemblage of the Karharbari Formation

comprises garnet (50.45%), biotite (15.55%), brown

tourmaline (1.05%), sillimanite (7.67%), apatite

(1.88%), zircon (11.40%), pyroxene (1.15%), epidote

(1.17%), rutile (2.09%), chlorite (2.60%), staurolite

(1.04%), sphene (1.95%), brookite (1.07%), hornblende

(0.37%), andalusite (0.35%), monazite (0.10%), spinel

(0.08%) and pink tourmaline (0.03%). The ZTR and

shape indices of heavy minerals of both the formations

are nearly same. In case of the Talchir Formation, the

ZTR index is mostly due to presence of brown

tourmaline, while in case of the Karharbari Formation, it

is due to higher proportions of zircon and rutile. Higher

density index of the Karharbari sandstones is due to the

presence of higher percentage of opaque minerals.

These heavy minerals can be linked with four types of

lithounits viz. acid and basic igneous and low- and high-

rank metamorphic rocks. Apatite, zircon, sphene,

monazite and pink tourmaline suggest the acid igneous

source. Basic igneous parentage is indicated by

pyroxene, rutile, brookite and spinel. Minerals like

biotite, chlorite and brown tourmaline are derived from

low-rank metamorphic rocks, whereas high-rank

metamorphic source is suggested by garnet, hornblende,

sillimanite, staurolite, epidote and andalusite. Though

these minerals represent diverse source rocks and of

variable mineral stability, their close association may be

due to high stability and preservation potential of certain

species and equal degree of mineral stability in case of

others (Pettijohn, 1984). Presence of unstable species

like pyroxene and sillimanite in appreciable quantity,

which would otherwise have been eliminated, may be

due to their abundance in the source rocks like

khondalite, basic granulite and charnockite, which

constitute bulk of the Eastern Ghats Supergroup and/or

existence of favourable geochemical conditions, which

inhibited complete dissolution. These observations

suggest deep weathering in the source area with

exposure of different types of rocks and favourable

hydrodynamic and geochemical conditions for

attainment of mineralogical maturity of the heavy

minerals as revealed by relatively higher ZTR index.

The palaeocurrent pattern of the Ong-river Gondwana

sediments indicates a predominantly northerly

palaeoflow and existence of the source area towards the

southern boundary of the basin (Hota et al. 2008b). The

southern margin of the Gondwana basin is bounded by

the Eastern Ghats Supergroup of rocks of Precambrian

age. These are composed of metamorphic rocks like

khondalite, basic granulite, charnockite, mica schist,

leptynite, quartzite, acid-gneiss and pyroxene granulite

intruded by granitic plutons and basic igneous rocks.

The observed heavy mineral suites are comparable with

heavy accessories of the Eastern Ghats Supergroup of

rocks and suggest their derivation form these rocks.

Euhedral and less amount of rounding of many heavy

mineral grains suggest short distance of transportation

and presence of the source rock in the vicinity of the

depositional basin. Large number of heavy minerals in




Karharbari sandstones in comparison to Talchir

sediments can be ascribed to exposure of diverse rock

types in the source area caused by deep chemical

weathering in humid climatic condition. The gradual

changeover of climate from cold glacial during Talchir

sedimentation to warmer humid climatic condition

during Karharbari sedimentation is attested by increase

of the percentage of opaque minerals from 51% to 63%.

Conclusions:

Following conclusions have been drawn in the present

study.

1. Garnet, biotite, zircon, sillimanite, apatite,

tourmaline, pyroxenes, epidote, rutile, chlorite,

staurolite, sphene, brookite, hornblende, andalusite,

monazite and spinel including opaques and leucoxene

constitute the heavy mineral suites of the Gondwana

sandstones of the Ong-river basin.

2. Garnet, sillimanite and biotite are the dominant

heavy minerals of both the formations. In addition to

these, the Talchir Formation contains appreciable

quantity of brown tourmaline, while the Karharbari

sandstones contain higher percentage of zircon.

3. Source area denudation, tectonism and action of

intrastratal solution played important roles to account

for the variation of the heavy mineral suites.

4. The observed heavy minerals are comparable with

the accessory constituents of khondalite, basic granulite,

charnockite, mica schist, leptynite, quartzite, acid-

gneiss, pyroxene granulite, granite, and basic igneous

rocks of the Eastern Ghats Supergroup. Proximity of the

source area is indicated by euhedral to subrounded

grains of most of the heavy minerals.

5. Progressive denudation, release of some heavy

minerals in the source region, climate change and/or

existence of favourable geochemical condition to escape

dissolution account for the variation of heavy mineral

frequencies of two formations of the Gondwana

Supergroup.

Acknowledgements:

The authors are thankful to the reviewers Prof. S. Das of

IIT, Kharagpur, India and Prof. W. Maejima of Osaka

City University, Japan for their critical and constructive

suggestions that enhanced the quality of the paper.

References:

[1] Flores, R. M. & Schideler, G. L. (1978). Factor

controlling heavy mineral variations on the South

Texas outer continental shelf gulf of Mexico.

Journal of Sedimentary Petrology, 48, 269-280.

[2] Goswami, S., Das, M. & Guru, B. C. (2006).

Permian biodiversity of Mahanadi master basin,

Orissa, India and their environmental countenance.

Acta Palaeobotanica, 46,101 – 118.

[3] Hota, R. N. (2011). Practical approach to petrology.

CBS publishers and Distributors, New Delhi, 150p.

[4] Hota, R. N., Dalabehera, L. & Nath, A. G. (2008a).

Textural characteristics and discriminant Analysis

of Talchir and Karharbari sandstones of the Ong-

River Gondwana Basin of Orissa. Journal of Indian

Association of Sedimentologists, 27 (No.1), 77 –

86.

[5] Hota, R. N., Maejima, W. & Nath, A. G. (2008b).

River transmutation during Lower Gondwana

sedimentation in Ong-river basin, Orissa. Journal of

Indian Association of Sedimentologists, 27 (No.2),

29 – 37.

[6] Hota, R. N., Nath, A. G. & Maejima, W. (2010).

Cyclic sedimentation of the Karharbari Formation,

Ong-River Gondwana basin, Orissa – Statistical

assessment from subsurface data. Journal of Indian

Association of Sedimentologists, 29, pp.1 – 12.

[7] Hubert, J. F. (1962) A zircon-tourmaline-rutile

maturity index and the interdependence of the

composition of heavy mineral assemblages with the

gross composition and texture of sandstones.

Journal of Sedimentary Petrology, 32, 440-450.

[8] Mishra, D. & Tiwari, R. N. (2005). Provenance

study of siliciclastic sediments, Jhura dome,

Kachchh, Gujrat. Journal of the Geological Society

of India, 65, 703-714.

[9] Pandya, K. L. (2006) Gondwanas. In: Mahalik, N.

K., Sahoo, H. K., Hota, R. N., Mishra, B. P.,

Nanda, J. K. & Panigrahi, A. B. (Eds) Geology and

Mineral Resources of Orissa. Society of

Geoscientists and Allied Technologists,

Bhubaneswar, pp. 91 – 103.

[10] Pettijohn, F. J. (1984). Sedimentary rocks, CBS,

New Delhi, 628p.

[11] Tewari, R.C. & Trivedi, G. S. (2001). Heavy

mineral assemblages viz-a-viz composition of

provenance of Gondwana rocks of peninsular India.

Indian Journal of Petroleum Geology, 10, 33-42.





Table 1: Stratigraphic Sequence of the Ong-River Basin, Orissa (Modified After Goswami et al. 2006

and Pandya, 2006)

Age Group Formation Dominant lithology

Recent to -

Sub-recent Soil, laterite, alluvium and gravel

Lower

Permian Lower

Gondwana

Group

Karharbari

Formation

Medium to very coarse grained sandstone with occasional

conglomerate, gray shale and coal stringer

Talchir

Formation Greenish sandstone and shale with diamictite at base

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Unconformity ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

Precambrian

Eastern

Ghats

Supergroup

Intrusives

Pegmatites, quartz vein and anorthosite

Metamorphics Khondalite, charnockite, garnetiferousgranite gneiss, biotite

schist and quartzite

Table 2: Average Transparent Heavy Mineral Content (In Number Percent) of the Gondwana Sandstones of the

Ong-River Basin and Possible Source Rock Types

Heavy Mineral Talchir

Formation

Karharbari

Formation Source rock type

Garnet 50.19 50.45 High rank metamorphic

Biotite 20.85 15.55 Low rank metamorphic, acid igneous

Tourmaline (brown) 12.02 1.05 Low rank metamorphic

Sillimanite 6.59 7.67 High rank metamorphic

Apatite 3.81 1.88 Acid igneous

Zircon 2.04 11.40 Acid igneous

Pyroxene 1.67 1.15 Basic igneous

Epidote 1.30 1.17 High rank metamorphic

Rutile 0.80 2.09 Basic igneous

Chlorite 0.60 2.60 Low rank metamorphic

Staurolite 0.09 1.04 High rank metamorphic

Sphene 0.04 1.95 Acid igneous

Brookite 0.00 1.07 Basic igneous

Hornblende 0.00 0.37 High rank metamorphic

Andalusite 0.00 0.35 High rank metamorphic

Monazite 0.00 0.10 Acid igneous

Spinel 0.00 0.08 Basic igneous

Tourmaline (pink) 0.00 0.03 Acid igneous

ZTR Index 23.33 21.66

Shape Index 0.32 0.31

Density Index 1.04 1.70




Figure 1: Gondwana Basins/ Outliers of Odisha (Modified after Pandya, 2006)





Figure 2: Photomicrographs of the Heavy Minerals. (a) Rounded Opaque, (b) Elongated opaque, (c) Leucoxene

with Partially Altered Ilmenite, (d) Colourless Garnet, (e) Pink Garnet, (f) Biotite, (g) Colourless Zircon, (h) Pink

Zircon, (i) Sillimanite, (j) Apatite, (k) Brown Tourmaline, (l) Pink Tourmaline, (m) Enstatite, (n) Hypersthene,

(o) Epidote, (p) Rutile, (q) Chlorite, (r) Staurolite, (s) Sphene, (t) Brookite, (u) Hornblende, (v) Andalusite,

(w) Monazite and (x) Spinel




Figure 3: Histogram Showing Transparent Heavy Mineral Frequency Percentages of the Lower Gondwana

Sandstones of the Ong-River Basin, Odisha

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1530-1534


Rating of Tunnel by Visual Field Inventory – A Case Study of

Punasa Tunnel, District Khandawa, M. P. India

GUPTA. M. C1, SINGH. B. K

2 and SINGH. K. N

3

1Water Resources Deptt. , Narmada-Tapti Area, (MP), India.

2Applied Geology, Deptt.of Civil Engg. Govt. Engg. College, Ujjain (MP) India 3School of Studies in Earth Science, Vikram University, Ujjain, (MP) India


Abstract: This paper proposes a new rating system for tunnel by visual field inventory method for Punasa tunnel a

part of Narmada Sagar Project. The new CRTVFI (Classification for Rating of Tunnel by Visual Field Inventory)

technique has been developed after modification in the methods proposed by Romana et al. (1983) considering other

geotechnical parameters given by Bieniawski, (1974, 1989) and Barton et al. (1974). The modified classification

has been used in the investigated area and paper focuses on design recommended on the basis of the classification

systems such as maximum span opening, lining and support recommendations. A very handsome amount of about

32.45% has been saved using CRTVFI classification against the proposed total cost of the tunnel. This classification

technique implied in the area, has not only saved money, but also saved precious time of construction.

Keywords: Tunnel rating, Tunnel stability, CRTVFI classification.

Introduction:

The rating of tunnel has been proposed by different

engineers and scientists from time to time for their

stability estimation stand up time and protection /

treatment measures. The various criteria for

classification of Rock Mass Rating (RMR) given by

Bieniawski (1974), "Q" classification by Barton et al.

(1974) and Romana et al. (1983) have been analyzed

and applied for concrete lining to check the stability in

Punasa tunnel (Gupta et al. 2011). The “Classification

based on Rating of Tunnel by Visual Field Inventory"

(CRTVFI) has been implemented at the time of

tunneling in basaltic lava flow for their longevity,

stability and finalization for stand up time with

treatment/ protection measures. Supporting system in

different condition i.e. dry and wet for their longer

stability has been applied for different rock mass

category i.e. in massive/dense basalt, vesicular

amygdular basalt and red bole. This classification has

helped in many ways especially in terms of saving the

time and money of the studied tunnel project.

Location of the Study Area:

The study area is part of Nimar Plateau, situated in

South Western part of Punasa Block, District Khandwa,

Madhya Pradesh belonging to Survey of India

Topographic Sheet Nos. 55 B/7 and 55 B/12 between

the longitude 76o20′ and 76

o32′ E and latitude 22

o and

22o13′ N.

Geology of the Tunnel Site:

The area of investigation is mainly covered with Deccan

Traps basaltic lava flows erupted during the Cretaceous

to Eocene period (Blandford, 1967). On the basis of

thirty two bore hole drill logs and resistivity survey

seven flows and aquifer zones have been recognized in

the area. The Punasa tunnel is covered with Middle

Trap, which forms a part of the Nimar Plateau. The

numerous dykes are present in the area and are running

parallel to the river Narmada. The top basaltic flows are

highly vesicular and amygdular; where as middle flows

are compact in nature. The river and nala are filled with

black colored alluvial soil.

Geotechnical Appraisal of Tunnel:

The tunnel appraisal has been carried out for the

estimation and support system for the project using

Rock Mass Rating System (RMR) and NGI index or Q

rating and Classification based on Rating of Tunnel by

Visual field Inventory (CRTVFI).

Joint Pattern in the Study Area:

The joint patterns/sets have been analyzed (Ahmed et al.

1999) for the estimation of tunnel quality with the help

of five sets of joints J1, J2, J3, J4 and J5 present in the

area under investigation. Joint J1 & J5 are horizontal to

sub horizontal in nature with 5 to 100cm spacing causes

crown fall in the tunnel. Joints J2, J3 and J4 are vertical

to sub vertical or columnar in nature with 20 to 50 cm.

spacing. 0.20 to 0.50 meters thick seams or weathered

1531 GUPTA. M. C, SINGH. B. K and SINGH. K. N



layer present in between surface of joints. More often

the surfaces of joints are smooth, plane or uneven filled

with silica and clay materials with a few noticeable

slickensides.

Table1: Geomechanical classification and required support in tunnel of Upper Punasa (After Bieniawski 1974;

modified after Gupta et al. 2011)

Rock mass Rock

Class RMR

Full-face or Heading and

benching method Rock bolt Shotcrete

Good rock

(massive/

dense basalt)

II 69 to

74

1 to 1.5m complete support

max. 20 meter at exit & inlet

portal ISMB/steel support

200mm with lagging

3m long +

spaced 3 m.

c/c 25 mm ,

shell type

50mm + 50mm in

crown & wall portion,

some patches are

covered with reinforced

shotcrete (in damp area)

Fair rock

(Vesicular

amygdular

Basalt)

III 52 to

55

Top heading and bench, 6.20

and 2.15 m. required support

system, 10 m. at exit & inlet

portal ISMB/steel support 200

mm with lagging

Systematic

rock bolt

4meters long,

spaced 1.5 -2

m. in crown

50-100mm in crown,

30mm in side walls.

Poor rock (Red

bole/

tuffaceous/

fractured zone/

shear zone)

IV 29

Top heading and bench, 6.20

and 2.15 m. required support

system, 10 m. at exit & inlet

portal ISMB/steel support

200mm with lagging

Systematic

rock bolt 4 to

5 meters long,

spaced 1-1.5

m. in crown

100-150mm in crown,

100mm in side walls.

Rock Mass Rating (RMR) System:

Rock mass rating system was developed by Bieniawski

(1973) and have been visualized for Punasa Tunnel Site

(Gupta et al. 2011). On the basis of the rock mass class,

Bieniawski (1989) has recommended a primary support

for tunnels of shallow depth having the diameter

between 5 - 12 meters with vertical stresses below 25

mpa (Table-1).

Q-System:

The Q - system of rock mass classification was

proposed by Barton, Lien and Lunde (1974) for the

determination of the tunnel quality of a rock mass. The

numerical value of this index Q is defined by quality

using different parameters, viz. Rock Quality

Designation (RQD), joint pattern or discontinuity of

joints, Joint number (Jn), joint roughness number (Jr),

joint alteration number (Ja), joint water reduction factor

(Jw) and stress reduction factors (SRF). After evaluating

various parameters the support system have been found

for the Punasa tunnel has been given in table-2.

( ) ( ) ( )RQD Jr JwQ x x

Jn Ja SRF=

Required De (Equivalent dimension) and ESR

(Excavation support ratio)

)(sup

)(,

ESRratioportExcavation

meterdiameterspanExcavationDe =

218.56.1

35.8==eD

(Value of ESR = 1.6) Above parameters have been used

in geological appraisal for different rock mass

categories. The value of excavation support ratio (ESR)

has been taken from Barton et al. (1974). The support

system for treatment/protection measures of tunnel has

been suggested as per Q-system & Rock mass

categories.

For massive/dense basalt or Grade I, II, II/I or hard

rock, the Q value is 13.33 where tunneling quality (Q) is

fair to good. For vesicular amygdular basalt or III, III/II,

II/III or soft rock, the Q value is 24.8 and Tunneling

quality (Q) is good. For red bole/ tuffaceous basalt or

III/IV, IV or soil, the Q value is 0.0825 where as

Tunneling quality (Q) is Very poor

Comparison of Support System:

The Upper Punasa tunnel has been excavated through

the basaltic lava flow which are competent than other

rock. The support system has been developed based on

the Rock Mass Rating (RMR) classification of

Bieniawski (1974) and Q system of Barton et al. (1974).

The Excavation Support Ratio (ESR) has been

considered to the support system in tunneling. The

comparison of various support systems modified after

Rutledge (1978), Wickham et al. (1972 & 74)

Bieniawski (1974) and, Barton et al. (1974). has been

shown in Table-3.

1532 Rating of Tunnel by Visual Field Inventory – A Case Study

of Punasa Tunnel, District Khandawa, M. P. India



Table 2: Support System Required for Underground Excavation (After Barton et al. 1974)

Rock mass

category/ Support

category

Q

Conditional

factor

RQD/Jn

P(kg/cm²)

(Approx.)

Span/

ER

(m)

Type of support

Massive/dense

basalt

/ 17 - 20

10 – 4 < 5 - > 30 1 3.5 – 9

Sb (utg), B (utg)1.5 to 2 m + S 2-3

cm.to B (tg) 1.5 - 2 m.+ clm or

S(Mr),10-15cm

Vesicular

amygdular basalt /

13 - 16

40– 10 < 10 - >15 0.5 5 – 23 Sb (utg) B (utg) 1.5 - 2 m. B (utg)1..5

– 2 m. + S 2 - 3 m. + S (mr) 5 - 10 cm.

Red bole /

tuffaceous / 29 - 32 0.4 - 1 1.4 - > 5 3 2.2 – 14.5

B (tg) 1 m. + S (mr) 5 - 12 cm. S (mr)

7.5 -25 cm. CCA – 20 - 40 m + B (tg)

1 m. CCA (sr) 30 -50 cm. + (tg) 1 m.

Index: Sb- Spot Bolting, B - Systematic Rock Bolting, S -Shotcrete, utg- Tensioned (Expanding Shell type for

competent Rock Mass, Grouted Post Tensioned in Very Poor Quality Rock Mass)

Table 3: Comparison of Support System for Different Rock Mass Category as Per Various I.S. codes of 1971,72, &

78 (Modified after Barton et al. 1974)

Index: B-Systematic Rock Bolting; Sb- Spot Bolting; (utg)- Untensioned, Grouted; (tg)- Tensioned, Grouted;

(mr)Mesh Reinforced; Clm - Chain Link Mesh; CCA- Cast Concrete Arch; (Sr) - Steel Reinforced;

R.S.- Reinforcedshotcrete; C/C - Center to Center; R.T.- Rock Type; R.M.R.- Rock Mass Rating; S.C.- Support

Category; R.M.Q.- Rock Mass Quality; S.S.- Support System. The Value of Q has been determined by Excavation

Support Ratio for Water Tunnel as 1.6.

Table 4: Classifications for Rating of Tunnel by Visual Field Inventory (CRTVFI) (Modified after Romana et al.,

1983)

Rock mass Gr. I, II

(1)

Gr. III, III/II

(2)

Gr. III/IV

(3)

Characterization

/rating Good Poor Very Poor

In dry condition/

Stability

Stable more than One

Year

Stable more than Six

Month Stable Three to Seven Days

Treatment Spot bolting(tg) Rock bolt(tg) + shotcrete

(50mm)

Systematic rock bolting(tg) +

Reinforced shotcrete(100mm)

In Wet condition/

Stability

Stable Six

Month(Maximum) Stable 15 to 30 Days Immediate or Twenty four hours

Treatment

Reinforced

shotcrete(100mm) +

Systematic rock bolt(tg)

Reinforced shotcrete

(100mm) + Systematic

rock bolt (tg)

Shotcrete (25mm)+Systematic

rock bolt (tg) + Steel support

(ISMB200mm) +Probe +back fill

concrete

Index: Gr.I and II: Massive dense basalt,Gr.II/III and III:Vesicular amygdular basalt, Gr. III/IV: Red bole,

tg: Tensioned grouted

1533 GUPTA. M. C, SINGH. B. K and SINGH. K. N



Classification for Rating of Tunnel by Visual Field

Inventory (CRTVFI):

Several classifications have been proposed by

geotechnical engineers and are followed for estimating

the nature of rock and treatment or supports in the

tunnel. The CRTVFI classification has been developed

for the field inventory of tunnel at the time of

monitoring by visual observation for tunnel stability.

The classification applied at the time of excavation of

tunnel is very simple and terminology used has been

widely accepted by geologist and engineers. The

classification of Romana et al. (1983) has been modified

on the basis of rating of tunnel by visual field inventory

and other parameters like joints, pattern, seepage points

and percolation quality. The property of rock reflects

the strength and stability of rock mass characterization.

The classification for rating of tunnel by visual field

inventory used in the study area in dry and wet

conditions with treatment and support system is given in

the table-4.

Conclusion:

The different rock type mainly Massive / Dense Basalt

or grade I, II and II/I, Vesicular Amygdular Basalt or

III, III/II and II/III and Red bole/Tuffaceous or III/IV,

IV/V has been applied to different types of support

system in dry and wet conditions. The classification of

rock mass rating and Q given by Bieniawski (1974;

modified in 1989) and Barton et al. (1974) utilized only

at an early stage of the projects for estimation and

support system. The author and his team executed

support system in the tunnel as per site conditions and

applied at the time of excavation and developed a new

Classification for Rating of Tunnel by Visual Field

Inventory (CRTVFI) including other geotechnical

parameters at the time of monitoring by visual

observation. The calculated numerical value as per

Bieniawski classification, the rock mass rating value for

massive/dense basalt is higher than vesicular amygdular

basalt but as per Q classification the value for vesicular

amygdular basalt is higher (24.8) and stable without any

treatment under the dry condition and stable with

nominal treatment in wet condition.

Acknowledgement:

The authors are highly thankful to the Narmada Valley

Development Authority & Water Resource Department,

Madhya Pradesh, Bhopal for providing the information

related to the work. Thanks are due to the team

members of the Narmada Valley Development

Authority for their help rendered during the field work.

References:

[1] Ahmed, M.J. & Soni, C.K., 1999: Geotechnical

Evaluation of deep rock cut tail channel, Narmada

(Indira) Sagar Project (M.P.) Indian Society of

Engineering Geology Bhopal. Geol. Vol. XXVII

pp.200-206

[2] Ahmed, M. J. and Pathak, S.K., 1999: Geotechnical

appraisal of river diversion arrangement of

Narmada Sagar Project, district Khandwa, (M.P.).

Indian Society of Engineering Geology Vol. VII

No.4, pp.126-130

[3] Barton, N., Lien, R. and Lunde, J., 1974:

Engineering classification of rock masses for the

design of tunnel support. Rock mechanics v16 (4),

pp.189-236

[4] Bieniawski, Z.T., 1973: Engineering classification

of jointed rock masses. Trans. S. African Inst. Civil

Engg. V 5, pp 335-343.

[5] Bieniawski, Z.T., 1974: The geomechanics

classification of rock masses and its application in

tunnelling proceeding, 3rd International Conf. on

rock mechanics, Denver, Vol. II A, pp. 27 - 32.

[6] Bieniawski, Z.T., 1984: Rock mechanics design in

mining and tunneling. A.A. Balkema, Rotterdam,

272p.

[7] Bieniawski, Z.T., 1989: Engineering rock mass

classification. John Willey & Sons., New York,

251p.

[8] Blandford, W.T., 1967 b. On the Traps and

intertrappean beds of Western and Central India

Mem. Geol. Surv. Ind. 6 pp. 137-162.

[9] Gupta,M. C.2008: Geotechnical and Env. Appraisal

of Upper Punasa canal, Khandwa District, M. P.

Unpublished Ph.D. Thesis, Vikram University,

Ujjain, M. P.

[10] Gupta, M.C., Singh, B.K. and Singh K.N.2011:

Engg. Geo. rock mass classification of Punasa

Tunnel site, Khandwa District, M.P.Geol. Soc.

India, Bangalore, vol. 77(3), pp 269-272.

[11] I.S. code 5878 (part II/Sec1) 1971: Indian

Standards code of practice for tunnel excavation in

rock, section 1- Drilling and blasting (Reaffirmed

2000).

[12] I.S. code 5878 (part II/Sec2) 1971: I S code of

practice for tunnel excavation in rock, section 2-

Ventilation, lighting, mucking and dewatering

(Reaffirmed 2000).

[13] I.S. code 5878 (part II/Sec3) 1971: IS code of

practice for Underground excavation in rock,

section 3- Tunneling method for steeply inclined

tunnels, shaft and underground powerhouse

(Reaffirmed 2000).

[14] I.S. code 5878 (part III) 1972: Indian Standards

code of practice for construction of tunnel.

Underground excavation in soft strata (Reaffirmed

2000).

[15] I.S. code 5878 (Part VII) 1972: Indian Standards

code of practice for Grouting (Reaffirmed 2000).

1534 Rating of Tunnel by Visual Field Inventory – A Case Study

of Punasa Tunnel, District Khandawa, M. P. India



[16] I.S. code 4756 -1978: Indian Standards code of

practice for safety code for tunneling 1978

(Reaffirmed 2007).

[17] Romana, M and Samuel Estefania, 1983: Field

inventory of tunnel for classification purpose. Vol.1

Tome/ Band2 (1983) International Society for Rock

Mechanics (ISRM), pp.207-210.

[18] Rutledge,T.C. and Preston,R.L.,1978: New Zealand

experiences with Engineering classification of rock

for the predicting of tunnel support . Proc.

International tunnel support, Tokyo, pp.23-29.

[19] Wickham, G.E, Tiedmann, H.R, and Skinner, E.H.,

1972: Support determinations based on geological

predictions. Proc. 1st North America Rapid

Excavation and Tunneling Conference, Chicago,

vol.1, pp 43-64.

[20] Wickham, G.E, Tiedmann, H.R, and Skinner, E.H.,

1974: Ground control prediction model RSR

concept. Proc. Rapid Excavation and Tunneling

Conference, AIME, New York, pp 691-707

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1535-1544


Geology and Characteristics of Metalimestone-Hosted Iron Deposit

near Negash, Tigray and Northern Ethiopia

SOLOMON GEBRESILASSIE, BHEEMALINGESWARA KONKA and FISEHA ADHANOM Department of Earth Science, CNCS, P.O. Box 231, Mekelle University, Mekelle, Ethiopia


Abstract: Recent exploration efforts in Ethiopia have helped in locating Mukuat iron deposit of about 300,000 tons.

The area forms part of Arabian Nubian Shield and consists of Neoproterozoic low grade N-S to NE-SW trending

basement rocks of Tsaliet Group (~860-750Ma) with metavolcanics, metavolcaniclastics and metasediments, and

younger Tambien Group (~740 Ma) with metasediments, slate, phyllite, metalimestone and pebbly slate (diamictite).

Apart from foliation, Tambien Group rocks show development of synclinal structures (Negash syncline). These are

intruded by post-tectonic granitoids (~600 Ma) and overlain unconformably by fluvial Paleozoic iron-rich Enticho

Sandstone and Edaga Arbi Tillite and by marine Mesozoic iron-rich Adigrat Sandstone, Antalo Limestone, Agula

Shale and Amba Aradom Sandstone. Dolerite dikes have intruded during uplift and faulting during Cenozoic time.

Iron deposit hosted by metalimestone of Tambien Group is confined to western limb of the Negash syncline. It

occurs in pockets (about 20-40m deep, 50-75m thick, 33-50 wt% Fe), and extends a kilometer along strike and

show red, brown and yellow colors. Hematite, goethite and limonite dominate the mineralogy and show columnar,

replacement and flow textures indicating ore remobilization. Presence of quartz and clay with iron ores, iron deposit

in discontinuous cavities, and at places dolerite dikes forming the wall of cavities suggest that the deposit is

younger, structurally controlled, and remobilized filling the solution cavities in limestone. Mineralogy and higher

Fe2O3 values (38-55%) compared to SiO2 (21-33%) and Al2O3 (1-4%) indicate development of iron-rich laterite as

part of weathering process and residual enrichment of Enticho Sandstone prior to remobilization possibly after

Eocene.

Keywords: Lateritic Iron Deposit, Cavity Filling, Metalimestone, Mukuat, Negash, Tigray, Ethiopia.

Introduction:

Ethiopia is one of the NE African countries with not

many known iron deposits except the few occurrences

such as the mafic intrusions hosted Bikilal P-Ti-Fe

deposit in western Ethiopia; iron-rich laterites e.g.

Melka Sedi, central eastern Ethiopia, Yubdo in Wollega

region, western Ethiopia and sulfide-related iron-rich

gossan deposits e.g. Rahwa (Tadesse et al., 2003;

Bheemalingeswara and Atakilty, 2012; Ottemann and

Augustithis, 1967). Recently increased exploration

activities have successfully delineated many metallic

(base metals sulfides) and non-metallic deposits (salt)

for further studies. Tigray region, northern part of

Ethiopia, which belongs to the southern part of the

Arabian-Nubian Shield (ANS) also, seen the surge in

the exploration activities as ANS is known to host many

gold, base metal and iron deposits elsewhere. As a result

of these concerted efforts presence of lateritic iron

deposits have been reported from different parts of

Tigray, northern Ethiopia e.g. at Shire, Sheraro and

Mukuat near Negash. The deposits at Shire and Sheraro

have developed on ferruginous sandstones (about 100mt

each) of Paleozoic/Mesozoic(?) age (Ebrahim, 2011,

Bheemalingeswara, 2012) while at Mukuat it is hosted

by metalimestones of Neoproterozoic age (Fig.

1)(Fiseha, 2012) with a calculated reserve of 300,000 t

(Ezana, 2011). Mukuat iron deposit is currently being

exploited and used as one of the ingredients to produce

cement in the region. Although it is being mined, many

questions such as ore mineralogy and paragenesis, mode

of occurrence and mechanisms of ore formation

remained obscure and unanswered. So, in this paper we

tried to provide answers to some of the questions on the

nature and characteristics of the iron deposit, its host

rocks, and other related geology by employing field,

petrographical and geochemical data.

2. Regional Geological Setting:

The Precambrian geological setting of northern Ethiopia

is mainly defined by the presence of low-grade

metavolcano-sedimentary domains in the southern part

of the ANS. The ANS is a collage of different terranes,

which were accreted from ~ 780 to 550 Ma (Johnson

and Woldehaimanot 2003). These basement rocks host

majority of the metallic and non-metallic deposits in the

region and in northern Ethiopia they are broadly divided

into two (Beyth, 1972): the Lower Tsaliet Group and

1536 SOLOMON GEBRESILASSIE, BHEEMALINGESWARA KONKA

and FISEHA ADHANOM



Upper Tambien Group. The Tsaliet Group mainly

consists of low-grade metavolcanic/ metavolcaniclastic

rocks, which were formed around 854 Ma (Teklay,

1997; Fig. 2a and b). They are of felsic to intermediate

to mafic in composition and affected by nearly N-S

trending foliation and NE-SW trending sinistral shear

zones (e.g. Tadesse et al., 1999). The Upper Tambien

Group mainly contains slightly metamorphosed

carbonate and metasedimentary rocks (slate, phyllite,

sericite-chlorite schist), which are dominantly exposed

in Maikenetal, Tsedia, Chehmit and Negash inliers (Fig.

2b; Alene et al., 2006). Their overall thickness reaches 2

to 3 km and were deposited between ~ 835 to 740 Ma in

a shallow marine environment (Alene et al., 2006;

Avigad et al., 2007). Younger diamictites cap the

Tambien Group at Negash (Alene et al., 2006). These

basement rocks are intruded by syn-tectonic (~ 800 to

735 Ma) and post-tectonic (620 and ~520 Ma)

granitoids (Gebresilassie, 2009). These basement rocks

are overlain by the fluvial Paleozoic Enticho Sandstone

and Edaga Arbi Tillite, which show interfingering

nature and angular unconformity. Enticho Sandstone is

dominated by poorly cemented well sorted medium size

quartz and Edaga Arbi Tillite by clay/silt bands and

boulders of different sizes. These sedimentary rocks are

overlain by marine Mesozoic Adigrat Sandstone, Antalo

Limestone, Agula Shale and Amba Aradom Sandstone

(from old to young). Both the Enticho and Adigrat

Sandstones are iron-rich and at places, iron values reach

up to 30 wt%, particularly in Adigrat Sandstone. These

Paleozoic and Mesozoic sedimentary rocks are overlain

by the Flood Basalts and intruded by the younger

dolerite dikes. The dikes have intruded during

upliftment and faulting in Cenozoic time.

Figure 1: Location Map of The Study Area: A) Map of Northeastern Africa Showing the Location Broader Northern

Ethiopia Region (Inset Rectangle) and B) Plan Map of Northern Ethiopia Showing the Location of the Study Area

3. Geology of the Study Area:

Outcrops of Tsaliet and Tambian Group rocks exist in

the study area (Fig. 3). The Tsaliet Group is exposed in

the western side of the area and defined by mafic to

felsic metavolcanic/volcaniclastic rocks. These rocks

are fine- to medium-grained and composed of

plagioclase and K-feldspars, sericite, epidote, chlorite,

tremolite and quartz with trace amounts of opaque iron

oxides. Metavolcaniclastic rocks are marked by the

presence of rounded to elongated clasts of 1 to 10 cm in

size embedded within volcanic rocks. Locally, they are

affected by propylitic alteration and malachite staining.

From west to east, Tambien Group contains

metasedimentary rocks of dolomite, slate/phyllite, black

limestone and pebbly slate (diamictite). Fine- to

medium-grained dolomite occurs in the western part of

the study area, which usually is bracketed by

slate/phyllite. Central part of the area is covered by

black to pink colored slate/phyllite domains. They show

fairly well developed foliation and locally lustrous

sheen on the surface (Fig. 3). Black limestone forms the

western and eastern limbs of Negash syncline. It shows

well developed karstic topography due to meteoric and

groundwater interaction and some of these dissolution

cavities host the iron ore (Fig. 3). The core of the

Negash syncline is occupied by pebbly slate

(diamictite). Negash granite (~600 Ma) and associated

aplitic dikes intrude the Tambien Group in the northern

part of the study area (Fig. 3). The granite is pink in

color with medium grain size and dominated by

1537 Geology and Characteristics of Metalimestone-Hosted Iron Deposit near

Negash, Tigray and Northern Ethiopia



orthoclase feldspar, quartz, plagioclase feldspar, biotite

and sericite, in the decreasing order. Remnants of

Paleozoic Enticho sandstone and Adaga Arbi Tillte

locally overlie these basement rocks. The younger

dolorite dykes of Cenozoic age that intruded these rocks

are particularly seen in the metalimestone rock (Fiseha,

2012).

Figure 2: A) Regional Geological Map of Northern Ethiopia Showing the Exposure of Tsaliet and Tambien Group

Rocks and B) Enlarged Regional Geological Map of The Area Around Negash where the Mukuat Iron Deposit is

Located. Circles with Number Shows Distribution of the Granitoid in the Region which Intruded the Basement

Rocks (After Gebresilassie, 2009 and References Therein)

4. Geological Structures:

The whole sequence is affected by different phases of

deformations. The first phase of deformation (D1) has

resulted in the development of foliation (particularly in

phyllites), which is pervasive and trending N-S to

N40°E. The second phase of deformation (D2) is

resulted in the development of a prominent synclinal

structure, known as Negash syncline with a fold axis

trending nearly N-S and forming 1 km wide core .

Grade of metamorphism being low it is well reflected

only in the metasedimentary rocks in the form of

foliation and slaty cleavage compared to mildly affected

black limestone and dolomite. Normal and dextral strike

slip faults and younger geological structures related to

E-W trending Wukro fault (not in the map) (related to

D3) are also present, displacing the metavolcanic and

metasedimentary rocks as well as the Paleozoic and

Mesozoic sedimentary sequences (Fig. 3) (Fiseha,

2012).


and FISEHA ADHANOM



Figure 3: Geological Map of the Mukuat Area Showing the Outcrops of Both Tsaliet and Tambien Groups along

with Iron Ore Lenses





5. Methodology:

Out of 8 samples selected for analysis, seven were

collected from mineralized zone and one from

weathered dolerite dike. All the rock samples were

analyzed powdered using jaw crusher and pulverizer at

Ezana Laboratory, Mekelle. The fine rock powders

about 100gm each were given for geochemical analysis

(major oxides) using XRF (PANAlytical, 2424). The

analysis was carried out at the laboratory of Mossobo

Building Materials Production Plc, Mekelle, using the

standard references. The precision of the results was

within the acceptable limits. The results are shown in

table 1. In addition to geochemical analysis, the

mineralized samples were cut and polished. The

polished sections were studied for ore petrography using

reflected light microscope at Petrography laboratory,

Department of Earth Science, Mekelle University,

Mekelle.

6. Iron Mineralization and Paragenesis:

Iron mineralization at Mukuat (Negash) is present

within the black limestone filling fractures and the

cavities developed by karstification and confined to the

western limb of Negash syncline (Fig. 4a-d). It displays

dark brown, reddish brown and yellowish brown colors.

The iron ore occurs as pockets especially at the contact

within the metalimestone wall rock and at places

parallel to and/or perpendicular to the bedding of the

weakly metamorphosed black limestone (Fig. 4a). A

pocket of an iron mineralized zone is about 20-40 m

deep and 50-75 m thick and extends a kilometer along

strike (Fig. 4d). At the contact between the black

limestone and slate/phyllite, the fractures developed in

the former are filled by iron-rich lateritic material and

clay (Fig. 4b) possibly mobilized from the overlying

Enticho Sandstone (Fig 3). This indicates that the iron

ore is remobilized from previously laterized overlying

Enticho Sandstone. The karst features, fractures,

bedding planes and dikes seem to have served as

channels, sites or conduits for the mobilization and

accumulation of iron ore. Ore microscopic study of the

polished sections from the iron deposit indicates

presence of hematite, goethite and limonite in different

proportions. They, together with trace amounts of pyrite

(Fig. 5a-f) within the matrix of quartz and clay form the

gangue. Hematite, the dominant ore mineral, that

varies from colloidal to fine and medium grain size is

present as aggregates, nodules, sub-rounded, irregular

grains etc. Hematite is not the primary mineral. It has

developed by replacing goethite. It has developed in the

interstices within the groundmass of goethite (Fig. 5a

and b), as collofom bands with goethite (Fig. 5c) and as

replacement to and/or discontinuous veinlets cutting

goethite (Fig.5e). Goethite is another abundant mineral,

which occurs in the form of lensoidal bodies (Fig. 5a)

and veinlets cutting the earlier formed hematite and

goethite grains (Fig. 5d). Traces of pyrite also occur

within the groundmass of iron minerals (Fig. 5f).

Limonite, though common could not be well

differentiated in the ore microscopic studies. It is

inferred to occur as dark grey grains where some of

them are showing replacement to goethite (Fig. 5a).

Pyrite is euhedral to subhedral in shape and does not

show alteration (Fig. 5f). The paragentic sequence of

Mukuat iron deposit is shown in figure 6. The

replacement of goethite by hematite and continuous

growth of the latter along grain boundaries of the former

suggest that goethite is the source for hematite.

Limonite is expected to have formed first from iron

oxide and hydroxide cement. It is the source for

goethite. So, it is a gradual process that has produced

these three minerals in suitable conditions. It has

resulted in the development of the iron minerals in

different phases. Occurrence of hematite as

discontinuous veinlets cutting across hematite and

goethite and forming alternate banding with goethite

suggests different phases/ generation of hematite and

goethite. The fact that goethite veinlets cut the earlier

generation hematite, limonite and goethite indicates that

goethite II formed after hematite I. The euhedral to

subhedral shape and absence of deformation in pyrite

shows that it is the youngest sulfide formed within the

groundmass of iron oxides and hydroxides.

7. Geochemistry:

Major oxide values for Fe2O3 vary from 38 to 55 wt%,

Al2O3 from 1 to 4%, SiO2 from 21 to 33% and CaO

from 13 to 27%. Others are present in trace amounts.

Highest values of Fe2O3 are reported from the center of

the mineralized zone, where the iron ore is filling the

cavities and fractures within the black limestone. Iron

mineralization does not indicate any alteration, similar

to one of the dolerite samples (N-07). It is not affected

by any alteration as indicated by its major oxide values

(Fiseha, 2012) and are similar to those of unaltered

basalts (Wilson, 1989) (Table 1).


and FISEHA ADHANOM



Figure 4: Plates Showing the Mode of Occurrence of Mukuat Iron Deposit at an Outcrop Level; a) Iron Ore

Occurring Parallel to the Bedding of Black Limestone, b) Iron Ore Interfingered with Enticho Sandstone, C) Some

of The Karsts Filled By Clay and Iron Ore and d) an Overview of Lensoidal Pocket of Iron Ore, Which is Structure

Controlled.

Table 1: Major Oxide Values (Wt%) of Iron Ore taken from Filled Cavities and Fractures within the Black

Limestone and Enticho Sandstone and Dolerite





Figure 5: Ore Microphotographs showing a) Hematite Grains Forming from Goethite Ground Mass, b) Continuous

Growth of Hematite along Goethite Grain Boundary and Interstices, c) Alternate Colloform Bands of Hematite and

Goethite, d) Younger Goethite Veinlets Cutting earlier developed Hematite and Goethite Grains, e) Hematite

Overgrowths Over Goethite and f) Pyrite Grain in Hematite and Goethite Groundmass [Note: ht = Hematite,

gt = Goethite, and py = Pyrite].


and FISEHA ADHANOM



Figure 6: Paragenetic Sequence of Iron Ore Minerals Identified in the Deposit. Roman Numbers Show different

Generations of Iron Oxides/Hydroxides. Abbreviations: P = Pervasive and V = Veinlet

8. Discussion:

The chlorite-epidote-sericite dominated mineralogy of

the metavolcanic, metavolcaniclastic rocks, presence of

grey to red colored slate/phyllite unit with lustrous

sheen, nearly N-S trending D1 foliation and D2 folds

affecting both the metasedimentary and

metavolcaniclastic rocks, and intrusion of the basement

rocks of the study area by post-tectonic granites and

aplitic dikes is consistent with situation of the rocks

elsewhere in Tigray region. The rocks have experienced

lower greenschist facies metamorphism typical of the

rocks elsewhere in ANS. Alteration patterns and their

nature such as presence of propylitic alteration and

silicification, affecting both the metavolcanic and

metasedimentary rocks, show cyclic flow of

hydrothermal fluids that resulted in the development of

base metal and gold mineralizations (low grade?) in the

region during Neoproterzoic (e.g. Hawzien, Abreha

Weatsbeha, Bheemalingeswara et al., 2012; Workamba,

Gebresilassie, 2009). Based on the available data the

geological history of the study area has been

constructed. After folding, uplift and shearing of Tsaliet

and Tambien Groups by D1 and D2 deformation, a

peneplation occurred by erosion and denudation

processes during late Proterozic and deposition of

fluvial and marine sedimentary rocks particularly iron-

rich Enticho and Adigrat Sandstones and others

(carbonate rocks- limestone and dolomite; and

argillaceous rocks- shale). These rocks together with

basement rocks are affected by the prominent E-W

trending Wukro fault, which caused upliftment of both

the basement rocks and the overlying iron-rich

sedimentary rocks. This has facilitated development of

minor structures and downward flow of water. The

percolating water circulating downwards and its

interaction with the carbonate rocks has created a

number of dissolution cavities of different sizes and

development of other karst –related features in black

limestone in the western limb of the Negash syncline.

The folded synclinal structure being overturned, dipping

west, made the western limb suitable for the

development of karst-related structures. During this

period, the iron-rich sandstone, particularly Enticho

Sandstone, after removal of the younger litho-units, has

experienced lateritization similar to the process that has

produced significant size lateritic iron deposits from

iron-rich sandstones near Shire and Sheraro north of

Negash in Tigray region. However, the iron

mineralization at Mukuat is a different type of

mineralization, though related to lateritization. It is a

transported one, occurring as pockets and confined to

the metalimestone unit. The ore mineralogy of the

deposit is constituted by hematite, goethite and limonite

and show oolitic/nodular shapes, colloform banding and

flow structures. Thus, it supports the hypothesis that this

small-scale iron deposit is a result of remobilization.

The occurrence of the iron mineralization as irregular

bodies filling cavities and fractures in black limestone,

shallow depth of mineralization and presence of clay

and quartz with iron ores are also other evidences of

mobilization. Petrography of the iron minerals and

megascopic field data reveal development of iron-rich

lateritic (or ferricrete) cover on Enticho Sandstone,

which is overlying few meters thick clay dominated

Edaga Arbi Tillite. Though they often show

interfingering nature and directly overlie the basement

rocks, in the study area the Tillite unit overlie the

basement metalimestone followed by Enticho

Sandstone, showing angular unconformity.

