CIRDAP

The Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP) is a regional, inter-
governmental and autonomous institution. It was established in 1979 at the initiative of the countries
of the Asia-Pacific region and the Food and Agriculture Organisation (FAO) of the United Nations
with support from other UN organisations and donor countries/agencies. The member countries of
CIRDAP are Afghanistan, Bangladesh (host state), India, Indonesia, Iran, Lao PDR, Malaysia,
Myanmar, Nepal, Pakistan, Philippines, Sri Lanka, Thailand and Vietnam.

The main objectives of the Centre are to: (i) assist national action, (ii) promote regional cooperation,
and (iii) act as a servicing institution for its member countries for promotion of integrated rural
development through research, action research, training and information dissemination.
Amelioration of rural poverty in the Asia-Pacific region has been the prime concern of CIRDAP.
The programme priorities of CIRDAP are set under four Areas of Concern: 1) Agrarian
development; 2) Institutional/infrastructural development; 3) Resource development including
human resources; and 4) Employment. Within these areas of concern, the thematic areas identified
for the Third Six-Year Plan (2002-2007) are: Poverty alleviation through participatory approaches
with emphasis on social sector development (e.g. health, education and nutrition); Employment
generation through microcredit support, infrastructure development and local resource mobilisation;
GO-NGO collaboration; Gender issues; Governance issues; and Environmental concerns for
sustainable rural development.

Operating through designated Contact Ministries and Link Institutions in member countries,
CIRDAP promotes technical cooperation among nations of the region. It plays a supplementary and
reinforcing role in supporting and furthering the effectiveness of integrated rural development
programmes in the Asia-Pacific region.

EDITORIAL BOARD

Chairperson Durga P. Paudyal

Editor K.A.S. Dayananda

Members T. Abdullah, Development Consultant

Q.K. Ahmad, Bangladesh Unnayan Parishad
Momtaz Uddin Ahmed, Dhaka University
Mohammed Farashuddin, East West University
B. Sudhakar Rao, CIRDAP

Editorial Assistant S.M. Saifuddin

 ISSN 1018-5291

Asia-Pacific Journal of
Rural Development
VOLUME XVII JULY 2007 NUMBER 1

CENTRE ON INTEGRATED RURAL DEVELOPMENT FOR

ASIA AND THE PACIFIC
ASIA-PACIFIC JOURNAL OF RURAL DEVELOPMENT
Volume XVII, July 2007, Number 1
ISSN 1018-5291
© CIRDAP 2007

The views expressed in this publication are those of the

author(s) and do not necessarily reflect those of CIRDAP

Published twice a year

Price including postage

Annual Subscription (Two Issues):
Tk.300 for Bangladesh
US$20 for Asia-Pacific Countries
US$35 for Other Countries

Advertisements are invited from professional bodies/institutions

The following tariff charges will apply

a) Inner back cover: US$200

b) Full page text: US$100
c) Half page text: US$50

Published by Durga P. Paudyal, Director General, on behalf of

Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP)
Chameli House, 17 Topkhana Road, GPO Box 2883
Dhaka-1000, Bangladesh

Printed by: Nizam Printers & Packages, Dhaka

 Asia-Pacific Journal of Rural Development
Volume XVII, July 2007, No. 1

CONTENTS

Articles Page
An Overview of Rural Development in Asia 1
Durga P. Paudyal

Factors Influencing Mobile Services Adoption in Rural India 17

Abhay Jain and B.S. Hundal

Adoption and Diffusion of Integrated Pest Management Technology: 29

A Case of Irrigated Rice Farm in Jogjakarta Province, Indonesia
Joko Mariyono

Exploring Predictive Strengths of Stochastic Pavement Deterioration 39

Models: A Case Study of Thailand’s Highway Network
Satirasetthavee Dussadee and Herabat Pannapa

Socio-Economic Study of Gender Role in Farm Production in 57

Nasarawa State of Nigeria
S.A. Rahman, H. Ibrahim and H. Ibrahim

Economic Evaluation of Rice-Prawn Gher Farming System on Soil 67

Fertility for Modern Variety of Rice Production in Bangladesh
Basanta Kumar Barmon, Takumi Kondo and Fumio Osanami