Development and presence of different horizons such as

nodular, mottled and pseudo-pisolitic horizons (formed

as part of lateritization on sandstone), is common in

Shire area (Ebrahim, 2011) and elsewhere (Nahod et al.,

1977; Tardy and Nahod, 1885) is not common in the

study area. Presence of lateritic material in the cavities





in metalimestone and absence of lateritic cover on the

surface of sandstone amply indicate its removal and

mobilization. Apart from the nature of rock, the factors

like pH and Eh which are themselves controlled by the

climate, vegetation and the morphology of the landscape

play an important role in defining the mobility of iron,

aluminum, silica etc. At the same time, the area has

experienced climatic condition (18.3-30oC and up to

300cm rainfall) suitable for the development of laterite

(Ebrahim, 2011). Presence of hematite, goethite,

limonite and ochre in the deposit suggests their presence

in the source lateritic cover. Hematite is fresh and does

not show any signs of alteration. It is formed by

replacing goethite. Petrographic study amply suggests

conversion of goethite to hematite but hematite does not

show any alteration to goethite as expected in an

oxidizing weathering condition. Such conditions are

common in arid conditions and the reason for such

development seems to be due to: a) transformation of

iron cement (iron hydroxide) to ferrihydrites (Combes et

al., 1990; Ramanaidu et al., 1996) similar to limonite

(crysptocrystalline) structure; b) limonite to metastable

goethite; c) the metastable goethite in the presence of

water into hematite during long burial digenesis of

sediments of Tertiary to Paleozoic in age (Tardy and

Nahod, 1985); this is noticed by not well developed

goethite in the polished sections; d) goethite does not

replace hematite because the temperature is relatively

high and does not favor hydration of hematite to

goethite in the lateritic profile particularly that occurs at

the top of the profile; e) the relative stability of goethite

and hematite depends on many factors such as grain size

effect. If the crystal of hematite is greater or equal to the

grain size of the crystal of goethite, hematite is stable

than goethite in water but in the reverse situation

goethite is stable. But in the case of concretion and

development of nodules in iron crust (ferricrete), both

minerals do not form into large crystals and appear as

very tiny particles (Tardy and Nahod, 1985). So, the

governing conditions for stability of these minerals are

mainly dehydration and water activity compared to the

less effective particle size; f) kinetics of precipitation, if

the iron is released from silicates and from other

primary source in water then it will form into goethite;

if the solubility product of goethite is excess than that of

ferrihydrite. So, the factor that favors ferrihydrite

formation will also favor the formation of hematite by

high temperature resultant dehydration. The factors

controlling the formation of ferrihydrite in solution

include rapid release of Fe, and low concentration of

organic compound (allowing concentration of inorganic

Fe3+). So, in ferruginous sandstone the iron release is

very high, as the large pore size of the quartz (in which

the iron serve as matrix) favors high activity of water

and release of iron. This may be one of the reasons for

the stability of hematite over goethite during

lateritization; and g) equilibrium condition involving

water activity, pore size and nodule formation. The

relative abundance of the major oxides especially SiO2,

Fe2O3, Al2O3 indicates increase of Fe2O3 (55 wt%)

compared to SiO2 (21 wt%) and Al2O3 (1.44 wt%) in

lateritic iron deposit. This clearly indicates

desilicification and removal of Al, though SiO2 removal

is relatively more. The suitable Eh and pH conditions

thus favoured residual concentration of Fe2O3.

Weathering and erosion of a laterite developed on top of

Enticho Sandstone is a continuous phenomenon.

Weathering and erosion resultant remobilization has

resulted in hematite filling the cavities in black

limestone. The concentration of iron ore in the fractures,

bedding and flanks of dolerite dikes suggest that these

structures served as conduits for the laterite migration

during remobilization. Generally, common in all

laterites, repetition of wet and dry seasons lead to

lateritization of the Enticho Sandstone. Leached iron

and other cations during the wet season are brought to

the surface by capillary action during the dry season.

The process of intrusion has no direct relation with the

iron deposit. It only has facilitated development of

fractures in the black limestone by upliftment and

helped in the formation of cavities by percolating rain

water. These cavities, inturn, have become sites for

accumulation of lateritic material. Based on the

stratigraphic and textural relationships, age of the

Mukuat (Negash) iron deposit is inferred to be much

younger compared to Enticho Sandstone, which formed

during Lower Ordovician (Garland, 1980).

Lateritization process was initiated during the exposure

of Enticho Sandstone after its uplift due to Wukro Fault.

Development of laterite and its mobilization to the site

of accumulation continued for considerable period

subsequent to dolerite dyke intrusion during Eocene and

later (Fiseha, 2012).

9. Conclusion:

Genetic mechanism associated with Mukuat iron ore

deposits is unique for arid tracts. Even though the

deposit is small in size and quantity, has brought into

light a series of processes responsible for metalimestone

cavities acting as hosting locales for lateritic iron

accumulation. Weathering and erosion of laterite and

resultant mobilization has led to iron ore minerals filling

the discontinuous cavities in metalimestone. Repetition

of wet and dry seasons facilitated mobilization of

leached iron through capillary action on to the surface

and its enrichment. One can use this input to explore

vast stretches of arid and semi arid tracts to locate such

isolated ore deposits elsewhere in similar terrains.

10. Acknowledgement:

We acknowledge Ezana Mining Plc and Messobo

construction materials Plc for providing necessary help


and FISEHA ADHANOM



during field work and sharing drill and chemical data.

Thanks are due to the reviewers Dr. Parvata Reddy

Ramachandra Reddy, Scientist, NGRI, Hyderabad,

India and Prof. Solomon Tadesse, Addis Ababa

University, Addis Ababa, Ethiopia for providing critical

comments and improving the quality of the paper.

9. References:

[1] Alene, M., Jenkin, G. R.T., Leng, M. J and

Darbyshire, F.D.P. 2006. The Tambien Group,

Ethiopia: An early Cryogenian (ca. 800-735 Ma)

Neoproterozoic sequence in the Arabian-Nubian

Schield. Precambrian Research, 147: 79-99.

[2] Avigad, D., Stern, R.J., Beyth, M., Miller, N and

Mc Williams, M.O. 2007. Detrital zircon U-Pb

geochronology of Cryogenian diamictites and

Lower Paleozoic sandstone in Ethiopia (Tigrai):

Age constraints on Neoproterozoic glaciation and

crustal evolution of the southern Arabian-Nubian

Shield. Precambrian Research, V.154, 88-106.

[3] Bheemalingeswara Konka., Solomon Gebresilassie

and Kassa Amare. 2012. Shear zone controlled base

metal mineralization around Abraha Weatsebaha -

Adidesta and Hawzein, Tigray region, northern

Ethiopia, Momena Journal of Ethiopian Sciences,

V.4(1), 3-28.

[4] Beyth, M. 1972. The Geology of Central-Western

Tigre. Ph.D. Thesis, Rheinische Friedrich-Wilhems

Universität, Bonn, Germany, 159p.

[5] Combes, J.M., Manceau, A and Calas, G. 1990.

Formation of ferric oxides from aqueous solutions:

a polyhedral approach by x-ray spectroscopy; II.

Hematite formation from ferric gels. Geochem

Cosmoch Acta, V.54, 1083-1091.

[6] Ebrahim, N.H. 2011. Geology and geochemistry of

lateritic iron deposit in Mydmu area near Shire

Indessilassie, western Tigray, Ethiopia. M.Sc

thesis, Mekelle Univ., Mekelle, Ethiopia, 136p.

[7] Ezana Mining Development plc. 2011. Mukuat Iron

mining near Negash (per. comm).

[8] Fiseha Adhanom. 2012. Geology and iron

mineralization in Mukuat, Wukro area, eastern

Tigray, northern Ethiopia, M.Sc Thesis, Mekelle

Univ., Mekelle, Ethiopia. 100p.

[9] Garland, C.R. 1980. Geology of the Adigrat area,

Memor No. 1. Ministry of Mines and Energy,

Addis Ababa, 51p.

[10] Gebresilassie, S. 2009. Nature and Characteristics

of Metasedimentary rock Hosted Gold and Base

metal Mineralization in the Workamba area,

Central Tigray, Northern Ethiopia, Ludwig-

Maximilans Univ., Munich, Germany, Ph.D.

Thesis, 134p.

[11] Johnson, P.R and Woldehaimanot, B. 2003.

Development of the Arabian-Nubian Shield:

perspectives on accretion and deformation in the

northern East African Orogen and the assembly of

Gondwana. In: M. Yoshida, B.F. Windley and S.

Dasgupta (eds.), Proterozoic East Gondwana:

Supercontinent Assembly and breakup. Geological

Society of London, Special Publication, V.206,

289-325.

[12] Nahod, D., Janot, C., Karpoff, A.M., Paquet, H and

Tardy, Y. 1977. Mineralogy, petrography and

structures of iron crusts developed on sandstones in

the western part of Senegal. Geoderma, V.19, 263-

277.

[13] Ramanaidu, E., Nahon, D., Decarreau, A and Melfi,

A.J. 1996. Hematite and goethite from durycrusts

developed by lateritic chemical weathering of

Precambrian Banded Iron Formations, Minas

Gerais, Brazil,Claysandclayminerals,V. 44(1),22-31

[14] Tadesse, T., Hoshino, M and Sawada, Y. 1999.

Geochemistry of low grade metavolcanic rocks

from the Pan African of the Axum area, northern

Ethiopia, Precambrian Research, V.99, 101-124.

[15] Tadesse, S., Milesi, J.P and Deschamps, Y. 2003.

Geology and mineral potential of Ethiopia: a note

on Geology and mineral map of Ethiopia. J. African

Earth Sciences, V.36, 273-313.

[16] Tardy, Y and Nahod, D. 1885. Geochemistry of

laterites, stability, Al-goethite, Al-hematite and Fe3+

kaolinite in bauxites and ferricretes, American

Journal of Science, V.285, 865-903.

[17] Teklay, M. 1997. Petrology, Geochemistry, and

Geochronology of Neoproterzoic Magmatic Arc

Rocks from Eritrea: Implications for Crustal

Evolution in the southern Nubian Shield. Eritrea

Department of Mines, Memoir 1, 125p.

[18] Wilson, M. 1989. Igneous petrogenesis: a global

tectonic approach. Unwin Hyman, London, 466p.

[19] Bheemalingeswara, K. 2012. Geological and

geochemical study of lateritic iron deposit near

Mentebteb, Shiraro, Northern Tigray, Ethiopia.

Project Report, CNCS, Mekelle University,

Ethiopia, 30p(unpubl.).

[20] Bheemalingeswara, K and Atakilt Araya, 2012.

Rahwa auriferous gossan, northern Ethiopia: A

strong indicatorfor subsurface massive sulfide

mineralization. International Journal of Earth

Sciences and Engineering, 5(3): 402-408.

[21] Ottemann, J and Augustithis, S.S. 1967.

Geochemistry and origin of “platinum-nuggets” in

lateritic covers from ultrabasic rocks and birbirites

of W. Ethiopia. Mineralium Deposita, 1(4): 269-

277.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1545-1551


Ore Fluids Associated With the Metasediment Hosted Central

Auriferous Zone of Gadag Gold Field, Karnataka

M. A. MALAPUR1, S. MANJUNATHA

2, B. CHANDAN KUMAR

1 and A. G. UGARKAR

1

1Department of Studies in Geology, Karnatak University, Dharwad 580 003 2Department of Geology, Karnatak Science College, Dharwad 580 001


Abstract: In Gadag gold field, three almost parallel and tabular auriferous zones, namely western, central and

eastern auriferous zones, each with a distinct lithological assemblage, occur within a sequence of metavolcanics and

a thick pile of metasediments. The western auriferous zone is hosted by mafic to felsic metavolcanics, the central

zone is hosted mainly by greywackes, just above the boundary with the metabasalt and the eastern zone is hosted by

arenite-argillites and chlorite phyllites. In this paper, the nature and composition of hydrothermal ore fluids

associated with the sediment hosted central auriferous zone are presented based on the fluid inclusion micro-

thermometry. In this zone, gold mineralization is structurally controlled epigenetic vein type, and is invariably

associated with wall rock alterations. In addition to gold, ore minerals that occur are arsenopyrite, pyrite, pyrrhotite,

chalcopyrite, sphalerite, galena and scheelite. The gangue minerals are quartz, chlorite, plagioclase, sericite,

carbonates and carbonaceous matter. Three types of fluid inclusions, namely CO2-rich inclusions, H2O inclusions

and CO2-H2O inclusions are recorded in the auriferous quartz veins of central auriferous zone. Due to tiny size

constraints, micro-thermometric determinations are made only on CO2-H2O inclusions. The hydrothermal fluids that

caused gold deposition in these zones were of low salinity (2.0 to 6.6 wt% NaCl equivalent), dominated by CO2-

H2O (about 30 mole % CO2 ) with moderate densities (0.7 to 1.04 g/cc) at a maximum depth 1.3 km, and gold

deposition occurred over a wide temperature range of 175 to 325oC. Similar to Archaean greenstone hosted vein

type lode gold deposits elsewhere, large volumes of low-salinity CO2-H2O and CO2-rich fluids were probably

produced by metamorphic devolatilization during prograde regional metamorphism at greenschist-amphibolite

facies boundary in Gadag Gold field. During retrograde greenschist facies metamorphism, interaction of gold

bearing H2O-CO2-NaCl fluids with the wall rocks at decreasing PT conditions (in the shear zone vicinity) might

have lead to fluid immiscibility and separated H2O–rich and CO2-rich phases, there by significantly changing the

gold solubility and causing its precipitation. The ore fluids of sediment hosted auriferous zone are comparable with

the volcanic hosted western auriferous zone.

Keywords: Gold, Ore fluids, Fluid inclusions, Gadag Gold Field, Karnataka

I. Introduction:

Significant contribution for world gold production has

been from mesohydrothermal vein type lode gold

deposits of Archaean granite greenstone terrains. Thus

such type of gold mineralization represents an important

genetic class. An understanding of their nature and

genesis is not only fundamentally important from an

academic viewpoint, but can help to formulate

exploration strategies and better define potential areas

of mineralization. Obviously, studies of fluid inclusions

in veins deposited during the mineralizing event have

proved to be inevitable to understand the nature and

source of the ore fluid, components and genetic aspects.

Although, there are several gold deposits/occurrences

distributed in the well-known granite-greenstone terrain

of Dharwar Craton, the fluid inclusion data comes from

very few gold deposits like, Kolar (Santosh, 1986;

Mishra and Panigrahi, 1999; Solankar et al., 2006),

Hutti (Pal and Mishra, 2002), Gadag (Ugarkar et al.,

2000) and Kuchiganahalu of Chitradurga (Mukherjhee

et al., 2001). Further, as far as the data on the fluid

inclusions of gold bearing vein quartz in Gadag Gold

Field is concerned, it is restricted to the auriferous zones

hosted within the metavolcanics (Ugarkar et al., 2000).

In this paper, the nature and composition of

hydrothermal ore fluids associated with the

metasediment hosted central auriferous zone is

presented based on the fluid inclusion

microthermometry.

II. Geological Setting:

Gadag greenstone belt, a northern continuation of the

well known Chitradurga belt (Dharwar Type) of middle

to late Archaean age (2600-2400 Ma, Swaminath and

Ramakrishnan, 1981) of Karnataka, is well known for

1546 M. A. MALAPUR, S. MANJUNATHA, B. CHANDAN KUMAR and A. G. UGARKAR



gold mineralization. The detailed geology and map of

Gadag gold field (Fig.1) are given by Narayanaswamy

and Ahmed (1963) and Ugarkar and Deshpande (1999).

The gold mineralization in Gadag gold field is

distributed in three main auriferous zones where group

of gold bearing lode-quartz veins are found in clusters.

These auriferous zones are namely Western, Central and

Eastern auriferous zones that run in a linear pattern

almost sub-parallel to regional foliation (i.e. NNW-

SSE). The various litho units of Gadag gold field

include metavolcanics (metabasalt to felsic volcanics)

and metasediments (conglomerate, greywacke, argillite,

arenite, arkoses, phyllite, banded iron formation and

limestone/dolomite). The entire sequence starting from

metavolcanics in the west and metasediments towards

east is well exposed on the western limb of the main

synclinal structure.

Although the central auriferous zone occurs in the

contact between metavolcanics and metasediments,

majority of the lodes of this auriferous zone are hosted

by greywacke-argillite suite of rocks, extending from

Beladhadi to NE of Kadkol passing through Nabhapur,

Kabuliyatkatti, Attikatti, Mysore mine and Sangli mine.

Figure 1: Geological Map of the Gadag Gold Field Showing the Western, Central and Eastern Auriferous Zones

(after Ugarkar et al., 2000)

Eastern auriferous zone, which lies to the southwest of

Singatrayankeri Tanda and east of Dindur Tanda, is

entirely hosted within metasedimentary rocks. This zone

occurs within greywacke-argillite assemblage.

The primary layering (bedding) in the sulphidic chert,

greywacke-argillite suite, chlorite phyllite and the

intercalated BIF bands dips towards east at angles

varying from 200 to 55

0. The schistosity dips at higher

angles ~700, most commonly towards northeast. The

trend of bedding varies from N200W to N40

0W and that

of schistosity N-S to N150W. The entire sequence is

younging towards northeast as indicated by way up

criteria like vesicular and convex surface in pillowed

metabasalt, graded bedding in greywacke. Therefore,

the volcanic unit found in the western side is considered

to be older. A thin band of thinly layered chert bed

occurs interbedded with chlorite-sericite phyllite

overlying the eroded surface of the volcanic substratum.

This chert bed acts as a marker horizon while mapping.

In the greywacke-argillite suite numerous interbedded

sequences of greywacke and argillite occur alternately.

The competent beds of greywacke and the incompetent

argillite have moved past each other due to the shearing

contemporaneous with the thrusting along the NNW-

SSE directions. These planes being low tension gashes

(areas) might have acted as channels for the upcoming

mineralizing hydrothermal solutions.

1547 Ore Fluids Associated With the Metasediment Hosted Central




III. Central Auriferous Zone:

The central auriferous zone of Gadag gold field is

hosted by shear zones within greywacke-argillite and

carbonaceous phyllites and located closer to the contact

between metavolcanics and metasediments. The

auriferous zone in the area of investigation is

characterized by hydrothermal wall-rock alterations

represented by chloritisation, sericitization,

carbonatisation, carbonate spottings, quartz mylonites

and iron sulphide porphyroblasts as disseminations and

stringers. Another important character is ubiquitous

presence of carbonaceous matter with mineralized zone

especially in central auriferous zone. In general, the

rock type of central auriferous zone of Gadag gold field

can be classified based on the mineral assemblages,

texture and physical properties in hand specimen, the

crudely schistose and banded mesocratic rock. These

rocks are strongly deformed, altered, chloritised,

carbonated and silicified. At places, the carbonaceous

matter contains euhedral pyrite grains as well as veins

of pyrite. Carbonaceous matter also occurs as thin layers

along schistosity planes. Shear zones/fractures, contacts

between different litho units with competency contrast,

schistosity planes are better places for hosting the gold

bearing milky white quartz veins and lenses. About 15

cm to 1.50 m thick quartz veins with encrustations of

rusty brown ankerite patches and inclusions of

chloritised host rock are often observed to occur along

the shear zones especially between competent

greywacke and incompetent phyllite beds. This quartz

contain fine to medium grained sulphides as

disseminations which often may be invisible to naked

eyes. Quartz is medium to fine grained and deformed

which in thin sections exhibits development of quartz-

mylonites with elongation and orientation of grains with

undulose extinction and highly sutured grain

boundaries. Sometimes, it exhibits cherty nature. Quartz

occurs as clots of crystals as well as grains with sutured

margins and exhibits wavy extinction. It also occurs as

irregular shaped branching veins. At places, quartz

veins are seen cutting across carbonate veins. Often,

quartz and carbonates occur as filling spaces between

fractured sulphides. Occasionally, minute discrete

grains of gold are seen within the vein quartz. These

features indicate that the shear zones hosting auriferous

zone are ductile type. The different ore minerals and

gangue minerals are gold, arsenopyrite, pyrite,

pyrrhotite, chalcopyrite, sphalerite, galena, scheelite,

quartz, chlorite, plagioclase, sericite, carbonates and

carbonaceous matter.

IV. Fluid Inclusions:

Twenty doubly polished sections of 0.3 to 0.5mm

thickness were prepared from quartz samples spatially

associated with gold mineralization from the gold-

quartz sulphide reefs hosted within the metagreywacke-

argillite suite of rocks in central auriferous zone. In

some of the sections, the inclusions were too small to be

studied for microthermometry. In fact, tiny sizes of

inclusions are a characteristic feature of inclusions in

mesothermal gold bearing quartz veins (see Wilkinson,

2001). Only eight sections with good number of

inclusions and suitable for study were finally chosen for

microthermometry. Microthermometric determinations

were carried out by using Linkam THMSG 600

heating/cooling stage in the Department of Geology,

Karnatak University, Dharwad.

In the microthermometry, the density of CO2 inclusions

was calculated from the P-T plots for CO2 in the low

temperature range as suggested by Angus et al. (1976).

In case of aqueous inclusions, the degree of fill is

determined by visual estimation with the help of

standard figures (Shepherd et al., 1985) and salinity

values were calculated from the T-X plots for the low

temperature part of the system NaCl–H2O given by

Potter et al. (1978). The density of aqueous inclusions

was calculated from standard figure compiled by

Ahmed and Rose (1980). The trapping depths were

calculated with the help of Haas (1971) NaCl boiling

curve diagram.

Three types of inclusion namely Type I, CO2-rich

inclusions, Type II, H2O inclusions and Type III, CO2-

H2O inclusions are observed. Microthermometric

determinations are made only on Type III, CO2-H2O

inclusions.

Type I, CO2-rich inclusions are less in population when

compared to CO2-H2O inclusions. They occurs either as

isolated inclusions, scattered within quartz or as arrayed

inclusions indicating their primary to secondary nature.

Type II, H2O inclusions are scattered in quartz grains.

Mostly they are of primary type. Type III, CO2-H2O

inclusions occur as isolated as well as arrayed within

quartz grains and vary in size from 10 - 20 µm. The

volume proportion of CO2 in these is about 30 percent.

Microthermometric determinations were carried out for

these CO2-H2O inclusions.

The results of microthermometric determinations of

H2O-CO2 fluid inclusions from auriferous quartz veins

hosted within metasediments in Central auriferous zone

of Gadag gold field are given in Table 1.




Figure 2: Frequency Histogram of Th-Co2 from Microthermometric Data of H2o-Co2 Inclusions in the

Ore Quartz Veins from Central Auriferous Zone of Gadag Gold Field

Figure 3: Frequency Histogram of Tmco2 from Microthermometric Data of H2o-Co2 Inclusions in the


Table 1: Microthermometric Data of H2o-Co2 Inclusions in Auriferous Quartz Veins from Central Auriferous Zone,

Gadag Gold Field

Sample

No.

Tm CO2 0C

Tm ice 0C

Th CO2 0C

Tht CO2+H2O 0 0C

Salinity

Wt. % NaCl

Density

gm/cc

G1 -56.60 -2 to -16 -42.60 293.00 2.0 to 4.2 0.88 to 0.91

G2 -56.80 -3 to -17 -42.70 325.00 2.0 to 6.6 1.02 to 1.04

G3 -56.70 -2 to -26 -51.60 185.00 2.1 to 3.8 0.72 to 0.74

G4 -56.70 -1 to -17 -55.20 to -57.10 175 to 185 3.1 to 5.2 0.70 to 0.74





Figure 4: Frequency Histogram of Calculated Thtco2 from Microthermometric Data of H2o-Co2 Inclusions in the


Figure 5: Plots of Salinity V/S Temperature of

Homogenization for H2o-Co2 Inclusions in the Ore

Quartz Veins from Central Auriferous Zone of Gadag

Gold Field

The temperature of homogenization (Th) of CO2 varies

from -42.6 to -57.10C. In Th histogram for CO2,

maximum Th values are recorded at -51.60, -42.60, -

42.70 and -57.100C (Fig. 2). The temperature of melting

(Tm) of CO2 varies from -56.60 to -56.80 0C with

maximum Tm values at -56.70 and -56.800C (Fig.3).

The temperature of melting of ice (Tm of ice) varies

from -2 to -260C.

The total temperature or final temperatures of

homogenization of CO2-H2O inclusions [(Tht)

CO2+H2O] range from 1750C to 325

0C. However,

maximum numbers of investigated inclusions are found

at 1850C. It is noteworthy to point out the pronounced

skewness of the frequency histogram is closer to the

homogenization temperatures at 1750C, 293

0C and

3250C (Fig. 4).

The salinity v/s homogenization temperature plot (Fig.

5) indicates an isothermal mixing trend for inclusions

homogenizing from 1750C to 185

0C. However, elevated

homogenization temperatures of some inclusions

between 2930C and 325

0C probably represent post-

entrapment changes (Mernagh and Wygralak, 2007).

Figure 6: NaCl Curve Diagram after Haas (1971), For

the H2o-Co2 Inclusions in the Ore Quartz Veins from

Central Auriferous Zone of Gadag Gold Field

The presence of halite crystals as inclusions in the ore

vein quartz indicates that the ore forming fluids were

saline. The calculated salinity of these inclusions is low




with 2.0 to 6.6 wt % of NaCl equivalent, while the

density varies from 0.7 to 1.04 g/cc. These inclusions

are mostly trapped at a maximum depth of 1.3 km

(Fig.6).

V. Discussion and Conclusion:

A review of data available on temperature, pressure and

composition of hydrothermal solution suggest that

deposition of gold and sulphide in the Archaean

mesothermal hydrothermal gold deposits in the granite-

greenstone terrain principally occur between 250 and

4000C, salinity being 2 to 6 wt % NaCl equivalent

(Brown and Lamb, 1986; Groves, 1993 and references

there in).

Phase separation in a H2O-CO2-CH4-NaCl fluid system

is a mechanism which cause gold deposition in a variety

of environments and fluid inclusion studies have

demonstrated the existence of low-saline, immiscible

H2O and CO2 rich fluids related to ore in a number of

gold deposits (Robert and Kelly, 1987; Guha et al.,

1991; Wilkinson and Johnston, 1996; Mishra and

Panigrahi, 1999; Barrie and Touret, 1999).

The microthermometric measurements of fluid

inclusions in gold-quartz veins of central auriferous

zone of Gadag field indicate that the auriferous

hydrothermal fluids were of low salinity (2.0 to 6.6 wt%

NaCl), dominated by CO2-H2O (about 30 mole % CO2)

with moderate densities (0.7 to 1.04 g/cc) at a maximum

depth of 1.3 kms. Gold deposition occurred over a wide

temperature range of 175 to 3250C.

Metamorphism of volcano-sedimentary host rocks at

greenschist–amphibolite facies boundary with chlorite-

calcite-quartz assemblage produces large volumes of

low-salinity CO2-H2O fluids, similar in composition to

those recorded in the Archaean greenstone hosted vein

type lode gold deposits in general (Groves, 1993;

Kerrich and Cassidy, 1994 and references there in).

Carbonatization is one of the prominent wall rock

alterations in the auriferous zone hosted in

metasediments in Gadag field. Gold-quartz vein type

mineralization here is consistent with a structurally

controlled (shear zone), hydrothermal epigenetic type

(Ugarkar and Deshpande, 1999). Further, it has been

suggested that there is a close relationship between gold

mineralization and retrograde greenschist facies

metamorphism (Ugarkar, 1998). Retrograde

metamorphism and corresponding mineral assemblages

can be maintained during metamorphism in presence of

CO2 dominance (Clark et al., 1986). Thus, CO2 seems to

be an almost universal constituent of the ore fluids

depositing gold and it forms a major constituent of most

of fluid inclusions in gold ores from the metamorphic

environment (Roedder, 1984). It has been suggested by

Hutchinson (1993) that in such environment gold might

have been carried as carbonyl or carbonate complex and

that the extraction of CO2 from the ore fluids by

reaction with divalent cations in the wall rock to form

carbonates, would result in the precipitation of gold

(Roedder, 1984) within suitable structural sites (shear

zones), through a combination of decreasing

temperature and fluid-wall rock interaction. Progressive

carbonization of wall-rocks with decreasing temperature

and pressure might lead to fluid immiscibility and

separate H2O-rich and CO2-rich phases. These physical

separations of two immiscible fluids significantly

change the solubility of gold and thus cause

precipitation (Sibson and Scot, 1998; Groves and

Foster, 1993) in the form of quartz veins.

Gold appears to have been precipitated in veins as a

result of fluid immiscibility at low to intermediate

temperature range of 175-3250C in the sediment hosted

central auriferous zone of Gadag field. The

hydrothermal fluids responsible for the gold

mineralization in the metavolcanic hosted western

auriferous zone of Gadag field were also of low salinity

(2 to 6 wt% NaCl), dominated by H2O-CO2 (20 to 40

mole % CO2) with moderate densities (0.8 to 1.0 g/cc)

at temperature range of 240-2500C (Ugarkar et al,

2000), which are comparable with the sediments hosted

central auriferous zone of present study, however, with

a wide temperature range of 175-3250C.

References:

[1] Ahmed, S.N. and Rose, A.W. (1980) Fluid

inclusions in porphyry and skarn ore at Santa Rita,

New Mixico: Econ. Geol., V.75, pp.229-250.

[2] Angus, S., Armstrong, B., DeReuck, K.M., Altunin,

V.V., Gadtskii, O.G., Chapela, G.A. and

Rowlinson, J.S. (1976) International

Thermodynamic Tables of the Fluid State, Vol.3,

Carbon Dioxide: Pergamon Press, Oxford,

England, p.385.

[3] Barrie, I. and Touret, J.L.R. (1999) Fluid inclusion

studies of gold bearing quartz veins from the

Yirisen deposit, Sula Mountains greenstone belt,

Masumbiri, Sierra Leone. Ore Geol. Rev. V.14,

pp.203-225.

[4] Brown, P.E. and Lamb, W.M. (1986) Mixing of

H2O and CO2 in fluid inclusions; geobarometry and

Archaean gold deposits. Geochem, Cosmochim.

Acta.50, pp.847-852.

[5] Clark, M. E., Archibald, N.J. and Hodgson, C.J.

(1986) The structural setting of the Victory Gold

Mine, Kambalda, Western Australia. In:

Macdonald, A.J. (Ed) Proceedings of Gold’86,

Toronto, pp.243-254.

[6] Groves, D.I. (1993) - The crustal continuum model

for late-Archaean lode-gold deposits of the Yilgarn





Block, Western Australia. Mineral. Deposita, V.28,

pp.366-374.

[7] Groves, D.I. and Foster, R.P. (1993) Archaean lode

gold deposits. In: Foster, R.P. (Ed.) Gold

Metallogeny and Exploration. Chapman and Hall,

London, pp.63-103.

[8] Guha, J., Lu, H.Z., Dube, B., Robert, F. and

Gagnon, M. (1991) Fluid characteristics of vein and

altered wall rock in Archaean mesothemal deposits.

Econ. Geol. V.86, pp.667-684.

[9] Haas, J.L. (1971) The effect of salinity on the

maximum thermal gradient of a hydrothermal

system at hydrostatic pressure. Econ. Geol. V.66,

pp.940-946.

[10] Hutchinson, R.W. (1993) A multistage

multiprocess genetic hypothesis for greenstone

hosted gold deposits. Ore Geol. Rev., v.8.pp.349-

382.

[11] Kerrich, R. and Cassidy, K.F. (1994) Temporal

relationship of lode gold mineralisation to

accretion, magmatism, metamorphism and

deformation-Archaean to present: A review. Ore

Geol. Rev., V.9, pp.263-310.

[12] Mernagh, T.P. and Wygralak, A.S. (2007) Gold

ore-forming fluids of the Tanami region, Northern

Australia. Mineralium Deposita, V.42, pp.145-173.

[13] Mishra, D.C. and Panigrahi, M.K. (1999) Fluid

evolution in the Kolar Gold Field-Evidence from

fluid inclusion studies: Mineralium Deposita. V.34,

pp.173-181.

[14] Mukherjhee, A., Roy, G. and Tripathi, A. (2001)

Fluid inclusions associated with gold mineralization

in Kunchiganahalu banded iron formation,

Chitradurga Schist Belt, Karnataka: A preliminary

Appraisal. Jour. Geol. Soc. India. V.58, pp.533-

537.

[15] Narayanaswamy, S. and Ahmed, M. (1963)

Geology of Gadag gold field, Dharwar district,

Mysore State. Geol. Soc. India. Mem. 1, pp.107-

116.

[16] Pal, N. and Mishra, A. (2002) Alteration

geochemistry and fluid inclusion characteristics of

the greenstone-hosted gold deposit of Hutti, Eastern

Dharwar Craton, India. Mineralium Deposita. V.37,

pp.720-736.

[17] Potter, R.W., Clynne, M.A. and Brown, D.L.

(1978) Freezing point depression of aqueous

sodium chloride solutions. Econ. Geol, V.73,

pp.284-285.

[18] Roedder E. (1984) Fluid inclusions. Mineralogical

Society of America, Revs. In Mineralogy, V.12,

644 p.

[19] Robert, F. and Kelly, W.C. (1987) Ore forming

fluids in Archaean gold bearing quartz veins at the

Sigma Mine, Abitibi greenstone belt, Quebec,

Canada, Econ. Geol.V.82, pp.1464-1482.

[20] Santosh, M. (1986) Ore fluids in auriferous

Champion reef of Kolar, South India. Econ. Geol.,

V.81, pp.1546-1552.

[21] Shepherd, T.J, Rankin, A.H. and Alderton, D.H.M.

(1985) A practical guide to fluid inclusion studies.

Blackie, Glasgow.

[22] Sibson, R.H. and Scott, J. (1998) Stress/fault

controls on the containment and release of over

pressured fluids: Examples from the gold-quartz

vein systems in Juneau, Alaska; Victoria, Australia

and Otago, New Zealand. Ore Geol. Rev. Special

Issue, 13, pp.293-306.

[23] Solankar, S.N., Ugarkar, A.G. and Vasudev, V.N.

(2006) Characteristics of fluids associated with

polymict conglomerate hosted gold mineralization

near Surapalli of eastern part of Kolar greenstone

belt, Dharwar craton. Indian Mineralogist, V.40,

No.2, pp.156-169.

[24] Swaminath, J. and Ramakrishnan, M. (1981)

Present classification and correlation. In

Swaminath, J. and Ramakrishnan, M. (Ed.), Early

Precambrian Supracrustals of Southern Karnataka.

Geol. Surv. India, Mem., 112, pp.23-38.

[25] Ugarkar, A.G. (1998) Implications of retrograde

metamorphism for gold mineralization in the

greenstone belts of northern Dharwar craton,

Karnataka, India. Gondwana Res. V.l, pp.215-219.

[26] Ugarkar, A.G. and Deshpande, M.P. (1999)

Lithology and gold mineralization of Gadag gold

field, Dharwar craton - Evidences for epigenesis of

gold in diversified host rocks. Indian Mineralogist,

V.33, pp.37-52.

[27] Ugarkar, A.G., Suresh, K.J. and Srikantappa, C.

(2000) Fluid inclusions in the western auriferous

zone of Gadag Gold field, Karnataka. Indian

Mineralogist, V.34, pp.91-99.

[28] Wilkinson, J.J. and Johnston, J.D. (1996) Fluid

pressure fluctuations, phase separation and gold

precipitation during seismic fracture propagation,

Geology, V.24, pp.395-398.

[29] Wilkinson, J. J. (2001) Fluid inclusions in

hydrothermal ore deposits. Lithos, V.55, pp.229-

272.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1552-1559


A Case Study of Particle Size Distribution of Paleosols around

Sargur Supracrustal Terrain, Dharwar Craton, South India

NARGES GOHARI RAD, PRAKASH NARASIMHA. K. N and MADESH. P Department of Studies in Earth Science, University of Mysore, Manasagangotri, Mysore -570006, Karnataka– India


Abstract: An investigation was carried out to know the quantitative mineralogical composition of various grain size

fractions in soil samples collected fromBettadabidu, Sargur and Doddakanya(magnesite mines) around Mysore

district. The methods used for determination of particle size distribution are sieve analysis and hydrometer test.

There are different types of soils present in these areas and classified according to Folk’s classification system. The

majority of rock types of the research area have been identified namely calc silicate, gneiss and meta-pelite (garnet

biotite schist), amphibolite, banded iron formation and ultramafic rock. Grain size greater than 2mm was considered

as gravel, between 2 to 0.05 mm was sand and in between 0.05 to 0.002 mm was silt. The grain size lesser than

0.002 mm was considered as clay (USDA Classification System). Geological map of the study area has been

prepared with help of GIS tools.

Keywords: Paleosols, Grain Size, Soil Classification, GIS

1. Introduction:

Soil is the top most layer of the earth's crust and it is an

organized mixture of organic and mineral matter. Soil is

created by geologic processes and responsive to

organisms, climate, and contain four main components

and many process acts on these components, four major

components of soil are inorganic material (clay, silt,

pebbles, and sand), soil water, soil air, microorganisms

and decaying organic matter [1, 2].

Soil plays a vital role in land ecosystems; its formation

begins with the weathering of bedrock or the transport

of sediments from another area. These small grains of

rock accumulate on the surface of the earth. This

formation process is very slow (hundreds to thousands

of years), and thus soil loss or degradation can be very

detrimental to a community.

Ancient soil formation is a kind of early diagenesis,

which can be obscured by additional changes after

burial of the soil (late diagenesis), these fossil soil types

could be identified within modern soil classification.

Fossil soils are the evidence of past soil forming factors,

such as climate, organisms, topographic relief and

parent material and time of formation [3].

Most of paleosols have been buried in the sedimentary

record, covered by flood debris, landslides, volcanic

ash, or lava. Some paleosols, however, are still at the

land surface but are no longer forming in the same way

that they did under different climates and vegetation in

the past [4].

Soils are formed in direct contact with the atmosphere,

and their chemistry is strongly affected by interaction

with atmospheric oxygen and carbon dioxide.

Therefore, paleosols should be a good record of former

atmospheric composition [5].

Undisturbed soil has its own distinctive profile

characteristics which are utilized in soil classification

and survey and are of great practical importance. The

grain size distribution is commonly used for soil

classification; however, there is also potential to use the

grain size distribution as a basis for estimating soil

behaviour and process of weathering in past. Particle-

size distribution is a basic physical property of mineral

soils that affects many important soils attributes [6].

The different types of soil are found developed on

metasedimentary, meta Igneous rocks and Peninsular

gneiss exposed around Bettadabidu, Doddakanya

(magnesite mines) and Sargur areas belonging to

ancient supracrustals suite, Sargur schist belt. Since

different types of soil are found developed on different

types of rocks, the present study is an attempt to classify

the soils associated with the rock types of the study

area.

1.1 Geology of the Study Area:

The study areas are located between latitude 11˚59' 00"

to 12˚12' 02" N and longitude 76˚26' 20" to 76˚38'00" E

falling in Survey of India toposheets numbers 57 D/11

and 58A/9 on 1:50000 scale (Fig.1). The entire areas

have a semi arid climate. Mean temperature is 16ºC and

1553 NARGES GOHARI RAD, PRAKASH NARASIMHA. K. N and MADESH. P



mean maximum temperature is 34.6ºC and average

rainfall is 783.1 mm.

Rock types of the study area have been subjected to

various types of deformation and regional high-grade

metamorphism [7, 8, 9, 10, 11]. The rock types belong

to the older Sargur Group of rocks of Precambrian age

[12]. Numerous shear and fault zones have been

recognized in the area which are the weak planes along

which development of thick soil cover is noticed.

Geomorphologically the area is an undulating terrain

with no significant break in geomorphologic feature.

Generally exhibits a peneplane feature [13]. This

undulating terrain show slopes ranging from 2 to 10

degrees.

Figure 1: Geological Map of the Area around Sargur, Mysore District

Materials and Methods:

Extensive field work has been carried out to identify

different types of soils developed on calc silicate and

meta-pelite (garnet biotiteschiest) around Bettadabedu

village, Amphibolites and banded iron formation around

Sargurarea and ultramafic rock near

Doddakanya,magnesitemine in Mysore district. The

district resource map published by Geological survey of

India has been used for the preparation of geological

map of the study area (Arc GIS 9.3). Soil and rock

samples were collected from the profile representing the

massive hard rock at the base followed by weathered

rock and topsoil layer.USDA standard charts and Folk’s

classification system are used to classify soil types in

the study area and compared with the grain size

distribution of soil on different bed rocks. Sieving

technique has been used for fractionation of soil

samples [14].

All soil samples fractionated by sieving technique

following 10 grain size fractions [15].

The grain-size distribution of fine grained soils is

usually determined by performing routine tests such as

hydrometer test followed by Stokes’ law [16],

Weatherly [17] and Brittain[18].

Results and Discussion:

Soil size analysis was carried out based on USDA

Classification with sieving methods (ten different sizes

(Table 1). The soils have been classified and identified

in Folks triangles based on the percentage of gravel,

sand and silt, those particles having less than 75 micron

have been subjected to hydrometer analysis

1554 A Case Study of Particle Size Distribution of Paleosols around




(Stoke’sLaw) and percentage of sand, silt and clay are

determined based on gravity. The studied soils have

been classified and identified in Folks triangles based on

the percentage of sand, silt and clay [19].

Three soil profiles of Bettadabidu village have been

studied and top soils developed on calc-silicate rocks

analysed (Fig. 2). Sieve and hydrometric investigation

of horizon “A” of soil profiles of Bettadabidu village

showed that the soil profile of a and c areclassified as

muddy gravel(mg) and loam soil respectively, whereas

profile b is classified as gravely mud(gm) and loam

(Table. 2,3 and Fig. 3,4). Results of particle size and

density of all soil profiles, reported as percentages of the

total sample weight (Table. 4 and Fig. 5). Based on

grain size distribution by sieve, the histogram graph has

been made and showed multimodal distribution on

profile a andc and bimodal distribution in b (Fig. 6).

Extensive field studies have been conducted in Sargur

area and four soil profiles which developed on three

types of parent rocks namely amphibolites, meta-pelite

and BIF are examined. Sieve and hydrometric study of

horizon “A” determined in profile d and e which

developed on amphibolite are muddy sandy gravel

(msg) and loam respectively. In profile f, the soil has

developed on meta-pelite rock, identified as sandy mud

(sm) and loam respectively. The soils developed on BIF

in profile g have been classified as gravely muddy sand

(gms) and sandy loam. Histogram distribution of profile

d, f and gare determined as multimodal distribution and

in e, asbiomodal distribution.

In Doddakanya area one profile has been studied and

soil developed on ultramafic rock is identified as muddy

sandy gravel (msg) and loam based on Folk’s

classification. Soil size distribution of profile h

determined as multimodal.

In the studied area soils developed on the meta

ultramafic and mafic rock are similar and they range

from muddy sandy gravel (msg) to loam, whereas

metamorphosed sedimentary rocks like pelites and BIF

have yielded soil which vary from sandy mud(sm) to

loam in meta pelites to gravely muddy sand (gms) and

sandy loam in BIF. However calc–silicates rock in

different profiles show muddy gravel (mg), gravelly

mud (gm) and loam soil. The studied profile of soil

developed on different parent rock indicates that the

derivation of soil is controlled largelyby the parent rock.

Acknowledgment:

The authors are grateful to the Department of Civil

Engineering, Sri Jayachamarajendra College of

Engineering, Manasagangotri, Mysore for providing

IOE facilities for the present investigation. Chairman,

Department of Earth Science, university of Mysore is

gratefully acknowledged for constant encouragement

and help.

References:

[1] D. D. Richter and D. Markewitz, How Deep Is

Soil? (1995). Vol.45No.9, pp. 600-609.

[2] J. F.Petersen, D. Sack and R. E. Gabler, Physical

geography, (2011),p. 646.

[3] G. J. Retallack, A paleopedological approach to the

interpretation of terrestrial sedimentary rocks: the

mid–tertiary fossil soils of Badlands National Park,

South Dakota, geological society of America

bulletin, (1983), Vol. 94, pp. 823-840.

[4] G.J. Retallack, Pedogenic carbonate proxies for

amount and seasonality of precipitationin paleosols:

Geology, (2005), Vol.33, pp. 333– 336.

[5] J. B. Maynard, Chemistry of Modern Soils as a

Guide to Interpreting Precambrian Paleosols,

Journal of Geology, (1993), Vol. 100,No. 3, pp.

279-289.

[6] L.M.Arya,F.J. Leij, M.T.van Genuchten and P.J.

Shouse. (1999). Scaling parameter to predict the

soil water characteristic from particle-size

distribution data, Soil Sci. Soc. Am. J. 63:510–519.

[7] JanardhanaRao, L. H. Srinivasarao, Sargur Group

of rocks from South India. J. Geol. Soci. Of India,

(1978), Vol.43, pp.67-89.

[8] C.Srikantappa, M. Raith, and K. Hormann,

Petrology and geochemistry of layered Ultramafic

to Mafic rocks from the Archaen, South Indias,

Archean Geochemistry,(1984), pp. 139-160.

[9] K.N.PrakashNarasimuha, S. Kobayashi, T. Shoji

and M. Sasaki, XRD, EPMA and FTIR studies on

garnet from Bettadabidu, Sargur

area, Karnataka, India. Journal of Applied

Geochemistry,(2009), Vol.11, No.1, pp. 1-11.

[10] K.N. PrakashNarasimha,Ramalingaiah. H,

KarelMelka, Krishnaveni, P.S.R Prasad, C.

Krishnaiah, K.S Jayappa and A.V. Ganesha,

Vermiculite mineralization associated with

ultramafics in Agasthyapura area, Mysore District,

Karnataka state, India-A mineralogical study.

ActaGeodynamicaetGeomaterialia,(2006), Vol.3,

No.4, pp. 19-31.

[11] K.N. PrakashNarasimha,Krishnaveni, P.S.R.

Prasad,Ramalingaiah.H and J. S.

Venugopal,Vermiculite in the Gopalpura area

Karnataka –A mineralogical study. Journal

ofGeological Society of India, (2006), Vol.67,No.2,

pp.159-163.

[12] Ramahrishnan, M. and Vydydhyanadhan, S,

Geology of India JGSI publication,(2008), p. 160.

[13] KS. Valdiya.,Geology, Environment and Society,

Universities Press, Hyderabad, (2004), p.72.

[14] S. O. Wanogho, The Forensic Analysis of Soils

with Particular Reference to Particle Size




Distribution Analysis, Ph.D. thesis, University of

Strathclyde, (1985).

[15] W. D. Kemper and R. C. Rosenau, Methods on soil

analysis. Part 1. Physical and mineralogical

methods- agronomy monograph. No. 9 (2nd Edn.)

ASSA., USA,(1986).

[16] G. Stokes, Mathematical and physical paper III,

Cambhhge University Press, Cambridge, MA.S,

(1891),p. 597.

[17] W. C. Weatherly, The Hydrometer Method for

Determining the Grain Size Distribution Curve of

Soils, Master of Science thesis, Department of Civil

Engineering, Massachusetts Institute of

Technology, Cambridge, (1929).

[18] H. G. Brittain, Particle-Size Distribution, Part II,

Determination by Analytical Sieving,

Pharmaceutical Physics, (2002), Vol.26, No.7, pp.

67–73.

[19] R.L. Folk, Petrology of Sedimentary Rocks.

HemphillPublishing Co., Austin, TX, (1974), pp.

182.

Table 1: Sieve Size and Grain Size Classification

(for coarse-grained soils with D > 75 mm)

Sieve

No.

Sieve Size/

mm Category

4 4.75 Gravel

10 2.00

Sand

16 1.18

30 0.600

40 0.425

50 0.300

60 0.250

100 0.150

200 0.075

pan 0.000 Fines (Silts&Clays)

Figure 2: Different Soil Profiles of Study Area; a, b, c) Bettadabedu, d, e, f, g) Sargur and

h) Doddakanya,magnesite mines (Nanjangud), Whereas A- Top Soil (Horizon A), B- Highly Weathered Rock

(Horizon B), C- Semi Weathered Rock (Horizon C), D- Less Weathered Rock (Horizon D), E- Bed Rock (Horizon E)





Figure 3: Folk’s Classification System Based on Gravel (G), Sand (S) and Mud (M)

Table 2: Profile ID and Top Soil Classification (Sieve Analysis): a, b, c) Bettadabidu, d, e, f, g) Sargur and

h) Doddakanya, Magnesite Mine (Nanjangud)

SOIL DATA

Profile

ID Horizon

Percentage from material passing Classification

Gravel Sand Silt

Be.a A 37.50% 30% 32.50% Muddy gravel

Be.b A 27.50% 25.50% 47% Gravelly mud

Be.c A 34% 27% 39% Muddy gravel

Sa.d A 25.50% 50.50% 24% Muddy sandy

gravel

Sa.e A 59% 24% 17% Muddy sandy

gravel

Sa.f A 5% 43% 52% Sandy mud

Sa.g A 27.50% 54% 18.50%

Gravelly muddy

sand

Nan.h A 37.50% 40% 22.50% Muddy sandy

gravel




Figure 4: Folk’s Classification System Based on Sand, Silt and Clay

Table 3: Profile ID and Top Soil Classification (Hydrometer Analysis): a, b. c) Bettadabidu, d, e, f, g) Sargur and

and h) Doddakanya, magnesite mine(Nanjangud)

SOIL DATA

Profile

ID Horizon

material passing

Percentage Classification

Sand Silt Clay

Be.a A 40% 39% 21% Loam

Be.b A 42% 41% 17% Loam

Be.c A 41% 43% 16% Loam

Sa.d A 44% 40% 16% Loam

Sa.e A 48% 40% 12% Loam

Sa.f A 47% 39% 14% Loam

Sa.g A 62% 28% 10% Sandy loam

Nan.h A 47% 40% 13% Loam

Table 4: Grain Size Distribution of Horizon A (Hydrometer Analysis): PF% value of (Percentage of Fine Grain),

a, b, c) Bettadabidu, d, e, f, g)Sargur, h)Doddakanya, Magnesite Mine (Nanjangud).

Grain Size

(mm)

PF%

(Be.a. A)

PF%

(Be.b.A)

PF%

(Be.c.A)

PF%

(Sa.d.A)

PF%

(Sa.e.A)

PF%

(Sa.f.A)

PF%

(Sa.g.A)

PF%

(Nan.h.A)

0.057 99% 96% 97% 91.50% 85% 92% 63% 97%

0.0045 35% 28% 27% 26% 19% 30% 17% 27%





Figure 5: Hydrometer Graph of Grain Size Distribution): a, b, c) Bettadabidu, d, e, f, g) Sargur and

h) Doddakanya, Magnesite Mine (Nanjangud). PF %) Precentage of Fine Grain

Table 5: Grain size distribution of horizon A (Sieve Analysis): a, b, c) Bettadabidu, d, e, f, g) Sargur and

h) Doddakanya, magnesite mine (Nanjangud)

Size of

Sieve

(mm)

Weight

percent

Be.aA

Weight

percent

Be.b.A

Weight

percent

Be.c. A

Weight

percent

Sa.d.A

Weight

percent

Sa.e.A

Weight

percent

Sa.f.A

Weight

percent

Sa.g.A

Weight

percent

Nan.h.A

0.053 24.4% 38.3% 25.2% 18.2% 15.3% 41.8% 12.8% 16.4%

0.075 8% 8.8% 12.9% 5.3% 2.1% 10.4% 5.7% 6.1%

0.1 3.8% 5.2% 8.1% 4.3% 2.1% 7.1% 4.6% 5%

0.2 3.2% 2.6% 2.7% 5.4% 2.1% 5.1% 5.7% 5%

0.3 4.8% 2.6% 5.4% 9.7% 2.7% 8.7% 10.3% 6.1%

0.4 2.7% 2.6% 2.7% 9.1% 2.7% 5.1% 9.8% 3.9%

0.6 4.8% 3.6% 3.8% 13.4% 3.7% 7.6% 14.4% 6.7%

1 10.7% 8.8% 4.8% 8.6% 10.6% 9.7% 9.2% 13.4%

2 18.8% 11.9% 10.2% 9.7% 16% 2.5% 10.3% 22.3%

4 18.8% 15.6% 24.2% 16.1% 42.7% 2% 17.2% 15.1%




Figure 6: Histogram Chart of Grain Size Distribution): a, b, c) Bettadabidu, d, e, f, g) Sargur and and)

Doddakanya, Magnesite Mine (Nanjangud)

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1560-1566


Effect of Ground Moisture on Spread of Contaminants in Sand

Deposit -An Experimental Study

E. SAIBABA REDDY1, SINA BORZOOEI

2 and G. V. NARASIMHA REDDY

1

1Department of Civil Engineering, J.N.T.U.H. College of Engineering, Kukatpally,Hyderabad (A.P) 500085, INDIA

2Department of Civil Engineering, Indian Institute of Technology Bombay, Powai, Mumbai-400076, India

Email: [email protected], [email protected], [email protected]

Abstract: Contaminant transport modeling needs the understanding of the factors affecting the movement of

contaminant through dry, partially saturated and saturated zone. When the contaminant reaches the ground water, it

primarily follows the pattern of ground water flow in sand deposits. While the contaminant passing through dry or

partially saturated soil, its movement depends on a number of parameters such as porosity, soil moisture and

contaminant properties. This paper presents the details of a laboratory investigation carried out to study the spread of

Glycerol and engine oil, in dry and wet sandy soil. Experiments were conducted, using ‘Pollutant Transport Test

apparatus’ which was designed and developed for the project. The test results indicated that : the spread of an

aqueous phase liquid is greater in wet sand than in dry sand and the vertical spread of nonaqueous phase liquid was

less in wet sand than in dry sand, but the lateral spread of that was more.

Keywords: Contaminant spread in sand, Laboratory tests on contaminant transport, LNAPL, Glycerol, Engine oil,

Pollutant transport test apparatus.

I. Introduction:

Soil and ground water contamination takes place due to

various sources such as underground oil/chemical

storage tanks, spillovers, leakage of industrial waste

from deep injection wells, septic tanks, use of fertilizers

and pesticides in agricultural field. These contaminants

get transported, through soil and ground water, to the

surrounding places. Studies related to the movement of

contaminant through ground and ground water is termed

as contaminant transport studies [Bear (1972), Anderson

(1984), Qusim and Chiang (1994), Rowe and Booker

(1985) and Reddi and Inyang (2000)].Contaminants are

classified as aqueous phase liquids (which dissolve in

water) and nonaqueous phase liquids (NAPL) which do

not dissolve in water. NAPLs are further classified,

depending on their densities, as light nonaqueous phase

liquids (LNAPL) and dense nonaqueous phase liquids

(DNAPL). The immiscible (NAPL) fluids exhibit

different behavior and properties in the subsurface than

dissolved contaminant plumes. Dissolved plumes are

transported by advection, diffusion, dispersion,

adsorption, biodegradation and chemical reaction

mechanism, however, the migration of NAPL is

governed by gravity, buoyancy, viscosity and capillary

forces. The study of contaminant transport can be

divided into two phases: 1) Movement of contaminant

before reaching ground water and 2) Movement of

contaminant after reaching the ground water. When the

contaminant reaches ground water the contaminant

transport is governed mainly by the characteristics of

ground water flow as shown in Fig.1 [Fetter (1993),

CONCAWE (1981), Bedientet.al. (1994) and Fried

(1975)]. As can be seen in Fig.1, the contaminant moves

through dry and partially saturated soil

Figure 1: Contaminant Spread in Ground Water Flow

The contaminant transport or spread in dry and partially

saturated soil depends on the properties of the soil and

contaminant [Brooks and Corey (1964), Van Genuchten

(1980), Nielson et. al. (1986), Davidson et. al.(1966),

1561 E. SAIBABA REDDY, SINA BORZOOEI and G. V. NARASIMHA REDDY



Stephens (1994), Cooke and Mitchell (1991), Lenhard

and Parker (1990)]. Among them the major properties

which affect the contaminant transport are:

a) Viscosity of contaminant

b) Soil sorption

c) Permeability of soil

d) Moisture content of soil

e) Mode of introduction of contaminant

(Instantaneous or continuous)

An experimental investigation was carried out to study

the effect of ground moisture on the spread

characteristics of water and Glycerol and engine oil

contaminants. These tests were conducted using

pollutant transport test apparatus. The design details of

the apparatus are presented elsewhere. [Ravi (1999),

Reddy (2003), Reddy and Peter (2003)]. This paper

presents a brief description of “Pollutant Transport Test

Apparatus”, the testing procedure and test program

adopted in the investigation. This is followed by the

analysis and discussion of the test results. Finally a set

of conclusions is drawn on the soil moisture affecting

the contaminant spread in sand.

II. Apparatus and Properties of the Material:

In this investigation, a series of the laboratory tests was

conducted to study the effect of soil moisture on

contaminant (plume) spread in sandy soil. In order to

conduct these experiments, an experimental apparatus

was developed. In the following sections the design

details of the apparatus and the properties of the

materials used in the investigation were presented.

II. a. Apparatus Design:

The objective of the experimental investigation was to

observe and study the spread of the plume when a fluid

is introduced vertically in to the soil. Fig. 2 shows

different components of the test apparatus, which

consist of:

1) Test tank

2) Inlet tubes

3) Tracing paper

Test Tank:

The test tank was made up of a perplex glass sheets with

inside dimensions of 600mm x 300mm x 305 mm deep

(Fig. 2). All the five faces of the test tank were provided

with 2mm diameter perforations to facilitate the free

movement of air and fluid through the soil in tank.

Inlet Tubes:

The test fluid (pollutant) was introduced vertically

downward through a thin circular ring (called inlet tube)

with an internal diameter of 51 mm.

Tracing Paper:

After introduction of fluid and waiting for the adequate

time, the spread of the plume was traced at different

level, using transparencies, by gently removing the

layers of soil and tracing the plan of plume at each

elevation. To trace the plume a transparent tracing paper

having dimensions close to the plan size of the test tank

(595mm x 295mm) was placed on the sand surface, the

transparency was placed on it and trace the plan of the

plume.

Figure 2: Overall View of Test Apparatus

ІІ.b Properties of Materials:

In the investigation silica sand was used in dry and wet

conditions. The wet sand was prepared by adding 1.5 %

(by weight of dry sand) of water. Different fluids were

used as pollutants.

1562 Effect of Ground Moisture on Spread of Contaminants in Sand Deposit

-An Experimental Study



Properties of Sand:

The sand was tested as per the British standard BS:1377

for its physical and engineering properties. The test

results are presented in Table 1.

Properties of Fluid:

Water, glycerol and engine oil were used as the

contaminant fluids. The properties of the fluids used are

presented in Table 2.

Table1: Properties of Sand

Property value

Grain size distribution

property, D10 0.35mm

Specific gravity of soil

solids, G 2.644

Minimum density(γmin) 13.89 KN/m3

Maximum density(γmax) 16.48 KN/m3

Density 16.41 KN/m3

Void ratio 0.578

Density index (ID) 97.95%

Permeability 1.10 x 10-5

m/s

Angle of internal friction (φ) 41o

Table2: Properties of Contaminants

Fluid

Property

Density

(KN/m3)

Viscosity

(N.s/m2)

Water 9.8 0.001

Glycerol (85%) 11.95 0.147

Engine oil 8.65 0.242

III. Experimental Procedure:

The test procedure involves compaction of the sand,

introduction of the fluid, waiting for a period to give

time for the fluid to spread into the soil, and tracing of

the plan of the plume at the different levels.

ІІІ.a Compaction of Sand:

Initially 20 kg sand was poured into the tank and

leveled. This sand layer was then compacted with a

rammer (4 kg weight with a square base of 10 cm x 10

cm) such that the complete plan area was compacted

with overlapping blows. The procedure was repeated for

three more layers totaling to80kg. At this stage another

10 kg of sand was poured and compacted with the same

energy as that of earlier layers. With this process the

tank got filled. A final layer of 2 kg of sand was then

placed to get the surface perfectly level with the top of

the tank. The 92 kg (4x20kg+10kg+2kg) of sand was

therefore used to fill the tank gave an average density of

1.675g/cc. To prepare the wet sand bed, 10 kg of dry

sand was weighed and poured into a wide tray and 1.5%

by weight (150cc) of water was added to it and mixed

thoroughly. This wet sand was then transferred into the

tank. After transferring two such weights of sand, it was

compacted in similar fashion that as of the dry sand.

The procedure was completed in five layers

(4x20kg+10kg) by using 90kg of dry sand and 1.35kg of

water.

ІІІ.b Preparation of Fluid:

As the tests were conducted with various fluids and

designated volume (100cc) of fluid as contaminant. The

glycerol samples were prepared with 85% concentration

a day before the testing so that a homogeneous solution

can be used in the test.

ІІІ.C Introduction of Fluid:

A 51mm diameter circular ring was used to introduce

the fluid into the sand. To introduce the fluid, the center

of the sand surface was initially located. The inlet ring

was placed over the center of the sand surface and then

the fluid was introduced.

ІІІ. D spread Period:

After completing the introduction of the fluid, a

spreading time was given for each liquid depending

upon the viscosity of the fluid. For example for water 5-

10 min was sufficient, but for engine oil an hour or

more was given. After completing the spread period, the

plan of the plume was traced.

III.E Tracing of Spread:

After introducing the fluid, the plume spread was traced

at different depths below the sand surface. For this, the

tracing plate (as shown in Fig. 2) was placed centrally

over the sand surface. The tracing sheet (transparency)

was placed centrally over the tracing plate and the sand

surface. The plan of the plume, which could be seen on

top of the sand surface, was then traced using a marker

onto the tracing sheet. After tracing the plume at the top

surface, the tracer plate was removed and a thin layer of

the sand was gently removed from the tank such that the

plume portion was not disturbed. The tracing plate was

then re-placed into the tank and a new transparency

sheet used to trace the plume at the new depth. The

depth at which it was being traced was marked on the

transparency. This procedure was repeated until the end

of the plume or to the bottom of the tank, whichever

was reached first.

IV. Test Program:

The test program was conducted on dry and wet sand.

The fluids used for the contamination included water,

glycerol (85%) and engine oil. For the sake of

convenience each test was identified with an




identification name such as T1, T2, T3, T4, and T5. The

details of the tests are also presented in Table 3.

Table3: Test Details (Volume of fluid=100cc)

Test ID Soil condition Fluid

T1 Compacted Dry sand Water

T2 Compacted Dry sand Glycerol (85%)

T3 Compacted wet sand Glycerol (85%)

T4 Compacted Dry sand Engine oil

T5 Compacted wet sand Engine oil

V. Analysis and Discussion of Test Results:

Initially test, T1 was conducted on dry sand using water

(100 cc) as fluid. This fluid was introduced using 14mm

diameter inlet tube. The top view of the plume at

different depths below the sand surface was traced and

shown in Fig.3.Fig. 3(a) indicates the tracing that was

obtained at zero depth (i.e. at the top surface of sand).

Similarly, Fig. 3(b) is the top view of the plume traced

at 4 cm depth (d = 4 cm) below sand surface. Fig.3(c),

(d), (e), (f) represent the top view of the plume at 8 cm,

11 cm, 13.5 cm, and 15.5 cm depth below the sand

surface respectively and Fig. 4 shows the combined top

view of the plume at different depths. The area of each

top view was calculated using AutoCAD software and is

tabulated in Table 4. The curves in Fig. 3 are further

processed to obtain the vertical section of the plume. To

obtain the vertical section the maximum spread of each

of the plots in Fig. 3 were measured in XX and YY

directions as shown in Table 4. After obtaining the

spread at each elevation (Table 4) the spread with depth

can be plotted as shown in Fig. 5. It can be observed

from Fig. 5 that the plume reaches a maximum depth of

15.5cm below the sand surface. The data used for Fig.

5(a) and (b) was combined to obtain a volume spread of

plume with AutoCAD 3D tools (Fig. 6) and the volumes

of plume were computed. Similar computations were

made for tests T2 to T5 and the volumes of spread were

calculated, then spread of contamination in sand deposit

is computed and comp the volume of introduced

contaminant and expressed as a volumetric ratio (Rv),

which is defined as ratio of volume of contaminated

sand deposit to the volume of introduced contamination

fluid.

The volumetric ratio of various contaminant fluids were

calculated to know the volume of spread and the results

are tabulated in Table 5.

Effect of Soil Moisture:

A set of experiments were conducted to study the effect

of moisture content of the sand on the pollutant spread.

To analyze the effect of soil moisture the tests T2, T3, T4

and T5 are analyzed. T2 and T3 are the tests conducted

on dry and wet sand respectively using glycerol (85%)

as the contaminant fluid. These test results are

combined in Fig.7 (a). Similarly, tests T4 and T5 were

conducted in dry and wet sand respectively using engine

oil as the contaminant fluid. The combined results of T4

and T5 are presented in Fig.7 (b).

Table 4: Spread Area of different Depth (Test T1)

Depth

(cm)

Spread in XX

direction (cm)

Spread in YY

direction (cm)

Area of

tracing

results

(cm²)

0 (-)0.7 to (+) 0.8 (+)0.4 to (-) 0.8 1.458

4 (-)2.8 to (+) 2.8 (+)2.0 to (-) 3.6 26.83

8 (-)3.3 to (+) 3.5 (+)2.8 to (-) 4.5 40.24

11 (-)3.3 to (+) 3.2 (+)2.8 to (-) 3.9 35.41

13.5 (-)2.0 to (+) 1.3 (+)0.8 to (-) 3.0 11.62

15.5 (+)0.2 (-)0.5 0.06

Table 5: Volume of Plume

Test ID Volume of contaminated

area(cm³)

Volumetric

ratio (Rv)

T1 380.24 3.8

T2 468.53 4.7

T3 1064.55 10.6

T4 637.12 6.4

T5 694.39 6.9

From Fig.7 (a) it can be observed that the spread of

Glycerol plume (both vertically and laterally) is more in

wet sand compared to dry sand. Further from Table 5 it

can be observed that , Volumetric ratio in dry sand is

4.7 and it increased to 10.6 in wet sand, resulting in

125% increase in volume of glycerol spread in wet sand

compared to dry sand. This is because, when glycerol is

introduced, it mixes with moisture content (water)

present in sand and results in decrease of viscosity of

fluid. This reduction in viscosity will result in increase

in permeability and hence free movement of fluid and

more spread.

The spread of the engine oil (which is light nonaqueous

phase liquid) is greater in the vertical direction for the

dry sand than the wet sand. This is because, in wet

condition NAPL acts as a nonwetting fluid that must

overcome capillary forces in order to squeeze through

the pores filled with wetting fluid (i.e. water), because

the capillary forces, repel the movement of NAPL and





results in decreasing downward migration. Whereas in

dry condition NAPL migrate downwards relatively

easily into the soil media, whose pores are largely filled

with air because capillary forces do not exist to repel the

movement of NAPL (Fig.7 (b)). Volumetric ratio in dry

sand is 6.4 and it increased slightly to 6.9 in wet sand,

resulting in just 8% increase in volume of engine oil

spread in wet sand compared to dry sand.

Figure 3: Tracing Results for Test T1




D = Depth below the Sand Surface

Figure 4: Tracing Results for Test T1 in Different

Depths from Sand Surface

Figure 5 (a): Vertical Spread (Xx Direction)

Figure 5 (b): Vertical Spread (Yy-Direction)

VI. Conclusion:

Based on the experimental investigation conducted and

the analysis of results following major conclusions are

drawn on the effect of moisture content on aqueous and

nonaqueous contaminant spread.

1. The vertical and horizontal spread of an aqueous

phase liquid is more in wet sand than in dry sand.

Volumetric spreading of aqueous phase liquid increases

with increase in the moisture contents of the soil.

2. The volumetric spread of Glycerol-85% (aqueous

phase liquid) in wet sand is 125 % more than that of in

dry sand.

3. The vertical spread of nonaqueous phase liquid is

less in wet sand than in dry sand, but the lateral spread

of that will be more.

4. Volumetric spread of Engine oil (nonaqueous phase

liquid) in wet sand is marginally (8%) more than

volumetric spread in dry sand.

Figure 6: Shape of Contaminated Plume (3D)

Figure 7 (a): Effect of Soil Moisture (Glycerol, T2 and

T3)

Figure 7 (b): Effect of Soil Moisture (Engine oil,

T4 and T5)

VII. Future Scope of Research:

The present paper presents the pollutant transport/spread

in soil in dry and wet conditions. The number of factors

influencing the pollutant spread includei) type of soil





cohesive and non-cohesive soil, ii) viscosity and density

of fluid, iii) density of soil. The spread in layered soil

also help in predicting the behavior in real life

conditions. Investigation of these parameters will

provide more understanding on the pollutant transport.

Further all these studies reported and suggested above

are for the point source of contamination. There is a

very large scope for study for the estimation of pollutant

spread with continuous source of pollutions. Hence this

paper is a binging of methodical study initiated to obtain

comprehensive knowledge on pollutant transport studies

and modeling.

Acknowledgements:

The authors express their gratitude to the learned

Reviewers speciallyDr.PR.Reddy,Scientist (Retd),

NGRI, for their valuable comments and suggestions

which improved the present work to a great extent.

Reference:

[1] Anderson, M. P. 1984. Movement of contaminants

in ground water: ground water transport-Advection

and dispersion in ground water contamination.

National Academy press, Washington, D. C. 37-45.

[2] Bear, J. 1988. Dynamics of fluids in porous media.

Dover publ., Inc., New York.

[3] Bedient. B. Philip, Rafi S. Hanadi and Newell

J.Charles 1994. Groundwater contamination-

transport and remediation. PTR prentice-Hall In.,

New Jersey, USA.

[4] Brooks, R. H., and Corey, A. T. 1964. Hydraulic

properties of porous media. Hydrology Papers,

Colorado State University, 27.

[5] CONCAWE 1981. Revised inland oil spill clean-up

manual. Report no: 7/81, Den Haag.

[6] Cooke, A. B., Mitchell, R.J. 1991. Evaluation of

contaminant transport in partially saturated soils.

Centrifuge '91, Boulder, Colorado, 503-508.

[7] Davidson, J. M., Nielson, D. R., and Biggar, J.

1966. The dependency of soil water uptake and

release upon the applied press. Increment. Soil

science society Amer. journal, 30 (3), 298-304.

[8] Fetter, C. W. 1993. Contaminant hydrology.

Macmillan Publishing Company, New York.

[9] Fried J. Jean 1975. Ground water pollution.

Elsevier scientific publishing company,

Amsterdam. Lenhard, R. J. and Parker, J. C. 1990.

Estimation of free hydraulic volume from fluid

levels in monitoring wells. Groundwater, 28, 57-67

[10] Nielsen, D. R., Van Genuchten, M. Th. and Biggar,

J. W. 1986. Water flow and solute transport

processes in the unsaturated zone. Water Resource.

Res., 22(9), 895-1085.

[11] Qasim, Syed R. and Chiang, Walter 1994. Sanitary

Landfill Leachate Generation, Control and

Treatment. Technomic Publishing A.G, Basel,

Switzerland.

[12] Ravi, K. 1999. Design and development of

laboratory test apparatus to study the behavior of

contaminant transport. M. Tech. thesis, submitted

to Jawaharlal Nehru Technological University,

Hyderabad, India.

[13] Reddy Saibaba , E. 2003. An experimental

investigation of contaminant spread in sand. Post

doctoral research report, submitted to the

department of civil Engineering, The University of

Birmingham, U. K.

[14] Saibaba Reddy, E. and Nirmala Peter, E.C. 2003.

Effect of soil moisture on pollutant spread in sandy

soil. Indian Geotechnical Conference, 18-20

December 2003, Indian Institute of Technology

Roorkee , 221-224

[15] Reddi Lakshmi N. and Inyang Hilary. I. 2000.

Geoenvironmental Engineering principles and

applications. Marcel Dekker Inc., New York.

[16] Rowe, R. K. and Booker, J. R. 1985. 1-D

contaminant migration in soils of finite depth.

Journal of Geotechnical Engineering, ASCE,

111(4), 479-499.

[17] Stephens. D. B. 1994. Hydraulic conductivity

assessment of unsaturated soils. Hydraulic

conductivity and waste contaminate transport in

soil, ASTM, STP 1142, David E. Daniel and

Stephen J. Trautweinn, Eds., ASTM, Philadelphia.

[18] Van Genuchten, M. Th. 1980. A closed form

equation for predicting the hydraulic conductivity

of unsaturated soils. Soil scientific society Amer.J.,

44, 892-898

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1567-1571


Identification of Rain/No-Rain Events using QuikSCAT

Scatterometer Data

PRITI SHARMA1, RAJESH SIKHAKOLLI

2, B. S. GOHIL

2 and ABHIJIT SARKAR

2

1National Centre for Medium Range weather Forecasting, New Delhi

2Space Applications Centre, Ahmadabad

Email: [email protected], [email protected], [email protected],

[email protected]

Abstract: The Ocean surface vector winds derived from Ku-band scatterometer measurements are known to be

affected by rain. For rain-free situations, the empirical relationship used for wind retrievals have the power-law and

the bi-harmonic dependencies of the radar backscatter on wind speed and wind direction, respectively, is modified

under rainy situations. This leads not only to the changes in the derived winds but also to the modifications of the

retrieval cost function in the presence of rains. Utilizing the change in the characteristics of the backscatter

measurements as well as derived winds from those under the rain-free situations, a new attempt has been made for

the identification of rain/no-rain events using QuikSCAT measurements. Based on the proposed rain-flagging

approach, rain identification for QuikSCAT Level 2A backscatter data over global oceans for the month of

November, 2009 has been compared with TRMM-PR Level 2A25 near surface rain data within 0.250×0.25

0 and

within 10 minutes. Similar comparison of QuikSCAT Finished Product (FP) rain-flags has Been Performed using

TRMM-PR Level 2A25 rain for the same period. For rain free situations, the false reporting of rain events has been

reduced to 0.4% from 0.8 % found for QuikSCAT FP rain-flags for inner swath. Performance for identification of

missing rain events under rainy situations has also been reported.

Keywords: QuikSCAT, TRMM-PR, Scatterometer, Rain Flagging, Performance

Introduction:

A microwave scatterometer measures the normalized

radar backscattering cross-section of the ocean surface

for deriving ocean surface wind vectors by making use

of suitable relationship between radar backscatter and

wind vector which is known as Geophysical Model

Function (GMF). In the presence of rain, retrieval of

ocean surface wind vector from scatterometer is

affected due to modifications in the measured radar

backscatter in many ways. Rain drops falling on the

ocean surface create crater, crown, stalk and ring waves.

Out of these different forms, ring waves are the main

contributor for the changes in the radar backscatter

through the mechanism of Bragg resonance [1]. Due to

the isotropic nature of the ring waves, backscatter

signals are depolarized causing diminishing difference

of radar backscatter in vertical and horizontal

polarizations. Apart from this, rain attenuates the radar

signal as well as contributes through the volume

backscattering. Exploiting this impact of rain on the

characteristics of measured radar backscatter and the

retrieval of wind vector, a preliminary attempt for

identification of rain/no rain situations has been made

by using QuikSCAT data over global oceans during

November, 2009 and the corresponding winds from

European Center for Medium-Range Weather Forecast

(ECMWF) model. The QuikSCAT Finished Product

(FP) rain flag is based on Multidimensional Histogram

(MUDH) algorithm [2], which utilizes various

parameters like rain probability, normalized beam

difference, and rank-1 wind vectors in a complex way

[3]. We have used TRMM-PR rain data to compare the

skills of rain flagging of present approach using

QuikSCAT data along with those of QuikSCAT FP.

Theoretical Basis:

As mentioned above, we have attempted rain-flagging

of scatterometer data primarily utilizing the changes

caused by rain in derived wind vector and the cost

function relative to rain-free situations. Before its

implementation, we have studied the impact of rain on

wind retrieval. For which, rain-affected radar

backscatter data has been simulated at QuikSCAT

observational geometry using simulated wind vectors,

QuikSCAT-1 GMF and the rain impact model [4].

Using simulated wind vectors and rain affected radar

backscatter (as mentioned above) wind vector solutions

have been derived by employing the Normalized

Standard Deviation (NSD) based wind retrieval

algorithm [5]. The set of wind vector solutions comprise

1568 PRITI SHARMA, RAJESH SIKHAKOLLI, B. S. GOHIL and ABHIJIT SARKAR



of wind speed, wind direction and the NSD cost

function and are prioritized according to increasing

NSD values. The rank-1 wind vector solution has

minimum NSD value. The impact of rain on wind-

retrieval has been studied which indicates that the Rank-

1 wind speeds below 10 m/s are over estimated due to

increase in backscatter caused by volumetric scattering

due to rain and the enhanced ocean surface roughness.

For higher intensity winds above 10 m/s, the derived

winds are under estimated due to decrease in backscatter

caused by attenuation due to rain with no additional

change in ocean surface roughness. Apart from this, it is

seen that retrieved wind directions get aligned toward

the direction normal to the sub-satellite track possibly

due to the directional symmetry of ring waves, the

surface roughness becoming increasingly isotropic. It is

found for the rain free situations that the trend of

variation of the Rank-1 SD cost function (product of

Rank-1 NSD value and rank-1 wind speed) initially

decreases slightly and then increases with increasing

rank-1 wind speed. For rainy situations, the Rank-1 SD

has been found to be much higher as compared to those

of rain free cases for the same Rank-1 wind speed.

However, at high wind speed there is not much

difference in SD values for rainy and rain-free cases

causing false or missing rain events, as discussed later.

Based on the aforementioned simulation study related to

the impact of rain on the radar backscatter and

associated wind retrieval, a set of five conditions for

rain-flagging has been proposed that are described in the

next section and are found to have potential of

identifying the rain/no rain situations in scatterometer

data. Moreover, based on the experiment comprising of

simulation of radiometric brightness temperature at Ku

band using a radiative transfer model using different

wind speed and rain rate, it has been found that in rain

free situations brightness temperature marginally

increases with increasing wind speed, but in the

presence of rain, brightness temperatures are found to

be much higher than those for rain free cases. More so,

H-pol brightness temperature shows higher variations as

compared to that of V-pol brightness temperature. Thus,

H-pol radiometric brightness temperature has been used

in our analysis.

Methodology:

To implement the rain-flagging scheme the rain-free

radar backscatter data has been simulated by using

simulated wind speed by varying it from 1 to 25 m/s

with an interval of 1 m/s, wind direction within a range

of 0 to 3500 within an interval of 10° at a varying

observational geometry of QuikSCAT for all the wind

vectors cells over the inner beam swath through

QuikSCAT-1 Geophysical Model Function (GMF) and

10 percent noise has been added (as real data is

erroneous). For rain-free conditions, using simulated

noisy radar backscatter data and the QuikSCAT -1 GMF

(as mentioned above), the wind vector solutions have

been derived by using NSD algorithm. Keeping in view

the impact of rain on retrieved wind vector solutions as

discussed before, for each Rank-1 wind speed bin

(typically 1 m/s), the average and the standard deviation

of SD parameter (modified cost function) have been

estimated for the Rank-1 wind vectors as depicted in

Figure 1. Binned mean and standard deviation of Rank-

1 SD denoted by SDavg and SDsd are modeled as

polynomial functions of Rank-1 ws respectively and are

expressed as

SDavg = ∑=

3

0K

Ak (ws)k ………..(1)

SDsd = ∑=

3

0K

Bk (ws)k

………..(2)

where, Ak and Bk are the respective polynomial

coefficients for SDavg and SDsd, k is the degree of the

polynomial. On the basis of simulation study discussed

earlier for rainy conditions the Rank-1 wind speed are

found to be higher than 7 m/s, the Rank-1 wind

direction is found to be about 15° around the sub-

satellite track and the Rank-1 SD is found to be higher

for rainy conditions for the same wind speed as

compared to those for rain-free conditions. Hence,

these criteria have been chosen to identify the data

indicating rain. The difference of the derived Rank-1

SD value (from test data) from the calculated rain-free

SDavg, using eq. (1) should be greater than a certain

threshold value calculated using SDsd (from eq. (2)), as

given below

|SD-SDavg | > 0.7. α.SDSD………..(3)

The SD threshold multiplier α is used for controlling the

false rain reporting and missing rain events under rain-

free and rainy conditions, respectively. After testing the

above three conditions on simulated data, comprising of

rainy and rain-free conditions, it has been observed for

the rain-free cases that the false reporting of rain cases

is about 2 percent of the total rain-free cases while for

the rainy cases, the missing rain events are about 30 %

of the rainy cases. The Occurrence of false and missing

rain events possibly for high wind speed is attributed

due to overlapping Rank-1 SD values for rainy and rain-

free cases. In order to further improve the rain-flagging

performance under rainy and rain-free conditions, we

have additionally used the parameters indicating the rain

impact on the characteristics of radar backscatter and

the brightness temperature.

Using QuikSCAT Level 2A radar backscatter data for

the period of November, 2009, under rain-free and rainy

conditions, categorized based on the wind vector quality

flag available in QuikSCAT Level 2B product it has

1569 Identification of Rain/No-Rain Events using QuikSCAT Scatterometer Data



been found that the measured backscatter for both H and

V-polarizations for rainy cases are higher as compared

to those of rain-free cases for the same wind speed

obtained from ECMWF model (figures not shown here).

So, based on these observations, the rain-flagging

scheme has been updated by making use of the root

mean square difference of azimuth averaged (fore and

aft beams) measured backscatter and those simulated

using ECMWF wind vectors. To achieve this, a set of

dual polarized backscatter for rain-free cases is

simulated using simulated wind vectors, QuikSCAT-1

GMF, and the QuikSCAT observational geometry. The

root mean square difference of radar backscatter σrms is

calculated as,

σrms=[0.5{(σavg vm-σ

avg vs )

2+(σ

avg hm-

avg hs )

2}]

1/2………..(4)

where σavg vm, σ

avg vs , σ

avg hm and σ

avg hs are the azimuth averaged

measured and simulated backscatter for V and H-

polarization, respectively. For rain-flagging, mean σavgrms

and standard deviation σsd rms of rain-free σrms are

calculated for all wind speed bins of 1 m/s each from of

ECMWF data. These parameters are modeled as

polynomial function of Model wind speed as given

below

σavgrms=∑

=

4

0I

CIP(mws)I………..(5)

σsd rms=∑

=

4

0I

DIP(mws)I………..(6)

where, CIP and DIP are the respective polynomial

coefficients of σavgrms and σ

sd rms, I is the degree of

polynomial and mws is ECMWF Model wind speed.

For any given measurement, the difference of σrms

calculated by using eq (4) and from eq (5) should be

greater than the threshold value calculated for the rain-

free cases as expressed below

| σrms-σavgrms | > β.σ

sd rms………..(7)

where β is a threshold parameter used to optimize the

rain flagging performance. Based on simulations, the

change in characteristics of σrms for rain-free and rainy

cases are shown in Figure 2(a) depicting its potential.

Furthermore, as mentioned in the previous section, the

change in H-pol brightness temperature (Tbh) observed

to be higher compared to V-pol brightness temperature,

the use of the H-pol brightness temperature has been

included in rain identification. For rain-free cases, the

variation of mean Tbavgh and standard deviation Tbsdh of

H-pol brightness temperature with ECMWF wind

speed, as obtained from QuikSCAT data, shown in

Figure 2(b), are calculated as

Tbavgh=∑=

3

0K

Ek(mws)k

………..(8)

Tbsdh=∑

=

3

0K

Fk(mws)k

………..(9)

where, Ek and FK are the polynomial coefficients of the

mean and standard deviation of the Tbh, k is the degree

of polynomial and mws is the ECMWF wind speed. For

any measured inner beam brightness temperature Tbh,

the difference of Tbh from the Tbavgh calculated using

eq. (8) should be greater than rain-free Tbsdh values

calculated using eq. (9) by a threshold value given as

| Tbh - Tbavgh | > γ. Tbsdh………..(10)

where γ is a threshold multiplier for H-pol brightness

temperature and is used to optimize the false reporting

and missing rain events under rain-free and rainy

conditions respectively.

Figure (1): Variations of Binned Mean and Standard

Deviation of Rank-1 Sd with Varying Wind Speed for

Rain-Free Conditions

Results and Discussions:

The aforementioned preliminary rain-flagging approach

has been tested with QuikSCAT data for the month of

November 2009 and its performance is evaluated by

using TRMM-PR data collocated within a grid of 0.25 o

×0.25

o and with a time difference of 10 minutes. The

grids of TRMM-PR are considered to be rainy in which

more than 50% data points indicate rain. The rain-

flagging scheme has been first trained by using

QuikSCAT data for the period of November 2009, for

calculating the polynomial coefficients for σavgrms, σ

sd rms,

Tbavgh, Tbsd , SDavg and SDsd parameters by using

QuikSCAT Level 2A data, ECMWF winds and wind

vector quality flag available in QuikSCAT L2B data for

the same period. Then, the rain-flagging scheme has

been tested with aforementioned QuikSCAT data over

1570 PRITI SHARMA, RAJESH SIKHAKOLLI, B. S. GOHIL and ABHIJIT SARKAR



inner beam swath and its performance for rain-free and

rainy cases has been evaluated in terms of false

reporting and missing rain events based on TRMM-PR

data with typical values of threshold multiplier α, β and

γ. As we increase the threshold multiplier α for σrms

from 0.5 to 2.5 keeping the values of β and γ constant

(within the range of 0.5 to 2.5), respectively, the false

reporting for rain-free condition increases with a

decrease in the missing rain events for rainy situation.

Since more than 90% of the global backscatter data

corresponds to rain-free situations hence our concerned

should be more towards reducing the false rain

reporting. The results related to false reporting and the

missing rain events are depicted in Table (1). Similarly,

rain-flags for QuikSCAT FP have also been compared

with TRMM-PR rain data. The comparison of the

performance of the proposed rain-flagging approach and

the QuikSCAT FP rain flags with TRMM-PR rain is

summarized in Table (1) indicating reduction in false

rain reporting over inner beam swath at the cost of

increasing missing rain rate as compared to that of

QuikSCAT FP and vice-versa. Therefore, the overall

success (considering both rain and rain-free cases) of

the presented rain flagging approach is more or less

similar to that of QuikSCAT FP. Thus, present approach

indicates its potential towards rain flagging of

scatterometer data and it is inferred that the proposed

rain-flagging approach tested with the QuikSCAT data

has yielded satisfactory results.

Conclusions:

The proposed approach comprising a set of conditions

provides improved identification of rain-free conditions

and thus facilitates larger statistics of wind retrieval

case. Scatterometer however is found to miss some

rainy events.

Acknowledgement:

The authors would like to thank Dr. R. R. Navalgund,

Former Director, Space Applications Centre, for the

constant encouragement. We thankfully acknowledge

NASA-JPL for QuikSCAT and TRMM-PR data used in

our analysis.

References:

[1] Contreras and Robert F., (2003), “Effects of rain on

Ku- band backscatter from the ocean”, J. Geophy,

Res., Vol. 108, C5, 3165, doi:

10.1029/2001JC001255, pp 34-1 to 34-15

[2] Huddleston, J. N. and B.W.Stiles, (2000),

“Multidimensional histogram rain-flagging

technique for Sea Winds on QuikSCAT”, Proc.

IGARSS2000, vol.3, pp.1232-1234

[3] QuikSCAT Science Data Product User’s Manual,

(2006), version 3.0, JPL Document D-18053-Rev

A, Jet Propulsion Laboratory, Pasadena, CA.

[4] Draper, D.W. and D.G. Long, (2004b), “Evaluating

the effect of rain on SeaWinds scatterometer

measurements”, J. Geophy. Res., Vol. 109,

C02005, doi: 10.1029/20021C001741, pp 1-12

[5] Gohil, B.S., Abhijit Sarkar and V.K. Agarwal,

(2008), “A new algorithm for wind vector retrieval

from scatterometer”, IEEE Geosci and Remote

Sensing Letters, vol. 5, No. 3, July, pp 387-391

Table 1: Rain Flagging Performance of the Qscat Data for the Period of November 2009 Collocated with TRMM

PR Data. Grids of 0.25 o ×0.25 o are Considered Rainy, If More than 50% Data Points Indicate Rain

False Rain Reporting

(%)

Missing Rain Events

(%)

Overall Success

(%)

FP Proposed FP Proposed FP Proposed

0.87 0.48 47.72 65.78 97.42 97.13

Total TRMM PR and QuikSCAT Collocated Data = 25948

Rain-free Cases = 25001, Rainy Cases = 947

1571 Identification of Rain/No-Rain Events using QuikSCAT Scatterometer Data



Figure (2): Trend of Mean and Standard Deviation of (a) RMS difference of Backscatter with ECMWF Wind Speed

(b) H-polarized Brightness Temperature with ECMWF Model Wind Speed

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012, P.P.1572-1577


Genesis of Thermal Springs of Odisha, India

S. C. MAHALA1, P. SINGH

2, M. DAS

2 and S. ACHARYA

3

1Hydrogeological Unit, RWS&S PMDI Circle, Lane-II, Unit-III, Bhubaneswar-751001, Odisha, India

2P. G. Dept. of Geology, Utkal University, Bhubaneswar-751004, Odisha, India 3Plot No- 155, VIP Colony, Nayapalli, Bhubaneswar-751015, Odisha, India


Abstract: Thermal springs are considered as energy centric natural resource emitting earth’s outlet mechanism .

The geothermal energy is outlined as a combination of heat, water, gas source, and mineral component. Thermal

springs of Odisha are more or less aligned in specific tectonic lineament of the region and may be related to

structural disturbance. They are mostly confined along the periphery of the Gondwana Graben of Odisha. The

chemical constituents of the water are usually related to interaction between water and rock. The source of water is

accepted as meteoric. The heating process is due to geothermal gradient, decay of radioactive materials and

exothermic reactions. The associated gaseous phase is related to the decay of radioactive elements and organic

matter. The paper discusses the genesis of the thermal springs of Odisha considering these various parameters.

Keywords: Thermal Spring, Genesis, Odisha

Introduction:

Thermal springs are the springs that discharge water at

temperatures noticeably higher than the atmospheric

temperature of the surrounding and ascend from great

depth. They are natural phenomena linked to earth’s

internal energy; geothermal energy. They are named as:

hot springs, mineral springs, magic water, geysers,

fumaroles etc depending on the nature, character and

mode of manifestation on the earth surface. In a

geothermal field, the water percolates into sub surface

fractures and then gets heated by the internal heat. The

natural geothermal gradient, decay of radioactive

materials and exothermic reactions influence the heating

process. Later on the hot water and steam escape to the

surface of the earth along the fractures. Origin of

thermal springs deals with source of heat, water, gas and

mineral constituents. Allen and Day (1935), Day (1939),

Barth (1950) and Ellis & Mahon (1977) have made

valuable contributions on the origin of thermal springs.

An attempt is made to discuss the origin of thermal

springs of Odisha, on the basis of data documented

during the present study. A conceptual model has been

proposed.

Geological Setting and Tectonic Framework:

Manifestations of thermal springs are traced at eight

places in Odisha till date. The springs are located within

the Eastern ghat mobile belt, North Odisha craton and

western Odishan Proterozoic sedimentary basins (Fig

1). The thermal spring water is discharged either from a

single spot at Attri, Taptapani, Magarmuhan, Bankhol

and Boden areas or from several spots, clustered

together at Tarabalo, Deuljhori and Badaberena areas.

The highest number of spots have been recorded at

Tarabalo thermal spring area.

The thermal springs are mainly confined to crystalline

schists or gneissic terrain of Eastern ghat Supergroup

and Iron Ore Supergroup.However, Boden is located

within Vindhyan Supergroup. The litho-types

encountered at Attri, Tarabalo, Taptapani and Deuljhori

are khondalite, charnockite, augen-gneiss (granitic)

from Eastern ghat Super group. These rocks have

suffered highest grade of regional metamorphism and

are intricately folded. Quartzites of Iron Ore Super

group are exposed around Magarmuhan and Bankhol

region. The Precambrian rocks are considered to be the

repository of radioactive minerals (Ghosh, 1954). They

do not have any volcanic association as there is no

recent/sub-recent igneous activity in and around these

springs in Odisha.

The thermal springs of Odisha are pigeon- holed into

two categories of geothermal environment-

1. Mahanadi valley geothermal province (Boundary

faults along Gondwana graben).

2. Archaean/Pre-Cambrian geothermal province-

(lineaments along Pre-Cambrian terrain).

The hot springs follow certain tectonic systems and tend

to occur in areas associated with tectonic movement

(Gupta, 1974). Regional geological investigation

established the relationship between the hot springs and

the mega lineaments, especially in Odisha. Thermal

1573 S. C. MAHALA, P. SINGH, M. DAS and S. ACHARYA



springs of Odisha are confined along the following

lineaments.

i. Mahanadi lineament (Attri, Tarabalo, Deuljhori),

southern fault of Gondwana basin.

ii. Lineament in Easternghat granulite terrain

(Taptapani), secondary lineament parallel to east coast

lineament.

iii. Lineament in Iron Ore Supergroup terrain

(Magarmuhan and Bankhol), secondary lineament

parallel to the northern fault of Gondwana basin.

iv. Lineament along boundary of Chhatisgarh basin

(Boden) and v.Lineament along Gondwana-

Precambrian boundary (Badaberena), northern fault of

Gondwana basin.

Hydrological and Chemical Characters:

The temperature of surface water of the thermal springs

ranges from 32oC (Boden) to 67

o C (Tarabalo). The pH

value of the thermal waters indicates a neutral to mildly

alkaline character. All the thermal springs discharge

colourless and clear hot water. Sulphur odour is

discernable throughout the year at Attri, Tarabalo,

Deuljhori, Taptapani and Badaberena. Though the

volume of water released remains more or less constant,

difference in the rate of discharge is observed.