Soil Hydraulic Characterisation under Different Cropping Patterns in 83

Sloping Agricultural Lands in Sri Lanka
A.A. Rivas, R.S. Clemente, S.L. Ranamukhaarachchi, A. Das Gupta,
M.A. Zoebisch, S. Thevachandran and M.S.D.L. De Silva

Practitioners’ Papers
Good Governance and HRD: Case Studies of User Managed Safe 95
Drinking Water and Health Projects in India
P. Durgaprasad and P. Sivaram

Land-use Suitability Evaluation Criteria for Precision Agriculture Adoption 113

in a Moderately Yielding Soya bean Cropping Area in Thailand
Kishore C. Swain and H.P.W. Jayasuriya

Index 126
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

An Overview of Rural Development in Asia

Durga P. Paudyal*

Abstract
The rural economy, including agriculture, has remained the mainstay in the countries of
Asia-Pacific region, in terms of both contribution to the national GDP and labour
absorption. Over the past few decades, there has been spectacular economic progress in
several CMCs. However, despite being a priority agenda of the national governments as
well as the main destination of international aid donors over the past few decades, the
achievements of rural development and poverty alleviation programmes have been lop-
sided, and meaningful headway are yet to make in several countries. In this background,
this paper attempts to examine the present state of rural development in CMCs of Asia-
Pacific region and suggests some new roles of national, regional and international
agencies for promoting sustainable rural development and achieving MDGs.

1.0 Introduction
Most countries of Asia-Pacific region are predominantly rural and agrarian, characterised
by unequal distribution of productive resources, high incidence of poverty, widespread
unemployment and under-employment, household and regional income inequality.
Hence, rural development encompassing growth and development of agriculture and non-
agriculture sector continues to be the priority agenda of the national development
strategies in these countries. In Asian countries, the concept of rural development
encompasses economic and social development, and aims at widening the opportunities
for gainful employment to the poorer section of the society. This is rightly so, because
food security, lack of employment and income-generating opportunities are among the
most important causes of poverty in the region.
2.0 IRD Concept
The concept of rural development (RD) went through a long process of evolution since
1950s, in tandem with the significant changes of development strategies nationally and
internationally. Table 1 shows the trend of major development paradigms and poverty
reduction-strategies since 1950s. It is seen that in 1970s integrated rural development,
popularly known as IRD, was launched to address the failure of a single sectoral
approach of 1960s. The main thrust of IRD approach was to develop an effective delivery
mechanism through integration of different rural development related sectors.
Several Asian countries initiated rural development programmes of various kinds and
adopted various approaches in planning and implementation. South Asian countries like
Bangladesh, India, Nepal and Pakistan initiated RD-related programmes and activities in
the name of IRD, while South-East Asian nations labelled it under different names such
as Integrated Agriculture Development Programme (IADP) in Malaysia, Integrated Area

*
Director General of CIRDAP, Dhaka, Bangladesh. E-mail: dg@cirdap.org

1
Development (IAD) in the Philippines and IAD in Thailand. IRD in Vietnam had the
main thrust on agricultural production. In Indonesia RD programmes aimed at integration
with national development goals along with improvement in the living standard and the
quality of life of the rural population.
During the initial period, most countries opted for a strategy of self-reliant growth by
mobilising resources for basic physical and social infrastructure, through planned
development approach. In their endeavours for self-reliance and for elimination of
poverty, the Asian countries accorded high priority to agriculture development through
seed-fertiliser-irrigation technology. At that time the active role of the national
governments was to develop physical and institutional infrastructure, and to evolve and
promote land augmenting or yield increasing technologies. These contributed to
strengthening the supply-side factors in rural areas, especially those largely beyond the
scope of the individual farmers and the private sector. By the early 1970s, it was realised
that increase in production was accompanied by growing income inequalities. The
production gains were in the ‘favourable areas’ and to ‘progressive farmers’. The
emphasis on output expansion accompanied by effective measures for reducing income
inequity emerged as a major policy issue. The ‘package approach’ clearly did not result
in a broad-based rural development nor did attack the growing problem of rural poverty.
Hence, new approaches like ‘target groups’, ‘basic minimum needs’, etc. came into
prominence later.
On the whole, the IRD concept could not gain the expected momentum, mainly due to
the poor performance by the national governments and IRD institutions. In the
contemporary administrative set-up with traditional values and norms and
compartmentalised sectoral working modalities, there is no wonder that the IRD model,
with multi-disciplinary approach and multi-sectoral operation, could not be successful to
improve the quality of life in the rural areas.
During 1980s, ‘self-reliance’ and ‘participatory endogenous development’ dominated RD
efforts, and during the 1990s, poverty reduction and broad based economic growth with
higher budgetary support for basic services and social sectors have been emphasised, in
which human capital formation, improvement of quality of life through income and
employment and market-based approaches have dominated the policy. In most countries,
labour intensive growth and improved social services were considered to offer a powerful
and viable route to poverty reduction.
The table also illustrates a constant shift of poverty alleviation strategies from supply side
to demand-driven interventions and the innovations attempted to create wealth, social
capital and markets for those at the bottom of the pyramid. Such strategies have acquired
a new sense of urgency to derive policy and programme support from the beneficiaries.
For example co-operative development in 60s and IRD programmes in 70s focussed
mainly on the supply side while the poverty alleviation approach in 80s, social
mobilisation and microcredit in 90s and the Millennium Development Goals (MDGs)
since 2000 have emphasised the activation and institutionalisation of the demand side to
make sustainable impact on the people themselves. This transition makes CIRDAP
Member Countries (CMCs) to review their own policies and institutional infrastructure at
national as well as regional levels to suit the changing circumstances.