Chemical nature of water of thermal spring often

provides clues to their origin. Chemistry of water in

some thermal springs of Odisha has been carried out but

to a limited extent (Swain and Padhi1986; Bhargav et. al

2001). The water and gas samples collected from these

hot springs (Fig.2) have been analysed in the laboratory

,by adopting standard techniques. The water quality of

the eight thermal springs is not identical. The total

dissolved solids indicate low mineral content.

On the basis of concentration of various chemical

constituents, the water of thermal springs have been

categorised mainly into three categories (i) Sodium

chloride (NaCl) rich, (ii) Sodium Bicarbonate

(NaHCO3) rich and (iii) Calcium Bicarbonate

(CaHCO3) rich. The thermal spring water of Attri,

Tarabalo, Deuljhori and Taptapani are sodium chloride

rich while that of Magarmuhan, Bankhol, Badaberena

are sodium bicarbonate rich. Boden water is calcium

bicarbonate rich. This indicates that thermal springs

lying south of Gondwana graben are predominantly

NaCl rich and those lying north of it are Na-HCO3 rich.

Analysis of gas samples reveals that nitrogen is the

main component of geothermal gases (88%-90.5%)

followed by oxygen (1.2-6.6%). Together they

constitute more than 95% in volume. Other gases

detected are helium, argon, traces of methane and

carbon dioxide.

Genesis of Thermal Springs:

Several views regarding the origin of the thermal

springs have been proposed [ Allen and Day (1935),

Day (1939), Barth (1950) and Ellis & Mahon (1977)].

Two schools of thought exist about the origin of hot

springs. One school has correlated the thermal springs

with the volcanic activity or molten magma at depth,

where large amount of circulating ground water are

mixed up and interacted with the components from

magmatic sources. The final products of such

combination are the sources of mineral constituents and

heated water of the hot springs. The other school

advocated for non-volcanogenic origin. However, it is

accepted that the heating of circulating ground water at

depth can be caused by the geothermal gradient,

disintegration of radioactive substances and exothermic

reactions.

Source of Water:

The water of the thermal spring is invariably accepted

as of meteoric . In a simple meteoric origin, the

meteoric water percolating downwards along deep

fractures can be heated up by the rise of temperature due

to normal geothermal gradient. Lindgren (1935) has

reported a rise of 1o C temperature for every 30 m of

vertical depth.

Source of Heat:

Heat is accepted as an essential physical property of the

thermal springs. As a matter of fact the water generating

from a thermal spring is heated by geothermal gradient,

i.e. heat from the earth’s interior. The water (meteoric)

percolating deep down into the crust through structural

breaks are heated up, while coming in contact with the

hot rocks . The water from thermal springs, in non-

volcanic areas, is supposed to be heated in this manner.

But normal regional heat flow is considered as the

source of heat for the hot springs in Indian shield

(Ravishanker et al., 1991). Five heat flow zones in

Indian sub-continent have been reported, where heat

flow values range between 30 to 468 mW/m2. Odisha

falls in zone-II (100-180mW/m2) and zone-III (70-

100mW/m2) on the basis of heat flow values. Northwest

of Eastern ghats in Odisha sector, Talcher coal field and

thermal spring belt of Attri and Tarabalo fall within

zone-II. The average temperature gradient is 61+/- 20oC

in this zone. Roy and Rao (1996) mentioned that

variation in heat flow is primarily due to the heat

generated by radioactive elements present in the crust.

Source of Minerals:

The dissolved mineral constituents are likely to have

been derived from water- rock interactions. It is the

mineral composition instead of rocks that has

contributed the elements to the thermal spring water.

1574 Genesis of Thermal Springs of Odisha, India



Dissolved CO2 in the water plays a key role in such

reaction through liberation of H+ ions that react with

various silicates of the wall rock to release the alkalies,

alkaline earths and chloride. As a result, the water gets

depleted in H+ resulting an increase in its alkalinity.

Prevalence of Na over K is due to the greater solubility

of sodium salts and their tendency to be derived from

reactions involving minerals like plagioclases,

pyroxenes and amphiboles in a high pH condition.

Differences in chemical composition of waters are

mainly due to differences in host rock composition

(Ellis and Mahon, 1977).

Source of Gas:

Thermal springs are normally associated with various

gas phases. The gases come out through the fissures in

rocks in the form of bubbles that create micro-rings of

waves on the water surface. Gases such as nitrogen,

oxygen, argon, methane and carbon dioxide are present

in decreasing order (Mahala et. al., 1997). The gaseous

composition of the thermal springs is different from the

present atmospheric level (PAL). Presence of

appreciable level of nitrogen in the geothermal gases,

suggests its derivation from the atmosphere along with

meteoric water. Besides, methane and ammonia of the

primitive atmospheric air under high pressure and

temperature might have got dissociated into nitrogen

and hydrogen that also explains high percentage of

nitrogen in the thermal spring gases (Datta Munshi et.al,

1984). The thermal spring gases of Odisha having traces

of carbon dioxide indicate a non-magmatic origin of

these springs. Less amount of oxygen (6.6%) in the

thermal spring gases reveals that the atmospheric air

containing about 16% oxygen might have been utilized

in the oxidation process. The rare gas helium is very

likely formed due to radioactive disintegration and

presence of methane may be attributed to the

decomposition of organic matter inside earth’s crust

(Ellis and Mahon, 1977). Sulphurous smell of the

thermal springs is seemingly due to formation of

hydrogen sulphide, due to inter action between sulphide

mineral-bearing rocks and/or organic matter.

Discussion:

Genesis of thermal spring is an important aspect of

present investigation. The paradigm on the genetic

aspect of components (water, heat and gas) of thermal

springs has been drawn on the basis of database

generated during present study. Thermal springs of

Odisha are mainly confined to the crystalline schists and

gneissic terrains of Precambrian age. Most of them are

located in or close to the boundary (contact zone) of the

Precambrian crystalline and Gondwana sedimentary

rocks. These springs usually emerge along the deep

faults or fissures. Field studies supplemented by satellite

data indicate that the thermal springs lie along

lineaments. This infers close relationship between

tectonism and thermal activity.

It is accepted that the meteoric water is the main source

of the water of these thermal springs (Singh et.al, 1996,

Swain & Padhi, 1986). Since the water of the thermal

springs in Odisha has low TDS, the heat is inferred to be

derived from a non-magmatic source as the thermal

systems associated with volcanism usually have very

high TDS (White, 1957, Saxena & Gupta, 1982).

The higher temperature of the spring water is because of

rise of temperature by geothermal gradient, exothermic

reactions, and disintegration of radio-active elements (

Baranwal et.al ,2006). As there is no igneous activity,

rise of temperature is not related to magmatism. In fact

the geothermal gradient has been responsible in adding

heat vis-a –vis rise of temperature of the spring water.

The metamorphic reactions (exothermic) are responsible

for rise of temperature of the water. The Precambrian

terrain is enriched in radioactive elements (Ghosh,

1954). The quartz-pebble-conglomerate (QPC) of Iron

Ore Supergroup also reported to have radioactive

elements (Ray et.al, 1990). Since these springs

discharged water through the Precambrian metamorphic

terrain, the storehouse of radioactive minerals, might

have released temperature through disintegration of

radons. Thus, high temperatures of these thermal

springs are results of multiple processes involved

therein.

The chemical elements present in the minerals/ rocks

get dissolved in the hot water while coming in contact

during circulation. It has already been reported that

rocks of Eastern Ghat Granulites contain sulphide

bearing minerals like pyrite, pyrrhotite etc. These

minerals, while coming in contact with hot water,

impart sulphur smell.

As cited by Acharya (1966), Gautier’s suggestion of

foundering of sialic blocks and consequent squeezing

out of water from the mass resulted in the formation of

water vapour. This is because of the down faulting of

Gondwana basin against the fault contact between two

tectono-stratigraphic terrain; Iron Ore Super Group

(IOSG) in the north and Eastern Ghat Super Group

(EGSG) in the south. A conceptual model on the genesis

of thermal springs of Odisha along Gondwana graben is

given in Fig.3. The middle part of Odisha is down

faulted (rifted) and the rift valley is filled with

sediments. The northern fault is termed as Sukinda

thrust and the southern one is designated as north-

Khurda fault. The two faults form the boundary of the

Gondwana graben and got filled up later by sediments.

These two mega lineaments are responsible for the

occurrence of the thermal springs.




Conclusions:

Thermal spring is a natural phenomenon and

manifestation of earth’s internal source of energy that is

received on the surface of the earth, in the form of

water, heat, mineral and gas. The thermal springs of

Odisha are situated in different litho-stratigraphic areas.

They are associated with mega lineaments, especially

along the contacts of Gondwana Graben and Pre-

cambrian crystalline rocks. The source of water is of

meteoric origin, which gets heated up by deep

circulation through fissures and faults. Hydro chemical

characters of the hot spring water have been influenced

significantly by water - rock interaction. The mineral

composition and radioactive character is influenced by

the rock assemblages through which the water

circulated. Presence of helium, the noble gas, is

indicative of radioactive disintegration. The atmospheric

air, litho-types and water-rock inter action together

contribute to the thermal spring gas composition.

Even though the conceptual model basically projects the

role of the two crustal scale faults, it is essential to note

that hot water circulation along the faults do suggest the

entire area in between the two faults also has hidden hot

spring chambers and they may emerge, in future,on the

surface through secondary faults associated with the two

main faults, as and when a structural disturbance

occurs.And as such, it is advisable to delineate the

subsurface conductivity pattern using an integrated

magneto telluric and deep resistivity surveys.

Acknowledgements:

We are thankful to Dr.P.R.Reddy, Retd. Scientist,

NGRI, Hyderabad for valuable suggestions and

objective reviewing/ editing.

References

[1] Acharya, S (1966) the thermal springs of Odisha.

The Explorer. pp.29-35.

[2] Allen, E.T. and Day, A.L., (1935) Hot springs of

Yellowstone National Park. Carnegie Inst.

Washington. Publ. No.466, pp.1.

[3] Baranwal, V.C., Sharma, S.P., Sengupta, D.,

Sandilya, M. K.., Bhaumik, B. K., Guin, R.K., and

Saha, S.K. (2006) A new high back ground

radiation area in the geothermalregion of

Easternghats mobile Belt(EGMB) of Orissa, India.

Radiation Measurements, 41,pp;602-610.

[4] Barth, T. F. W., (1950) volcanic geology, hot

springs and geysers of Iceland.Carnegie Institute,

Washington, publ. 587, pp. 1748.

[5] Bhargav, J. S., Srivastav, S. K. and Naik, P. K.,

(2001) A study on the geochemistry of Attri hot

spring area, Odisha.Bhu-Jal, (CGWB) vol.16, No-

324.

[6] Day, A. L., (1939) Presidential address to Geol.

Soc. America on hot spring problem. Bull. Geol.

Soc. Amer. Vol. 50.

[7] Datta Munshi, J.S., Billgrami, S. K., Verma, P. K.,

Datta Munshi, J. and Yadav, R. N., (1984) Gases

from thermal springs of Bihar,

India.National.Academy Science letters” India,

vol.7, No-9, pp. 291-293.

[8] Ellis, A.J. and Mahon, W.A.J. (1977) Chemistry

and geothermal systems. Academic press.

Newyork.

[9] [9] Ghosh, P. K., (1954) Mineral springs of India.

Rec. Geol. Surv. Of India. Vol. 80. pp. 541-558.

[10] Gupta, M. L., (1974) Geothermal resources of some

imalayan hot spring areas. Himalayan Geology,

Vol. 4, pp. 492-515.

[11] Lindgren, W (1935) Mineral deposits. 3rd edition.

McGraw Hill , New York

[12] Mahala, S., Singh,P., Das, M., Ray, P., and

Acharya, S. (1997) Trait on Geothermal Gases from

thermal Springs of Odisha. Vistas in Geol

Res.No.2, pp. 221-225.

[13] Ray, P., Singh, P., and Acharya, S., (1990) A note

on quartz pebble conglomerate around Sikheswar-

Burhaparbat in Dhenkanal-Keonjhar district,

Odisha. Proc.77th Ind. Sci. cong. Assoc. Cochin,

p.33. (Abst).

[14] Roy, S. and Rao, R. U. M., (1996)- Regional heat

flow and the perspective for the origin of hot

springs in the Indian shield. GSI. Spl. Publ. No-45,

pp. 39-40

[15] Saxena, V.K. and gupta, M.L (1982) Geochemistry

of thermal and cold waters of Godavari valley.

Jour. Geol.scc.Ind. Vol.23, No-II, pp;551-560.

[16] Shanker, Ravi., Guha, S.K., Seth, S.K., Ghosh, A.,

Ghosh, S., Nandy, D.R., Jangi, B.L. and

Muthuraman, K., (1991) Geothermal Atlas of India.

GSI Spl. Publ. No-19.

[17] Singh, P., Mahala, S., Ray, P., Das, M. and

Acharya, S.(1996) a study on hydrology and

geochemistry of thermal springs in Orissa, India.

Jour. Ind. Acad. Geosci. Vol.39(1), pp:31-35.

[18] Swain, P. K. and Padhi, R. N. (1986) Geothermal

fields of Ganjam and Puri districts, Odisha. GSI,

Records Vol. 114, Pt. 3, pp. 41-46

[19] White, D. E., (1957) Magmatic, connte and

metamorphic water.Geol. Soc. of America, vol.68,

pp.1659-1682.

1576 Genesis of Thermal Springs of Odisha, India



Figure 1: Outline Map of Odisha showing location of Thermal Springs. 1. Deuljhori, 2. Tarabalo, 3. Attri,

4. Bankhol, 5. Magarmuhan, 6. Badaberena, 7. Taptapani, 8. Boden




Figure 2: Photographs Showing Thermal Springs at A. Tarabalo B. Taptapani C. Deuljhori D.Attri

Figure 3: Conceptual Model Showing Location of Thermal Springs along Gondwana Graben

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1578-1583


Remote Sensing Studies in Delineating Hydrogeological Parameters

in the Drought-Prone Kuchinda-Bamra Area in Sambalpur District,

Odisha

NANDITA MAHANTA1 and H. K. SAHOO

2

1Department of Geology, Bonaigarh College, Bonaigarh, Sundargarh -770038

2P. G. Department of Geology, Utkal University, Bhubaneswar -751004


Abstract: The hydrogeomorphological map of Kuchinda-Bamra area was prepared from IRS IA (LISS – II)

satellite imagery in the scale of 1:250,000 by visual interpretation method for delineation of fresh groundwater

potential zones using remote sensing techniques. The various hydrogeomorphic units such as structural hills,

residual hills, intermontanne valleys, pediments, buried pediplains and denudational hills are identified. The

groundwater potential zones have been classified into low, moderate and good which is also corroborated from

geophysical studies and field checks.

Keywords: Hydrogeomorphic units, Darjing Group, Kuchinda-Bamra area

Introduction:

The Kuchinda-Bamra area in Sambalpur district of

Odisha is a drought prone area and forms a part of the

hard rock terrain. Because of erratic nature of rainfall

and unscientific management of water resources, the

area frequently experiences drought. The demand of

groundwater has increased due to growth of population,

rapid industrialization and development in agricultural

activities. The development of groundwater resources

can play an important role in the economic development

of the area. Keeping the above facts in view an attempt

has been made in the present work to analyze the remote

sensing data systematically and to draw meaningful

inferences with additional inputs from geophysical data

in the Kuchinda- Bamra blocks of Kuchinda subdivision

lying in the northern part of Sambalpur district of

Odisha. The present study shall help in understanding

the geomorphological, lithological and hydrogeological

factors, which govern the occurrence and distribution of

groundwater in the study area.

Study Area:

The study area consists of two blocks namely Kuchinda

and Bamra of Kuchinda subdivision in northern part of

Sambalpur district and forms north-west upland of

Odisha. It is situated 85 km away from the district

headquarter, Sambalpur and lies between latitude 210

37’ N to 220 13’N and longitude 84

0 15’E to 84

0 42’ E

in the Survey of India Toposheet Numbers 73 B/8, 73

C/2, 73 C/5, 73 C/6, 73 C/9, 73 C/10, 73 B/12. The

total population of the study area is 155,585 as per 2001

census. The area experiences a subtropical climate with

maximum temperature of 43°C during summer and

minimum temperature of 10°C during winter. The area

receives most of the rain from south west monsoon. The

location map of the study area is given in Fig. 1.

Physiography and Drainage:

Physiographycally, the area is highly undulated with

very large elevated hills with maximum height of 736 m

above MSL (Bamlo Pahar) and flat areas of altitude 221

m above MSL. Some parts of the area are characterized

by gently undulating plains with isolated residual hills

and mounds. The major drainage system is controlled

by Bheden River with streamlets and nallas like Purtala

nala, Kharla nala, Lamdora, Tabko Jharan, Lohranga

nala and Sian Jhor etc. The drainage pattern is dendritic

and influent in nature. Because of the predominance of

granitic and metasedimentary rocks, the area

experiences medium drainage density. The drainage

map of the area is shown in Fig. 2.

Geological Setting:

The study area comprises a variety of rock types

varying in age from Precambrian to Recent.

Garnetiferous biotite-quartz schist of the Older

Metamorphic Group is the oldest rock type in the study

area which is overlain by metavolcanics and quartzites

belonging to Deogarh Group. These are succeeded by

granitoids (Bamra granite). The rocks of Darjing Group

comprising of Birtola Formation(mostly quartzite and

conglomerate) and Jalda Formation (mostly staurolite

schist, calc-gneiss, garnetiferous mica- schist) overlie

the granitoids with an unconformity at the base which

1579 NANDITA MAHANTA and H. K. SAHOO



has been assigned an age of 2700 Ma (Mahalik, 1987;

Mahalik and Nanda, 2006). The rocks of Darjing Group

are intruded by mafic and ultramafic sills and dykes.

During the Meso-Proterozoic period, another basin

(Kunjar basin) has been developed over the 2700 Ma

granitic basement (Chaki et al. 2005) now represented

by quartzite. Laterite and alluvium of Recent age of

considerable thickness form the surficial cover. The

rocks of the Precambrian age are affected by atleast

three generations of deformation with development of

folds, faults, foliations, joints etc creating secondary

porosities in the hard rocks that help in the storage and

movement of groundwater.

Hydrogeological Setting:

Presence of a variety of rock types in the study area

with different degrees of deformation is responsible for

different hydrogeological condition in various parts with

wide variation in groundwater potential. About 85% of

the total area is covered by hard (consolidated) rocks

where groundwater is stored and moves in secondary

porosities developed due to weathering and fracturing.

Groundwater in Kuchinda-Bamra area is found to exist

in semi-confined to confined conditions in the fractured

and jointed rocks at deeper depths but under unconfined

condition in shallow weathered zones. The

unconsolidated formations are constituted by laterite

and alluvium. The yield of groundwater in the

weathered granitoids varies from very low (1 lps) to

moderate (7 lps) in Bamra block and from 1 lps to 9.2

lps in Kuchinda block. The depth to water table during

post-monsoon period varied from 0.91 to 6.7 mbgl in

2008 and from 2.44 to 8.23 mbgl during pre-monsoon

period in 2009. The net annual utilizable groundwater

resource of Kuchinda and Bamra blocks are 5470 Ham

and 6642 Ham respectively. Groundwater development

is relatively less in Bamra block (8.93%) in comparison

with Kuchinda block (11.83%). However, both the

blocks of the study area fall under safe/ (white)

category.

Remote Sensing Studies:

Remote sensing technique provides an effective modern

tool in delineating groundwater potential zones. It is one

of the most useful tools for watershed characterization,

planning and management of groundwater resources in

recent times. The integrated approach of remote sensing

plays an important role in the systematic analysis of

various lithological, geomorphological, soil,

hydrological, and land use characterization of an area.

The surface information of a particular area is necessary

for interpretation of the subsurface geological and

hydrogeological features. Information on various

aspects of surface and subsurface can be known from

remote sensing studies, geophysical surveys and various

logging techniques. Remote sensing method is the

quickest method for the interpretation of various

hydrogeomorphological units and groundwater

conditions. In recent years, remote sensed data have

been increasingly used in groundwater studies viz: Hari

Narayan et al. (2000), Mahapatra and Maharana (2000),

Per Sander et al. (1997), Sankar (2001), Sankar and

Venkataraman (2002), Rokade et al (2007), Chakrabarty

and Paul (2004), Gopinath et al. (2003), Sakram et

al.(2011) etc. The combination of landform – cum –

lineament mapping is very much useful in areas where

the overburden and / or weathered rock as well as partly

weathered / fractured rock contain adequate quantities

of groundwater (Beeson & Jones, 1988).

Data used and Methodology:

The present study was carried out by visual

interpretation of the satellite imageries of IRS- IA (LISS

– II) in the scale of 1:250,000. The Survey of India

toposheets were referred for the preparation of the base

maps. With the help of satellite data, based on image

characteristics, the study area has been divided in to

number of geomorphological units. Various geological

formations, landuse, landcover classifications and

lineaments were delineated and verified during the field

visits. The hydrogeomorphological map prepared from

setellite imagery(Fig.3) provide integrated information

on several factors which directly or indirectly control

the movement and occurrence of groundwater.

Result and Discussion:

The analysis of IRS- IA LISS-II data based on various

image interpretations led to the identification of six

hydrogeomorphological units differing in groundwater

occurrences. These are named as structural hills,

denudational hills, residual hills, intermontanne valleys,

pediments and buried pediplains. Different

hydrogeomorphic units with their lithostratigraphic

classification, structural features and groundwater

occurrence is presented in Table 1. The structural hills

are represented by both Darjing Group and Deogarh

Group of rocks which are well foliated, folded and

faulted but sometimes also are massive and their

groundwater occurrence is poor. Denudational hills

comprise of rocks of Deogarh Group mostly where

groundwater occurrence is poor but can be available

only in intermontane valleys. Residual hills comprising

of Darjing Group of rocks and granitoids are found as

isolated hills in the flat terrain but their groundwater

content is very limited. The most important groundwater

host in the area of study is the buried pediplains which

are flat terrains with thickness varying from 5 to 20m

comprising of alluvium and weathered residues of

granitoids. In this unit, the groundwater occurrence is

moderately good to very good. The pediments which are

gently undulating areas with fractures and joints

comprise of granitoids and metavolcanics where

1580 Remote Sensing Studies in Delineating Hydrogeological Parameters in the

Drought-Prone Kuchinda-Bamra Area in Sambalpur District, Odisha



groundwater occurrence is poor to moderate. Among all

these geomorphic units, the buried pediplains can be

exploited for getting good groundwater supplies.

Lineaments are linear features commonly associated

with dislocations and deformations. They provide the

pathway for groundwater movement and are

hydrogeologically very important. Excellent

groundwater prospects are expected at intersections of

lineaments (Sankar, 2001). In the study area, two large

lineaments trending NE-SW intersect at a point which

represent a good source of groundwater. A fault

trending ENE-WSW near Kusumi has the potential of

storage and movement of huge amount of groundwater

resources. Resistivity survey by Vertical Electrical

Sounding was conducted at 26 locations in the area of

study with a maximum current electrode separation of

250m. This indicates presence of 3 to 5 layer situations

where a soil or lateritic horizon represents the first layer.

The thickness of this layer varies from 0.4 to 6.3m and

resistivity varies from 8.0 to 1130 ohm-m. This layer

contains water in very limited quantities with some parts

indicating a dry soil zone. The second layer represents

highly weathered granitoids in most places but in a few

locations it contains clay, sandy loam and sandy clay.

The thickness of this layer varies from 0.9 to 16.7m and

the resistivity varies from 3.0 to 169.6 ohm-m.

Groundwater occurrence in this layer is very good. The

third layer in some cases represents highly weathered

and highly fractured granitoids and granulites but in

other cases it represents hard and consolidated rocks.

The thickness of this layer varies from 1.9m to infinity

and the resistivity varies from 29 ohm-m to more than

2000 ohm-m. The fourth layer represents partly

fractured and partly weathered granite in some cases

and hard and unconsolidated rocks in other cases. The

thickness of this layer varies from 6.2m to infinity. The

resistivity varies from 12.5 ohm-m to more than 6500

ohm-m. The groundwater occurs only in fractured and

weathered zones in limited quantities. The isoresistivity

map of the groundwater rich second layer (Fig.4)

indicates resistivity value varying from 90 to 120 ohm-

m at Mantrimunda, Niktimal, Gorpos etc corresponding

to buried pediplain which are very good locales of

groundwater. The interpretation of lineaments in the

southern part of the area also corresponds to areas

representing values of around 60 ohm-m representing

good sources of groundwater. This is further

corroborated from field observation.

Conclusion:

The present study has helped to identify the

groundwater prospect zones by using the remote sensing

techniques. In various hydrogeomorphological units

such as structural hills, denudational hills and residual

hills, the groundwater is of limited occurrence and

intermontanne valleys in these units can be searched for

to get groundwater. In pediments the groundwater

potential is found to be poor to moderate but sometimes

good. The buried pediplains are of moderate thickness

varying from 5 to 20 m which are flat terrain and

contain significant quantity of groundwater. The

groundwater from this can be exploited by dug wells

and shallow tube wells. Presences of lineaments in the

buried pediplains are suitable sites for groundwater

exploitation.

References:

[1] Beeson,S. and Jones,C.R.C.(1988) The combined

EMT/VES geophysical method for siting bore

holes, Groundwater, Vol.26, pp.54 – 63

[2] Chaki A, Bhattacharya D, Rao J.S. , Chaturvedi

A.K. and Bagchi A.K. (2005), Geochronology of

the Granitoids of the Kunjar area, Sundargarh

District, Orissa : Implication to the Regional

Stratigaphy. Journal Geological Society of India,

Vol. 65, PP, 428 – 440.

[3] Chakraborty,S. and Paul,P.K.(2004). Identification

of potential ground water zonesin the Baghamundi

block of Purulia district of West Bengal using

remote sensing and GIS, Jour. Geol. Soc.Ind,

Vol.64, No.1, pp.69 – 75.

[4] Gopinath,G.,Seralathan,P.,Ramasamy,S.and

Unnikrishnan,K.(2003) Delineation of groundwater

potential zones in the hard rock terrains of

Madurai-Dindigal district, Tamilnadu using remote

sensing technique, Jour. Of Applied Hydrology,

Vol.XVI, No.1, pp.49 – 55.

[5] Harinarayanan, P., Gopalkrishna, G.S. and

Balasubramanian, A. (2000) Remote sensing data

for groundwater development and management in

Keralapura watersheds of Cauvery basin, Karnatak.

In: National Seminar on Earth Resources,

Mangalore University, pp. 7-8 (abs).

[6] Jaiswal, R.K., Mukharjee, S., Krishnamurthy, J. &

Saxena, R. (2003) Role of remote sensing and GIS

techniques for generation of groundwater prospect

zones towards rural development- an approach.

Intern. Jour. Remote Sensing, Vo. 24, No. 5, pp.

993-1008.

[7] Mahalik, N.K. & Nanda, J.K. (2006) Geology &

mineral resources of Orissa, Precambrian, SGAT

Publication, pp. 45-90.

[8] Mahalik N.K., (1987), Geology of rocks lying

between Gangpur Group and Iron Ore Group of the

Horse shoe syncline in North Orissa, Indian Journal

of Earth Sciences, Vol. 14, No 1 PP. 73 – 83.

[9] Mahapatra,K.C. and Maharana,R.C. (2000).

Integrated survey on groundwater potential in

Turva town of Balangir district, In Proc. Of

Seminar on “Prospect of groundwater development




and management in Orissa”, Organised by CGWB,

Bhubaneswar, pp.38 – 50.

[10] Per Sander, Minor, T.B. & Chesley, M.M. (1997)

Groundwater exploration based on lineament

analysis and reproducibility tests. Groundwater,

Vol. 35, pp. 888-894.

[11] Rokade,V.M., Kundal,P. and Joshi,A.K.(2007)

Groundwater potential modeling through remote

sensing and GIS: A case study from Rajura Taluka,

Chandrapur district, Maharastra, Jour. Geol. Soc.

India, Vol.69, No.5, pp.943 – 948.

[12] Sakram,G., Rajita Sanda, Sudhakar, A.and Sexena

Praveen Raj(2011) Groundwater resources and its

management in the lateritic terrain around

Zaheerabad, Medak district, Andhra Pradesh,India

– A remote sensing and GIS approach, International

Jour. of Earth Sciences and Engineering, Vo.04,

N0.6, pp.995 – 999.

[13] Sankar, K. (2001) Remote sensing study of

groundwater occurrence in Kanyakumari District,

Tamilnadu, India, Jour. Appl. Hydrology. Vol.

XIV, No. 4, pp. 16-26.

[14] Sankar,K. and Venkataraman,S.(2002) Geological

and geomorphological mapping in and around

Bharatidasan University: a remote sensing

approach, Jour of Applied Hydrology, Vol. XV,

No.1, pp. 15 – 22.

Table 1: Various Geomorphic Units of the Study Area with their Groundwater Potential

Sl.

No.

Geomorphic

units

Lithostrati-

graphic units

Major

structure

Description Groundwater condition

1 Structural Hills

(SH) a)Darjing Group

–Quartzite, schist

and phyllite

b)Deogarh

Group - Quartzite

Massive,

Well

foliated &

folded

Folded and

faulted

A group of linear to

curvilinear hill ranges

at times with

intermontanne valleys

Groundwater occurrence

is poor but good in

intermontanne valleys but

of limited occurrence

2 Denudational

Hills (DH) a)Deogarh

Group –

Quartzite

b)Bamra Granite

Folded and

faulted

Intrusive

into Country

rocks,

Fractured

A group of linear to

curvilinear hill ranges

at times with

intermontanne valleys

Group of hill ranges

with domal shape

Poor Groundwater

occurrence but good in

intermontanne valleys but

of limited occurrence

Groundwater occurs in

weathered and fractured

zones, Ground water

occurrence is moderate

3 Residual Hills

(RH)

a)Granite and

gneisses

b)Darjing Group

–Quartzite, schist

and phyllite

c)Bamra Granite

Forms

basement,

sometimes

foliated

Massive,

Well folia

ted & folded

Massive,

also foliated

Isolated massive hills

of less areal extension

surrounded by plains

Isolated hills


Isolated mounds


Limited ground water

occurrence


occurrence


occurrence

4 Intermontanne

valley (IV)

In between

various rock

groups forming

hills

May be

structurally

controlled

A broad or linear

valley between

mountains filled with

alluvium

Moderate with limited

quantity

5 Pediment(P) Bamra granite

and

metavolcanics

Gently

undulating

with

fractures/

joints

Gently undulating

with weathered zone

and fractured zone

Poor to moderate,

sometimes good

6 Buried Pediplain

(BPP)

Alluvium of

recent age

underlain by

granite, gneiss,

meta volcanic and

quartzites

Fractures at

deeper

levels

Flat terrain with 5 to

20 m thickness of

overburden with

extensive occurrence

of alluvium

Moderate to good,

sometimes very good

1582 Remote Sensing Studies in Delineating Hydrogeological Parameters in the

Drought-Prone Kuchinda-Bamra Area in Sambalpur District, Odisha



Figure 1: Location Map of the Study Area

Figure 2: Drainage Map of the Study Area

Figure 3: Hydrogeomorphological Map, RH– Residual Hill, DH-Denudational Hill, SH – Structural Hill,

IV-Intermontanne Valley, P-Pediment, BPP-Buried Pediplain




Figure 4: Isoresistivity Map of the Second Layer

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1584-1589


Impact Analyses of Industrial and Mining Activities on

Groundwater Regime -Case Studies in Goa

A. G. CHACHADI Department of Earth Science, Goa University,


Abstract: In the present study analysis of impacts of industrial and mining activities on groundwater regime has

been carried out. It is found that mining activities in specific locations have been impacting both shallow

groundwater as well as moderately deep groundwater systems. The industrial activity at one site studied has not

indicated significant change in groundwater levels during the study period. The geological data indicate that the

shallow aquifers located in the peripheral areas of the Plateau region are in continuity with plateau laterites and

behave as a single geological unit.

Keywords: Hydrograph, Mining, Piezometer, Verna

Introduction:

The measurement and analysis of ground water level

fluctuations in piezometers and observation wells is an

important facet of groundwater studies. Water-level

fluctuations can result from a wide variety of

hydrologic, meteorological, and hydrogeological

phenomena. Some are natural phenomena like

groundwater recharge, evapotranspiration,

phreatophytic consumption, bank storage effects near

streams, tidal effects near oceans, change in

atmospheric pressure, earthquakes etc. and some

induced by man such as external loading of confined

aquifers, artificial recharge, groundwater pumping,

deep-well injection, return flows from irrigation,

geotechnical drainage and open pit mine pumping, land

use changes etc. In many cases, there may be more than

one mechanism operating simultaneously.

Well hydrographs even under natural undisturbed

conditions show seasonal and diurnal fluctuations. Only

the long term fluctuation (secular) is absent. Reduced

rainfall or changed climatic conditions will lead to

changes in groundwater levels even in the absence of

anthropogenic causes. On the other hand under human

induced conditions this balance can still be retained by

well-managed recharge and discharge conditions in an

area. This means in an area where tremendous

anthropogenic activities are witnessed involving the

groundwater regime and still the well hydrograph do

maintain their undisturbed original trend could only

mean that the groundwater regime is balanced. In other

words the system has reached a steady state condition.

Aquifer response to recharge to and discharge from it is

indicated in the water level changes measured at

different time periods. Unconfined aquifers, which are

in continuity with the ground surface and the

atmospheric pressure, are quicker in responding than

confined and semi confined aquifers. The magnitude of

change in water level below ground at a given point

depends not only on magnitude of groundwater

withdrawals and recharges but also on the intrinsic

characteristics of geological matrix both of saturated

and unsaturated zones besides initial water table depths

and antecedent soil moisture conditions. For example

long and steady spell of rain on a loamy saturated soil

with deep water table condition in a highly permeable

geologic section could raise the water table to greater

height than an intense rainfall event of shorter duration

on dry clayey soil with shallow water table condition in

a low permeable geologic section. Qualitative

information on the saturated – unsaturated media, nature

of aquifer recharges and discharges can be ascertained

from the study of well hydrographs. For example

identical features in rising and falling limbs of the

hydrographs from different locations in the watershed

should indicate similar aquifer hydraulic conditions. In

order to achieve such objectives the observation wells

and piezometers should have to be located carefully so

that each of them represents a sizable hydrogeological

regime in space and depth and should be well

distributed over the entire watershed area.

Objective of the Present Study:

Groundwater regime in the State of Goa has been

subjected to variety of anthropogenic stresses such as

mining, industrialization and urbanization. Mining and

industrialization have been on rise in the recent past and

have impacted groundwater regime in several ways. It is

therefore intended in the present study to evaluate

1585 A. G. CHACHADI



impacts if any of mining and industrialization on the

groundwater regime at selected locations through study

of well hydrographs both from shallow as well as deep

aquifers.

Methodology:

In order to achieve the set objectives two locations have

been selected one representing the industrial Estate and

other representing open cast iron ore mining. In the

industrial Estate 24 open wells were identified and the

shallow groundwater levels were monitored on monthly

basis for three years and in the mining area 4

piezometers were installed and groundwater levels were

monitored over a period of three years. Continuous

water levels measurements were also made on one of

the piezometers. Rainfall data for the corresponding

period were collected. The locations of these monitoring

wells are shown in the base map. After completion of

the data collection the appropriate data was plotted as

well hydrographs and these were subjected to analysis

and interpretation. The results of these are presented in

this paper.

Locations of the Study Area:

1. Study around the Industrial Estate:

In Goa there are as many as eighteen industrial Estates

and many are proposed. Verna Industrial Estate is one

of the largest industrial hubs located between Margao

and Panaji. It is located about 20km from Panaji city

side by the highway connecting Panaji to Margao. The

area under study is a part of Marmagoa and Salcete

taluka of South Goa district. The area is represented on

Survey of India (SOI) toposheet numbered 48 E/15 and

48 E/ 15/5 of 1:25000 scale. It is situated between

Latitudes: N 15° 17’ 30’’ to N 15° 25' 00’’ and

Longitudes: E 73° 54’ 30’’ to E 74° 00' 00’’ and covers

an area of about of 46 km2. The north-eastern region of

this area is bounded by the Zuari River. There are

several villages on the periphery of this industrial estate

(Fig.1). There are about 200 bore wells drilled to a

depth ranging from 60m to 100m below ground to

abstract ground water for various uses. These bore wells

generally tap confined to semi confined aquifers. On the

plateau there are few open wells and few filter points

taping shallow perched aquifer in local depressions, on

the foot hill regions shallow dug wells are commonly

dug for domestic use by the villagers. The plateau on

which the Verna Industrial Estate stands act as rainfall

recharge area to the shallow aquifers located in the

periphery of the plateau. Any change in the land use on

plateau may affect the recharge to these shallow

aquifers extent the wells dry.

Figure 1: Location of the Verna Industrial Study Area

The study area comprises of Barcem and Sanvordem

Formation of Goa group of rocks and includes

essentially metagrey-wackes, conglomerates and meta-

basalts with subordinate metasediments. In most of the

places these formations are capped by laterites and in

the lower reaches by sands and sandy clays.

Results and Discussions-Industrial Case Study Area:

In and around the Verna industrial estate 24 open wells

were established and the groundwater levels were

monitored in these wells on monthly basis for three

years. The monitoring well locations were transferred

on the base maps (Fig.2). After completion of the data

collection the appropriate data was plotted as well-

hydrographs and these were subjected to analysis and

interpretation. Rainfall data is also plotted for the

corresponding time. Two typical well hydrographs for

the study area are shown in Figs. 3 and 4.

The rainfall variation in the area during the groundwater

level monitoring is such that from 2006 to 07 it

increased by 12% followed by about 17% decrease

during 2007 to 08. However, from 2008 to 09 and 2009

to 10 the rainfall increased by 9% and 18% respectively.

In 2006 the rainfall started in May and it increased in

June while the peak to groundwater levels reached in

June. During 2008 the rainfall started in June but

decreased in July causing the groundwater levels to

reach peak levels only during August. Where as in 2009

the rains started in June but increased in July causing

the groundwater levels to reach peak heights during July

itself. Drastic fall in rainfall during August 2009 is

reflected in the corresponding fall in groundwater

levels. There was sizable rainfall during October and

November during years 2009 and 2010. This late

rainfall is again reflected in the less rapidly falling limb

1586 Impact Analyses of Industrial and Mining Activities on




of the hydrograph during 2009-10 compared to 2008-09

especially in well number 23.

W1

W2

W3

W4 W5

W6

W7

W8

W9

W10

W11W12

W13

W14

W15

W16

W17

W18

W19

W20

W21

W22

W23

W24

Spring

Spring

Kesarwal

Mhalsa

Siphrem

Nuvem

Mogarupoy

Umorem

Mardhwal

Verna

Verna

Arosim

Kansavlim

Konsua

Mosil

Karvona

Mudli

Nakordem

Lotli

Ambora

Kelosim

Navtawada

Sindoli

Kortali

Sankval

Verna Industrial Estate

0 1000 2000

SCALE

m m

Zuari River

W8Well number

Drainage network

Spring

Area of ground level

Sub-watersheds

Index

25m & more amsl

Figure 2: Location of the Groundwater Level

Monitoring Wells in and around the Study Area

Figure 3: Groundwater Level Hydrograph for Well no.

23 in the Study Area

Figure 4: Groundwater Level Hydrograph for Well no.

18 in the Study Area

As seen from the above figures the rainfall during the

period of record has increased to some extent steadily

except during the year 2008. The groundwater levels

show a seasonal fluctuation in their temporal behavior

during those three years of recording. The distribution

of rainfall intensity and duration influences the behavior

of the groundwater level fluctuations.

It is observed from the hydrographs that there is no

significant change in the groundwater levels over the

period of recording and the fluctuations are only

seasonal due to rainfall recharge and various

abstractions during non monsoon season. The following

conclusions can be drawn:

1. The groundwater levels show fluctuations mainly

due to rainfall recharge and various abstractions. With

existing short term water level data and in the absence

of pre industrial groundwater level records it shall be

unwise to conclude on the long term behavior of the

phreatic groundwater levels in the area. However, it is

learnt from the field inquiries from the well owners that

the groundwater levels in the area in lateritic aquifers

have not changed significantly in the past.

2. The geological data indicate that the shallow

aquifers located in the peripheral areas of the Verna

Plateau are in continuity with plateau laterites and

behave as a single geological unit.

3. The peripheral shallow aquifers around the Verna

Plateau are also fed by rainfall recharge that could be

taking place on the Plateau area.

2. Study around Mining Lease Area:

Mining in Goa is considered as backbone of Goa’s

economic activity. Opencast iron ore mining has been

on rise with mechanized means. The recent trends of

iron ore markets have attracted many to enter into

mining activity. This has led to widespread unscientific

mining activity causing multiple impacts on other

resources including groundwater. In the present study it

has been attempted to evaluate the possible influences

of open cast mining on groundwater regimes in the

neighborhood of the mining activity. Two locations

have been chosen for this study one is located in the

close proximity to the Arabian sea and the other in the

interior area.

Study Area Location:

This study area falls in the Survey of India Toposheet

No. 48 E/4. The mine is situated at latitiude N 150 28

’

20’’ and longitude E 73

0 35

’ 44

’’ in Bardez taluka, North

Goa. The lease area is located on the northeastern flank

of the plateau extending northwest southeast (Fig.5).

The plateau is elevated to about 128m above sea level

and the drainage network is almost negligible. The

surface run-off moves on north and south along the

slopes. A village is located along the southeastern

boundary of the mine lease. The present study area is an

iron ore mine where in the mine pits are located at the

hill bottom. The slope of the hill is fairly steep and

made up of detrital laterites and siliceous clays lying

above the sequence of clays. The laterites are blocky in

nature and have gaps between the blocks which provide

avenue for rain water percolation and saturation. Four

piezometers were installed to a depth varying from 77m

to 105m below ground as shown in figure 5. The fourth

1587 A. G. CHACHADI



piezometer could not remain intact due to collapse

inside and groundwater levels recording could not be

continued.

The details of the lithological logs recorded during the

drilling of the four piezometers in the study area are

given in Table 1. As seen from the lithological sections

the laterite is found upto 19m depth in the area. It is

generally hard and compact on the surface and becomes

gradually soft and gravelly with depth. Except at

piezometer no. 2 remaining sections do not have

phyllitic clay layer under laterite. Siliceous and chlorite

schists are very often encountered in this area.

Figure 5: Location of the Piezometers around the

Mining Lease

Seasonal Behavior of Groundwater Levels:

Of the 83 days of southwest monsoon recorded up to

12.8.2009 there was no rain for 26 days, and rainfall

was less than 10mm for 16 days. Only two days on

6/6/09 and 1/7/09 it rained about 300mm on each day.

Remaining days witness moderate rains. The uneven

distributions of rainfall events have increased in the last

one decade and the number of rainy days has come

down drastically. Although the total monsoon rainfall in

Goa does not change substantially but its time

distribution has become uneven. Most of the time it

rains heavily within a short time spans leaving rest of

the monsoon days dry. The total southwest monsoon

rainfall up to 30/6/2009 just before the second highest

rainfall event was 760.20mm (Fig.6).

The groundwater level hydrographs at piezometer 1 and

2 (Fig. 6) have not shown significant rising levels

during this period. This indicates that during this period

the effective rainfall infiltration was used to saturate the

soil moisture and the zone of aeration in the area besides

interflows.

The second rainfall peak occurred on 1/7/09 and all the

hydrographs soon showed a rising trend which is quick

in time and has a steep rising limbs. This indicates that

the aquifers tapped by the piezometers are mainly

unconfined in nature. The steep falling limb of

hydrographs of piezometer 1 and 2 further indicate that

the aquifer is fairly permeable thereby diffusing the

raised groundwater levels quickly. In piezometer 3 the

falling of groundwater level is fairly slow and it decays

over a period of time. No assessment could be made for

piezometer 4 as the data was not complete. Except in

piezometer 1 the groundwater levels in piezometer 2

and 3 continue to fall despite occurrence of rainfall

events post 1/7/09 rainfall event indicating rapid aquifer

drainability.

Table 1: Lithological Details of Piezometers in the

Study Area

Bore hole no: PZ-1

Date 19/5/09

Depth :105.40m

Water Table:84m

Depth Range(m) Lithology

From To

0.00 12.00 laterite

12.00 45.00 siliceous chlorite

45.00 61.00 Chlorite Schist

61.00 90.00 Clay & Chlorite Schist


Bore hole no: PZ-2

Date : 29/5/09

Depth : 77 m

Water Table:84m


From To

0.00 18.00 laterite

18.00 27.00 phyllitic clay

27.00 48.00 Chlorite Schist & Clay


Bore hole no: PZ-3

Date: 20/05/09

Depth :95.70m

Water Table:62m


From To

0.00 19.00 laterite


Bore hole no: PZ-4

Date :22/05/09

Depth: 80m

Wate Table:60m


From To

0.00 18.00 laterite

18.00 47.00 Siliceous chlorite

47.00 57.00 Siliceous phyllite

57.00 65.00 Friable silica

65.00 80.00 powdery iron ore

1588 Impact Analyses of Industrial and Mining Activities on




Figure 6: Short Term Variation of Groundwater Levels

and Rainfall

Long Term Behavior:

The long term three years groundwater level

hydrographs are shown in Figs. 7, 8 and 9 respectively

for piezometers 1, 2 and 3. In piezometers 2 and 3

several sharp spikes, both positive and negative are seen

in the non-rainy season of 2009-10 as well as 2010-11.

These spikes are also seen in piezometer 1 but less

pronounced. The mine pit is dewatered during non-rainy

season which would cause cone of depression in the

ground water levels. As the piezometers 2 and 3 are

close to the mine pit than piezometer 1 the water level

fluctuations are more conspicuous in the piezometers 2

and 3. These non-rainy season water level fluctuations

indicate that the mine pit has hydraulic connection with

the surrounding aquifers. The water levels respond very

moderately to normal rainfall events. On the other hand

if the rainfall is exceptionally high the spiked behavior

in the water levels is conspicuously seen as shown in

2009 monsoon season. The water level in piezmeter 1

kept on rising between the two spiked rainfall events on

6/6/2009 and 1/7/2009; however, the water levels in the

other two piezometers fell during the corresponding

time. This might be due to hydraulic disposition and

variable hydraulic parameters of the aquifer tapped by

piezometers. Groundwater levels have remained within

the limits of seasonal fluctuations during the period of

observation.

Figure 7: Ground Water Level Hydrograph At

Piezometer-1 along with Rainfall

Figure 8: Ground Water Level Hydrograph At


Figure 9: Ground Water Level Hydrograph at


The impact of rainfall runoff on the mine pit water

levels is shown in Fig.10 The mine pit water levels have

been rising gradually till the second rainfall event that

took place on 1/7/09. There is spike in the pit water

level due to surface runoff contribution from the rainfall

event on 1/7/09. The fact that the general pit water level

is around a steady 40m amsl during the entire period

except for spikes indicate normal draining of pit water

by way of outflow. The gradual rise in pit water levels

from June to July can be attributed to base flow and

interflow contributions to the mine pit.

Figure 10: Variation of Pit Water Level due to Rainfall

Conclusions:

The Vernal plateau at Verna industrial estate act as

groundwater recharge area to the peripheral shallow

aquifers. The groundwater levels have not shown

significant change in their trend during the period of

record. Although long term water level data is necessary

for drawing conclusions on the trends of groundwater

levels however, inquiries from the owners of the wells

1589 A. G. CHACHADI



show that there is no much change in the groundwater

levels in the area. The fact that the mine pit water level

remains constant at around 40m amsl indicates that the

surrounding aquifer continuously contributes to the

mine pit water and hence mining activity certainly leads

to groundwater runoff in the area. Also the groundwater

levels are much higher than the mine pit water levels

therefore ground water flow occurs to the pit due to

favourable hydraulic gradient towards the pit.

References:

[1] Chachadi, A.G. (2004). Well hydrographs as tools

for impact assessment of open cast mining on

groundwater regime in Goa- Journal of Applied

Hydrology, No.2 &3, pp.20-26.

[2] Chachadi, A.G., B.S. Choudri, Ligia Naronha,

Lobo Ferreira, J.P. (2004). Estimation of surface

run-off and groundwater recharge in Goa mining

area using daily sequential water balance model-

BALSEQ. Journal of Hydrology (IAH), 27 (1-2),

pp.1-15.

Acknowledgements:

The financial assistant from UGC under SAP is greatly

acknowledged.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1590-1598


Identification of Artificial Recharge Sites in a Hard Rock Terrain

using Remote Sensing and GIS

S. SARAVANAN Department of Civil Engineering, National Institute of Technology Tiruchirappalli Tiruchirappalli 620 015,

Tamilnadu, India


Abstract: In order to demonstrate, a case study was conducted in lower Palar basin of Tamilnadu, India using

remote sensing and GIS. The study focuses on the identification of the potential groundwater zones and suitable

sites for artificial recharge zone in hard rock areas. Utilizing the remote sensing data and GIS techniques, thematic

maps, such as geomorphology, geology, land use, slope, drainage density, lineament density and soil type were

prepared on 1:50,000 scale. Each unit in the thematic maps are assigned to a suitable rank and weightages depending

on the groundwater occurrence. These maps were overlaid two at a time and integrated coverage was classified into

four categories based on the groundwater potential and artificial recharge; viz very good, good, moderate and poor.

Later on, the final results were compared with the groundwater level data from the bore wells collected by

Groundwater Board, Tamilnadu, India. It was found that the integrated remote sensing and GIS techniques are the

most suitable methods for delineating groundwater potential zone and artificial recharge zone in hard rock terrain.

Keywords: Artificial Recharge, GIS, Groundwater Potential, Remote Sensing.

Introduction:

Precipitation in India is confined to only about three or

four months in a year and varies from 100 mm in the

western parts of Rajasthan to over 1000 cm in north-

eastern part. Due to the topographic and climatic effect

the availability of water is highly uneven in both space

and time. The assessment of potential zone for ground

water is essential for effective management of water

supplies, irrigation and industries. Remote sensing and

Geographical Information System (GIS) are playing an

important role in the field of hydrology and water

resources development. The technique of integration of

remote sensing and GIS has proved to be extremely

useful for groundwater studies (Ravindran et al., 1993;

Krishnamurthy et al., 1996; Saraf and Chaudhury,

1998). Groundwater exploration is essentially a

hydrologic inference operation and is dependent on the

correct interpretation of the hydrological indicators and

evidence. The occurrence of groundwater in hard rock

regions is erratic and its analysis needs information of

geomorphology and other associated parameters (Rao,

1991; Pradeep, 1998; Kumar et al., 1999). Assessment

of groundwater availability in hard rock regions

primarily a function of fracture-controlled permeability

and zone of weathering (Dhokariker, 1991). The depth

of fracture zone and weathering determine the

availability of groundwater. Present study attempts to

identify the potential zones for groundwater exploration

and artificial recharge using an integrated approach of

GIS and remote sensing for a hard rock terrain in Lower

Palar Basin, Tamilnadu, India.

The Study Area:

The study area is bounded by longitudes E 79o57’10”

and E 80o12’10” and latitudes N 12

o28’30” and N

12o41’05”, Kanchipuram district in Tamilnadu, India.

The total geographical area of watershed is 360.5 km2

and shown in Figure 1.1. This basin is getting rainfall by

northeast monsoon, which starts in last week of

September and extends until the end of December. The

average annual rainfall is 957.7 mm and temperature

ranges from 20.9oC to 34.5

oC. The climate of the region

varies between arid and semi-arid region. The

topography of the area is characterized by uniform and

very gentle slope and the eastern part shows undulating

rolling. The general elevation of the area ranges from 60

m to 240 m above mean sea level with a gentle gradient

from west to east. The area is mainly drained by Palar

River flowing towards the eastern parts and finally

confluences in to the Bay of Bengal. There are few

small drainages in the central part of the block

contributing to the numerous tanks. Buckingham canal

flows from northeast to south east part. All the tanks in

this block are mainly fed by rainfall only.

Methodology and Data used:

The input map was generated from the satellite image,

based on colour, tone, texture, pattern and association of

features. The geomorphology and lineaments maps have

1591 S. SARAVANAN



been prepared by visual interpretation on 1:50,000 scale

using IRS 1C LISS-III data. The Survey of India

toposheet no. 57P/14, 66D/2, and 66D/3 have been used

for preparation of drainage and base map. The

geological map of the study area has been delineated

from the geological map prepared by Geological Survey

of India (1978). All primary input maps such as

geomorphology, lineament, geology, soil, land use, and

drainage, were digitized in Arc/Info, GIS software from

ESRI. Features in each of the thematic layers were

labelled and categorised separately. Suitable weightages

were assigned depending on the geospatial

characteristics to delineate groundwater prospective

zones (Krishnamurthy et al., 1996, 1997). The

groundwater potential zone was delineated into four

classes such as (i) very good (ii) good (iii) moderate and

(iv) poor, by grouping the corresponding weightages.

Classification was made by assuming normal

distribution. The artificial recharge zone was delineated

with the help of Boolean logic and conditional methods.

This method is to identify the criteria and to formulate

the set of logical conditions to extract the suitable zones.

Finally the output has classified into (i) suitable and (ii)

unsuitable. The overall methodology adopted for the

present study has been depicted in Figure 1.2.


Geological Setup:

Geologically, the area consists of charnockites of

Achaean age and Alluvium deposit of recent age, found

along the western part and southern part (Figure 1.3) of

the study area. Small pockets of Pegmatite, Laterite and

lime stones also identified. Marine formations of

Quaternary age are found along the south-eastern part.

Charnockites shows medium grey tone and medium to

coarse texture. Alluvium occurs along the river course

the thickness of alluvial aquifer ranges from 10-15 m. It

consists unconsolidated materials composed of various

proportions of sand, silt and clay, which forms fertile

soil strata that support agriculture. This deposit forms

very good water bearing horizon.

Geomorphology:

Geomorphology is the highly influencing parameter of

groundwater occurrence. The study area predominantly

consists of buried pediment deep, followed by alluvial

plains, pediments, buried pediments shallow and

residual hills. The buried pediment is formed due to

weathering of the hornblende, biotite gneisses under

semi-arid climatic conditions (Sankar, 2002). In this

zone infiltration is moderately good. The thickness of

weathered zone varies from 10-15 m and favours a good

amount of water to circulate within this zone before

reaching the deeper fractured zone. Groundwater

potential of this zone is very good. Alluvial plains are

along the river course of the Palar River; it is the

youngest geological unit and including various

landforms formed by fluvial action. The flood plains

have deposition of loose unconsolidated materials such

as sand and silt, which is having more pore spaces and

highly permeable zone. It is very good in ground water

prospect. Pediments are isolated residual hillocks that

consists very low weathered zone with thickness of

varying up to 4 m. These areas have high runoff

potential, but poor ground water recharge. Residual hills

are the end products of the process of pediplanation,

which reduces the original mountain masses into series

of scattered knolls standing on the pediplains

(Thornbury, 1990). Groundwater prospect is very poor

in this area. The mud flat, beach and coastal plains are

found in the eastern part of the study area. Figure 1.4

depicts the geomorphologic features of the study area.

Lineament:

Lineament is defined as topography of the underlying

linear structural features. Lineaments provide the

pathways for ground water movements and are hydro

geologically very important. The lineaments

intersections are considered as good ground water

potential zones. Numerous lineaments have been

mapped and shown in Figure 1.5. The lineaments are

trending in the direction of northwest to southeast; also

a few of them are oriented in northeast to southwest

direction. The lineaments are more in north, south

central part and the length of the lineaments varying

from few km to several km. The lineament present in

the hard rock region is very significant for ground water

occurrence. In the present study area, it is a function of

a fracture - induced secondary permeability and

hydraulic conductivity.

Drainage Density:

Drainage density provides a direct measure of surface

runoff and permeability. Increasing drainage density

reduces the recharge capabilities of ground water

regime. It was seen that the least drainage density

occurs in the alluvial plains of the study area, which

have the maximum ground water prospect.

Slope:

Infiltration and runoff capacities are controlled by slope

of the terrain, soil and landuse practices. Slope with

lesser gradient tend to spread the overland flows thus

favouring the infiltration and ground water prospects.

The areas with 0-1% and 1-2% of slope are excellent

from the point of occurrence of ground water, while the

areas with slope greater than 30% have poor ground

water prospect.

1592 Identification of Artificial Recharge Sites in a Hard Rock

Terrain using Remote Sensing and GIS



Landuse and Soil Type:

Figure 1.6 shows the landuse map of the study area. The

study area predominantly has agricultural land, other

than, settlement, degraded forest, fallow harvested land,

salt affected land, upland with or without scrub, water

logged area, and few tanks. To note that agricultural

land and forest allows more water to infiltrate than

barren land.

The study area predominantly contains hydrological soil

group B, followed by A and C. The porous soil (such as

A and B) allows more water to infiltrate than clay soil

(such as D). In order to produce the final map for

ground water potential and artificial recharge zone, GIS

model has been used, to integrate the above thematic

maps such as geomorphology, geology, land use, slope,

drainage density and soil type. Each thematic layer

consists of a number of polygons, which correspond to

different features. The polygons in each of the thematic

layers have been categorized, depending on the

suitability/relevance to the ground water potential, and

suitable weights were assigned based on Krishnamurthy

et al. (1996; 1997). Subsequently, Table 1 shows the

weightages assigned for the features. Finally, all the

thematic layers were integrated and the ground water

potential zone map was arrived, based on the statistical

analysis of final weights. Figure 1.7 presents the final

map of ground water potential zone of the Lower Palar

basin with four classes viz., very good, good,

moderately good, and poor.

Selection of Artificial Recharge Zones:

For Selection of artificial recharge zones the first task

was to identify the factors facilitating recharge to take

place (Saraf and Choudhury, 1998). The existing

artificial recharge system in the area has been studied

with respect to its hydrogeology, topography and

response in the water level of the wells. The prime task

in this method is to identify the criteria and to

Formulate the set of logical conditions to extract the

suitable zones. In this case, the output will have only

two classes i.e. suitable or unsuitable. The areas in

which the defined conditions of the information layers

are fulfilled together, a value of 1 is given whereas, the

remaining part will have a zero value. This analysis is

suitable for objective criterion but is not suitable to

show gradational values (Kundu, 2000). Suitability

analysis has been done using Boolean logic model. The

criteria for site selection are:

1. The sites should be over a slope of 2°-5°.

2. The sites should be on channel fills of alluvial plains.

3. Geologically, the area should be covered with sand,

silt or weathered zone.

4. The drainage networks in the sites should be on the

2nd or 3

rd order stream.

Although an attempt has been made for selection of

sites for artificial recharge by an integrated analysis

through a combination of weighted indexing and

Boolean logic method in the GIS platform, but detailed

litholog information and geophysical data of the

proposed sites and the surrounding areas are likely to

improve the results of this analysis. Based on the

weighted indexing and Boolean logic analysis figure 1.8

presents the final map of suitable zone for artificial

recharge of the Lower Palar basin with two classes viz.,

suitable, and unsuitable.

The summary of the results shows that almost all

alluvial plains and valley fills associated with high

density of lineaments and lineament intersections have

been classified as relatively high groundwater

prospective zone. The low groundwater potential zone

matches with steeply mountainous areas underlain by

granite with low lineament density. Furthermore, it

shows that in hard rock areas, the groundwater potential

is high with high lineament density and low drainage

density. Similarly for the artificial recharge zone, the

alluvial flood plain region having mild slope shows

suitable. Finally to verify the ground water potential

zone map, it is compared with the ground water level

fluctuation map (Figure 1.9) generated from the water

level measured from the bore well of the study area.

Visual comparison of both maps proves that the analysis

yield reliable ground water potential zone map for the

study area.

Conclusion:

This study has successfully utilized the remote sensing

data on the GIS environment to obtain a detailed

understanding of the potential zones for groundwater

and artificial recharge zone in a watershed of a hard

rock terrain. Most of the thematic maps necessary for

groundwater prospecting zones in the study area were

directly generated from the processing of digital remote

sensing data using ERDAS Imagine software. Various

algorithms necessary for hydrological application

incorporated in Arc/Info GIS were useful in creating

stream network, watershed, sub-basin, lineament and

river gradients maps. Further, the present study

concludes the following:

The present result show that integrated remote sensing

and GIS techniques is the most suitable method for

groundwater potential prediction zoning.

(1) The analysis of the surface water bodies,

lineaments, and intersection of lineaments are directly

influencing the ground water potential. The deep buried

pediments and alluvial flood plains are the most

prospect zone for ground water.

(2) The methods and results of this study were

effective only for groundwater zone prediction in hard

1593 S. SARAVANAN



rock terrain, but were less effective in the alluvium

environment.

(3) The suitable zone for artificial recharge has

been made through a combination of weighted indexing

and Boolean logic method in the GIS platform, but

detailed litholog information and geophysical data of

the proposed sites and the surrounding areas are likely

to improve the results of this analysis.

Furthermore, the comparative analysis with the ground

water level fluctuation map generated from water level

data of bore well in the study area, shows that the

analysis gives satisfactory results for delineation of

potential ground water zone.

References:

[1] Dhokarikar, B.G. (1991). Groundwater resource

development in basaltic rock terrain of Mahastra.

Water Industry Publication, Pune, 275p.Geological

Survey of India. (1978). Know your district

(published), Tamilnadu and Pondicherry,

Geological Survey of India.

[2] Krishnamurthy, J., Arul Mani, M., Jayaraman, V.,

and Manivel, M. (1997). Selection of Sites for

Artificial Recharge Towards Groundwater

Development of Water Resource in India.

Proceeding of the 18th Asian Conference on

Remote Sensing, Kuala Lumpur. 20 - 24 October.

[3] Krishnamurthy, J., Kumar, V.N., Jayaraman, V.,

and Manivel, M. (1996). An approach to demarcate

ground water potential zones through remote

sensing and GIS. International Journal of Remote

Sensing, 17(10), 1867-1884.

[4] Kumar, A., Tomar, S., and Prasad, L.B. (1999).

Analysis of fractures inferred from DBTM and

remotely sensed data for ground water development

in Godavari sub-watershed, Giridith, Bihar. Journal

of Indian Society Remote Sensing, 26(2), 105-114.

[5] Kundu, P 2000,Integrated remote sensing and GIS

based Hydrologic modeling for groundwater

recharge investigation. M. Tech dissertation report,

University of Roorkee (Unpublished).

[6] Pradeep, K.J. (1998). Remote sensing techniques to

locate ground water potential zones in Upper Urmil

river basin, District Chatarpur, Central India.

Journal of Indian Society Remote Sensing, 26(3),

135-147

[7] Rao, U.R. (1991). Remote sensing for national

development. Curr. Sci., 61(3&4), 121-128.

[8] Ravinderan, K.V., and Jayaram, A. (1997). Ground

water prospects of Shahbad Teshil, Baran District,

Eastern Rajasthan: A remote sensing approach.

Journal of Indian Society of Remote Sensing, 25(4),

239-246.

[9] Sankar, K. (2002). Evaluation of Ground water

potential zones using Remote sensing data in upper

Vaigai river basin,Tamil nadu, India. Journal of

Indian Society of Remote Sensing, 30(3), 119-129.

[10] Saraf, A.K., and Chaudhury, P.R. (1998).

Integrated Remote Sensing and GIS for

Groundwater Exploration and Identification of

artificial recharge sites. International Journal of

Remote Sensing, 19(10), 1825-1841.

[11] Thronbury, W.D. (1990). Principle of

Geomorphology. Wiley Eastern Limited, New

Delhi, 594p.

Figure 1.1: Study Area Map of Lower Palar River Basin, Tamilnadu, India





Figure 1.2: Overall Methodology of the Present Study

Figure 1.3: Geology of Lower Palar River Basin, Tamilnadu

1595 S. SARAVANAN



Figure 1.4: Geomorphology of Lower Palar River basin, Tamilnadu

Figure 1.5: Lineament of Lower Palar River basin, Tamilnadu





Figure 1.6: Landuse of Lower Palar River Basin, Tamilnadu

Figure 1.7: Groundwater Potential Zone of Lower Palar River Basin, Tamilnadu

1597 S. SARAVANAN



Table 1 Ranking and Weightages of Various Themes.

Theme Class

Weights for

Groundwater

potential zone

Weights for

Artificial

recharge zone

Geomorphology

BuriedPediment (Deep), FloodPlain,

Bazada, Valleyfill Buried Channel,

Palaeo Channel,

4 4

Medium Dissected Plateau, Buried

Pediment (Shallow), 3 3

Inselberg, Highly Dissected Plateau,

Pediment, 2 2

Denudational Hills Deflection Slope,

Ridges Structural Hills 1 1

Hydrological Soil

Group

A 4 4

B 3 3

C 2 2

D 1 1

Slope

0-3% 4 3

3-5% 3 3

5-10% 2 2

>10% 1 1

Lineament

Density

High 4 4

Moderate 3 3

Less Moderate 2 2

Low 1 1

Geology

Alluvium, Schist with low clay

content, Weathered Gneiss 4 4

Charnockite, Kankar, Limestone 3 3

Pegmatite, Laterite 2 1

Landuse

Wet crop, Plantation 4 3

Dry crop, Fallow, Harvested land 3 3

Scrub, Barren 2 2

Rock outcrops, Forest & Others 1 1

Drainage Density

Less 4 4

Less Moderate 3 3

Moderate 2 3

High 1 1





Figure 1.8: Artificial Recharge Zone of Lower Palar River Basin, Tamilnadu

Figure 1.9: Groundwater Level Fluctuation Map of Lower Palar River Basin, Tamilnadu

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1599-1608


Drinking and Irrigation Water Quality in Jalandhar and

Kapurthala Districts, Punjab, India: Using Hydrochemsitry

P. PURUSHOTHAMAN1, M. SOMESHWAR RAO

1, B. KUMAR

1, Y. S. RAWAT

1, GOPAL KRISHAN

1,

S. GUPTA2, S. MARWAH

2, A. K. BHATIA

2, Y. B. KAUSHIK

2, M. P. ANGURALA

2 and G. P. SINGH

2

1National Institute of Hydrology, Roorkee- 247 667, India

2Central Groundwater Board (North Western Region), Chandigarh, India


Abstract: The physical and chemical parameters of groundwater play a significant role in classifying and assessing water quality. Hydrochemical study reveals quality of water that is suitable for irrigation, agriculture, drinking and industrial purposes. In this study groundwater samples from different aquifers, shallow, medium and deep, were collected and analysed for major ions. The analysed samples were used for classifying water type, source and quality for irrigation purpose. The major ionic abundance in the area shows that trend Ca++>Mg+>Na+>K+ (Shallow and Medium aquifer) Na+>Ca++>Mg+>K+ (Deep aquifer) and HCO3

->Cl->SO4--. The dominant hydrochemical facies in

shallow and medium aquifer is CaMgHCO3 type and in deep aquifer is NaHCO3 type. The drinking water quality is very good with the water quality for irrigation purpose is good in the study area. The study reveals that there is a chance of precipitation of carbonate minerals which may poses risk for soils in the study area.

Keywords: Drinking water Quality, Groundwater Chemistry, Irrigation water Quality, Jalandhar and Kapurthala

Districts, Punjab

Introduction:

Groundwater is one of the most important resources for human life. It is estimated that approximately one third of the world’s population use groundwater for drinking [1]. Generally, groundwater quality depends on the quality of recharged water, atmospheric precipitation, inland surface water and subsurface geochemical processes [2]. The domestic and industrial contaminants discharged in to the environment deteriorate the quality of groundwater and resulting in the pollution of irrigation and drinking water [3]. As the groundwater contains a wide variety of dissolved inorganic species in various concentrations, as a result of chemical and biochemical interactions between groundwater and geological materials through which it flows; and to a lesser extent because of contributions from the atmosphere, surface water bodies and anthropogenic activities [4]. Each groundwater system in an area is known to have a unique chemistry, which is acquired as a result of chemical alteration of the meteoric water recharging the system [5,6]. The chemical alteration of meteoric water depends on several factors such as soil-water interaction, dissolution of mineral species, duration of rock-water interaction and anthropogenic sources. Thus, the groundwater chemistry could reveal important information on the geological characteristics of the aquifers and the suitability of groundwater for domestic, industrial and agricultural purposes [7].

Importance of hydrochemistry of groundwater has led to a number of detailed studies on geochemical evolution of groundwaters [4, 7-29]. In which the chemistry of the groundwater have been used to understand the hydrogeochemical characteristics of the region. The Jalandhar and Kapurthala districts in the state of Punjab, India depends mainly on groundwater for agricultural and domestic purpose despite the presence of two perennial rivers, R. Satluj and R. Beas, and well connected canal network. The groundwater level show declining trends particularly around the areas of Phagwara in Kapurthala district and Shahkot and Nakodar of Jalandhar district [30,31]. Groundwater is used more than 80% for day to day activities in the urban regions of Jalandhar and Kapurthala districts [32]. The irrigation activity in these districts is done through canal systems and groundwater. The usage of canal system is very minimal for irrigation purpose with about 11.37% in Jalandhar and none in Kapurthala districts resulting in more usage of groundwater. But there has been decrease of over 35% in the use of canal system in the state for irrigation purpose with increase in usage of groundwater [32]. The main crop production in these districts is paddy and wheat (fig. 1a,b) and there is an increase in the consumption of water for agricultural activities from 1981- 2007 (Fig. 1c) [33]. Due to these intense usage of groundwater the groundwater in all the blocks of Jalandhar and Kapurthala districts has been experiencing high groundwater draft and have resulted

1600 P. PURUSHOTHAMAN, M. SOMESHWAR RAO, B. KUMAR, Y. S. RAWAT, GOPAL KRISHAN, S. GUPTA, S. MARWAH, A. K. BHATIA, Y. B. KAUSHIK, M. P. ANGURALA and G. P. SINGH


in overexploitation with high groundwater stage of 254% and 204% in Jalandhar and Kapurthala districts respectively [30,31]. But, there is a lack of knowledge about the groundwater chemistry and suitability of groundwater for various purposes of this area. Hence,

the present study is focussed to understand the hydrochemical characteristics and groundwater quality for both drinking and irrigation purpose in these districts.

Figure 1: Water Utilized for Two Major Crops A: Rice, b: Wheat and c: Irrigation Purpose in the Jalandhar and

Kapurthala Districts (Source: Dept of Agriculture Punjab, 2007 [33])

Study Area:

The study area, Jalandhar and Kapurthala districts lies between latitude 30°59′ to 31°39′N and longitude 74°55′ to 75°57′ with a total geographical area of about 4000 sq.km [30,31] (Fig. 2). The study area is part of Bist Doab Tract, which is inter- alluvial plain between Beas and Satluj River. The district Kapurthala comprises of two non-contiguous parts with Tehsil Phagwara as separated portion. Jalandhar District is bounded in the south by R. Satluj and Kapurthala, Hoshiarpur and Nawanshahar districts in W-NW, N-NE and Eastern part respectively [31]. Kapurthala District is bounded partly in the North and wholly in the West by the Beas River. The Phagwara block is surrounded on three sides, the NW, W and SW by Jalandhar District, on the NE and east by Hoshiarpur District and by Nawan Shahar in the South. Climate in the study region is classified as tropical and dry sub humid. The study area receives average annual precipitation 701mm and 779 mm in Jalandhar and Kapurthala districts respectively with almost 70% of rainfall occurring during the southwest monsoon [30,31]. The Jalandhar district and Phagwara

and Kapurthala areas of Kapurthala districts are well connected with the R. Satluj canal system and its tributaries for irrigation purpose. Physiographically, the study area comprises of recent alluvium belonging to Indo- Gangetic alluvial plain of Quaternary age [30,31]. The Jalandhar district is mainly drained by the R. Satluj and its tributaries East (White) and West (Black) Bein and the Kapurthala district by the R. Beas. The groundwater flow in the Jalandhar district is towards South-West direction while the Kapurthala district is from north to southeast direction [30,31].

Sampling and Analysis:

The sampling of groundwater at different depths from peizometer (shallow (<30m), medium (30-60m) and deep aquifer (>60m)) was carried out during 2007- 2009 from various locations in the Jalandhar and Kapurthala Districts (Fig. 2) by Central Groundwater Board, Northwest Region (CGWB(NWR)), Chandigarh. The collected groundwater samples were analysed for major cations and anions at the chemistry lab in the CGWB (NWR), Chandigarh.

1601 Drinking and Irrigation Water Quality in Jalandhar and Kapurthala Districts, Punjab, India: Using Hydrochemsitry


PUNJAB

INDIA

K

Figure 2: Sampling Location Map of the Study Area (Jalandhar and Kapurthala Districts)

Result and Discussion:

A. Major Ion Chemistry:

The pH of the groundwater in the study regions shows alkaline nature irrespective of the aquifer sampled. The shallow and medium aquifer show less alkalinity with pH values ranges from 7 to 8 whereas, the deeper aquifer show higher alkalinity with values ranging between 7.6 and 8·2. The pH of groundwater in the study area is within the limits of Indian Standard (IS) [34] and WHO standard [35] for drinking water quality. The electrical conductivity (EC) of the groundwater samples in the shallow aquifer shows values higher than desirable limit (782 µS/cm) and lesser than the permissible limit (2340 µS/cm) of WHO Standard. The other two aquifers (medium and deep) show lower EC with the values less than the desirable limit of Indian and WHO standards.

The total dissolved cations and anions are well balanced within ±5% of normalized inorganic charge balance (NICB). The calcium ions dominate the overall cation concentration in the groundwater followed by magnesium, sodium and potassium ions (Table 1) in shallow and medium aquifers. In deep aquifer the sodium ion dominates followed by the calcium ions the overall cation concentration in the groundwater. The concentration of sodium (Na+) increases than other ions in the medium and deep aquifer when compared with shallow aquifer. The sodium (Na+) and potassium (K+) does not have any prescribed limit for drinking water quality, but high concentration of sodium ion results in salty nature of water. The calcium (Ca++) concentration in the shallow groundwater shows that all the samples except that of Kartarpur and Bholath are within the Desirable limit of both WHO and IS standards of drinking water (<75Mg/l). The locations kartarpur and Bholath show calcium (Ca++) concentration within the permissible limit of WHO and IS standards (<200Mg/l).

The other two aquifers, medium and deep, show calcium (Ca++) concentration within the desirable limit of WHO and IS standards. The magnesium (Mg++) concentration in the groundwater of the study area is well within the permissible limit of the WHO and IS standards (<100Mg/l). Among the anions, bicarbonate ion dominates the total anion concentration in all the aquifers followed by chloride. The chloride concentration in the groundwater in all the aquifers are well within the desirable limit of the WHO and IS standards (<250Mg/l) except the shallow groundwater at Kartarpur (445Mg/l) which is below the desirable limit. The concentration of chloride in the shallow aquifer at Kartarpur is very high (445 mg/l) indicating high evaporation activity in this part of the study area. The sulphate concentration increases at some locations making it as the second most dominant species in the shallow (Shahkot) and deeper aquifer (Shahkot and Sarih). The sulphate and nitrate concentration in the groundwater is well within the desirable limit of the WHO and IS standards. This shows that the groundwater in the study area is less affected with anthropogenic activities and is suitable for drinking purposes.

B. Classification of Groundwater:

To understand the similarities between the groundwater in the study area they have been classified hydrochemically using the major cations and anions with the conventional Piper trilinear diagram [36] and Chadha’s diagram [37]. To evaluate the hydrochemistry of the groundwater the major cations (Ca++, Mg++, Na+, K+) and anions (HCO3

-, Cl-, SO4--) in the groundwater

samples were plotted on the Piper trilinear diagram. The plot shows that the groundwater in the study area is rich in alakaline earths with the location Kartarpur showing alkalis rich water in the shallow aquifer. The diamond domain explains the variations in major cations and



anions. The groundwater in shallow aquifer samples fall mostly on the CaMgHCO3 field with the samples at locations Goraya, Bholath fall on the field and at Kartarpur falls on the CaMgCl field (Fig. 3). The groundwater in the medium aquifer shows mixed trend with the samples fall in the CaMgHCO3 and NaHCO3 fields (Fig. 3). The plot shows that deep groundwater is of NaHCO3 field (Fig. 3). The two triangle plot shows that the water samples are of bicarbonate type except that of Kartarpur which is of chlorine type. Among cations the water samples fall mostly on the sodium field and with few samples falling on the mixed water field. The Chadha’s diagram (Fig. 4) clearly shows that most of the water samples at shallow groundwater fall on CaMgHCO3 field with samples at location Goraya and Bholath falling at NaHCO3 field and with Kartarpur falling at NaClSO4 field. The medium groundwater shows mixed characteristics with the samples falling both at CaMgHCO3 and NaHCO3 field. The groundwater at deeper aquifer shows the water is mainly of NaHCO3 type with the samples at locations Adampur and Phillaur showing CaMgHCO3 type.

The water type in the study area can be stated as follows: Shallow aquifer: CaMgHCO3 + NaHCO3 (Goraya +

Bholath) + CaCl (Kartarpur) Medium Aquifer: CaMgHCO3 + NaHCO3 Deep Aquifer: NaHCO3 + CaMgHCO3 (Adampur,

Shahkot)

C. Water Quality Classification for Irrigation:

As the study area mainly utilizes the groundwater for irrigation purposes it is necessary to know the quality of the groundwater for irrigation purposes. Sodium concentration is an important criterion in irrigation-water classification because sodium reacts with the soil to create sodium hazards by replacing other cations. The extent of this replacement is estimated by Residual Sodium carbonate (RSC) and sodium percentage (Na %). The diagram used for understanding the suitability of ground water for irrigation purposes is based on the sodium adsorption ratio (SAR) and electrical conductivity of water. Hence, to understand the water quality for irrigation purposes all the above said parameters have been analysed (Table 2).

1. Sodium Adsorption Ratio:

This is also expressed as salinity hazard. This index quantifies the proportion of sodium (Na+) to calcium (Ca2+) and magnesium (Mg2+) ions in a sample. Sodium hazard of irrigation water can be well understood by knowing SAR. The SAR values for each water sample were calculated by using the equation [38] SAR= Na+/ √ (Ca+++Mg++)/2)

Where, the concentrations are in meq/l. SAR is an important parameter for the determination of the suitability of irrigation water because it is responsible for the sodium hazard [39].The SAR value of groundwater in all the aquifers of the study region is less than 10 showing excellent nature of soil for irrigation purpose. The SAR shows excellent (Table 2) nature of water quality for irrigation purpose.

2) Salinity and Sodium Hazard:

The water quality for irrigation can be classified using the salinity and sodium hazard diagram [41]. The diagram is divided in to C1 to C4 with increase in salinity hazard and S1 to S4 with increase in SAR (Fig. 5). In the present study the groundwater in different aquifers of the study area falls mostly on the C2S1 field indicating good quality. Three samples in the deeper aquifer fall on C2S2 field (Sarih, Sultanput Lodhi and Shahkot) shows moderate quality and almost all the samples in shallow aquifer and one sample in medium aquifer falls on C3S1 filed also indicates moderate quality. Overall, the groundwater in the study area is suitable for irrigation purpose.

3) Sodium Percentage:

The high concentration of sodium in the soil releases calcium and magnesium in the soil particles due to absorption of sodium in clay particles. Sodium concentration is important in classifying irrigation water because sodium reacts with soil to reduce its permeability. The role of sodium in the classification of groundwater for irrigation was emphasised because of the fact that sodium reacts with soil and as a result clogging of particles takes place, thereby reducing the permeability [39, 40]. Wilcox [41] proposed a classification of irrigation waters based sodium percentage which is calculated using the formula given below. Na%= (Na++K+)*100/ (Ca+++Mg+++Na++K+) The EC vs Na% (Fig. 7) plot shows that groundwater samples in the shallow aquifer show good to permissible nature. The medium and most of the deep aquifer samples show excellent to Good nature with few deep aquifer samples showing good to permissible nature (Fig. 6).

4) Residual Sodium Carbonate:

The relative abundance of sodium with respect to alkaline earths and boron, and the quantity of bicarbonate and carbonate in excess of alkaline earths also influence the suitability of water for irrigation. This excess is denoted by ‘Residual sodium carbonate’ (RSC) and is determined as suggested by Richards [38]. The water with high RSC has high pH and land irrigated by such waters becomes infertile owing to deposition of



sodium carbonate as known from the black colour of the soil [43]. The RSC can be denoted as given below:

Residual sodium carbonate (RSC) = (HCO3- + CO3

-) - (Ca+++Mg++) Where, the samples are in Meq/l.

Table 1: Major Ion and Stable Isotope Composition of Groundwaters in Jalandhar and Kapurthala Districts, BIST

DOAB Region

Location pH EC

(µS/cm)

Ca 2+

(meq/l)

Mg 2+

(meq/l)

Na+

(meq/L)

K+

(meq/l)

HCO3-

(Meq/l)

Cl-

(Meq/l)

SO42-

(meq/l)

NO3-

(meq/l)

F-

(meq/l)

NICB

(%)

Shallow Kartarpur (J) 7.2 1595 8.63 6.17 2.10 0.20 3.23 12.54 0.83 0.01 0.02 1.41 Goraya (J) 7.7 1100 1.80 2.30 6.90 0.18 9.48 0.99 0.35 0.32 0.04 0.01

Nakodar (J) 7.0 750 2.24 1.56 3.49 0.10 6.38 0.54 0.21 0.32 0.02 -0.38

Shakot (J) 7.0 1213 3.64 5.35 3.49 0.20 7.31 2.06 2.60 0.71 0.01 0.01

Phillaur (J) 7.8 1244 3.34 4.69 4.59 0.18 10.97 0.96 0.52 0.39 0.05 -0.33

Bholath (K) 7.1 795 4.74 1.48 1.75 0.15 6.33 1.38 0.29

0.02 0.62

Kapurthala (K) 7.2 700 3.54 1.73 1.88 0.16 4.92 0.99 0.81 0.90 0.02 -2.21

Sultanpur Lodhi (K)

7.2 1006 2.34 2.96 5.24 0.12 7.82 1.24 1.25 0.02 0.01 1.62

Phagwara (K) 7.5 850 2.44 2.63 3.06 0.18 6.21 1.49 0.50 0.90 0.01 -4.62

Medium

Adampur (J) 8.0 710 1.45 1.32 4.93 0.10 5.25 1.75 0.81

0.06 -0.41

Jalandhar (J) 7.2 510 2.34 0.62 2.01 0.08 4.46 0.39 0.12

0.04 0.27

Sarih (J) 7.0 670 1.75 1.56 3.23 0.15 6.02 0.39 0.29

0.02 -0.21

Kartarpur (J) 7.2 958 2.69 0.91 6.81 0.13 6.80 2.06 1.37 0.04 0.04 1.13

Goraya (J) 8.1 784 2.19 3.46 1.75 0.16 6.92 0.56 0.07

0.03 -0.13

Nakodar (J) 7.1 702 1.55 2.88 2.97 0.17 6.69 0.42 0.19

0.03 1.58 Bholath (K) 7.7 455 1.85 0.66 2.10 0.10 4.08 0.56 0.00

0.03 0.33

Deep Adampur (J) 7.8 400 1.50 0.82 2.10 0.08 4.11 0.24 0.10 0.00 0.01 0.24 Sarih (J) 8.0 565 0.49 0.39 5.07 0.08 4.30 0.37 1.25 0.00 0.01 0.84

Jalandhar (J) 8.0 550 0.60 0.74 4.02 0.05 4.34 0.87 0.19 0.02 0.04 -0.59

Kartarpur (J) 7.6 510 1.05 0.62 3.10 0.05 4.34 0.28 0.15 0.01 0.01 0.21 Goraya (J) 7.9 660 1.10 0.82 4.67 0.10 5.64 0.28 0.56

0.02 1.47

Nakodar (J) 7.5 685 0.70 0.99 5.20 0.07 6.08 0.54 0.17

0.03 1.05 Shakot (J) 7.6 704 0.60 0.30 6.20 0.05 4.21 1.10 1.66

0.03 1.10

Phillaur (J) 7.6 690 2.04 2.30 2.53 0.19 6.62 0.28 0.12

0.02 0.18 Bholath (K) 8.0 410 1.05 0.41 2.53 0.05 3.26 0.56 0.19

0.02 0.13

Kapurthala (K) 7.6 510 0.65 0.46 4.06 0.05 4.33 0.48 0.33 0.03 0.02 0.32 Sultanpur Lodhi (K)

8.0 503 0.50 0.20 4.45 0.04 3.20 1.01 0.83 0.03 0.02 0.97

Phagwara (K) 7.9 655 1.30 1.56 3.89 0.14 6.00 0.28 0.37

0.03 1.51



Figure 3: Piper Plot of the Groundwater at different Aquifers in Jalandhar and Kapurthala Districts

of Bist- Doab Region

-80 -40 0 40 80

(Ca+Mg)-(Na+K)

-80

-40

0

40

80

(HCO3)-(C

l+SO4)

-80 -40 0 40 80

-80

-40

0

40

80

Shallow

Medium

Deep

Figure 4: Chadha’s Classification of Groundwater at different Depth in the Jalandhar and Kapurthala Districts

of Bist- Doab Region



Table 2: Classification of Water for Irrigation Purposes in Jalandhar and Kapurthala Districts

Parameters Range classification Sh. Med. Deep

(Number of

samples)

SAR

<10 Low 9 7 11 11-18 Medium 18-26 High >26 Very High

Salinity Hazard (EC)

(µS/cm)

<250 Low 250- 750 Medium 4 5 11 750- 2250 High 5 2

>2250 Very High

Na%

<20 Excellent 1 20- 40 Good 5 3 1 40- 60 Permissible 1 2 2 60- 80 Doubtful 2 2 5

>80 Unsuitable 3

RSC <1.25 Good 3 1

1.25- 2.50 Doubtful 1 3 3 >2.50 Unsuitable 5 3 8

Sh.: Shallow; Med: Medium Aquifer; Deep: Deep Aquifer

Sodium H

azard

Low

Very High

High

Medium

Salinity Hazard

Low Medium High Very High

100 1000

EC(µS/cm)

0

4

8

12

16

20

24

28

32

SAR

100 1000

0

4

8

12

16

20

24

28

32

Shallow

Medium

Deep

Figure 5: Water Classification According to SAR and EC [42]



UnsuitableExcellent to Good

Good to Permissible Permissible to

Unsuitable

1000 2000 3000 4000 5000

EC(µS/cm)

0

10

20

30

40

50

60

70

80

90

100

Sodium Percent

1000 2000 3000 4000 5000

0

10

20

30

40

50

60

70

80

90

100

Shallow

Medium

Deep

Figure 6: Classification of Irrigation Waters (Wilcox, 1955)

Almost all groundwater samples in the study region show high percentage of sodium and high RSC values (Table 2) indicating high concentration of sodium precipitation of carbonates. When the concentration of sodium ion is high in irrigation water, Na+ tends to be absorbed by clay particles, displacing magnesium and calcium ions. This exchange process of sodium in water for Ca2+ and Mg2+ in soil results in precipitation of calcium and magnesium bicarbonate with high concentration of bicarbonate resulting the reduction of permeability and eventually results in soil with poor internal drainage [27]. This shows that the soil in the study region suffers high risk in precipitation of carbonate that will result in the decrease in permeability.

Conclusion:

The study shows that groundwater quality in the area varies with different aquifers. The groundwater quality in the study area is suitable for drinking purpose with the concentration of most of the ions falling below desirable limit and all values lying within the permissible range of WHO and IS standards. The Piper plot and Chadha’s Diagram shows that groundwater in the shallow and medium aquifer are similar i.e. Ca+Mg+Na bicarbonate type whereas the deep aquifer shows NaHCO3 type of water. The study also shows that water quality for irrigation purpose differs widely. The sodium adsorbtion ratio (SAR) shows that water quality in all the aquifers is excellent. The SAR vs EC and Wilcox classification of water shows that the

groundwater in the study area is suitable for irrigation purpose. The sodium percentage and residual sodium carbonate (RSC) values in the groundwater shows possible precipitation of clay particles or Ca, Mg carbonates in the soil resulting in the deterioration/ degradation of soil. In summary, it is evident from the study that the water quality for drinking and irrigation purpose is suitable in the study area.

Acknowledgement:

The authors are thankful to Chandigarh regional Directorate of Central Ground Water Board for providing groundwater chemistry data for Jalandhar and Kapurthala Districts, Punjab, India. The present work is part of the research project funded by the World Bank under Purpose Driven Study of HP II, Ministry of Water Resources, Government of India, (GoI), initiated by the Hydrological Investigations Division of NIH, Roorkee, India. Hence, the financial support from the World Bank is acknowledged.

References:

[1] Nickson, R.T., Mc Arthur, J.M., Shrestha, B., Kyaw- Nyint, T.O. and Lowry, D. 2005. Arsenic and other drinking water quality issues, Muzaffargarh District, Pakistan. App. Geochem. 20, 55-68.

[2] Twarakavi, N. K. C. and Kaluarachchi, J. J. 2006. Sustainability of groundwater quality considering land use changes and public health risks. J. Environ. Manag. 81, 405-419.



[3] Forstner, U. K. and Wittman, G. T. W. 1981. Metal Pollution in the Aquatic Environment. Springer Verlag, Berli, Heidelberg, pp- 255.

[4] Obiefuna, G. I. and Orazulike, D. M. 2011. The Hydrochemical Characteristics and Evolution of Groundwater in Semiarid Yola Area, Northeast, Nigeria. Res. J. Environ. Earth Sci. B, 400-416.

[5] Back, W. 1966. Hydrochemical facies and groundwater flow pattern in northern part of Atlantic Coastal Plain. US Geological Survey Professional Paper 498A.

[6] Drever, J. J. 1982. The geochemistry of natural waters. Prentice-Hall, Englewood Cliffs, p. 388.

[7] Aghazadeh, N. and Mogaddam, A. A. 2010. Assessment of Groundwater Quality and its Suitability for Drinking and Agricultural Uses in the Oshnavieh Area, Northwest of Iran. J. Environ. Protect. 1, 30-40.

[8] Garrels, R. M. 1967. Genesis of some groundwaters from igneous rocks. In: Abelson, P. H. (ed) Researches in geochemistry. Wiley, New York, 405-420.

[9] Garrels, R. M. and MacKenzie, F. T. 1967. Origin of the chemical compositions of some springs and lakes. In: Equilibrium Concepts in Natural Waters. American Cancer Society, Washington, DC.

[10] Paces, T. 1973. Steady state kinetics and equilibrium between groundwater and granite rock. Geochim. Cosmochim. Acta. 37, 2641- 2663.

[11] Frape, S. K., Fritz, P. and McNutt, R. H. 1984. Water-rock interaction and chemistry of groundwaters from the Canadian Shield. Geochim. Cosmochim. Acta. 48, 1617-1627.

[12] Stuyfzand, P. J. 1986. A new Hydrochemical classification of water Types: Principles and application to the coastal dunes aquifer system of the Netherlands. Proceedings of the 9th Salt water Intrusion Meeting, Delft, pp- 641-655.

[13] Hem, J. D. 1989. Study and interpretation of the chemical characteristics of natural water. US Geol. Surv. Water-Supply Paper 2254.

[14] Sarin, M. M., Krishnaswami, S., Killi, K., Somayajulu, B. L. K. and Moore, W. S. 1989. Major ion chemistry of the Ganga-Brahmaputra River basin system: weathering processes and fluxes to the Bay of Bengal. Geochim. Cosmochim. Acta. 53, 997-1009.

[15] Walraevens, K. 1990. Hydrogeology and Hydrogeochemistry of the Ledo-Paniselian semi confined aquifer in East and West Flanders. Acad. Analecta 52, 11-66

[16] Stuyfzand, P. J. 1993. Hydrogeochemistry and hydrology of the coastal dune area of the western Netherlands. Ph.D. Thesis, Free University Amsterdam, Netherlands.

[17] Garcia, M. C., Hidalgo-Medel, V., Apella, M. C. and Blessa, M. A. 1998. Chemical quality of deep groundwater in Yerba Buena and San-Miguel de Tucuman Cities (Abstract). Vol. 2, School of Environment Sciences and Technology, Buenos- Aires, pp: 498.

[18] Garcia, M. C., Hidalgo-Medel, V., Apella, M. C. and Blessa, M. A. 1998. Chemical characteristics of deep groundwater in the cities of San Miguel de Tucuman and Yerba Buena Province of Tucuman. Proc. 10th Lat.-Am.Geol. Cong. and 6th Natural Cong. of Eco. Geol., Buenos- Aires, Vol. 3, 342-347.

[19] Mclean, W., Jankowski, J. and Lavitt, N. 2000. Groundwater Quality and Sustainability in an Alluvial Aquifer, Australia. In: Silio, O. and Balkema, A. A. (Eds.), Groundwater, Past Achievements and Future Challenges. Rotterdam, 567-573.