2
Table 1: Development Paradigms and Poverty-Reduction Strategies

Period Dominant development paradigm Poverty-reduction strategies

1950s Growth through industrialisation Community development

1960s Agricultural intensification,
human capital development
1970s Redistribution with growth
1980s Structural adjustments, private
sector led development
1990s Human development, broad-
based growth
2000s Market-led-globalisation, broad
based growth, millennium
development goal
Decentralisation, Co-operatives, Rural
Development
Basic needs, integrated rural development
Growth, human resources development, safety
nets, NGO involvement
Microfinance, NGO/INGO, civil society, private
sector, targeted programmes and safety nets.
Microfinance, NGO/INGO, civil society, private
sector, MDGs

The past development experiences in the Asian countries suggest the need to go beyond
the economic perspective, to take explicit account of the socio-cultural concerns and to
focus more on poverty at the grassroots level and to ensure meeting their demands.
Moreover, the continuing high incidence of rural poverty in the countries reflects the
need to further intensify the efforts at poverty reduction in the years to come. The
prevention of environmental degradation and achievement of other social objectives are
also closely linked to success in reducing poverty. This calls for further emphasis on
increasing agricultural growth, encouraging non-farm activities and accelerating the
diversification and growth of the economy particularly in the rural areas. It is also
important that investment in human capital should increase significantly in order to
improve the skills of the labour force and enable them to work with the new generation
of technologies. This would require institution building, reallocating public expenditures
to priority sectors and ensuring involvement of non-government organisations (NGOs)
and private sector in rural development and poverty alleviation.
2.1 Development experience of Asia-Pacific region
The Asian and Pacific region attained unprecedented economic growth over the past
decades, which is also expected to continue in the coming years, despite some setback in
the mid-1990s. There have been some interesting patterns of development across the
countries, territories and communities across the region. Firstly, it is observed that some
countries in the region have made spectacular progress, while others lagged behind. The
economic growth rates and per capita income, which are often counted as determining
factors for measuring stages of development, have been faster in some countries than
ever before. Despite the social and political conflicts in several countries causing
economic declines and extensive damage to infrastructure and productive capacity, the
growth of the economies of several Asian countries, except Afghanistan, Nepal etc., have
been impressive. Among the low income countries, India and Vietnam achieved
considerable GDP growth rates of 6 per cent or more during the first half of the 2000s.