[20] Pannatier, E. G., Broers, H. P., Venema, P. and Van Beusekom, G. 2000. A new Process- Based Hydrogeochemical classification of Groundwater: Application to the Netherlands National Monitoring System. INO report, NITG 00-143-B, 38, NITG, Utrecht.

[21] Galindo, M. C., Vece, M. B., Perondi, M. E., Moserrate Araoz, M., Garcia, M.G., Hidalgo-Medel, V., Apella, M. C. and Blesa, M. A. 2001. Chemical behaviour of the Sali-River Province of Tucuman, Argentina. Environ. Geol. 40, 847-852.

[22] Walraevens, K., Mahauden, M. and Coetsiers, M. 2003. Natural background Concentration of trace element in aquifers of the Flemish region, as a reference for the government sanitation policy. FZK/TNO Conference on contaminated soil (Consoil), Ghent, 215-224.

[23] Hartman, J., Berna, Z., Stuben, D. and Henze, N. 2005. A statistical procedure for the analysis of seismotechtonically induced hydrochemical signals: a case study from the Eastern Carpathians, Romania. Tectonophy. 405, 77-98.

[24] Nagaraju, A., Suresh, S., Killham, K. and Hudson-Edwards, K. 2006. Hydrogeochemistry of Waters of Mangampeta Barite Mining Area, Cuddapah Basin, Andhra Pradesh, India. Turk. J. Eng. Environ. Sci. 30, 203-219.

[25] Jezersky, Z. 2007. Hydrogeochemistry of a deep gas storage cavern, Czech Republic. Hydrogeol. J. 15, 599-614.

[26] Zhu, G. F., Su, Y. H. and Feng, Q. 2008. The hydrochemical characteristics and evolution of groundwater and surface water in the Heihe River Basin, northwest China. Hydrogeol. J. 16, 167-182.

[27] Belkhiri, L., Boudoukha, A. and Mouni, L. 2010. Groundwater quality and its suitability for drinking



and agricultural use in Ain Azel plain, Algeria. J. Geogr. Reg. Plann. 3, 151-157.

[28] Karmegam, U., Chidamabram, S., Sasidhar, P., Manivannan, R., Manikandan, S. and Anandhan, P. 2010. Geochemical Characterization of Groundwater’s of Shallow Coastal Aquifer in and around Kalpakkam, South India. Res. J. Environ. Earth Sci. 2, 170-177.

[29] Tiri, A., Belkhiri, L., Boudoukha, A. and Lahbari, N. 2011. Characterization and evaluation of the factors affecting the geochemistry of surface water of Koudiat Medouar Basin, Algeria. African J. Environ. Sci. Tech. 5, 355-362.

[30] CGWB. 2007. Ground Water Scenario of Jalandhar District, Punjab. Central Ground Water Board North Western Region, Chandigarh, pp- 15.

[31] CGWB. 2007. Groundwater Information Booklet Kapurthala District Punjab. Central Ground Water Board North Western Region, Chandigarh, pp- 20.

[32] Tiwana, N. S., Jeerat, N., Ladhar, S. S., Singh, G., Paul, R., Dua, D. K. and Parwana, H. K. 2007. State of Environment: Punjab- 2007. Punjab state council for science and technology, pp. 243.

[33] BIS. 1991. Specification for drinking water IS: 10500: 1991. Bureau of Indian Standards. New delhi.

[34] WHO (World Health Organization). 2004. Guidelines for drinking water quality. 2nd edn, Vol. 1, Geneva, WHO, pp- 130.

[35] Piper, A. M. 1944. A graphic procedure in the geochemical interpretation of water analyses. Trans. Am. Geophys. Union 25, 914-928.

[36] Chadha, D. K. 1999. A proposed new diagram for geochemical classification of natural waters and interpretation of chemical data. Hydrogeol. J. 7, 431-439.

[37] Richards, L. A. 1954. Diagnosis and Improvement of Saline and Alkali Soils Agriculture Handbook. 60, Department of Agricultural, Washington DC, US, pp- 160.

[38] Todd, D. K. 1980. Groundwater Hydrology. 2nd ed., John Wiley and Sons, New York, p. 535.

[39] Domenico, P. A. and Schwartz F. W. 1990. Physical and Chemical Hydrogeology. John Wiley and Sons, New York.

[40] Wilcox, L. V. 1955. Classification and use of irrigation water. US Geological Department Agri. Circ. 969: 19.

[41] U.S. Salinity Laboratory. 1954. Diagnosis and Improvement of Saline and Alkaline soils. U.S. Dept. of Agriculture. Hand Book, No. 60, 160.

[42] Eaton, E. M. 1950. Significance of Carbonate in irrigation water. Soil Sci. 69, 123-133.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1609-1618


Groundwater Quality Assessment around Talabasta Area, Banki

Sub-Division, Odisha, India

ROSALIN DAS1, MADHUMITA DAS

2 and SHREERUP GOSWAMI

3

1Department of Geology, Banki Autonomous College, Banki-754008

2Department of Geology, Utkal University, Vani Vihar, Bhubaneswar-751004, Odisha

3Department of Geology, Ravenshaw University, Cuttack-753003, Odisha


Abstract: The study area forms part of the Athgarh basin of upper Gondwana formation consisting of a thick pile of

fresh water fluviatile and lacustrine sediments. The deeper aquifers are the main source of drinking water supply for

the local community. Twenty groundwater samples have been collected during the pre-monsoon and post-monsoon

periods of 2010 and analyzed for the water quality parameters such as pH, electrical conductivity, total dissolved

solids, calcium, magnesium, sodium, potassium, bicarbonate, sulphate, chloride etc. Suitability of the ground water

for different purposes such as drinking and irrigation is evaluated following various classification schemes and water

quality standards. Piper’s Trilinear diagram reveals that water of the study area belongs to the Ca-Mg- HCO3 facies.

Keywords: Talabasta, Athgarh Formation, Groundwater quality, Odisha

Introduction:

The study area is located in Banki sub-division of

Cuttack district, Odisha (Fig. 1) bounded by 200 15’ N

to 200 29’N latitudes and 85

0 20’ to 85

0 45’ E longitudes

and covers a total area of 58.5 hectors. The study area

belongs to the Athgarh basin of upper Gondwana

formation. It consists of a thick pile of fresh water

fluviatile and lacustrine sediments. The area enjoys

humid tropical climate. The nearest perennial source of

water is Rana river. The rock types of this area are

laterite, conglomerate, sandstone (feldspathic and

ferruginous), sandy clay, ironstone shale and fireclay.

Fireclay occurs interstratified with sandstone having

limited down dip extension. Ironstone shale appears to

be a marker horizon above fireclay bed. The occurrence

of fireclay is formed by weathering of feldspathic

constituents in the zone of oxidation.

Materials and Method:

20 water samples have been collected during pre and

post-monsoon periods of 2010. Out of 20 samples 10

samples have been collected from tube wells and 10

samples from dug wells. Electrical conductivity (EC)

and pH have been measured using digital meters

immediately after sampling. Water samples have been

analysed for chemical constituents, such as sodium,

potassium, calcium, magnesium, chloride, bicarbonate,

carbonate, sulphate, nitrate, fluoride and Total

Dissolved Solids (TDS) in the laboratory using the

standard methods as suggested by the American Public

Health Association (APHA, 1995); Trivedi and Goel

(1984) and Vogel (1964). The groundwater quality is

assessed with respect to WHO (2004) and BIS (1991)

standards.


Groundwater Chemistry:

Understanding the quality of groundwater with its

temporal and seasonal variation is important because it

is the factor that determines suitability for drinking,

domestic and agricultural purposes. The pH of water is

an indicator of its quality and geochemical equilibrium

(Hem, 1985). pH of the pre-monsoon samples are acidic

and ranged from 5.25 to 6.89 with an average value of

6.07, while in the post-monsoon an increase is observed

and ranged from 6.05 to 8.21 with an average 7.13

(Tables 1, 2). The EC values ranged from 235µs/cm to

998µs/cm during pre-monsoon and 268µs/cm to

1090µs/cm during post-monsoon period. Higher values

are generally noticed in the southeastern part of the

study area. The TDS values ranged from 177mg/l to

739mg/l and 211mg/l to 827mg/ during pre and post-

monsoon periods respectively. Total Hardness values

ranged from 97mg/l to 466mg/l during pre-monsoon

and 87mg/l to 300mg/l during post-monsoon (Tables 1,

2). Water samples of 14 locations are hard whereas in 2

locations the waters are very hard and water samples of

4 locations are moderately hard in nature.

The chemical parameters of different groundwater

samples are given in the Tables 1 and 2. The

concentrations of chloride ranged from 20.7mg/l to

127.8mg/l during pre-monsoon and 2.53mg/l to

1610 ROSALIN DAS, MADHUMITA DAS and SHREERUP GOSWAMI



142.81mg/l during post-monsoon period. The

concentrations of fluoride ranged from 0.101mg/l to

1.76mg/l during pre-monsoon and 0.04mg/l to 1.73mg/l

during post-monsoon period. Sulphate concentrations

ranged from 0.27mg/l to 63.7mg/l during pre-monsoon

and 0.062mg/l to 56.3mg/l during post-monsoon period.

Nitrate values ranged from 0mg/l to 23.5mg/l and

0.01mg/l to 39.8mg/l during pre and post-monsoon

periods respectively. The calcium concentration ranged

from 26mg/l to 127mg/l and 11.7mg/l to 112mg/l during

pre- and post-monsoon periods respectively. The

magnesium content of the water samples ranged from

7.63mg/l to 36.23mg/l during pre-monsoon and 6.48mg/l

to 34.26mg/l during post-monsoon period. Sodium

concentrations ranged from 1.7mg/l to 88.3mg/l during

pre-monsoon and 9.76mg/l to 97.0mg/l during post-

monsoon period. Potassium concentrations of the study

area ranged from 0.68mg/l to 59mg/l and 1.0mg/l to

65mg/l during pre and post-monsoon periods respectively

(Tables 1, 2).

Drinking Water Suitability:

To assess the suitability of groundwater for drinking

purpose, the concentrations were compared with the

WHO standards (2004) for drinking water, which shows

that most of the surface and subsurface water of the

study area are suitable for the drinking and domestic use

with few exceptions, as most of the parameters are

within the permissible limits (Table 3). Based on TDS

classification, 20% of water samples are excellent in

both the seasons (Table 4). 10% samples of pre-

monsoon and 5% samples of post-monsoon are

classified under poor class on the basis of total hardness

(Table 5).

Piper Trilinear Diagram:

The groundwater samples are plotted in the Piper (1944)

Trilinear Diagram (Fig. 2). The plot shows that most of

the groundwater samples fall in the field of mixed Ca–

Mg-HCO3 type of water. The groundwater samples of

pre-monsoon and post-monsoon fall in the fields 4 and

5, (Fig. 2), which suggest that alkaline earth exceeds

strong acids. Calcium and Sodium are the major cations

in the study area forming 75-90% of the cations.

Bicarbonate is the major anion in the study area.

Irrigation Purpose:

The quality of water suitable for irrigation purpose is

judged on the basis of Sodium Adsorption Ratio (SAR),

Percent Sodium (Na %), Residual Sodium Carbonate

(RSC), Potential Soil Salinity (PS), Permeability Index

(PI) etc.

Sodium Adsorption Ratio (SAR):

Excessive sodium content relative to the calcium and

magnesium reduces the soil permeability (Kelley, 1995;

Tijani, 1994) and thus, inhibits the supply of water

needed for the crops. The excess sodium or limited

calcium and magnesium are evaluated by SAR which is

expressed as,

( ) 2/MgCa

NaSAR

++++

+

+=

Where concentrations are expressed in equivalent per

million (epm). The sodium or alkali hazard in the use of

water for irrigation is determined by the absolute and

relative concentrations of cations and is expressed in

terms of sodium absorption ratio (SAR), proposed by

the U.S.Salinity Laboratory (USSL, 1954). The

calculated values of SAR in the study area ranged from

0.339 to 14.79 in pre-monsoon and 2.717 to 16.69 in

post-monsoon. From SAR value (Table 6) quality of

water is detected. A more detailed analysis for the

suitability of water for irrigation is made by plotting the

data on U.S. Salinity Laboratory diagram (Richards,

1954) (Fig. 3).

In the U.S. Salinity diagram (Richards, 1954) EC is

plotted against SAR (Fig. 3). The analytical data plotted

on the USSL diagram (Richards, 1954) illustrates that

10% of the groundwater samples from pre-monsoon and

20% from post-monsoon fall in the field C3S1,

indicating high salinity and low sodium water, which

can be used for irrigation for almost all types of soil

with little danger of exchangeable sodium (Fig. 3).

Groundwater samples that fall in the low salinity hazard

class (C1) can be used for irrigation of most crops in

majority of soils. However, some leaching is required,

but this occurs under normal irrigation practices except

in soils of extremely low permeability. In pre and post-

monsoon periods 45% and 35% of groundwater samples

respectively fall in medium salinity hazard class (C2).

They can be used if a moderate amount of leaching

occurs. High salinity water (C3) cannot be used in soils

with restricted drainage. Even with adequate drainage,

special management for salinity control is required and

crops with good salt tolerance should be selected. Such

areas need special attention as far as irrigation is

concerned.

Percent Sodium (Na %):

In all natural waters percent of sodium content is a

parameter to assess suitability for agriculture (Wilcox,

1948). A maximum of 60% sodium in ground water is

allotted for agricultural purposes (Ramakrishna, 1998).

Wilcox (1948) defined Sodium percentage in terms of

epm of the common Cations.

++++++

++

+++

+=

KNaMgCa

100)KNa(%Na

In pre-monsoon the sodium percentage (%Na) in the

study area ranges from 10.29% to 58.16%. The highest

1611 Groundwater Quality Assessment around Talabasta Area, Banki Sub-Division, Odisha, India



percentage of sodium is found in the dug well water

sample of Pakhalakhala. And the minimum value is

located in the tube well water sample of Talabasta. In

post-monsoon the value of sodium percentage (% Na)

varies within 11.06 % to 70.53%. Wilcox (1955) used

%Na and specific conductance in evaluating irrigation

waters using the Wilcox diagram as given in (Fig. 4).

Most of the water samples fall in excellent to good and

good to permissible categories indicating their

suitability for irrigation. 25% of samples in post-

monsoon are doubtful, but no water sample is strictly

unsuitable for irrigation (Table 7). The water of

excellent to good and good to permissible can be used

for irrigation purpose.

Residual Sodium Carbonate (RSC):

In addition to the SAR and %Na, the excess sum of

carbonate and bicarbonate in groundwater over the sum

of calcium and magnesium also influences the

suitability of groundwater for irrigation. An excess

quantity of sodium bicarbonate and carbonate is

considered to be detrimental to the physical properties

of soils as it causes dissolution of organic matter in the

soil, which in turn leaves a black stain on the soil

surface on drying. This excess is denoted by Residual

Sodium Carbonate (RSC) and is calculated as follows

(Raghunath, 1987)

RSC= (CO3--

+ HCO3-) – (Ca

+++ Mg

++)

Where all concentrations are expressed in epm.

The classification of irrigation water (WHO, 2004)

according to the RSC values is presented in (Table 8).

The residual sodium carbonate varies from -4.91 to 2.52

in pre-monsoon, while it varies from -3.88 to 1.94 epm

in post-monsoon water samples.

Potential Soil Salinity (PS):

Doneen (1962) proposed a criterion based on the

salinity of the irrigation water, which is an improvement

over the U.S. soil salinity. The potential soil salinity

(PS) is given by the concentration of chloride added to

half of sulphate expressed in epm.

PS=Cl +1/2 SO4

Potential soil salinity varies from 0.66epm to 4.21epm

in pre-monsoon where as in post-monsoon it varies from

0.65epm to 4.11epm. From the PS classification, the

groundwaters of the study area come under excellent to

good category (Table 9).

Permeability Index (PI):

The soil permeability is affected by long-term irrigation

influenced by Na+, Ca

++, Mg

++ and HCO

-3 contents of

the soil. The permeability index (PI) values also indicate

the suitability of groundwater for irrigation. It is defined

as (Doneen, 1962)

The concentration of cations and anions are in epm. The

groundwater samples of the study area fall in class I and

II (Doneen, 1962) on the basis of Permeability Index

(PI) values (Fig.5). According to these values, 30% of

the groundwater comes under class-I category and 70%

comes under class-II category in pre-monsoon. In post-

monsoon period 25% samples fall in class-I and 75%

samples fall in class-II category (Fig. 5). According to

Doneen chart (Domenico and Schwartz 1990) the

ground water of the study area is of good quality for

irrigation purpose based on the permeability index. The

increased percentage of groundwater samples under

class–I1 is due to dilution and subsequent lower values

of permeability index.

Magnesium Ratio (MR):

Generally calcium and magnesium maintain a state of

equilibrium in most waters. More Mg2+

present in

waters will adversely affect the soil quality, converting

it to alkaline and decreases crop yields. Rocks in the

watershed contain more magnesium and most waters

contain Mg2+

more than Ca2+

.

The magnesium ratio can be expressed as,

Where all the ions are expressed in epm. The

Magnesium ratio in all most all the pre-monsoon and

post-monsoon samples is below 50, except in four

locations,where the ratio is more than 50.

Gibb’s Diagram:

Based on aquifer lithology, the mechanism, controlling

chemical relationship of ground water has been studied

by Gibb (1970). In the Gibb’s diagram, three kinds of

fields are recognized namely precipitation dominance,

evaporation dominance and rock water dominance. In

case of anion, all the samples pointed towards rock

dominance in the chemistry of ground water (Fig. 6).

This reflects the influence of aquifer lithology vis-à-vis

ground water chemistry. Similarly, in case of cations, all

the samples show rock dominance (Fig. 6). Hence, the

groundwater of the study area is mainly dominated by

the lithology of aquifer of the region.

Conclusion:

From the chemical analysis, it is revealed that the

groundwater of Talabasta area is acidic or / and weakly

acidic in nature. Calcium and Magnesium are the major

cations and bicarbonate is the major anion in the study

area. The higher concentrations of the bicarbonate

indicate chemical weathering of the rocks. The

groundwater belongs to Ca-Mg-HCO3 facies. The Ca,




Mg, HCO3 indicate the temporary hardness, alkalinity

and dominance of alkaline earth and weak acids. All the

chemical parameters indicate that the groundwater of

the study area is safe for drinking. Graphical

representation of the chemical data on the irrigation

suitability diagram (USSL, 1954) shows that medium

salinity-low sodium (C2S1) and high salinity-low

sodium (C3S1) waters are present, which need adequate

drainage to overcome the salinity problem. Chemical

composition of ground water is a reflection of rock

weathering, decomposition and anthropogenic

interventions, causing changes in time and space. The

chemical quality of the study area is well within the

permissible limit meant for drinking and irrigation

purpose. As water quality is interlinked with water

quantity, regular monitoring should be undertaken to

maintain the quality of ground water, as ground water

consumption is bound to increase due to erratic

monsoon pattern and higher demand due to population

growth.

Acknowledgements:

The authors thank to Groundwater Survey and

Investigation (GWSI), Bhubaneswar for chemical

analysis. The authors are also thankful to revered

reviewers for critically going through our manuscript

and suggesting many alterations to improve the standard

of the paper.

References:

[1] APHA 1995Standard methods for the examination

of water and wastewater, 19th edn. American

Public Health Association, Washington, D.C., pp

1467

[2] BIS 1991. Indian Standard for Drinking Water as

per BIS specifications (IS 10500-1991)

[3] Domenico, P.A. and Schwartz, F.W. 1990. Physical

and chemical Hydrogeology. John Wiley and Sons,

New York, pp 410-420

[4] Doneen, L.D. 1962. The influence of crop and soil

on percolating water. Proc. 11961 Biennial

conference on Groundwater Recharge, pp 156-163.

[5] Gibbs, R.J. 1970. Mechanism controlling world

water chemistry. Science 170, 1088-1090

[6] Hem, J.D. 1985. Study and interpretation of the

chemical characteristics of natural water. USGS,

Water Supply Paper 2254, 264

[7] Kelley, W.P. 1995. Alkali soils-Their formation

properties and reclamation, Reinold Publ.Corp.,

New York

[8] Piper, A.M. 1944. A graphical procedure in the

geochemical interpretation of water analysis. Am

Geophys Union Trans 25, 914–928

[9] Ragunath, H.M. 1987. Groundwater. Wiley

Eastern, New Delhi, pp 563

[10] Ramakrishna 1998. Ground water, Hand book,

India, pp 556

[11] Richards, L.A. 1954. Diagnosis and improvement

of saline and alkaline soils U.S. Deptt. Agri. Hand

Book 60, pp160

[12] Tijani, J. 1994. Hydrochemical assessmrnt of

groundwater in Moro area, Kwara State, Nigeria,

Environment Geology 24, 194-202

[13] Trivedi, R.K. and Goel, P.K. 1984. Chemical and

biological methods for water pollution studies. Env.

Publ. Karad, India. pp 215

[14] Vogel, A.L. 1964. A text book of qualitative

inorganic analysis, 3rd

edition, pp 438

[15] WHO 2004. Guideline of Drinking water quality in

Health criteria and other supporting information.

World Health Organization 2:336

[16] Wilcox, L.V. 1948. The Quality of water for

irrigation, Use U.S. Dept. of Agriculture, Tech,

Bull, 1962, Washington, DC, pp 19

[17] Wilcox, L.V. 1955. Classification and use of

irrigation water, USDA, Circular, 969, Washington

D.C., U.S.A., pp 19




Figure 2: Piper’s Trilinear Diagram

Figure 3: U.S. Salinity Laboratory Diagram Showing Classification of Water Samples for Irrigation Purpose




Figure 4: Wilcox diagram for Classification of Groundwater Based on EC and Na%

Figure 5: PI Diagram




Figure 6: Gibb’s Diagram

Table 1: Chemical Parameters of Groundwater (Pre-monsoon)




Table 2: Chemical Parameters of Groundwater (Post -Monsoon)

Table 3: Comparison of Water Quality of the Study Area with WHO (2004) Standard and BIS (1991) Standard

Parameter Range WHO

IS 10500

Pre-monsoon Post-monsoon Higest Desirable Maximum Permissible

pH 5.25-6.89 6.05-8.21 6.5-8.5 6.5 8.5

EC (mhos/cm2) 235-998 268-1073 400-2000

TDS (mg/l) 177-739 211-827 500-1000 500 2000

Calcium (mg/l) 26-127 11.7-112 100-200 75 200

Magnesium (mg/l) 7.63-36.23 6.48-34.26 30-50 30 100

Sodium (mg/l) 1.7-88.3 9.76-97 20-1756

Potassium (mg/l) 0.68-59 1.0-65 10-12

Bicarbonate (mg/l) 108.6-437.5 0-451.4 200 600

Sulphate (mg/l) 0.27-63.7 0.062-56.3 25-250 200 400

Chloride (mg/l) 20.7-127.8 2.53-142.81 25-600 250 1000

Nitrate (mg/l) 0-23.5 0.01-39.8 50 45 100

Fluoride (mg/l) 0.101-1.76 0.04-1.73 1.5 1.0 1.5

Alkalinity (mg/l) 95-366 98-370 75 to 400 200 600

Total Hardness (mg/l) 97-466 87-421 No guideline 300 600

Table 4 Suitability of Groundwater for Drinking Based on TDS Classification

TDS (mg/l) Water class Pre-monsoon Post-monsoon

<300 Excellent 20% 20%

300-600 Good 55% 40%

600-900 Fair 25% 40%

900-1200 Poor 0% 0%

>1200 Unacceptable 0% 0%




Table 5: Suitability of Groundwater for Drinking Based on Total Hardness (TH) as CaCO3 (mg/L)

TH(mg/l) Water class Pre-monsoon Post-monsoon

<75 Excellent 0% 0%

75-150 Good 25% 50%

150-300 Fair 65% 45%

>300 Poor 10% 5%

Table 6: Quality of Water Based on SAR Values

Water classes for Irrigation SAR value No. of sample (out of 20)

Pre-monsoon Post-monsoon

Excellent Up to 10 15 13

Good 18-Oct 5 7

Medium 18-26 Nil Nil

Bad >26 Nil Nil

Table 7: Water Class for Irrigation Based on Na%

Water class for irrigation % of Na No of sample Pre-monsoon

(out of 20)

Post-monsoon

(out of 20)

Excellent to Up to 20 3 1

Good 20-40 11 6

Permissible to 40-60 6 8

Doubtful 60-80 Nil 5

Unsuitable >80 Nil Nil

Table 8: Rating of Waters Based on Residual Sodium Carbonate

RSC(epm) Water category Pre-monsoon Post-monsoon

No. of wells % of sample No. of wells % of sample

<1.25 Safe 17 85% 16 80%

1.25-2.5 Marginally 02 10% 04 20%

>2.5 Unsuitable 01 5% Nil Nil

Table 9: Classification of Ground Water Based on P.S. (Potential Soil Salinity)

P.S Class No. of Sample (out of 20)

Pre-monsoon Post-monsoon

<5 Excellent to Good 20 20

5-10 Good to Injurious Nil Nil

>10 Injuries to Satisfactory Nil Nil

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1619-1629


Terrain Analysis and Hydrogeochemical Environment of Aquifers

of the southern West Coast of Karnataka, India

S. S. HONNANAGOUDAR1, D. VENKAT REDDY

1 and MAHESHA. A

2

1Dept of Civil Engineering, National Institute of Technology Karnataka, Surathkal, Mangalore - 575025, Karnataka

2Department of Applied Mechanics and Hydraulics, National Institute of Technology Karnataka, Surathkal,

Mangalore - 575025, Karnataka


Abstract: Dakshina kannada district is situated in peninsular region. The peninsula is composed of geologically

ancient rocks of diverse original and most of them have undergone metamorphism. The early Precambrian tonalitic

gneisses invaded by granites, granulites and dolerite dykes. Granulites are mostly restricted to areas south of

Mangalore. High grade alumina rich (corundum bearing) metamorphic schists have been encountered and younger

alkaline intrusive rocks like Aegerine syenites have been reported. There are five rivers and estuaries. Number of

lineaments cut across each other and some lineaments are parallel to each other. The Arabian sea class is the largest

among other land cover features in the study area. The river/tidal creek land cover appear as long irregular and

sinous in outline. Mulki river, Netravati river, Gurupur river at southern terrain. The qualities of groundwater at

sandy aquifer are good, lateritic/weathered gneissic rocks it is sweet.

Keywords: Groundwater, Rock Types, Lineaments, Water Quality

Introduction:

The Dakshina Kannada district which is a coastal

district of Karnataka spreads along the west coast of

India covering coastal track of about 40 km. It has been

found that two-third of the population lives within a

narrow belt, directly landward from the ocean edges.

Also, the rivers of district are being seasonal and tidal in

nature; seawater intrusion into adjoining aquifers during

the non-monsoon period is greatly felt up to several

kilometres inland along and on either side of the river

courses. Hence, the present work is aimed at

groundwater related investigation involving aquifer

characterization and seawater intrusion of coastal

Dakshina Kannada district, Karnataka between

Talapady to Mulki towns. The focus of the present work

is to create awareness on water quality degradation and

its management in rural coastal areas between the rivers

Talpady, Gurupur, Pavanje, Mulki, and Netravati. The

fresh water partly met by surface water resources.

Geology:

Dakshina kannada district is situated in peninsular

region. The process of erosion, transportation and

deposition, the shape of the coastline changes, often

slowly but sometimes rapidly. The intersection between

rock, water, wind and vegetation are responsible for the

dramatic landforms along the coast. The peninsula is

composed of geologically ancient rocks of diverse

original and most of them have undergone

metamorphism. It represents a stable land of the earth’s

crust. The geological formations of the district are

similar those of Karnataka state but for the coastal

sedimentary deposits and laterites. The geological

section across Netravati river of dakshina kannada

district is shown in figure (Figure1)

The early Precambrian tonalitic gneisses invaded by

granites, granulites and dolerite dykes. Granulites are

mostly restricted to areas south of Mangalore; however

tongues of granulites can be seen in gneissic granite

quarries north of mangalore. High grade alumina rich

(corundum bearing) metamorphic schists have been

encountered and younger alkaline intrusive rocks like

Aegerine syenites have been reported (Ravindra and

Janardhan 1981). Marine expeditions conducted by

Geological Survey of India (GSI) identified the

existence of sandy paleobeach ridges some 25 km

offshore of Mangalore coast. Earlier Murty (1977)

reported occurrences of lignite in clays from the

foundation wells of Netravati bridge, south of

Mangalore. Black clays occur in some of the paleo

fluvial courses such as Baikampadi, Kuloor and Kottar

usually admixed with plant remains (peat or lignite).

The geology of the terrain comprises peninsular gneissic

basements, laterites, coastal alluvium, granite-migmatite

complexes and intrusions of acid, basic and ultramafic

into the basement rocks. Sparsely developed quaternary

formations are also seen along the coastal tract. The

deposition of number of rocky and Sandy Island of the

1620 S. S. HONNANAGOUDAR, D. VENKAT REDDY and MAHESHA. A



coast are also important landscapes of the terrain. The

different drainage patterns found in the study area are

dendratic, trellis, parallel, rectangular radial and

complex patterns (Figure 2)

Stream density over the terrain and the distribution

pattern clearly envisages 3 distinct categories as given

in Table 1.1.

Table 1.1: Stream Density

Sl No. Category Stream density

(Km/unit area)

1 Low density 0.0 - 3.0

2 High density 3.0 - 5.0

3 Extremely

high density > 5.0

(Source: Sreedharamurthy, et. al., 2002)

Table 1.2: Stream Frequencies

Sl

No. Category

Stream frequency

(No. of stream unit area)

1 Low frequency 0.0 - 5.0

2 High frequency 5.0 - 15.0

3 Extremely high

frequency 15.0 – 25.0

(Source: Sreedharamurthy, et. al., 2002)

Northern Coastal Terrain: Most of the western part of

the terrain is characterized by low density where the

coastal study plain occurs. This plain is also

characterized by lateritic plateaus and sand dunes.

Southern Coastal Terrain: Major part of the southern

coastal terrain is characterized by low stream density. In

general there is trend to increase in stream density from

west to east (Figure 3)

Spatial distribution pattern of stream frequency over the

terrain three distinct categories are identified by the

stream frequency distribution patterns in Table 1.2.

Southern Coastal Terrain: The entire area shows

drastically low stream frequency occurrences. (fig 4)

Lineaments:

Lineaments have been detected in northern coastal

terrain. Number of lineaments cut across each other and

some lineaments are parallel to each other. These

lineaments represent faults, fractures, escarpments,

dykes, ridges etc. The length of the lineaments varies

from 1.5 km to 18 km (B.R.Raghavan, 2002). The

orientation of the lineaments is east to west and some

lineaments show nearly north south orientation. (Figure

5).

Geomorphology:

The Arabian sea class is the largest among other land

cover features in the study area. The river/tidal creek

land cover appear as long irregular and sinous in

outline. Mulki river, Netravati, Gurupur river at

southern terrain. Along with the rivers a number of tidal

creeks also mark the study area. However the extent of

the tidal creeks is restricted only to the coastal plains.

The occurrence of marsh land is restricted to the coastal

plains. The marsh lands are observed Netravati and

Gurupur rivers. The occurrence of the sand bodies has

also been observed along either bank of Gurupur and

Netravati rivers as intermittent patches. The built-up

lands are increasing all along the sides of the National

Highway (No. 66.) (Udayashankar, 1994). The coastal

plain, valley fills and either side of the rivers; there are

two types of cropping practices differ from each other in

season and cropping pattern. Kharrif cropping season is

from June to September while rabi cropping season is

from October to March. The minor forests are mainly

observed along the foothills of western ghats.

Figure 1: Geological section map of Netravati River, Dakshina Kannada district

(after Ravindra and Venkat Reddy, 2011)

1621 Terrain Analysis and Aquifer Characteristics of West Coast

of Dakshina Kannada District, Karnataka, India



Figure 2: Drainage Pattern: Southern Coastal Karnataka

Figure 3: Distribution of Stream Density

Figure 4: Distribution of Stream Frequency




Figure 5: Lineaments of Southern Coastal Karnataka

Land use:

Land use refers to human activities for various use

carried out on land and cover refers to natural

vegetation, water bodies, rock, soil, artificial cover and

others resulting due to land transformations. Change in

land use pattern influences the hydrological regime of

the basin (James et, al., 1987). Various hydrological

processes such as infiltration, evapotranspiration, soil

moisture status etc., are influenced by land use

characteristics of water shed. The deforestation may

cause peak flows in the river due to reduced infiltration

leading to floods, due to increased soil erosion, removal

of too fertile soil layer may result in drop in soil

fertility.

The spatial information on land use and their pattern of

changes are essential for planning, management and

utilization of land for agriculture, forestry, urban

industries, environmental studies etc.

Infiltration:

Infiltration rates are very widely depending on the

condition of the land surface. Infiltration has an

important place in the hydrological cycle. Detailed

study of hydrological process helps planners,

hydrologists, farmers and decision makers in number of

ways. The infiltration rate of the area which helps in

estimating peak rate and volume of runoff, estimation of

surface runoff and overland flow, estimation of

groundwater recharge, estimation of soil moisture

deficits. Typical infiltration rates are at the end of one

hour after the commencement of the storm are presented

in Table 1.3.

Table 1.3: Infiltration Rate

Sl.

No. Soil Type

Infiltration

rate cm/hr

1 High (Sandy soil) 1.25 – 2.54

2 Intermediate (loam,

silt, clay) 0.25 -1.25

3 Low (clay, clay-loam) 0.025 – 0.25

From east to west the majority of the sedimentary units

consists of a wedge of permeable sand sediments

deposited in shallow marine water, littoral sands formed

in the fore shore and the adjacent beach and sand-dune

systems (Antonelline et al. 2008). In the westernmost

are fine continental alluvial deposits (silt and clay)

overlay the littoral sands (Amorosi et al. 2002;

Bondesan et al. 1995).





Groundwater Quality:

The qualities of groundwater at sandy aquifer are good,

lateritic/weathered gneissic rocks it is sweet. The dug

wells in the alluvial area are saline water during summer

months and in the monsoon period fresh water they are

getting. There are 30 open well samples are collected

and test the well water samples to detect the presence of

drinking water. The drinking water quality parameters

for analysis are TDS, pH, Turbidity, Total hardness, Fe,

Nitrate, Sulphate and Coliform. The well location is

mentioned in Table no. 1.4. The results are calculated in

Table 1.5 to Table 1.13 in the three seasons April,

August and October 2012.

Table 1.4: Well Inventories

Well

No.

Location Well

No.

Location

1 Hotel Nisarga, Talapady 16 NMPT, Panambur

2 Uchilla 17 Hotel Vishwasagar, Panambur

3 Kotekar 18 Jyothi service station, Hosabettu

4 Thokkuttu 19 KEB Quarters, Surathkal

5 S.S. Bricks, Thokkuttu 20 Veerabhadra Temple, Marigudi

6 Adam kadru school 21 NITK, Surathkal

7 Kalapu 22 Gerald Kuvelo Nivas, Mukka

8 Shaikh cottage 23 K.A. Abdulkadar, Mukka

9 Mogaru (Near Netravathi Bridge) 24 “Kiran” Haliyangadi

10 Ronsun service centre, Mogaru 25 Kikanda House, Near Pavanje Bridge

11 Karnataka Bank, Near pumpwell 26 Venkappa shetty, Padambadur

12 Kuntikana 27 Nani House, Kolnad

13 VRL, kuloor 28 Sri Guru, Kolnad

14 Vivekanand, Panjioruguru, Kuloor 29 Srinivas, Bappanadu temple, Mulki

15 Thokuru 30 Gopalkrishna Naik, Mulki

Table 1.5: Water Quality Test Results for pH

Well No. pH

April

pH

August

pH

October Well No.

pH

April

pH

August

pH

October

1 6.63 5.86 7.67 16 5.97 7.55 6.52

2 6.36 6.69 6.9 17 6.19 6.48 6.13

3 6.17 7.13 7.56 18 6.22 6.51 6.32

4 6.33 6.53 6.82 19 6.08 6.41 6.23

5 6.49 6.06 6.6 20 5.71 6.44 6.13

6 6.66 6.6 6.87 21 5.91 7.87 6.17

7 6.54 6.54 6.8 22 6.11 7 5.93

8 5.89 6.16 6.5 23 6.36 6.31 5.83

9 6.47 5.94 6.25 24 6 7.01 6.78

10 6.58 6.45 6.65 25 6.51 6.29 5.94

11 6.63 6.2 6.39 26 6.83 6.51 6.34

12 6.16 6.28 6.54 27 6.32 6.91 6.53

13 6.38 6.1 6.17 28 5.69 6.45 6.01

14 6.63 6.48 6.45 29 6.2 6.37 5.96

15 6.38 6.19 6.43 30 6.38 6.6 6.19




Table 1.6: Water Quality Test Results for TDS mg/l

Well No. TDS

April

TDS

August

TDS

October Well No.

TDS

April

TDS

August

TDS

October

1 6.63 5.86 7.67 16 5.97 7.55 6.52

2 6.36 6.69 6.9 17 6.19 6.48 6.13

3 6.17 7.13 7.56 18 6.22 6.51 6.32

4 6.33 6.53 6.82 19 6.08 6.41 6.23

5 6.49 6.06 6.6 20 5.71 6.44 6.13

6 6.66 6.6 6.87 21 5.91 7.87 6.17

7 6.54 6.54 6.8 22 6.11 7 5.93

8 5.89 6.16 6.5 23 6.36 6.31 5.83

9 6.47 5.94 6.25 24 6 7.01 6.78

10 6.58 6.45 6.65 25 6.51 6.29 5.94

11 6.63 6.2 6.39 26 6.83 6.51 6.34

12 6.16 6.28 6.54 27 6.32 6.91 6.53

13 6.38 6.1 6.17 28 5.69 6.45 6.01

14 6.63 6.48 6.45 29 6.2 6.37 5.96

15 6.38 6.19 6.43 30 6.38 6.6 6.19

Table 1.7: Water Quality Test Results for Turbidity

Well

No.

Turbidity

April

Turbidity

August

Turbidity

October

Well

No.

Turbidity

April

Turbidity

August

Turbidity

October

1 6.63 5.86 7.67 16 5.97 7.55 6.52

2 6.36 6.69 6.9 17 6.19 6.48 6.13

3 6.17 7.13 7.56 18 6.22 6.51 6.32

4 6.33 6.53 6.82 19 6.08 6.41 6.23

5 6.49 6.06 6.6 20 5.71 6.44 6.13

6 6.66 6.6 6.87 21 5.91 7.87 6.17

7 6.54 6.54 6.8 22 6.11 7 5.93

8 5.89 6.16 6.5 23 6.36 6.31 5.83

9 6.47 5.94 6.25 24 6 7.01 6.78

10 6.58 6.45 6.65 25 6.51 6.29 5.94

11 6.63 6.2 6.39 26 6.83 6.51 6.34

12 6.16 6.28 6.54 27 6.32 6.91 6.53

13 6.38 6.1 6.17 28 5.69 6.45 6.01

14 6.63 6.48 6.45 29 6.2 6.37 5.96

15 6.38 6.19 6.43 30 6.38 6.6 6.19





Table 1.8: Water Quality Test Results for Chloride mg/l

Well No. Chloride

April

Chloride

August

Chloride

October Well No.

Chloride

April

Chloride

August

Chloride

October

1 6.63 5.86 7.67 16 5.97 7.55 6.52

2 6.36 6.69 6.9 17 6.19 6.48 6.13

3 6.17 7.13 7.56 18 6.22 6.51 6.32

4 6.33 6.53 6.82 19 6.08 6.41 6.23

5 6.49 6.06 6.6 20 5.71 6.44 6.13

6 6.66 6.6 6.87 21 5.91 7.87 6.17

7 6.54 6.54 6.8 22 6.11 7 5.93

8 5.89 6.16 6.5 23 6.36 6.31 5.83

9 6.47 5.94 6.25 24 6 7.01 6.78

10 6.58 6.45 6.65 25 6.51 6.29 5.94

11 6.63 6.2 6.39 26 6.83 6.51 6.34

12 6.16 6.28 6.54 27 6.32 6.91 6.53

13 6.38 6.1 6.17 28 5.69 6.45 6.01

14 6.63 6.48 6.45 29 6.2 6.37 5.96

15 6.38 6.19 6.43 30 6.38 6.6 6.19

Table 1.9: Water Quality Test Results for Hardness mg/l

Well No. Hardness

April

Hardness

August

Hardness

October Well No.

Hardness

April

Hardness

August

Hardness

October

1 300 278 276 16 50 46 45

2 160 157 156 17 190 189 190

3 40 41 45 18 80 76 78

4 90 87 86 19 50 47 44

5 580 569 589 20 30 31 37

6 120 122 128 21 50 48 54

7 50 46 47 22 110 109 109

8 50 47 40 23 70 65 65

9 370 375 379 24 30 32 32

10 80 81 83 25 350 346 346

11 280 275 278 26 80 76 76

12 20 22 23 27 130 126 126

13 250 245 249 28 50 48 48

14 80 76 79 29 280 267 267

15 200 201 208 30 70 65 65




Table 1.10: Water Quality Test Results for Nitrate mg/l

Well No. Nitrate

April

Nitrate

August

Nitrate

October Well No.

Nitrate

April

Nitrate

August

Nitrate

October

1 4.76 4.7 4.98 16 15.17 15.89 14.8

2 13.21 12.41 12.95 17 29.25 29.84 28.45

3 39.64 36.61 35.61 18 18.52 17.58 18.89

4 9.58 9.89 9.09 19 14.26 13.2 14.27

5 11.26 10.23 11.29 20 19.31 20.39 19.56

6 1.09 1.02 1.98 21 8.66 8.9 9.12

7 0.81 0.4 1 22 19.76 18.78 19.23

8 23.16 22.12 22.94 23 6.78 6.7 6.12

9 2.41 2.19 2.86 24 5.26 5.98 5.23

10 10.25 9.26 10.01 25 9.53 9.53 8.92

11 1.09 1.67 1.41 26 7.62 6.89 6.21

12 20.44 21.32 21 27 4.53 5.56 5.12

13 10.85 11.85 11.01 28 12.82 12.86 11.86

14 17.9 16.76 15.23 29 5.64 5.9 6.24

15 5.57 6.5 5.59 30 4.51 4.98 4.89

Table 1.11: Water Quality Test Results for Coliform Count CFU/ml

Well

No. Coliform Coliform Coliform

Well

No. Coliform Coliform Coliform

A B A B A B A B A B A B

1 1 - 1 - 2 - 16 - - - - - -

2 - - - - - - 17 2 - 2 - 1 -

3 - - - - - - 18 - - - - - -

4 1 - 1 - 1 - 19 - - - - - -

5 - - - - 1 - 20 1 - - - - -

6 - - - - - - 21 - - - - - -

7 - - - - - - 22 5 - 4 - 5 -

8 1 - 2 - 1 - 23 1 - 1 - 1 -

9 - - - - - - 24 - - - - - -

10 - - - - - - 25 3 1 2 1 1 1

11 1 - 1 - 1 - 26 - - - - - -

12 - - - - - - 27 1 - - - - -

13 1 - 1 - - - 28 - - - - - -

14 - - - - - - 29 3 - 1 - 1 -

15 4 - 5 - 3 - 30 1 - 2 - 1 -





Table 1.12: Water Quality Test Results for Sulphate as SO4 mg/l

Well No. Sulphate

April

Sulphate

August

Sulphate

October Well No.

Sulphate

April

Sulphate

August

Sulphate

October

1 - - - 16 18.59 17.21 33.62

2 31.56 30.54 65.08 17 30.99 31.9 65.2

3 7.438 7.41 16.82 18 23.26 24.16 50.34

4 7.53 7.51 11.02 19 20.99 22.29 46.5

5 - - - 20 18.19 19.59 18.89

6 24.02 24 50 21 12.08 13 14.3

7 36.3 35.1 64.4 22 29.27 28.25 29.85

8 19.55 19.45 37.9 23 35.82 36.84 38.74

9 - - - 24 13.38 14.14 16.18

10 13.17 14.15 32.2 25 30.67 30.21 32.11

11 20.7 20.1 45 26 20.86 20.06 23

12 13.16 11.19 27.58 27 14.03 13.08 12.07

13 23.37 22.27 46.58 28 28.67 27.6 28.9

14 13.7 14.5 31.6 29 31.83 32.14 37.1

15 39.68 37.64 79.38 30 13.13 14.1 17.16

Table 1.13: Water Quality Test Results for Dissolved Iron Fe mg/l

Well No. Iron (Fe)

April

Iron (Fe)

August

Iron (Fe)

October Well No.

Iron (Fe)

April

Iron (Fe)

August

Iron (Fe)

October

1 0.837 0.887 0.91 16 0.489 0.47 0.49

2 0.878 0.8 0.9 17 0.535 0.55 0.54

3 0.809 0.84 0.89 18 0.967 0.99 0.89

4 0.86 0.9 0.96 19 0.942 0.98 0.79

5 0.926 0.91 0.87 20 0.995 0.95 0.85

6 0.829 0.88 0.83 21 0.542 0.59 0.58

7 0.803 0.89 0.8 22 0.541 0.56 0.49

8 0.887 0.89 0.78 23 0.478 0.49 0.48

9 0.829 0.88 0.8 24 0.3 0.52 0.57

10 0.808 0.84 0.9 25 0.546 0.65 0.58

11 0.818 0.89 0.68 26 0.451 0.55 0.52

12 0.828 0.85 0.86 27 0.517 0.57 0.55

13 1.025 1.02 1.1 28 0.474 0.64 0.87

14 0.977 0.91 0.93 29 0.584 0.5 0.9

15 0.64 0.69 0.68 30 1 1.09 1

Results:

The pH in the month of April 2012 well no. 2, 3, 4, 5, 8,

9, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24,and

27, 28, 29, 30 are 6.54 to 6.83 is the highest and

remaining wells having the pH is 5.87 to 6.33. In the

month of August 2012 the pH varies 6.51 to 7.87. The

well numbers are 2, 3, 4, 5, 7, 16, 18, 21, 22, 24, 26, 27

and 30. The remaining wells of the water quality pH is

5.86 to 6.48 in the month of October 2012 the pH varies

5.93 to 6.45 the well numbers are 9, 11,13, 14, 15, 17,

18, 19, 20, 21, 22, 23, 25, 26, 28, 29 and 30. The pH

varies from 5.93 to 6.45. The remaining wells pH is

6.50 to 7.67 which is acceptable. TDS in the month of

April 2012 well number 1, 15 and 25 are more than

acceptable limit and in the month of August and

October 2012 all the wells are within the acceptable

limit. Turbidity in the month of April 2012 the well nos.

are 1,5, 15, 24 and 30 are more than 5 NTU, in the

month of August well nos. 5 and 25 are more than 5

NTU and in the month of October 2012 well nos. are 5

only. Well nos. 5, 9 and 25 are more than acceptable

limit for Hardness. Chloride in the month of August

2012 well nos. are 1, 5, 25 are more than 1000 mg/l.

Nitrate and sulphate are within the limit. Coliform




count is found in the well nos.1, 4, 5, 8, 11, 15, 17, 22,

25, 29 and 30 for all the season.

This contamination is more pronounced in wells along

the stream courses up to the distance where tidal effect

extends. Further, Ground water in proximity to stream

course is contaminated with seepage of domestic waste.

As a general rule, pumpage must be distributed in time

and space and there should not be any concentration of

wells to avoid saline water ingress.

Conclusions:

The intersection between rock, water, wind and

vegetation are responsible for the dramatic landforms

we see along the coast. Dakshina kannada district is

situated in peninsular region.

The slope density and stream frequency of the study

area are characteristically high along the eastern part of

the terrain because the streams originate on the scarps of

the Western Ghats.

The land use and land cover pattern of the terrain is

suggestive of probable stress on the terrain in future in

terms of expansion of built-up land and loss of

agricultural land due to a probable increase in the

number of industries and population.

From the field study some of the area is affected by

saltwater intrusion during the summer period and some

of the areas are affected almost throughout the year.

Ground water in proximity to stream course is

contaminated with seepage of domestic waste. As a

general rule, pumpage must be distributed in time and

space and there should not be any concentration of wells

to avoid saline water ingress.

References:

[1] Adyalkar, P. G., and Shrihari Rao, S. (1979).

Hydrodynamic method of assessing groundwater

recharge by precipitation in Deccan trap region-

case study. Jour. Geol. Soc. Ind., V. 20, pp. 134-

137.

[2] Adrian D. Werner, James D. Ward, Leanne K.

Morgan, Craig T. Simmons, Niville I. Robinson,

and Michael D, Teubner., (2012).“Vulnerability

indicators of seawater intrusion.” Ground Water 50

(1), pp.48-58.

[3] Aghashe, R.M. (1994). Artificial recharge of

groundwater in critical and semi-critical areas.

Proc. Natl. Workshop Development of groundwater

in critical and semi-critical areas: Alternatives,

Options and Strategies, NewDelhi, pp. 75-88.

[4] Alley,William M. and C. J.

Taylor.(2001).“Groundwater level monitoring and

the importance of long term waterlevel data.”

U.S.Geol. Survey Circular: pp.1217.

[5] Anbazhagan, S., and Ramasamy, S.M., (1993).

Role of Remote Sensing in geomorphic analysis for

water harvesting structures, In: S.M. Ramasamy,

(Ed.) Trends in Geological Remote Sensing, Rawat

Publications, New Delhi, pp. 208-212.

[6] Anning, D., (2008). “Dissolved solids in basin-fill

aquifers of the southwest”. Hydrogeol. J., pp.18-19.

[7] Bloom, A. L. (1978). Geomorphology - A

systematic analysis of late Cenozoic Landforms.

Prentice Hall, pp. 1-510.

[8] Boulton (1970). “Analysis of data from pumping

tests in unconfined anisotropic aquifers.” J.

Hydrol., 10, 369-381.

[9] Chitea. F. (2011). “Geophysical detection of marine

intrusions in Black sea coastal areas (Rommania)

using VES and ERT data”, Geo-Eco-Marina,

pp.95-102.

[10] Coates, D. R. and Vitek, J. D. (1980). Threshold in

Geomorphology, George Allen, London, pp. 223.

[11] Kelly W.E., (1977).“Geoelectric sounding for

estimating aquifer hydraulic conductivity”, Ground

Water, 15(6), pp.420-425.

[12] Mahesha A., (1996).“Control of seawater intrusion

through injection – extraction well system”, J.

Irrig. Drain. Eng., ASCE, pp.314-317.

[13] Manjunath B.R. and Harry N.A. (1994). “Geology

of Western Coastal Karnataka, GeoKarnataka”,

MGD Centenary Volume, pp. 109-116.

[14] Moench, A.F., (1994). “Specific yield as

determined by type curve analysis of aquifer test

data”, Ground Water, 32(6), pp.949-957.

[15] Moench, A.F., (1997). “Flow to a well of finite

diameter in a homogeneous, anisotropic water table

aquifer”, Water Resour. Res., 33 (6), pp.1397-1407.

[16] Mohsen Sheriff., (1999). “Seawater intrusion in the

Nile delta aquifer: An overview”,

http://aguas.igme.es/igme/publica/tiac-02/EGIPTO-

I.pdf. pp.296-308 (April 10, 2012).

[17] Polemio. M.V., Dragone and Limoni, P.P., (2009),

“Monitoring and methods to analyse the

groundwater quality degradation risk in coastal

karstic aquifers (Apulia, Southern Italy)”, Environ.

Geol., 58, pp. 299-312.

[18] Ravindra, B. M., Venkata Reddy, D. (2011).

“Neotectonic Evolution of Coastal Rivers of

Mangalore Karavali Karnataka, India”. Int. Jour.

Of Earth Science and Engineering, V.04. No. 04,

pp. 561-574.

[19] Reddy, N. and Prakasam, P. (1982). “Evaluation of

artificial recharge projects in Granitic Terrains of

Andhra Pradesh”. Workshop on Artificial Recharge

of Groundwater in Granitic Terrain, pp. 194-210.

[20] Singh, O. P. (1981), “Morphometric evalution of

landforms in Palamau Upland.” In: H. S. Sharma

(Ed) Perspectives in Geomorphology”, Concept

Publishing Co., New Delhi, V. 4, pp. 109-131.





[21] Singh (2006). “Semi-analytical model for

drawdown due to pumping a partially penetrating

large diameter well”, J.Irrig. Drain., ASCE, 133(2),

pp.155-161.

[22] Sridharan. K, D. Sathyanarayana, A. Siva Reddy.,

(1990), “Analysis of flow near a dug well in an

unconfined aquifer, J.Hydrol., 119, pp.89-103.

[23] Sreedharamurthy, T. R., Raghavan, B. R. and Bhat,

V.T., (2002). “Terrain Analysis of Coastal

Karnataka”, Sponsored project by Ministry of

Environment and forests, Mangalore University,

Mangalore.

[24] Sung-Ho Song, Jin Yong Lee and Namsik Park.,

(2006). “Use of vertical electrical sounding to

delineate seawater intrusion in coastal area of

Byunsan, Korea”, Environ. Geol. (2007), Vol 52:1,

pp.1207-1219.

[25] Todd, D. K. (1980). “ Groundwater Hydrology.”

John Willey and Sons, Newyork, pp. 1-535.

[26] Ward, R. C. (1968), “Some hydrological

characteristics of British Rivers”, Jour. Hydrol., V.

pp. 358-372.

[27] Way, D. A. (1973). “Terrain Analysis,” Hutchinson

and Ross, Stroudsburg, Pasadina, Dowden, pp. 72-

124.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1630-1638


Water Quality Modeling and Management of Karanja River in

India

BASAPPA. B. KORI1, SHASHIKANTH MISE

1 and SHASHIDHAR

2

1Dept. of Civil Engineering, PDA Engineering College, Gulbarga, Karnataka, India

2Department of Civil Engineering, IIT- Hydrabad, India


Abstract: The river Karanja, a major drinking water source for the Bhalki municipality is under threat by the

disposal of effluents from the industries such as Sugar, Paper and distilleries. To study the effect of effluent disposal

on this river, one dimensional steady-state stream water quality model was developed using QUAL2Kw. In this

model pre-monsoon data is used for calibration. From the sensitivity analysis, it was observed that the model is

sensitive for temperature, bottom width, discharge and biochemical oxygen demand. Simulations were carried out

for several scenarios such as bottom algae, head water release, shade and temperature under dry weather flow

condition. All these profile did not meet the requirement of minimum DO concentration, to get it bottom algae is to

be reduced to 75%, head water BOD is to be reduced to 8mg/L, and shade is to be increased up to 30%.

Keywords: Stream Water Quality, Water Quality Management and QUAL2Kw

1. Introduction:

Environmental pollution is one of the serious threats

faced by mankind. This has accelerated the discovery of

environmental problems during the past few decades.

Rapid population growth, urbanization, industrialization

and land development along the stream have increased

the stress on river pollution and have resulted in its

deterioration (Surindrasuthar, 2009). The deterioration

of aquatic ecosystem results from decrease in the

dissolved oxygen concentration and also, due to

pollutant degradation by micro-organisms, chemical

oxidation, plant, algal and phytoplankton respiration

(Drolc and Konkan, 1996). It also results in unpleasant

odors, and other aesthetic damage (Arruda

Camargo,2010). Hence, it is essential to monitor water

quality changes in the entire river, but it is tedious, time

consuming and un-economical.

The mathematical models are the alternative way to

describe the relation between waste loads and water

bodies. QUAL2Kw is one-dimensional steady state

stream water quality model; it uses unequally-spaced

reaches. It is well documented and is freely available.

QUAL2Kw includes many new elements (Pelletier and

Chapra, 2005).

A conventional sensitivity analysis is performed by

varying important parameters that which has effect on

the model output (Nikolaos P. Nikolaidis et al. 2006).

For the management of water quality, several scenarios

are studied by changing model input parameters during

the dry period (Ritu Paliwal and Prateek Sharma 2007).

considering i) Bottom algae modification, ii)

temperature modification iii) head water modification

and iv)Shade modification.

2. Material and Methods:

2.1. Study Area:

The Karanja river is one of the tributaries to the

Godavari river. It originates near Kohir village of

Andhra Pradesh state of India and joins another

tributary of Godavari i.e. Manjera river at 122 km

downstream. This river has a dam called Karanja which

in near Bhyalhalli village. Fig.1 shows the study area,

which has spread between N 17o 49

׀ , E 77

o 20

׀ and N

18o

02׀ , E 77

o 12

with an altitude of 554-575m above

MSL. The average annual rain fall is 830mm and

average temperature ranges between 35oc to 42

oc.

2.2. Water Quality Monitoring Sites:

Based on the topographical, nature of water flow, nature

of river bed, and disposal of effluent and considering the

accessibility for sampling, a total of four sampling

points were selected covering 21.85km stretch of

Karanja river for the present investigation. The samples

were collected at various locations along the river.

2.3. River Descretization:

The stretch of the river between karanja reservoir and

Bhalki pump station was selected for the study. The

length of 21.85km was deseretization into 19 reaches

with unequal lengths, based on the geography of the

study area. The releases from the Karanja reservoir were

1631 BASAPPA. B. KORI, SHASHIKANTH MISE and SHASHIDHAR



taken as head water data. Fig. 2 shows the river system

unequal segmentation along with the reservoir in which

effluents are being dischared.

2.4. Sampling and Analysis:

The studies were carried out on hourly basis for 24

hours pre-monsoon (30th

June 2010) and analyzed for

the parameters such as temperature, pH, dissolved

oxygen(DO), (BOD)5, nitrate nitrogen (NO3-N),

ammonia nitrogen (NH4-N), conductivity, velocity, and

water depth were measured along the river at sampling

locations. All the activities such as sample collection,

preservation, transportation and analysis were carried

out as per the standard methods (APHA-1995).

2.5. Modeling Tool:

The modeling tool QUAL2Kw has a general mass

balance equation for a constituent concentration ci

(Fig.3) in the water column (excluding hyporheic) of a

reach i (the transport and loading terms are omitted

from the mass balance equation for bottom algae

modeling) as (Pelletier et al., 2006).

Where Qi= flow at reach i(L/day), Qab,i= abstraction

flow at reach i(L/day), Vi= volume of reach i(L), Wi=

the external loading of the constituent to reach i

(mg/day), Si= sources and sinks of the constituent due to

reactions and mass transfer mechanisms (mg/L/day),

Ei= bulk dispersion coefficient between reaches

(L/day), Ei−1, Ei are bulk dispersion coefficients between

reaches i−1 and i and i and i + 1 (L/day),ci=

concentration of water quality constituent in reach

i(mg/L)and t= time (day). Figure 4 Represents the

Schematic Diagram of Interacting Water Quality State

Variables

For auto-calibration, the model uses genetic algorithm

(GA) to maximize the goodness of the fit of the model

results compared with measured data by adjusting a

large number of parameters. The GA maximizes the

fitness function f(x) as:

Where Oi,j= observed values, Pi,j= predicted values,

m=number of pairs of predicted and observed values, wi

= weighting factors and n=number of different state

variables included in the reciprocal of the weighted

normalized RMSE. Detailed descriptions of auto-

calibration method can be found in Pelletier et al.

(2006).

3. Model Calibration:

3.1. Model Input:

The model allows the use of input data for the hydraulic

characteristics of various reaches. Each reach is

idealized as a trapezoidal channel. Under condition of

steady flow, the Manning’s equation was used to

calculate mean velocity and depth as a function of the

stream width, bottom slopes and manning’s roughness

co-efficient. The Karanja river is a natural stream

channel with weeds, windings and pools. For such a

stream, manning’s roughness co-efficient may be

assumed between 0.06-0.07 (Pelletier et al. 2006).

(3)

Where Q = flow rate [m3/s], So = bottom slope [m/m],

n = the manning’s roughness coefficient, Ac = the cross-

sectional area [m2], and P = the wetted perimeter [m In

the head water boundary condition sheet of the model,

the required parameters such as flow rate, temperature,

conductivity, dissolved oxygen, ultimate carbonaceous

biochemical oxygen demand, ammonia nitrogen, nitrate

nitrogen, alkalinity and pH are given as input. The

phytoplankton and pathogen were not measured and the

inputs were left blank. During sampling, it was observed

that the maximum surface of the river bed was covered

by algae and bottom sediment; therefore the algae cover

and bottom-sediment oxygen demand were both

assumed to be 100%. The sediment/hyporheic zone

thickness, sediment porosity and hyporheic exchange

flow were assumed to be 10 cm, 40%, and 5%

respectively.

3.2. Kinetic Parameters:

The ranges of model kinetic parameters were obtained

from QUAL2Kw user manual (Pelletier and Chapra,

2005) and documentation for the enhanced stream water

quality model QUAL2E and QUAL2E-UNCAS (Brown

and Barnwell, 1987).To calculate re-aeration rate

coefficient, Owens–Gibbs formula (Owens et al., 1964)

was applied. The other parameters were set as default in

QUAL2Kw.The model was calibrated by using pre-

monsoon data. Hence, the pre-monsoon calibrated

model was used for further study and the calibrated

kinetic rate parameter for pre-monsoon was presented in

Table 1.

1632 Water Quality Modeling and Management of Karanja River in India




Figure 2: Discritization of Karanja River

Figure 3: Mass Balance in a Reach Segment i. Source: (Pelletier and Chapra, 2005)

(Note: ab: Bottom Algae, ap: Phytoplankton, mo: Detritus, cs: Slow BOD, c f : Fast BOD, cT: Total Inorganic

Carbon, o: Oxygen, no: Organic Nitrogen, na: Ammonia Nitrogen, nn: Nitrite and Nitrate Nitrogen)

Figure 4: Schematic diagram of interacting water quality state variables Source: (Pelletier and Chapra, 2005)




Table 1: Calibrated Kinetic Parameters for the Karanja River Water Quality Modeling in 2010

Parameters Values Units Min.

Value

Max.

Value

Carbon 40 gC 3 50

Nitrogen 7.2 gN 3 9

Phosphorous 1 gP 0.4 2

Dry weight 100 gD 100 100

Chlorophyll 1 Ga 0.4 2

Inorganic Suspended Solids

Setting Velocity 0.8705 m/d 0 2

Slow CBOD Hydrolysis Rate 2.59355 /d 0 5

Slow CBOD Oxidation Rate 0.06117 /d 0 0.5

Fast CBOD Oxidation Rate 0.873 /d 0 5

Organic N Hydrolysis 4.3187 /d 0 5

Organic N Setting Velocity 1.1737 m/day 0 2

Ammonia Nitrification 0.3379 /d 0 10

Nitrate de-nitrification 1.77236 /d 0 2

Sediment de-nitrification Transfer

Coefficient 0.22872 m/d 0 1

Organic Hydrolysis 1.86105 /d 0 5

Organic P setting Velocity 0.28358 /d 0 2

Inorganic P setting velocity 1.40378 m/d 0 2

Inorganic P sediment oxygen

Attenuation half saturation

constant

1.97752 mgO2/L 0 2

Bottom Plants:

Maximum Growth Rate 49.628 gD/m2/d 0 100

Excretion Rate 0.416735 /d 0 0.5

Death Rate 0.31285 /d 0 0.5

External nitrogen half Saturation

constant 242.451 mgN/L 0 300

External phosphorous half

Saturation constant 97.458 mgP/L 0 100

Inorganic carbon half saturation

constant 7.54E-05 moles/L

1.30E

-06 1.30E-04

Light constant 88.15366 Layers/d 1 100

Ammonia preference 10.9891 mgN/L 1 100

Subsistence quota for nitrogen 1.0466244 mgN/L 0.072 72

Subsistence quota for phosphorous 7.2045982 mgN/L 0.01 10

Maximum uptake for nitrogen 363.662 mgN/gD/d 350 1500

Maximum uptake for phosphorous 111.9995 Mg P/gD/d 50 200

Internal nitrogen half saturation

ratio 4.1545502 1.05 5

Internal phosphorous half

saturation ratio 2.7959 1.05 5

Detritus dissolution rate 1.16205 /d 0 5

Detritus setting rate 1.13735 m/d 0 5




Table 2: Water Quality Measurement at Monitoring Station along the Karanja, River as on 30th June 2010

Station Length

(km)

Cond.

(mg/L)

D O

(mg/L)

UBOD5

(mg/L)

NH4

(µg/L)

NO3

(µg/L)

Alk

(mg/L)

PH

HBB 21.85 140 3.8 11.24 510 1800 100 7.9

RB 17.65 170 4.52 10.00 440 1520 100 7.6

DV 7.75 170 5.00 7.40 380 750 105 7.7

HBR 0.00 160 5.95 6.42 320 630 108 7.7

3.3. Model Implementation:

The measured data of pre-monsoon season (30thJune

2010) were used for calibration. The calculation step

was set at 5.625 minutes. The solution of integration

was done with Euler’s method, as Euler’s method is

suggested. For the pH modeling, we used the Brent

method. To obtain the best adjustment, the modeling

system assigns standard weights to various parameters.

The weight for dissolved oxygen was given as 50 and is

justifiable as it is the most influenced parameter. Weight

5 was given for BOD and weight 1 was given to other

parameter. The model was run until the system

parameters were appropriately adjusted and the

reasonable agreement between model results and field

measurements were achieved. The model was run for a

population size of 100 with 100 generations. (Pelletier

et al., 2005).

4. Result and Discussion:

The results for the water quality parameters are shown

in Table 2(water quality measurement as on 30th

June

2010). Calibration results are shown in Fig. 5. The

various scenarios of water quality for Karanja river are

shown in Fig. 6.

4.1. Calibration:

The model was calibrated by using pre-monsoon water

quality data. Hence, the pre-monsoon calibrated model

was used to simulate water quality with different

scenarios. The model calibration results for the water

quality data at four monitoring stations are shown in

Fig.5. The model calibration result is well in agreement

with the measured data with some exceptions. As seen

in Fig.5 dissolved oxygen increase continuously as

distance from head water increases, but at 16.33km and

9.65km dissolved oxygen slightly decreases due to

constant washing clothes, goat, cattle by villagers,

which probably decreases the dissolved oxygen at

16.33km and 9.65km. Biochemical oxygen demand

continuously decreases up to a distance of 9.65km, from

this point onwards BOD remains constant with slight

variation may be due to conversion of algal death to

carbonaceous biochemical oxygen demand (park and

ucbrin 1997). Ammonia nitrogen was increasing up to

15km and decreasing slowly towards the end, and

nitrate nitrogen rapidly decreasing from head water

towards downstream. pH is slightly decreasing at

16.33km and 9.65km, may be due to constant human

activity, and again slightly increases with distance;

Finally alkalinity remains constant from head water to

towards the of river.

4.3. Sensitivity Analysis:

Model sensitivity analysis is carried out in order to

identify the parameters of river water quality that have

the greatest effect on the model output. The analysis

was performed for six parameters (Table 3), by keeping

all the parameters constant, one being increased or

decreased by 20%. It was found that the model was

sensitive for temperature, bottom width, discharge and

biochemical oxygen demand.

4.4. Scenario for Water Quality:

In order to identify what strategies should be adopted to

protect water quality in the study area, the calibrated

model was applied to develop several management

scenarios. Dissolved oxygen profiles obtained from

different management scenario are shown in Fig.6.

i)The simulated dissolved oxygen profile was produced

by different bottom algae modification, in which the

bottom algae was kept below 75% to maintain minimum

dissolved oxygen (i.e.5 mg/L) along the selected length

of river. ii) The biochemical oxygen demand from the

head water is maximum (i.e.17.52 mg/L) during

summer, due to which the dissolved oxygen level along

the river is less than minimum permissible limit. From

this it was observed that, the biochemical oxygen

demand is to be reduced to minimum (i.e.8 mg/L) in

summer from head water to ensure minimum dissolved

oxygen along the study area. Hence to achieve this, the

load from head is to be reduced (i.e.8 mg/L). iii) During

the study period the maximum air temperature observed

is 450c, and minimum air temperature is 13.5

0c, whereas

average temperature is 290c, Temperature affects the

physiology and behavior of fish and other aquatic life.

Highest temperature typically occurs in the month of

April and May, which is critical for both air and water

temperature. iv)To minimize the temperature of water,

shade is an important parameter that controls the stream

heating resulting from solar radiation, from this, it was

observed that the minimum percentage of shade

(i.e.30%) is to be maintained to minimize the

undesirable water temperature and to maintain

minimum dissolved oxygen (i.e.5 mg/L).




Figure 5: Model Calibration Results for the Water Quality Parameters in Karanja River




Figure 6: Various Scenario of Karanja River for Water Quality Management




Table 3: Sensitivity Analysis for the Data on Karanja River in 2010

%DO change

Sl. No. Parameters + 20%Parameters -20% parameters

1 Temperature -3.38 +3.70

2 Bottom width -2.73 +2.89

3 Discharge +2.09 -3.22

4 Biochemical

oxygen demand -0.97 +0.97

5 Nitrate Nitrogen -1.61 +0.00

6 Ammonia Nitrite 0.000 +1.61

5. Conclusion:

The steady- state stream water quality model

QUAL2Kw, was calibrated by using pre-monsoon

(30th

June-2010) data. Model sensitivity analysis was

carried out in order to identify the parameters of river

water quality which have the greatest effect on the

model output. The calibrated model was applied to

develop several scenarios by changing the model input

parameters during pre-monsoon period. They are i) by

bottom algae modification, ii) head water modification

iii) temperature modification and iv)shade modification.

The results show that the bottom algae should be

maintained below 75% to maintain minimum dissolved

oxygen of 5mg/L along the selected length of river, and

in summer, biochemical oxygen demand of the head

water should be minimized (i.e.8mg/L) to maintain

minimum dissolved oxygen level along the river. To

minimize temperature of water, shade is an important

parameter that controls the stream heating resulting

from solar radiation, Trees provide shade to stream and

minimize the undesirable water temperature and

increase the dissolved oxygen level. The combination of

bottom algae modification, head water modification and

minimization of water temperature is necessary to

ensure minimum dissolved oxygen concentration along

the study area.

Acknowledgement:

The authors express their deep sense of gratitude to

Greg.Pelletier, Department of Ecology, Olympia, WA

for the insight given by him in this subject. The authors

also wish to express their gratitude to prof. Shesharao

M. Wanjerkhede, professor of Computer Science and

Engineering, GNDEC Bidar, for his technical

assistance, especially in preparaion of the manuscript.

At the same time they express their gratitude to shri

V.S. Suryan the former Principal of C.B.College Bhalki

for having made emendations in the final draft of this

article.

References:

[1] Ambrose, R.B., Wool, T.A., Connolly, J.P., Shanz,

R.W., 1987. WASP5, A Hydrodynamic and Water

Quality Model. U.S. Environmental Protection

Agency, Athens, GA, EPA/600/3-87/039.0).

[2] APHA., 1995. Standard method’s for the

examination of water and wastewater (19th

edition).American public Health association.

[3] Brown, L.C., Barnwell, T.O. Jr., 1987. The

Enhanced Stream Water Quality Models QUAL2E

and QUAL2E-UNCAS: Documentation and User

Manual. USEPA, Environmental Research

Laboratory, Athens, GA, EPA/600/3-87/007.

[4] Chapra, S.C., Pelletier, G.J., 2003. QUAL2K: A

Modeling Framework for Simulating River and

Stream Water Quality (Beta Version):

Documentation and Users Manual. Civil and

Environmental Engineering Dept., Tufts University

[5] Drolc, A., Konkan, J.Z.Z., 1996. Water quality

modeling of the river Sava, Slovenia. Water Res. 30

(11), 2587–2592.

[6] FláviaBottino, IveCiolaFerraz, Eduardo Mario

Mendiondoand Maria do CarmoCalijuri.,

2010.Calibration of QUAL2K model in Brazilian

micro watershed: effects of the land use on water

quality. Acta Limnologica Brasiliensia vol. 22, no.

4, p. 474-485.

[7] Ghosh, N.C., Mcbean, E.A., 1998. Water quality

modeling of the Kali river, India. Water Air Soil

Pollut. 102, 91–103.

[8] Nikolaos P. Nikolaidis, et al. 2006.Circulation and

nutrient modeling of Thermaikos Gulf, Greece.J.of

marine systems. 60, 51-62.

[9] Owens, M., Edwards, R.W., Gibbs, J.W., 1964.

Some reaeration studies in streams. Int. J. Air

Water Pollut. 8, 469–486.

[10] Ritu Paliwal, et al., 2007.Water quality modeling of

the river Yamuna (India) QUAL2E-UNCAS. J.

Env.management. 83,131-144.

[11] Park, S.S., Lee, Y.S., 2002. A water quality

modeling study of the Nakdong River, Korea. Ecol.

Model. 152, 65–75.

[12] Park, S.S., & Uchrin, C.G., 1990. Water quality

modelingof the lower south branch of the Raritan

River, New Jersey.Bulletin of N.J. Academy of

Science, 35(1), 17–23.




[13] Park, S.S., & Lee, Y.S., 1996. A multiconstituent

movingsegment model for the water quality

predictions in steep andshallow streams.Ecological

Modeling, 89, 121–131.

[14] Park, S.S., & Uchrin, C.G., 1997. A stoichiometric

model for water quality interactions in macrophyte

dominated water bodies.Ecol. Model. 96, 165–174.

[15] Prakash R. Kannel. Seockheon Lee., 2007.

Application of QUAL2Kw for water quality

modeling and dissolved oxygen control in the river

Bhagamati. Environ.Monit. Assess. 125, 201-217

[16] Prakash Raj Kannel.S Lee., 2007. Application of

automated QUAL2Kw for water quality modeling

and management in the Bhagamati river, Nepal.

Ecological modeling 202, 503-517.

[17] Pelletier, G.J., Chapra, S.C., 2005. QUAL2Kw

theory and documentation (version 5.1), A

Modeling Framework forSimulating River and

Stream Water Quality, retrieved 10 May2005 from:

http://www.ecy.wa.gov/programs/eap/models/.

[18] Pelletier, G.J., Chapra, C.S., Tao, H., 2006.

QUAL2Kw, A framework for modeling water

quality in streams and rivers using a genetic

algorithm for calibration. Environ. Model Software

21, 419–425.

[19] Rodrigo de Arruda Camargo, Maria LúciaCalijuri,

Aníbal da Fonseca Santiago,Eduardo de Aguiar de

Couto and Marcos DornelasFreitas Machado e

Silva., 2010.Water quality prediction using the

QUAL2Kw model in a small karstic watershed in

Brazil. ActaLimnologicaBrasiliensia, vol. 22, no. 4,

p. 486-498.

[20] Surindarsuthar, et al. 2009. Assessment of metals in

water and sediment of Hindan river; impact of

industrial and urban discharges. J. of hazardous

materials. 171, 1088-1095.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1639-1644


Prediction of Penetration Rate and Sound Level Produced during

Percussive Drilling using Regression and Artificial Neural Network

S. B. KIVADE, Ch. S. N. MURTHY and HARSHA VARDHAN Dept. of Mining Engg, National Institute of Technology Karnataka (NITK,) Suratkal, Srinivasnagar-575025,

Mangalore-D.K, India


Abstract: The main objective of this investigation is to develop a general prediction model and to study the effect of

predictor variables such as uniaxial compressive strength, air pressure and thrust on penetration rate and sound level

produced during percussive drilling of rocks. The experiment was carried out using three levels Box-Behnken

design with full replication in 15 trials. Modeling was done using artificial neural network (ANN) and

multipleregression analysis (MRA). These techniques can be utilized for the prediction of process parameters.

Comparison of artificial neural network and multiple linear regression models was made and found that error rate

was smaller in ANN than that predicted by MRA in terms of sound level and penetration rate.

Keywords: Percussive Drill; Artificial Neural Network; Multiple Regression Analysis; Uniaxial Compressive

Strength; Sound Level; Penetration Rate.

1. Introduction:

In this study, two mathematical models, multiple regress

ion analysis (MRA) and artificial neural network (AN-

N) were used to predict the penetration rate and sound

level produced during percussive drilling of rocks. The

effect of input parameters i.e., uniaxial compressive

strength (UCS), air pressure (AP) and thrust (T) on the

sound level and penetration rate (output parameters)

was observed. The experiment was carried out using

three levels Box-Behnken design with full replication in

15 trials. Neural network is a massively parallel

distributed processor that has a natural propensity for

storing experimental knowledge and making it available

for use. Also, nonlinearity and input – output mapping

are the two most important benefits in the use of neural

networks. Hence neural networks have been adopted to

model the input – output relationship of non-linear and

interconnected systems. The most common network

used is back propagation, which is essentially a

stochastic approximation to non-linear regression.

In this investigation, the sound level and penetration

rates of percussive drills were measured in the

laboratory and correlated with the rock properties for

the development of reliable equations. Noise

measurements were carried out in open space (outdoor

location) to reduce the effect of reflecting noise. Integral

steel chisel bits of 30, 34, and 40 mm diameters and 42,

43, and 62 cm length were used.

2. Previous Investigations:

The physicomechanical properties of rocks are the most

important parameters in the design of ground workings

and in the classification of rocks for

engineering purposes. The measurement of rock

strength has been standardized by both the American

Society for Testing and Materials [ASTM 1984] and

the International Society for Rock Mechanics [ISRM

1981]. Standard sample preparation is time

consuming and expensive. ANN models are suitable for

complex problems where many factors influence the

mechanism and the result. In this technique many

competing correlations can be examined using massive

parallel networks composed of many computational

elements connected by links of variable weights

[Kalogirou, 2000]. To overcome inaccuracy resulted

from application of empirical methods, it is necessary to

adopt newly developed scientific concepts [Lu Y 2005].

Neural networks may be used as a direct substitute for

auto correlation, multivariable regression, linear

regressi-on, trigonometric and other statistical analysis

and techniques [Singh et al. 2003]. A trained neural

network can be thought of as an ‘‘expert’’ in the

category of information it has been given to analyze.

Thus, the neural network can act as an expert. The

particular network can be defined by three fundamental

components: transfer function, network architecture and

learning law [Simpson, 1990]. It is essential to define

these components, to solve the problem satisfactorily.

Neural networks consist of a large class of different

architectures. Multi Layer perceptron (MLP) and Radial

1640 S. B. KIVADE, Ch. S. N. MURTHY and HARSHA VARDHAN



Basis Function (RBF) are two of the most widely used

neural network architecture in literature for

classification or regression problems [Cohen and

Intrator, 2002, 2003; Kenneth,et al; 2001; Loh and Tim,

2000].

Some of the recent research on estimating UCS using

Multiple Regression (MR), artificial neural network

(ANN) and ANFIS models was carried out by Yilmaz

and Yuksek [2008, 2009]. A higher prediction

performance of adaptive neuro-fuzzy inference system

(ANFIS) over MR and ANN models was reported by

Yilmaz and Yuksek [2009]. Majdi and Beiki [2010]

used Genetic Algorithms (GA) in design and optimizing

the Back Propagation Neural Network (BPNN) structure

and applied the GA–ANN to predict the modulus of

deformation of rock masses. In general, the models

developed are successful in predicting the mechanical

properties of rocks with index properties. However,

there have been few attempts found in the literature on

the identification of the parameters and functions which

play a vital role in defining strength properties of rocks

with regard to the aforementioned rock properties.

3.0 Laboratory Investigations:

3. 1 Experimental Work:

The experiments were designed based on a three level

Box-Behnken design with full replication. Uniaxial

compressive strength, air pressure and thrust are the

independent input variables. Sound level and

penetration rate are the dependent output variables.

Fifteen experimental trials were carried out and the

output data were recorded. Table 1 shows the uncoded

and coded levels of the independent variables.

Table 1: Uncoded and Coded Levels of the Independent

Variables

Independent variables Coded levels

-1 0 1

Uniaxial compressive

strength (MPa) 30.8 81.2 129.7

Air pressure (kPa) 392 490 588

Thrust (N) 100 550 1000

3.2 Percussive Drill Machine:

The experimental set up used in the present work was

the same as given by Vardhan and Murthy (2007) and as

shown in Fig 1.The important specifications of the

jackhammer are: (a) Weight of the machine – 28 kg (b)

Number of blows per minute - 2200 (c) Type of drill

rod- Integral steel chisel (30 mm, 34 mm and 40 mm)

(d) Recommended optimum air pressure- 589.96 kPa.

Figure 1: Jackhammer Drill Setup for Drilling Vertical

Holes in Rock Samples

3.3 Rock Samples Used In the Investigation:

For this study, rocks were collected from different

localities of India. During sample collection, each block

was inspected for macroscopic defects so that it would

provide test specimens free from fractures and joints.

The different rocks used in the investigation and their

physical properties are given in Table 2. The size of the

rock blocks was approximately 30 cm × 20 cm × 20 cm.

3.4 Instrumentation for Noise Measurement:

Sound pressure levels were measured with a CENTER

make Model 320, IEC 651 Type II sound level meter.

The instrument was equipped with a CENTER make

windscreen for minimizing the sound effect produced

from wind, ½ inch electret condenser microphone,

digital display, time weighting and level ranges. The

microphone and the preamplifier assembly were

mounted directly on the sound level meter. The sound

level meter was calibrated before taking up any

measurement using an acoustic calibrator available in

the institute. For all measurements, the sound level

meter was hand held. The instrument was set to measure

A – weighted sound pressure levels in the range of 30

dB to 130 dB.

4. Multiple Regression Analysis (MRA):

The mathematical models for the mechanical properties

with parameters under consideration can be expressed

as Y = f (x1, x2, x3, …) + ∈ where Y represents the

response and x1, x2, x3 are the independent process

variables and ∈ is fitting error. The second order

polynomial equation used to represent response surface

for the factor is given by

1641 Prediction of Penetration Rate and Sound Level Produced during




2

0

1 1 1

n n n

i i ij i i j i j

i i ii j

Y b b x b x b x x= = =

<

= + + + + ∈∑ ∑ ∑ (1)

When there is a curvature in the response surface the

first-order model is insufficient. A second-order model

is useful in approximating a portion of the true response

surface with parabolic curvature. The second-order

model includes all the terms in the first-order model,

plus all quadratic terms like and all cross product

terms like b13 x1i x3j.The second-order model is flexible,

because it can take a variety of functional forms and

approximates the response surface locally. Therefore,

this model is usually a good estimation of the true

response surface.

The above second order response model can be

expressed as

Y = bo + b1 × UCS + b2 × AP + b3 × T + b11 × UCS2 +

b22 × AP2 + b33 × T

2 + b12 × UCS × AP + b13 × UCS

× T + b23 × AP × T (2)

Where b0 is the constant of the regression equation,

coefficients b1, b2, b3, are linear terms, the coefficients

b11, b22, b33, are quadratic terms and the coefficients b12,

b13 and b23 are interaction terms. The value of the

coefficients of the polynomial equation for depth of

penetration and sound level were calculated using

Minitab 15 software. After determining the coefficients,

the mathematical models were developed and are given

in equation Eq. (3) and (4). All the data is used to

generate the regression equations for the prediction of

penetration rate (PR) and sound level (SL).

PR = - 2.10 - 0.0239 × UCS + 0.0158 × AP + 0.00607 ×

T - 0.000051 × UCS2 - 0.000022 × AP

2 - 0.000005 × T

2

+ 0.000047 × UCS × AP - 0.000018 × UCS × T +

0.000006 × AP ×T ……(3)

SL = 115 + 0.0378 × UCS - 0.0020 × AP + 0.00895 × T

+ 0.000054 ×UCS2 + 0.000016 × AP

2- 0.000008 × T

2 -

0.000039 × UCS × AP + 0.000001 × UCS × T +

0.000001× AP × T ……(4)

4.1. Analyzing the Adequacy of the Developed

Model:

The adequacy of the model is tested using the analysis

of variance technique (ANOVA). As per the technique,

if the calculated F-ratio values of the model exceeds the

standard tabulated value i.e., 3.48 of the F-ratio for a

desired level of confidence (say 95%), then the model

may be considered adequate within the confidence limit

( Myer, and Montgomery, 2002). Table 3 shows

analysis of variance for testing adequacy of models and

infers that the model is adequate.

F-ratio (9, 5, 0.05) = 3.48

The R2 value obtained for penetration rate (PR) and sou-

nd level (SL) were nearer to 0.934 and 0.953, which

shows the adequacy of the model. Table 4 shows the

comparison of measured value, ANN simulated and

MRA value.

5. Artificial Neural Network (ANN):

Artificial neural networks are composed of simple

elements operating in parallel. These elements are

inspired by biological nervous systems. The ANN

modeling is carried out in two steps; the first step is to

train the network, whereas the second is to test the

network with data, which were not used for training

neural networks, or artificial neural networks (ANN) to

be more precise, represent the emerging technology

rooted in many disciplines. The behavior of a neural

network is defined by the way its individual computing

elements are connected and by the strength of those

connections or weights. The weights are automatically

adjusted by training the network according to a

specified learning rule until it performs the desired task

correctly. The neural networks consists of a set of

sensory units (source nodes) that constitute the input

layer, one or more hidden layers of computation nodes

as shown in Fig. 2. The back propagation algorithm is

based on the error correction learning rule, and it has

been used successfully to solve some difficult and

diverse problems. The term back propagation refers to

the manner in which the gradient is compared for non-

linear multilayer networks. The back propagation

consists of two passes through the different layers of the

network, and its effect propagates through the network,

layer by layer.

Back propagation is most often applied in the modeling

of non-linear and interconnected parameters systems.

Neurons in the input layer do not have transfer function,

while in the hidden layers a logistic sigmoid (logsig)

transfer function and in output layer linear (purelin)

transfer have been used. MATLAB platform is used to

train and test the ANN. In the training, varied numbers

of neurons (3-7) and the hidden layer (3-9) have been

used in order to predict accurately the output process

parameter. Figure 3 shows the decrease of the mean

square error (MSE) during the training process of

1HX3N. In this study, the input layer consists of three

vector elements, which are UCS, AP, and Thrust.

Levenberg – Marquardt (LM) Back- propagation

training algorithm is used to train the network. In this

network, hidden layer with ‘Tansig’ transfer function

are chosen initially.




Table 2: Uniaxial Compressive Strength of different Rock

Sample Shale Dolomite Sand

stone

Lime

stone Hematite Dolerite

Soda

granite

Black

granite Basalt Gabbros

Uniaxial

compressive

strength

(MPa)

30.8 38.9 63.6 68.1 73.6 81.2 98.4 112.3 119.8 129.7

Table 3: Analysis of Variance for Testing Adequacy of Models

Sum of Squares Degree of freedom

F-ratio P value Models Regression Residual Regression Residual

PR 16.2735 1.9684 9 5 4.59 0.054 adequate

SL 37.3253 2.2371 9 5 9.27 0.012 adequate

Table 4: Design Matrixes with Process Variables and Experimental Measured Responses

Expt.

no

UCS Air

pressure

Thrust Measured value ANN

Simulated value

MRA

Simulated value

SL PR SL PR SL PR

1 -1 -1 0 119.40 2.61 119.39 2.6 119.4 2.61

2 1 -1 0 123.55 0.765 123.52 0.763 123.53 0.764

3 -1 1 0 121.50 4.065 121.48 4.065 121.49 4.066

4 1 1 0 125.25 0.925 125.22 0.923 125.24 0.924

5 -1 0 -1 119.50 0.78 119.49 0.76 119.49 0.77

6 1 0 -1 122.30 0.2 122.26 0.19 122.28 0.21

7 -1 0 1 120.40 2.87 120.37 2.84 120.39 2.85

8 1 0 1 123.30 0.69 123.3 0.68 123.31 0.69

9 0 -1 -1 119.90 0.37 119.87 0.35 119.89 0.36

10 0 1 -1 122.10 0.62 122.08 0.62 122.09 0.63

11 0 -1 1 120.70 0.95 120.71 0.94 120.71 0.96

12 0 1 1 123.0 2.18 122.98 2.16 122.99 2.17

13 0 0 0 122.80 2.35 122.78 2.33 122.77 2.34

14 0 0 0 122.80 2.35 122.78 2.33 122.77 2.34

15 0 0 0 122.80 2.35 122.78 2.33 122.77 2.34

Figure 2: Neural Network Architecture of 3 Input Neurons and Two Output Neurons with Three Hidden Layer

The ‘Purelin’ transfer function is used in the output

layer for the prediction of SL and PR. The logistic

sigmoid transfer function used for hidden layer is given

in Eq. 5.

……(5)

Were x is the weighted sum of the input. The general

form of ‘Purelin’ transfer function used for output layer

is given in Eq. 6.

F(x) =x ……(6)

Training of the neural network is initiated by selecting

number of neurons, momentum correction factor,

learning rate and activation function. The network is

validated using the validation set as input to the network

and the predicted output (SL, PR) is compared with the

actual values. If the predicted values and actual values

are not in close agreement then the network retained by

varying the number of neurons, and hidden layers. The

1643 Prediction of Penetration Rate and Sound Level Produced during




decision as to the number of neurons used in the hidden

layer usually depends on the arithmetical mean of the

number of inputs and outputs.

Figure 3: Training Results based on the 3-3-2

Configuration

In this application 3 to 12 neurons in the hidden layers

are employed. The computations are performed using

the NN toolbox of MATLAB. In order to calculate the

SL and PR, mathematical formulations can be derived

from the resulting weights and the activation functions

used in the ANN. Ten networks are trained and

validated to obtain the results by varying the number of

neurons and hidden layers. Of the ten networks selected

for training (1HX3N, 2HX3N, 3HX5N, 1HX7N,

2HX9N, 1HX12N, 2HX9N, 3HX9N, 1HX12N,

3HX12N) the network with one hidden layer with three

neurons give the best coefficient of multiple

determination (R2- value) around 0.9998.

Hence the network 1HX3N is selected for testing the

data.

The effect of process parameters UCS, AP, and T on the

SL and PR are analyzed. The major parameters which

influence the SL and PR are UCS and AP. It is found

that thrust is having less influence on SL and PR. Fig. 4

and 5 shows the cross-correlation graph between

estimated and actual penetration rate and estimated

sound level and actual sound level based on

multiple regression models. Fig. 6 and 7 shows the

cross-correlation graph between estimated and actual

penetration rate and estimated sound level and actual

sound level based on artificial neural network models.

6. Conclusions:

In this paper, use of ANN and MRA for prediction of

sound level and penetration rate of different rocks was

described and compared. The multiple regression

equations for the prediction of the penetration rate and

sound level were developed. R2 value of 0.934 and

0.953 for penetration rate and sound level respectively

were obtai-ned. UCS and AP are the major parameters

influencing the sound level and penetration rate and

thrust is having less influence on these.

Figure 4: Cross-Correlation Graph between Estimated

and Actual Penetration Rate based on Multiple

Regression Model


and Actual Sound Level based on Multiple Regression

Model

The input and output parameters are analyzed using

neural networks model (NN) with different network

configurations. Ten networks are trained and validated

to obtain the results by varying the number of neurons

and hidden layers. Of the ten networks selected for

training, the network with one hidden layer with three

neurons gave the best coefficient of multiple

determinations (R2 value) around 0.9998. Hence the

network 1HX3N is selected for testing the data.

Expressions for sound level and penetration rate are

extracted from the neural network model. Comparison

of artificial neural network and multiple linear




regression models was made and found that error rate

was smaller in ANN than that predicted by MRA in

terms of sound level and penetration rate and also

observed that R2 value obtained by ANN model is

higher than MRA.


and Actual Penetration Rate based on ANN Model


and Actual Sound Level Based on ANN Model

7. References:

[1] ASTM, (1984); “American Society for Testing and

Materials, Standard test method for unconfined

compressive strength of intact rock core

specimens”, Soil and Rock, Building Stones:

Annual Book of ASTM Standards, vol. 4.08.

Philadelphia, Pennsylvania.

[2] Brown ET. (1981); “Rock characterization testing

and monitoring”, International Society of Rock

Mechanics (ISRM) suggested methods. Oxford:

Pergamon.

[3] Yilmaz, I.,Yuksek, A.G.,(2009); “Prediction of the

strength and elasticity modulus of gypsum using

multiple regression, ANN,ANFIS models and their

comparison. International Journal of Rock

Mechanics and Mining Sciences 46(4), 803–810.

[4] Yilmaz, I.,Yuksek, A.G.,(2008); “An example of

artificial neural network application for indirect

estimation of rock parameters”. Rock Mechanics

and Rock Engineering 41(5), 781–795.

[5] Majdi, A., Beiki, M., (2010); “Evolving neural

network using a genetic algorithm for predicting the

deformation modulus of rock masses”. International

Journal of Rock Mechanics and Mining Sciences

47 (2), 246–253.

[6] Lu Y. (2005); “Underground blast induced ground

shock and its modeling using artificial neural

network”. Comput Geotech. 32:164–78.

[7] Kalogeria S. (2000); “Applications of artificial

neural networks for energy systems”. Appl Energy

67(2):17–35.

[8] Simpson, P. K. (1990); “Artificial neural system

foundation, paradigm, application and

implementation”. New York: Pergamon Press.