3
Bangladesh and Sri Lanka recorded a growth of above 5 per cent during the same period
(Table 2).
Moreover, the information provided in Table 2 indicates that the annual variation in
growth rates in these countries has been much lower, and the trend is upward. The Indian
economy, which faced severe balance-of-payments crisis during 1990s, recovered its
growth momentum after the economic reforms initiated since 1991. The GDP growth
accelerated to over 8 per cent in 2005. Economic growth in Pakistan averaged 6.1 per
cent per year during 1981-1990 which, however, accelerated to 5 per cent in 2003. In Sri
Lanka, the growth rate increased from an average of 4.3 per cent in 1981-1990 to 5.5 per
cent in 2003 again revived to the previous position in the early period of 2005. As a
result of this varied growth, there has been huge divergence of income levels and other
important indicators of socio-economic progress. Besides diversity and uniqueness at the
national levels among countries, the region has experienced remarkable diversity of
economic growth and social progress also within the countries. For example, the per
capita GDP is US$4,187 in Malaysia during 2003 compared to US$237 in Nepal (Table
4). Between these two extremes, wide differences exist among the remaining CMCs. In
terms of GDP per capita criterion adopted by the World Bank for classifying countries,
eight CMCs – Afghanistan, Bangladesh, India, Lao PDR, Myanmar, Nepal, Pakistan and
Vietnam – are classified as low-income countries having per capita GNI of US$825 or
less; five countries – Indonesia, Iran, Philippines, Sri Lanka and Thailand belong to the
lower-middle income category (income range US$826-3255) while Malaysia is included
in the upper-middle income category (income range US$3256-10065).
However, it must be noted that macroeconomic data such as GDP growth rate and per
capita are somewhat deceptive and hiding many social malaises in CMCs. In the recent
years, as shown in Table 3, that the remittance of the migrant workers has been a new
source of capital in the form of hard currency. Indeed, in some countries such as Nepal
and Sri Lanka, the remittance provided a cushion to prevent economic breakdown during
the difficult time of internal conflict. However, due to weak economic structures of the
countries such as Sri Lanka, Philippines, Pakistan, Bangladesh etc. the excessive
dependence on foreign remittances from expatriate workers rather than on FDI make
them vulnerable at macro- and micro-level. The subsistence agriculture based household
economy has been too dependent on remittance when the able household members are
migrant workers. Similarly, one cannot undermine the psychological trauma of
their dependents back home in the difficult geo-political situations, such as in Iraq,
Lebanon etc.
The social cost of such remittances is another important hidden malaise. Nearly 800,000
Sri Lankan workers are working in the Middle-East and about 500,000 of them are house
maids. Although they earn around US$150 per month and the country’s (population 20
million) economy is heavily dependent on their remittances, the country has to bear a
heavy social cost due to the lack of job security and welfare package at home and in the
workplace. There is rampant family breakdown, alcoholism, neglect and abuse of
children. Needless to say that almost all the workers come from the lowest strata of the
society.

4
Table 2: Recent Economic Performance: Annual GDP Growth Rate at Constant Prices
Percentage per annum

Country 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Per capita GNI
(US$) 2005
Low-income
Afghanistan … … … … … … … … … … … 190
Bangladesh 4.9 4.6 5.4 5.2 4.9 5.9 5.3 4.4 5.3 6.3 5.4 440
India 7.3 7.8 4.8 6.5 6.1 4.4 5.6 4.3 6.8 6.9 8.5 620
Lao PDR 7.1 6.9 6.9 3.0 6.1 8.3 5.8 5.8 6.1 … … 390
Myanmar … 6.4 5.7 5.8 6.1 4.7 … … … … … 220
Nepal 5.7 5.0 3.3 4.5 4.2 3.9 -0.5 3.5 5.0 … … 250
Pakistan 5.1 6.6 1.7 3.5 3.3 4.5 2.2 3.0 5.0 6.4 7.8 600
Vietnam 9.5 9.3 8.2 5.8 4.8 6.8 6.9 6.2 6.8 … … 540
Middle-income
Indonesia 8.1 7.7 4.5 -13.1 1.7 5.1 3.4 3.7 4.1 5.1 5.9 1140
Iran 3.7 6.6 3.4 3.6 8.3 5.8 5.4 6.7 6.2 5.6 5.9 …
Philippines 4.7 5.9 5.2 -.0.6 4.3 6.0 3.2 … … 6.1 5.1 1170
Sri Lanka 5.5 3.8 6.3 4.7 4.5 4.7 -1.5 4.0 5.5 5.4 5.3 1010
Thailand 9.2 5.9 -1.4 -10.5 4.5 4.7 1.9 5.3 6.4 6.2 4.5 2490
Upper-middle
income
Malaysia 9.8 10.0 7.3 -7.4 10.9 6.2 0.5 4.1 4.6 7.1 5.0 4520
… data not available.
Source: The World Bank, World Development Indicators 2004, 2006; Asian Productivity Organisation, Asia-Pacific Productivity Data and Analysis, 2003; Asian
Development Bank, Asian Development Outlook, 2002.

5
Table 3: Movement of People in the CMCs

Country Net migration Workers’ remittance received

(thousands) ($ millions)
1990-95 1995-2000 % change 1990 2004 % change
Afghanistan 3313 -397 -111 … … …
Bangladesh -260 -300 15 779 3584 360
India -1407 -1400 -0.5 2384 21727 811
Indonesia -725 -900 24 166 1866 1024
Iran -1512 -456 -70 1200 1032 -15
Lao PDR -10 -7 -30 11 1 -91
Malaysia 230 390 70 325 987 204
Myanmar -126 60 -147 6 118 1867
Nepal -101 -99 -2 0 823 …
Pakistan -2611 -41 -98 2006 3945 97
Philippines -900 -900 0 1465 11634 694
Sri Lanka -182 -160 -12 401 1590 296
Thailand -88 -88 0 973 1622 67
Vietnam -270 -200 -26 … 3200 …
… data not available.
Note: Net migration refers to the annual number of immigrants less the annual number of emigrants.
Source: The World Bank, World Development Indicators 2006.

6
Table 4: Trends in per Capita Income and Human Development

Country GDP per capita (1987 $) Human Development Index (HDI) value
1980 1990 1995 2003 Changes 1980 1992 1995 2000 2003 Changes
over over
1980 (%) 1980 (%)
Bangladesh 144 179 202 376 161.0 0.234 0.309 0.371 0.506 0.520 122.22
India 262 374 425 564 61.6 0.296 0.382 0.451 0.577 0.602 103.38
Indonesia 349 537 720 970 177.9 0.418 0.586 0.679 0.680 0.697 66.75
Iran … … 2066 … … 0.570 0.650 0.694 0.721 0.736 29.12
Lao PDR … 310 363 375 20.9* … … 0.465 0.522 0.545 …
Malaysia 1688 2301 3108 4187 148.0 0.687 0.794 0.834 0.790 0.796 15.87
Myanmar … … … … … 0.356 0.406 0.481 … 0.578 62.36
Nepal 148 182 206 237 60.1 0.209 0.289 0.351 0.499 0.526 151.67
Pakistan 259 350 381 555 114.2 0.287 0.393 0.453 … 0.527 83.62
Philippines 679 628 630 989 45.7 0.557 0.621 0.677 … … …
Sri Lanka 328 438 512 948 189.0 0.552 0.665 0.716 … 0.751 36.05
Thailand 718 1291 1843 2305 221.0 0.551 0.798 0.838 … 0.779 41.37
Vietnam … 610 816 482 -21.0 … … 0.560 0.695 0.704 …
… data not available; * with respect to 1990
Source: UNDP, Human Development Report 1998, 2005.

7
Table 5: Inequality in CIRDAP Member Countries

Country Real GDP per capita Adult literacy rate Life expectancy at birth Access to safe water Access to hygienic
(PPP$) ( %) ( %) sanitation (%)
Rural Urban National Rural Urban National Rural Urban National Rural Urban National Rural Urban National
Bangladesh 1,133 2,444 1,382 36.7 44.1 38.1 56.1 59.0 56.9 96.0 99.0 97.0 44.0 79.0 48.0
India 1,047 2,435 1,422 45.0 70.9 52.0 59.0 68.5 61.6 79.0 85.0 81.0 14.0 70.0 29.0
Indonesia 3,073 5,568 3,971 78.8 92.7 83.8 60.0 71.2 64.0 54.0 79.0 62.0 40.0 73.0 51.0
Lao PDR 1,766 5,600 2,571 50.6 79.1 56.6 49.5 62.4 52.2 41.0 60.0 44.0 16.0 98.0 28.0
Malaysia 6,808 11,926 9,572 75.0 90.7 83.5 68.2 74.1 71.4 56.0 96.0 78.0 88.6 98.6 94.0
Myanmar 1,014 1,478 1,130 81.4 88.2 83.1 58.5 59.9 58.9 50.0 78.0 60.0 36.0 56.0 43.0
Nepal 1,004 2,287 1,145 25.6 42.9 27.5 54.0 71.3 55.9 60.0 88.0 63.0 12.0 58.0 18.0
Pakistan 1,767 3,030 2,209 28.0 56.0 37.8 60.1 67.8 62.8 69.0 82.0 74.0 22.0 77.0 47.0
Philippines 1,678 3,649 2,762 92.5 96.3 94.6 64.6 69.7 67.4 80.0 92.0 84.0 63.0 89.0 75.0
Sri Lanka 2,793 5,590 3,408 88.1 97.6 90.2 70.0 81.4 72.5 52.0 88.0 57.0 62.0 68.0 63.0
Thailand 2,959 15,286 7,742 90.8 98.6 93.8 66.9 73.6 69.5 88.0 94.0 89.0 95.0 98.0 96.0
Vietnam 1,055 1,917 1,236 92.5 98.2 93.7 63.8 76.2 66.4 30.0 70.0 43.0 17.0 47.0 24.0
Note: Data for this table were used from the Human Development Report 1995 and manipulated in some cases where segregated data for rural and urban areas
were not available.
Source: CIRDAP, Rural Development Report 1999.