[9] Singh, T. N., Kanchan, R., Verma, A. K., and

Singh, S. (2003); “ An intelligent approach for

prediction of triaxial properties using unconfined

uniaxial strength”. Mining Engineering Journal, 5,

12–16.

[10] Cohen, S., and Intrator, N. (2002); “Automatic

model selection in a hybrid perceptron/ radial

network. Information Fusion.” Special Issue on

Multiple Experts, 3(4), 259–266.

[11] Cohen, S., and Intrator, N. (2003); “A study of

ensemble of hybrid networks with strong

regularization”. Multiple Classifier Systems, 227–

235.

[12] Loh, W., and Tim, L. (2000); “A comparison of

prediction accuracy, complexity, and training time

of thirty three old and new classification

algorithm”. Machine Learning, 40(3), 203–238.

[13] Kenneth, J., Wernter, S., and MacInyre, J. (2001);

“Knowledge extraction from Radial Basis Function

networks and Multi Layer Perceptrons”,

nternational Journal of Computational Intelligence

and Applications, 1(3), 369–382

[14] Vardhan, H. and Murthy, Ch. S. N., (2007); “An

experimental investigation of Jack hammer drill

noise with special emphasis on drilling in rocks of

different compressive strength”; Noise control

Engineering Journal; may- June: Vol. 55(3); pp.

282-293.

[15] Myer, R. and Montgomery, D.C. (2002); “Response

surface methodology”, John Wiley, New York.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1645-1651


Utility of Red Mud as an Embankment Material

SUBRAT KUMAR ROUT1, TAPASWINI SAHOO

2 and SARAT KUMAR DAS

2

1Civil Engg Department, ITER, SOA University, Bhubaneswar

2Civil Engg. Department, NIT Rourkela


Abstract: Globally, aluminum industries are producing approximately 75 million tones of red mud every year. Less

than half of this is used. Storage of this unutilized red mud apart from covering vast tracks of usable land also

pollutes land and water in the vicinity. Embankments as part of construction of highways, expressways, railways

and remote area connectivity are using large quantities of natural resources. This paper discusses use of red mud as

an alternate embankment material, based on laboratory findings and finite element analysis. The geotechnical

properties such as specific gravity, classification, compaction characteristics, triaxial shear strength and dispersion

properties of red mud are discussed. A comparison is made with the properties of fly ash. A method to prevent

dispersion of red mud and stability analysis of the embankment using finite element analysis with static and

dynamic load is also presented.

Keywords: CBR, Compaction Characteristics, Finite Element Method, Grain Size Classification, Plasticity Index,

Red Mud, Unconfined Compressive Strength.

1. Introduction:

Aluminum industries are producing huge quantity of

industrial waste known as red mud. Globally there are

approximately 70 million tones of red mud being

produced every year with less than half of this is used.

The red mud is highly alkaline (pH ranges from 10.5 to

13) and caustic.It is typically deposited as slurries with

15 to 40% solids. Storage of this unutilized red mud

takes vast tracks of usable land. This also pollutes the

environment in terms of ground/ surface water and land

contamination. Various efforts have been made to use

red mud such as removal of sulfur compounds from

kerosene oil [1], pozzolanic pigment [2], purification of

waste waters from nuclear power plants [3], plasma

spray coatings (ceramic/cermet) on metal substrates [4].

But the volume of utilization in these cases is limited.

Construction of embankment has become an integral

part of highways, expressways and other connectivities.

Presence of expansive soils and shortage of borrow area

soil creates hindrance to such projects. From

environmental consideration, vast use of top soil is also

a matter of concern. It takes thousands of years to form

the natural top soil, which is essential for agriculture

and growth of vegetation. Hence, there is a great

necessity to use alternate/waste material in place of

natural top soil.

Very limited efforts have been made in various parts of

the world regarding utilization of red mud as an

embankment material. Some of the initial efforts in

geotechnical characterization of red mud are presented

as follows. Vogt [5], observed that the in-situ undrained

shear strengths of red mud vary highly compared to

uncemented clayey soils. It has very high friction angles

(φ), varying from 380-42

0. Somogyi and Gray [6], Fahey

and Newson [7] observed that red mud has a

compression indices (Cc ) of 0.27-0.39 (similar to silty-

clay soils), coefficient of permeability (k) of 2-20 x10-

7cm/s and coefficient of consolidation ( Cv) of 3 – 50

x10-3

cm2/s. Red mud tends to have low plasticity [e.g.,

WL = 45%, IP = 10%] and relatively high specific

gravity (GS = 2.8-3.3). There is lack of clay mineralogy

and these wastes show many geotechnical properties

similar to clayey tailings found in other mineral

processing [8](Vick 1981).

Llimited study has been carried out to find geotechnical

properties of red mud particularly information about

Indian red mud. For the high embankment, stability

analysis is very important, which is generally found by

slope stability analysis. The limit equilibrium method

with circular or wedge/planer slip surface is assumed for

this analysis. But, in limit equilibrium method, it is not

possible to find out the stress and strain. In embankment

analysis it is also important to study the dispersion

properties of material to protect the embankment against

weathering effect.

Hence, in this study an attempt has been made to

characterize red mud as an alternate embankment

material. Accordingly, necessary geotechnical

laboratory investigations have been carried out. Finite

Element Method (FEM) is used to study the stability of

embankment,based on above geotechnical properties.

1646 SUBRAT KUMAR ROUT, TAPASWINI SAHOO and SARAT KUMAR DAS



Present study is expected to help the engineers, planners

to use red mud as alternate material, particularly for

difficult soil in borrow area and/or to avoid the

environmental degradation of top soil.

3. Experimental Programme:

3.1 Materials and Test Programme:

The red mud used in the experimental work was

collected from National Aluminium Company Ltd.

NALCO, Damanjodi Koraput, Odisha and a typical

discharge point is shown in Figure 1.

Figure 1: Discharge of Red Mud as Slurry into the

Pond

The geotechnical properties of red mud like specific

gravity, plasticity index, swelling index, linear

shrinkage, grain size classification, compaction

characteristics, and triaxial shear tests were investigated

as per relevant IS codes.

Table 1 Geotechnical Properties of Red Mud

Sl.

No

Properties Red mud

1 PH value 11.4

2 specific Gravity 3.34

3 Plasticity

characteristics

Liquid limit (%) 24.75

Plastic limit (%) 17.5

Plasticity index (%) 7.25

4 Volumetric

shrinkage(%) 1.6

5 Linear

shrinkage(%) 5.26

6 IS classification ML,CL

The basic physical properties of red mud are shown in

Table 1. It can be seen that the red mud is highly

alkaline with PH value of 11.4 and the specific gravity

(3.34) is also very high compared to soil. Figure 2

shows the X-ray diffraction pattern of red mud and it

was observed that red mud has hematite as

major mineral and goethite, gibbsite, rutile, boehmite,

sodanite as minor minerals. The high specific gravity of

red mud is due to presence of iron rich minerals.

It has low plasticity and low volumetric and linear

shrinkage. As per BIS soil classification it can be

classified as silty soils of low plasticity as well as clayey

soils of low plasticity (ML-CL). However, it may be

mentioned that as shown in Figure 2, it does not contain

standard soil minerals. Hence, as discussed below its

behaviour is not matching with that of soil.

10 20 30 40 50 60 70

0

50

100

150

200

250

300

Intensity(arb unit)

2Theta(arb unit)

RED MUD

B

Gb

Gb

Gb RGb

H,Go

Go,R,H

Go,R,H

Gb,Go,H

H,Go,R

GbB,H

Gb

Gb

R,Go,H

S,Gb,H

S,Go,Gb

S,H,R,Go

R,S,B,H,Gb

H=hematite, B=Boehm ite

Gb=Gibbsite, R=Rutile

Go=Goethite, S=Sodalite

Figure 2: X-Ray Diffraction Pattern of Red Mud

4. Results and Discussion:

The results of other laboratory tests are presented in the

following section.

4.1 Grain Size Classification:

The grain size distribution curve of red mud is presented

in Figure 3. It is observed that more than 90% of

particle sizes of red mud are fine grained (< 0.075mm).

Figure 3: Grain Size Distribution Curves of Red Mud

with other Soils

1647 Utility of Red Mud as an Embankment Material



The grain size distribution of a local soil and fly ash is

also presented for comparison in Figure 3. It is observed

that grain size distribution of red mud and fly ash is

comparable and finer than local soil. However, the

particles of fly ash are spherical but that of red mud is

angular. The low plasticity of fly ash is due to spherical

particles. Though red mud is angular, the plasticity is

low due to absence of clay minerals. This low plasticity

may help to use red mud as a subgrade material.

4.2 Compaction:

The compaction curve for red mud using light

compaction and heavy compaction is shown in Figure 4.

Figure 4 also describes the compaction characterisation

of fly ash and a local sandy soil for comparison. It can

be seen that red mud has higher maximum dry density

(MDD) in comparison to other materials. This high

MDD value is due to high specific gravity value of red

mud.

Figure 4: Compaction Characterisation of Red Mud

with Fly Ash and Other Soil

4.3 Triaxial Shear Strength:

The stress- strain curves for the compacted sample of

red mud and fly ash are shown in Figure 5.

Figure 5: Stress- Strain Curve of Red Mud, Fly Ash and

Soil

The shear strength value of the red mud is found to be

more as compared to fly ash. The cohesion of red mud

is found as 28.8 kN/m2 and the angle of internal friction

(φ) as 34.830. The corresponding values for the fly ash

are found to be 19.68 kN/m2 and 24.37

0 respectively. It

may be mentioned here that as shown in Figure 2, the

grain size distribution of fly ash and red mud are

comparable and are fine grained. This high value of

φ of red mud is partially due to particle morphology

and partly due to mineralogy.

4.4 Dispersive Test:

Soils or any kind of waste product like red mud, slag,

crusher dust, coal ash get dispersed and washed away in

water.This process is termed as dispersiveness. Non-

plastic nature of particle and its inadequate inter particle

attraction causes dispersiveness. This is the main

property which used to come into consideration during

the construction of dykes, embankments and different

kinds of water retaining structures.

Figure 6a: Prepared Samples for Crumb Test

Figure 6b: Sample in Distilled Water after Seven

Minutes

Dispersive property cannot be identified by the standard

laboratory index tests , such as visual classification,

grain size analysis, specific gravity or Atterberg’s

limits. Therefore, other laboratory tests have been

derived for this purpose. In the present study crumb test,




double hydrometer test and turbidity tests [9] have been

performed to identify dispersive clays. The crumb test

has been done [9] and the results are shown in Figure 6.

It is observed that red mud is highly dispersive. Based

on double hydrometer test, dispersion ratio of red mud

is found to be 94%, which is extremely dispersive [5].

Turbidity test also has been done as per [10] also

showed red mud as highly dispersive material.

5. Finite Element Analysis:

In the present study the finite element analysis of the

embankment is done using FEM based software-

PLAXIS [11].

5.1 Finite Element Analysis using PLAXIS:

PLAXIS [11] is a finite element program for

geotechnical applications in which different soil models

are used to simulate the soil behavior. It’s

implementation consists of three stages, known as input

stage, calculation stage and post processing (curves)

stage. Input stage contains model design, assigning the

material parameters, boundary conditions, loading and

meshing. In the present analysis 15-node triangular

element is considered for meshing, which contains 12

stress points. In PLAXIS, stresses and strains are

calculated at individual Gaussian integration points

rather than at nodes. In the calculation stage, analysis

type is chosen such as Plastic, dynamic, consolidation

and phi-c reduction. The assigned loads are activated in

this stage and analyzed. In the post processing stage,

curves are plotted between various calculated

parameters such as load Vs displacement.

To compare with the limit equilibrium method in

addition to stress-strain calculation, the factor of safety

(FOS) of slope is calculated using Phi-c (φ-C) reduction

method.

5.1.1 Φ-C Reduction Method:

Phi-c reduction is an option available in PLAXIS to

compute FOS for the stability problems. This option can

be selected as a separate calculation type in the general

tab sheet. In the Phi-c reduction approach the strength

parameters tanφ and c of the soil are successively

reduced in the same decrement. The total multiplier

∑Msf (MSF) is used to define the value of the soil

strength parameters at a given stage in the analysis:

reduced

input

reduced

input

c

cMsf ==Σ

φ

φ

tan

tan

Where the strength parameters with the subscript ‘input’

refer to the properties entered in the material sets and

parameters with the subscript ‘reduced’ refer to the

reduced values used in the analysis. ∑Msf is set to 1.0 at

the start of calculation to set all material strengths to

their unreduced values. The variation of MSF with

displacement is presented to find out the FOS. The

material properties considered for the FEM analysis is

shown in Table 2.

Table 2: Soil Properties for Mohr-Coulomb Model

Mohr-Coulomb

parameters

Local

soil Red mud

Unit weight (kN/m3) 16 19.8

Permeability (m/day) 5 x 10-3

5.832 x 10-4

Cohesion (kN/m2) 30 28.8

Internal friction (degree) 15 34

Young’s modulus

(kN/m2)

3500 1771

Poisson ratio 0.3 0.34

5.2 Stability Analysis of Embankment Slope:

For the embankment slope analysis the top width of a

typical 2-lane road is taken as 14m (7m carriage

way+1.5X2 paved shoulder+2X2 earth shoulder).

Similarly for 4-lane road it is taken as 26m. The

embankment slope considered are with slope 1:2 (1

vertical: 2 horizontal). The embankment heights

considered are 10m and 15m. The stability analysis of

embankment with red mud only first analysed.

5.2.1 Example 1:

In the 1st attempt to construct the embankment, only red

mud has been used as the base material. The slope

height is kept at 10m with the slope inclination of 1:2.

Model diagram with its deformation mesh is shown in

Figure 7. Figure 8 shows the shear failure results of

PLAXIS analysis. The factor of safety is described in

terms of MSF. The variation of the MSF with

displacement is shown in Figure 9 and the FOS of the

slope is found to be 2.97.

Figure 7: The PLAXIS Model for the Slope Using Only

Red Mud

Some more studies are made for other slopes and the

results are shown in Table 3. It was found that for all the

cases te FOS is found to be more than 1.5.




Figure 8: The Shear Failure Surface of Slope as Per

Plaxis Model for the Slope using Only Red Mud

0 500 1e3 1.5e3 2e3 2.5e3

1

1.5

2

2.5

3

Sum-Msf

Figure 9: The Factor of Safety of Slope as Per Plaxis

Model for the Slope Using Only Red Mud(X-Axis for

Defomation and Y-Axis for Factor of Safety)

Table 3: FOS Based on Its Embankment Height and

Width

(Width, height) FOS

14, 10 (RM) 2.96

14, 15 (RM) 2.56

26, 10 (RM) 2.96

26, 15 (RM) 2.53

Erosion and Drainage:

Another important aspect of high embankment is the

erosion due to rain and flooding on one side or both

side. The flood water may lead to hydraulic fracturing

[12] and may ultimately lead to failure. Hydraulic

fracturing by réservoir water acting on the upstream

face of the dam core causes concentrated leaks of water

to enter the core. Hydraulic fracturing due to high water

pressure has caused leakage or failure of many

embankments/ dams [12]. Identification of hydraulic

fracturing also helps in positioning of filter bed. Such a

study is also made to find the effect of water logging

during flood.

AS discussed above It is observed that red mud is a

dispersive (erodible) material. Hence, following [13]

attempt has been made to provide soil cover with local

soil, similar to that of fly ash embankment. The

properties of local soil for the present analysis are also

presented in Table 2.

5.2.2 Example 2:

In this example an attempt has been made to analyze

embankment by covering the red mud with the local c-φ

soil. The slope geometry as described above with only

red mud is analyzed with cover material of 3.0m on

sides 1.0m on top as shown in Figure 10. Figure 11

shows the shear failure results of PLAXIS analysis. The

variation of the MSF with displacement is shown in

Figure 12 . The FOS of the slope is 3.0.

Figure 10: The PLAXIS Model with Its Deformation

Mesh for the Slope using Red Mud and 1.0m Vertical

Soil Cover

Figure 11 The Shear Failure Surface of Slope as Per

Plaxis Model for the Slope using Red Mud And 1.0m

Vertical Soil Cover

To study the effect of water logging, the next attempt

same soil cover by phreatic line is taken into

consideration. The PLAXIS model with its deformation

mesh with its position of phreatic line and effective

stress diagram in Z-Z direction is shown in Figure 13

and 14 respectively. From the effective stress diagram it

is observed that little tensile stress occurred at the

surface of the embankment, which is much above the




phreatic line. Hence, there is less chance of hydraulic

fracture.

0 500 1e3 1.5e3 2e3 2.5e3

1

1.5

2

2.5

3

3.5

|U| [m]

Sum-Msf

Figure 12: The Factor of Safety of Slope as Per Plaxis

Model for the Slope using Red Mud and 1.0m Vertical

Soil Cover

Figure 13: The PLAXIS Model with its Deformation

Mesh for the Slope using Red Mud and 1.0m Vertical

Soil Cover With its Phreatic Line

Figure 14: Effective Stress Diagrams for the Slope

using Red Mud and Soil Cover of 1.0m Vertically in Zz

Direction

Study has also been made for different heights and

widths of dam with only red mud and with red mud core

and soil cover. The comprehensive results have been

presented in Table 4. It is observed that in all the cases

the FOS is more than 2.5, and [13 and 14].Since more

than 1.5 is sufficient high embankment can be

constructed using red mud.

Table 4: FOS at different Width and Height with Soil

Cover

Width, height (Meter) FOS

14, 10 (RM, soil cover) 3.00

14, 15 (RM, soil cover) 2.57

26,10 (RM, soil cover) 2.99

26, 15 (RM, soil cover) 2.54

The embankments are also been analysed putting the

axle load on it. The results for axle load with static and

dynamic action are shown in Table 5. It can be seen that

with load also the FOS is more than or close to 1.5.

Table 5: FOS at different Width and Height for

Concentrated Load and due to Vibration

Width, height

(red mud, Soil cover) FOS

14, 10 (RM) concentrated load 1.77

14, 10 (RM, soil) concentrated load 1.48

14, 10 (RM) due to vibration 1.78

14, 10 (RM, soil) due to vibration 1.49

Conclusion:

This paper described the design of high embankment

using red mud based on the laboratory geotechnical

investigation and the stability analysis using FEM.

Based on the observations and discussions thereof

following conclusions can be made:

1. The specific gravity of Red mud is very high

compared to soil, due to presence of iron compound

minerals. It has low plasticity and low volumetric and

linear shrinkage with 90% of particle finer than

0.075mm.

2. Red mud has high maximum dry density (MDD) in

comparison to other materials, due to high specific

gravity.

3. The angle of internal friction value of the red mud is

found to be more as compared to fine grained fly ash.

4. Based on crumb test, double hydrometer tests and

turbidity test red mud is found to be highly dispersive.

5. To protect the dispersive red mud against external

weathering it is recommended to cover it with local soil.

The embankment with soil cover is found to have more

than required FOS value (as per standard

specifications).

7. The embankments are also analyzed considering

concentrated load and the load with vibration. The FOS

is found to decrease with static and vibration load.

8. This limited study shows that red mud has the

potential to be used as an embankment material. More

study is being conducted regarding economy,




convenience in construction and safety during

construction.

References:

[1] Singh, A.P., Singh, P.C., Singh, V.N (1995)

“Cyclohexanethiol separation from kerosene oil by

red mud”, J. Chem. Technol. Biotechnol. 56 (1993)

167.

[2] Pera, J., Boumaza, R., and Ambroise, J.,

“DEVELOPMENT OF A POZZOLANIC

PIGMENT FROM RED MUD” (Received

December 11, 1996; in final form June 2, 1997)

[3] Mikelic Luka, Orescanin Visnja, Lulic Stipe.

“Utilization of Red Mud for the Purification of

Waste Waters from Nuclear Power Plants.” TA-7

“waste management and decommissioning”.

[4] Mishra, B., Staley, A., Kirkipatrick, D., (2002),

“Recovery of value added products from red mud”,

Minerals and Metallurgical processing. Vol(19),

87-94.

[5] Vogt, M. F. (1974). “Development studies on

dewatering of red mud.” 103rd

Annual Meeting of

AIME, Dallas, Tex., 73-91.

[6] Somogyi, F., and Gray, D. (1977). “Engineering

properties affecting disposal of red mud.” Proc.,

Conf. on Geotechnical Practice for Disposal of

Solid Waste Materials, ACSE, 1-22.

[7] Fahey, M., Newson, T. A., and Fujiyasu, Y. (2002).

“Engineering with tailing.” Invited Lecture, Proc.,

4th

Int. Conf. On Environmental Geotechnics. Rio

de janeiro, Brazil, 2, 947-973, Balkema, Lisse.

[8] Vick, S. G. (1981). Planning, design and analysis of

tailing dams, Wiley, New York, 369.

[9] Bhuvaneshwari, S., Soundra, B., Robinson, R.G.,

and Gandhi, S.R., “Stabilization and

Microstructural Modification of Dispersive Clayey

Soils”, First International Conference on Soil and

Rock Engineering, Columbo, Srilanka, August 5-

11, 2007.

[10] Sridharan, A., Pandian, N.S. and Subranya Prasad,

P. (2001g), Vol. 7: Pin hole studies on Indian coal

ashes, Technical report of task force on

Characterization of fly ash submitted to

Technology Mission-Fly Ash Disposal and

Utilisation, Dept. of Science and Technology, Govt.

of India.

[11] Brinkgreve. R.B.J, Broere.W, Waterman.D (2008).

“PLAXIS -2D (Version 9.0)”, Delft University of

Technology and PLAXIS b.v., The Netherlands.

[12] Sherard, J.L (1986) “Hydraulic fracturing in

embankment dams,” Journal of Geotechnical

Engineering, Vol. 112, No. 10, pp 905-927.

[13] IRC: 58-2001. “Guidelines for use of fly ash in

road embankments”, The Indian Road congress,

Jamnagar house, Shahjahan Road, New Delhi-

110011.

[14] IRC: 75-1979. “Guidelines for the design of high

embankments”, The Indian Road congress,

Jamnagar house, Shahjahan Road, New Delhi-

110011.

www.cafetinnova.org

Indexed in



ISSN 0974-5904, Volume 05, No. 06

December2012,P.P.1652-1658


Rapid Chloride Penetration Test on Geopolymer Concrete

SHANKAR H. SANNI and R. B. KHADIRANAIKAR Department of Civil Engg., Basaveshwar Engg. College, Bagalkot, Karnataka


Abstract: This paper presents the experimental investigation done on permeability characteristics of geopolymer

concrete. The grades choosen for the investigation were M-30, M-40, M-50 and M-60 with the molarity of 8M. 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, 100x200 mm cylinders heat-cured at 60°C in an

oven. The variation was studied on the specimens subjected to ambient air as well as oven heat curing. The test

results indicate that the geopolymer concrete in 6 hours allowed 1560 to 1980 Coulombs giving low rating as per

ASTM 1202. The RCPT values indicate improved resistance for chloride permeability and low rate of corrosion risk

level.

Keywords: Geopolymer Concrete, Rapid Chloride Penetration Test, Molarity, Sodium Hydroxide, Sodium Silicate

Introduction:

In the context of increased awareness regarding the ill-

effects of the over exploitation of natural resources, eco-

friendly technologies are to be developed for effective

management of these resources. 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 [Narasimhan, et

al., 1997].

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 [Rangan, 2006]. 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

[Davidovits, 2008]. Fly ash is the most common source

material for making geopolymers. Normally, good high-

strength geopolymers can be made from class F fly ash

[Schmucker, et al., 2004]. The alkaline activation of

materials can be defined as chemical process that

provides a rapid change of some specific structures,

partial or totally amorphous, into compacted cemented

frame works [Palomo, et al., 2003].

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

[Davidovits, Fernandez-Jimenez, van Deventer, 1999,

2000]. Durability of concrete is the ability of concrete to

remain fully functional over an extended period under

prevailing service conditions for the purpose for which

it has to be designed. The durability of concrete is

closely related to its permeability. The permeability

indicates the rate at which aggressive agents can

penetrate to attack the concrete and the steel

reinforcement. The chloride permeability of GPC was

varying from ‘Low’ to ‘very low’ as per ASTM 1202

[Rajamane, et al., 2011].

Experimental Investigations:

Materials:

The following materials have been used in the

experimental study [Shankar, et al., 2012]

a) Fly Ash (Class C) collected form Raichur Thermal

power plant having specific gravity 2.00.

b) Fine aggregate: Sand confirming to Zone –III of

IS:383-1970 having specific gravity 2.60 and fineness

modulus of 2.70.

c) Coarse aggregate: Crushed granite metal confirming

to IS:383-1970 having specific gravity 2.72 and fineness

modulus of 6.40.

d) Water : Clean Potable water for mixing

e) Alkaline Media: Specific gravity of

1653 SHANKAR H. SANNI and R. B. KHADIRANAIKAR



i) Sodium Hydroxide (NaOH) = 1.16

ii) Sodium Silicate (Na2SiO3) = 1.57

f) Superplasticizer : Conplast (SP-430)

Tests were conducted on specimen of standard size as

per IS: 516-1959. Details of tests conducted and

specimens used are given in Table 1.

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 [Rangan, 2008]. The details of mix design

and its proportions for different grades of GPC are

given in Table 3.

Mixing, Casting, Compaction and Curing of

Geopolymer Concrete:

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.

(Figure 1 and 2). In preparation of NaOH solution,

NaOH pellets were dissolved in one litre of water in a

volumetric flask for concentration of NaOH (12M).

Alkaline activator with the combination of NaOH and

Na2SiO3 was prepared just before the mixing with fly

ash. The ratio of alkaline liquid to fly ash by mass varies

with the grade of concrete [Rangan, 2008]. The alkaline

liquid (Na2SiO3 / NaOH) used in the current study was

2.5 for all the mixes. This ratio was selected, since it

produced the highest compressive strength [Shankar, et

al., 2012]. The chloride penetration inside the concrete

may cause the corrosion of steel hence it was mandatory

to check the pH of alkaline solution [Fig. 6]. The fly ash

and alkaline activator were mixed together in the mixer

until homogeneous paste was obtained. This mixing

process can be handled within 5 minutes for each

mixture with different molarity of NaOH. Fresh fly ash

based geopolymer concrete was usually cohesive. The

workability of the fresh concrete was measured by

means of conventional slump test [Fig. 5]. Heat curing

of GPC is generally recommended, both curing time and

curing temperature influence the compressive strength

of GPC [Hardjito, et al., 2004]. After casting the

specimens, they were kept in rest period for two days

and then they were demoulded. The demoulded

specimens were kept at 60°C for 24 hours in an oven as

shown in Fig. 4.

Rapid Chloride Penetration Test (RCPT):

Whiting in 1981 developed the rapid chloride

permeability test under research sponsored by the

Federal Highway Administration (FHWA). This test

was adopted by the AASHTO in 1983 and was given

the designation AASHTO T-277. In 1991, this test was

adopted by ASTM and was given the designation

ASTM C 1202 ‘Electrical Indication of Concrete’s

ability to resist chloride Ion penetration’ is referred to as

Rapid Chloride Permeability Test (RCPT).

Figure 1: Mixing of Alkaline Solution

Figure 2: Mixing of GPC

Figure 3: View of GPC Specimens

1654 Rapid Chloride Penetration Test on Geopolymer Concrete



Figure 4: Curing of GPC Specimens

Figure 5: Slump Cone Test

Figure 6: Ph of Alkaline Solution

The RCPT determines the electrical conductance of the

concrete specimen to provide a rapid indication of its

resistance to the penetration of chloride ions. The

instrument used is designed conforming to the ASTM C

1202.

The RCPT is table mounted unit, which measures the

electrical charges passed through a 2-inch thick slice of

4-inch nominal diameter concrete core or cylinder

during a period of 6 hours. A potential difference of 60

V D.C is maintained across the ends of the specimens.

One end of the specimen is immersed in a sodium

chloride (3% Nacl, -ve node) solution while the other in

sodium hydroxide solution (0.3N, NaOH, +ve node)

solution. The total charge passed, in Columbs, is related

to the resistance of the specimen to chloride ion

penetration. RCPT cell is shown in Fig. 7. The total

charges passed are automatically recorded by RCPT

software which stores charges passed of every time.

Figure 7: RCPT Sample Holding Cell

Specimen Preparation:

The 50 mm thick concrete disk was sliced from 100mm

x 200mm height cylinder (Fig. 8)

Conditioning of Specimen:

The sliced specimen disk was kept in vacuum

desiccators to remove air present in pore for 3 hours.

Then water was allowed into desiccators for 1 hour

under vacuum so that disk gets water saturated. The

specimen was kept for 18 hours in water under

atmospheric pressure.

Measurement of Charge Passing:

In this step well conditioned disk was mounted in

testing cell and chambers were filled with NaOH and

NaCl solution. Then the testing cell was applied to 60 V

D.C through measurement unit as shown in Fig. 8, the

charges were recorded by the computer.

Results and Discussions:

Workability:

The workability of the geopolymer concrete decreases

with increase in the grade of the concrete as presented in

Table 2, this is because of the decrease in the ratio of

water to geopolymer solids. As the molarity of the

NaOH solution increases the workability of the




geopolymer concrete decreases, because of the decrease

in the water content. Thus we can say that as the grade

of the concrete increases, the mix becomes stiffer

decreasing the workability.

Figure 8: RCPT Working Procedure




Table 1: Details of Specimen used and Tests Conducted

Type of tests conducted Size of specimen No. of specimen cast for

different grades

Compressive strength 150x150x150mm 5

Split tensile strength 100x200mm 5

Table 2: Slump Values for different Grades of Gpc

Grade Na2SiO3/NaOH Slump (mm)

M-30 2.5 135

M-40 2.5 130

M-50 2.5 110

M-60 2.5 95

Table 3: Mix Proportions of GPC Mix with Molarity of 12M (Na2SiO3/ NaOH as 2.5)

Materials

Mass (kg/m3)

M-30 M-40 M-50 M-60

Coarse

aggregates

20 mm 277.20 277.20 277.20 277.20

14 mm 369.60 369.60 369.60 369.60

7 mm 646.80 646.80 646.80 646.80

Fine sand 554.40 554.40 554.40 554.40

Fly ash 380.69 394.29 408.89 424.62

Na2SiO3/ NaOH 2.50 2.50 2.50 2.50

SiO2/Na2O 2.00 2.00 2.00 2.00

Sodium hydroxide solution 48.95 45.06 40.89 36.4

Sodium silicate solution 122.36 112.65 102.22 91

Super Plasticizer 5.70 5.91 6.13 6.37

Extra water 38.06 39.42 40.88 42.46

Table 4: Chloride Ion Penetrability Based on Charge Passed (ASTM C1202 –2007)

Charge Passed (Coulombs) Chloride Ion Penetrability

> 4000 High

2000 – 4000 Moderate

1000 – 2000 Low

100 – 1000 Very Low

< 100 Negligible

Table 5: Strength for different Grades of Geopolymer Concrete Mixes

S. No Grade Compressive Strength (N/mm

2) Split tensile strength (N/mm

2)

7 Days 28 days 7 Days 28 days

1 M 30 25.50 38.30 3.06 4.23

2 M 40 33.48 44.78 3.43 4.81

3 M 50 41.66 56.46 4.70 5.73

4 M 60 49.40 66.45 5.10 6.13

Table 6: RCPT Values for different Grades at Heat Curing

Sl. No. Grade

RCPT

Charges Passed at 30

minutes

Extrapolated values for 6 hrs

RCPT

(Coulombs)

Chloride Ion permeability

7 Days 28 days 7 Days 28 days 7 Days 28 days

1 M 30 488 164 5856 1968 HIGH LOW

2 M 40 457 142 5484 1704 HIGH LOW

3 M 50 438 136 5256 1632 HIGH LOW

4 M 60 413 129 4956 1548 HIGH LOW




Ph of Alkaline Solution:

The amount of chloride required for initiating corrosion

is partly dependent on the pH value of alkaline solution.

The pH value of Alkaline solution in the present study

was 13, which is less than the pH computed for the

activator solution (pH = 14). The value of pH more than

14 is due to the molarity of NaOH solution which was

more than unity as more than 40 grams of NaOH was

dissolved per litre of the solution. The results as seen in

the present case were also reported [Rajamane, et al.,

2011].

Compressive Strength and Split Tensile Strength of

Geopolymer Concrete:

The details of compressive strength and split tensile

strength for various grades of concrete are shown in

Table 5.

• As the grade of concrete increases, the compressive

and split tensile strength of geopolymer concrete also

increases, this is quite relevance with the behavior of

ordinary Portland cement concrete.

Rapid Chloride ion Penetration Test:

Rapid Chloride ion Penetration Test (RCPT) is

essentially a measure of concrete’s electrical

conductivity which depends on both pore structure

characteristics and pore solution chemistry (Caijun Shi,

2004). Chloride diffusion is one of the major reasons

for causing corrosion to steel reinforcement inside

concrete. Therefore it is necessary to study concrete for

its chloride ion permeability. Alumino-silicates is the

binder in the GPC, but in the conventional concretes

calcium silicate hydrate (C-S-H) gel is the main binding

system. The results of GPC specimens subjected to

RCPT are presented in Table 6.

During the RCPT the solution temperature was

increasing beyond 80°C hence the RCPT charges passed

at 30 minutes has been extrapolated to 6 hrs RCPT to

decide the chloride ion penetrability as mentioned in

Table 4. Since GPC make use of alkaline activator

solutions that are highly conductive, the unreacted

residual alkaline activator solutions could sometimes

give a false alarm, leading to conclude that GPC are

highly permeable to chlorides.

• As the grade of GPC increases the RCPT charges

decreases. Thus we can say that RCPT of GPC is having

‘low’ chloride ion penetrability for higher grade of

concrete.

• The maximum and minimum charges passed for 28

days was 1968 and 1548 Coulombs for M30 and M60

grade GPC respectively.

• The maximum and minimum charges passed for 7

days was 5856 and 4956 Coulombs for M30 and M60

grade GPC respectively.

Conclusions:

Based on the experimental investigations done the

following conclusions can be drawn:

• The fly ash can be used to produce geopolymeric

binder phase which can bind the aggregate systems

consisting of sand and coarse aggregate to form

geopolymer concrete (GPC).

• Conventional methods of mixing, compaction,

moulding and demoulding can be adopted for GPC’S

also. The only precaution needed is in handling of

catalytic liquids systems (CLS) which is highly alkaline

in ratio (often pH>13).

• As the GPC do not have any Portland cement, they

can be considered as less energy interactive. The GPC

utilize the industrial wastes such as fly ash for

producing the binding system in concrete. Therefore

these concretes should be considered as eco-friendly

materials.

• As the grade of concrete increases, the RCPT

charges decreases. Thus we can say that permeability is

very less in case of GPC. The results of 7 days allow

high charge to pass in concrete specimen due to

incomplete geopolymerisation reaction. Hence 28 days

results were considered for RCPT test in view of

DURABILITY aspects.

References:

[1] Caijun Shi, (2004) Effect of mixing proportions of

concrete on its electrical conductivity and the rapid

chloride permeability test (ASTM C1202 or

ASSHTO T277) results, Cement and Concrete

Research, 2004, 34, pp. 537-545.

[2] Davidovits, J., (2008) Geopolymer chemistry and

application, Institute Geopolymer, France, pp. 585.

[3] Davidovits, J. (1999) Chemistry of Geopolymeric

systems Terminology, 99 International Conference,

Saint-Quentin, France, 30 June-2 July.

[4] Fernandez-Jimenez, A., Palomo, A and Puertas, F.,

(1999) Alkali activated slag mortars, mechanical

strength behavior, Cement and Concrete Research,

29, pp. 1323-1329.

[5] Fenandez-Jimenez, A and Palomo, A (2003)

Characteristics of fly ashes, Potential reactivity as

alkaline cements, Fuel, pp. 2259-2265.

[6] Hua Xu, van Deventer, J.S.J., (2000) The

Geopolymerisation of Alumino-Silicate Minerals,

International Journal of Mineral Processing, 59(3),

pp. 247-266.

[7] Hardjito, D. Wallah, S.E., Sumajouw, DMJ and

Rangan, B.V., (2004) On the development of flyash

based geopolymer concrete, ACI Materials Journal,

101(52), pp. 467-472.

[8] Narasimhan, M. C. Patil, B. T, and Shankar H.

Sanni (1999) Performance of Concrete with Quarry

Dust as fine aggregate – An Experimental Study,




Civil Engineering and Construction Review,

September, pp. 19-24.

[9] Rangan, B. V., (2006) Studies on low-calcium fly

ash based Geopolymer concrete, Indian Concrete

Institute, Oct-Dec, pp. 9-17.

[10] Rangan, B.V., (2008) Mix design and production of

flyash based geopolymer concrete, The Indian

Concrete Journal, Vol. 82, No. 5, May, pp. 7-14.

[11] Rajamane, N.P, Nataraja, M.C, Lakshmana, N,

Dattatreya, J. K, (2011) Rapid chloride

permeability test on geopolymer and Portland

cement concretes, The Indian Concrete Journal,

Oct., pp. 21-26.

[12] Schmucker, M, and MacKenzine, KJD (2004)

Microstructure of sodium polysialate

siloxogeopolymer, Ceramic International, pp. 433-

437.

[13] Shetty, M.S., (2002) Concrete Technology, S.

Chand and Company, Fifth Revised Edition.

[14] Shankar H. Sanni and Khadirnaikar, R. B. (2012)

Effect of molarity on geopolymer concrete,

International Conference on Sustainability,

Challenges and Advances in Concrete Technology,

(SCAT 2012), PSG College of Technology,

Coimbatore, May 1-3, pp. 203-208.

[15] Shankar H. Sanni et al., (2012) Permeability

characteristics of geopolymer concrete, B.E Project

Report, Basaveshwar Engineering College,

Bagalkot.

[16] Whiting, D., Rapid determination of chloride

permeability of concrete, Report No. FHWA/RD-

81/119, Washington DC.

[17] IS: 2386 (Part-IV)-1963, Methods of test for

aggregates for concrete-mechanical properties,

Bureau of Indian standard, New Delhi.

[18] IS: 456-2000, Code of practice for plain and

reinforced concrete, Bureau of Indian Standard,

New Delhi.

[19] IS: 383-1970, Specification for coarse and fine

aggregates from natural sources for concrete,

Bureau of Indian standard, New Delhi.

[20] IS: 516-1959, Methods of test for strength of

concrete, Bureau of Indian standard, New Delhi.

[21] ASTM C 1202, Standard Test Method for Electrical

Indication of Concrete’s ability to resist chloride

ion penetration, American Society of Testing and

Materials, Pennsylvania, 2007.

i

News Item:

NIT Professor spoke at an International Conference in

SanFransisco, USA.

Forensic engineering is the engineering discipline that focuses on the practices of failure investigations and ensuring

that the failure causes and impacts are reported in a professional and unbiased manner in the judicial proceeding.

The purpose of the Technical Council for Forensic Engineers (TCFE) is to develop practices and procedures to

reduce the number of such failures, to disseminate information on failures and their causes, to provide guidelines for

conducting failure investigations, and to provide guidelines for ethical guidelines for ethical conduct in forensic

engineering.

Dr. C. Natarajan, Professor of Structural Engineering in the Department of Civil Engineering at the National

Institute of Technology, Tiruchirappalli participated the International Conference on Forensic Engineering in

SanFransisco, USA, from October 31st to November 3

rd 2012 and also attended the meeting for the Committee on

Forensic Investigation (CFI). This conference was sponsored and organized by the American Society of Civil

Engineers. The conference was very well attended with over 200 international experts and engineers met and

presented at the conference. The topics of the conference covered from failure case studies to advanced

investigation techniques. The Forensic Congress is organized by a panel of experts lead by Mr. Anthony Dolhon of

Exponent. Dr.C.Natarajan spoke about the Indo – US joint collaboration in advancing Forensic Engineering and

cited the high lights of the Indo-US Forensic workshop conducted at NIT – Tiruchirappalli in December 2010. He

also discussed some of the case studies of structural failure analysis he investigated.

Dr. C. Natarajan is a very well known structural engineer, widely consulted in Tri-states, namely Tamil Nadu,

Kerala and Pondicherry in the design and construction of high rise buildings, multistoried luxury Hotels, Industrial

Parks, long span Highway Bridges, Transmission Towers and Elevated water Tanks. He is also widely referred for

structural failures and buildings collapses, for forensic investigation and failure analysis. He now works on a

publication with Dr. A.Rajaraman (Former Director grade Scientist) of Structural Engineering Research Centre,

Chennai a book on 100 forensic case studies and the lessons learned”.

During his visit to the Forensic Conference, he interacted with International Engineers, Educators and Researchers

and tried to build collaborative bridges with them to organize an International Conference on “Emerging Innovative

Technologies in Civil and Infrastructure Engineering” on the eve of NIT – Trichy Golden Jubilee Celebrations in

2014. In addition to new technologies, design methodologies, new smart materials developed in Civil Engineering

field, health monitoring of existing infrastructures and forensic engineering will also be topics of the conference.

The mission of the Committee on Forensic Investigation (CFI) was to identify needs related to the investigation of

failures of constructed facilities; define the problems related to the methods and process of failure investigation and

to develop effective means of disseminating information on the methods and processes. CFI was previously headed

by Mr. Ron Anthony, a wood material expert. Mr. Anthony is now a member of the Executive Committee of TCFE.

Prof. Shen-en-Chen, Department of Civil Engineering, University of North Carolina at Charlotte, and Dr. Danielle

D. Kleinhans of the American Concrete Reinforcing Steel Institute is now responsible for CFI activities.

Mr. Gregory Di Loreto, the 2013 ASCE (American Society of Civil Engineers) president, was the opening speaker

at the conference. He has addressed the critical issues of financing the US national infrastructure, which is rapidly

deteriorating. Mr. Raymond “Paul” Giroux of the Kiewit Infrastructure West Company was the featured speaker

and gave an outstanding keynote speech on the history of the famous Golden Gate Bridge. His Power point

presentation showed an animated sequence of cranes in action during the bridge construction.

Mr. R. N. Reikar of Mumbai was announced at the sixth Forensic Congress as the recipient of the 2009 Forensic

Engineer of the Year, for his significant contributions to the forensic community. He was nominated by Professor

Ken Carper of Washington State University and the editor of the Journal of Performance of Constructed Facilities.

ii

Dr. Subhash C. Yaragal, is conferred with Prof. Satish Dhawan

Young Engineer State Award -2011

Dr. Subhash C. Yaragal, Faculty, Department of Civil Engineering, National Institute of Technology Karnataka,

Surathkal, is conferred with the prestigious Prof. Satish Dhawan Young Engineer State Award for the year 2011, for

his outstanding contributions in the field of Engineering Science. This award is instituted by Department of

Information Technology, Bio-Technology, and Science & Technology, Government of Karnataka. The award

carries a cash prize of Rs. 50,000, a memento and a citation.

INTERNATIONAL JOURNAL OF BIOLOGICAL SCIENCES AND ENGINEERING

Published by

CAFET-INNOVA Technical Society

The International Journal of Biological Sciences and Engineering (IJBE) ISSN 0976-

1519 is an international forum providing up to the minute information on all scientific

and clinical investigations together with their applications. The journal includes full

length and mini reviews, original articles, book reviews, announcements and scientific

reports to the editor. The journal hopes to attract the most important and highly

innovative papers from the current research from academia, research communities,

most importantly from young investigators and students. IJBE features scholarly

articles in rapidly moving Biosciences related research areas like Botany, Zoology,

Agriculture, Ethno botany, Microbiology, Genetics, Molecular Biology, Bio-Technology,

Bio-Chemistry, Bio-Chemical Engineering, Medical Electronics, Medical

Instrumentation, Bio-Informatics, Pharmacology, Pharma Chem, Pharma Bio,

Placentology, Physiology, Clinical Trials, Crop Research, Agricultural Ornithology and

Organismal sciences, which will be published subject to acceptability after referee and

editorial assessment.

Professors / Academicians / Scientists / Researchers / Engineers / Scholars / Students

can send their papers directly to: [email protected]

CONTACT:


D. Venkat Reddy (Editor-in-Chief)

Professor

NIT-Karnataka, Surathkal, INDIA

� +91-9739536078




� +91-9866587053


Raju. A (Executive Editor)

� +91-7411311091


INTERNATIONAL JOURNAL OF ELECTRONICS, INDUSTRIAL CONTROL & ROBOTICS

Published by

CAFET–INNOVA Technical Society

In the recent years, a number of fruitful research and development activities have been

pursued around the world exploiting advances in industrial electronics, control theory,

soft computing techniques etc. The main theme of the International Journal of

Electronics, Industrial Control & Robotics, ISSN 2231-4903 lies on the current trends

and impacts of these emerging technologies towards industrial automation and modern

societal needs. Some of the topics covered in this journal theme include areas of Control,

Automation, Robotics, Industrial Electronics, Signal Processing, Sensors & Actuators,

Computer Vision, etc.

Professors / Academicians / Scientists / Researchers / Engineers / Scholars / Students

can send their papers directly to: [email protected]

CONTACT:


Prof. Bidyadhar Subbudhi

(Editor-in-Chief)

Professor & Head

Department of Electrical Engineering

NIT-Rourkela, Odhisa, INDIA



� +91-9866587053


INTERNATIONAL JOURNAL OF ENGINEERING AND SCIENCE STUDENTS

Published by

CAFET–INNOVA Technical Society

The International Journal of Engineering and Science Students (IJES) aims at providing

platform to young researchers in the field of Electrical, Mechanical, Civil and Chemical

Engineering along with interdisciplinary branches like Agricultural Engineering,

Applied Engineering, Biological Engineering, Computer Science Engineering,

Electronics & Communication Engineering, Energy Engineering, Industrial

Engineering, Mechatronics, Nano-Engineering, Nuclear Engineering, and Petroleum

Engineering as well as Natural Sciences, Life Sciences and Applied Sciences.

It is one of the first journals to exclusively highlight the research contribution of

undergraduate and graduate students in the field of engineering and science. The

journal believes that the true research potential of students can be realized by

encouraging them to participate in research and acknowledging their achievements. The

journal publishes original research work conducted by students, which is peer reviewed.

However, the papers may not be of the technical rigor as expected in advanced scientific

journal.

The authors can submit their manuscripts directly to: [email protected]

CONTACT:


Hafeez Basha (Editor-in-Chief)

K L University, A.P., INDIA

� +91-9866587053



Raju. A (Executive Editor)


� +91-7411311091


Utility of Red Mud as an Embankment Material

Documents

Transcript of Utility of Red Mud as an Embankment Material