8
Secondly, the present state of development in the countries of Asia-Pacific region reflects
the uneven socio-economic impact across countries as well as between the rural and
urban areas. Impressive gains have been achieved in many countries, but these have not
been shared equitably by different socio-economic groups or by geographic regions.
Table 5 shows that the average per capita rural income of Bangladesh, India, Lao PDR,
Nepal, Philippines, Sri Lanka and Thailand, is less than half of urban income. The
literacy rate in rural areas is only half of that in urban areas in Pakistan. Similarly, glaring
disparities exist in terms of access to basic facilities between rural and urban areas in
most countries. The figures suggest that, even with similar levels of national
achievements, the rural urban disparities that exist in the CMCs differ both in nature and
magnitude.
Such diversity in the development experience, both among and within countries, reflects
that the poor and the vulnerable groups particularly in the rural areas have been receiving
disproportionately smaller shares of the benefits of development while often bearing the
major burden of the associated costs. On institutional part, area specific problems and
potentials required decentralised framework with a participatory mode, while
institutional strengthening and empowerment and participation of the rural people
continue to remain formidable challenges towards achieving equitable and sustained rural
development. Despite several models of decentralisation practised in Asian countries,
effective policy with local capacity building is yet to emerge. A concerted public policy
action through both macrosupport and microintervention is required to tackle numerous
structural and institutional barriers impeding rural development. This involves the
crucially important task of mainstreaming rural development and poverty alleviation
efforts into the overall national development policies. This is evident in higher incidence
of poverty in the rural areas and the growing rate of rural-urban migration to seek
economic opportunities.
Thirdly, on the demographic front, CMCs accounted about 2.02 billion population in
2003, about 32 per cent of the world total, which is projected to increase to 2.53 billion
by 2020. Due to the rapid urbanisation process, the urban population in the CMCs is
projected to increase from a total of 0.8 billion in 2003 to around 1.13 billion in 2020,
representing significant increase of the urban population. It means that nearly 80 per cent
of the estimated increase of the total population in the CMCs during 2003-2020 will have
to be absorbed mostly in the urban areas. Such a significant increase in the level of
urbanisation is likely to be accounted for by high rates of rural-urban migration and
reclassification of selected rural areas as urban in addition to natural increase in urban
population. As a result the share of rural population may decline from 58 per cent in 2003
to 55 per cent by 2020. There is also an interesting trend in working-age-population.
Current estimates show that the youth population (>18) is going to rise to around 40 per
cent by 2005, meaning job creation to be on the top of the agenda in non-farm/service
activities, as the educated youths generally avoid farm-based income/employment and
other traditional activities. With adequate emphasis on investment in health, education,
skill up-gradation and other areas to develop the human capital, an accelerated pace of
rural development can be generated and sustained. The demographic transition can thus
be used to accelerate the pace of economic development through creating an enabling

9
economic and policy environment to ensure functioning labour markets and expanding
productive employment opportunities.
Finally, there are increasing evidences across the countries that ethnic and indigenous
communities and tribal groups are marginalised in the development process. The Human
Development Report 2004 of UNDP has emphasised on this issue with hard facts and
glaring examples. Similarly, the issues and concerns of women, children and elderly
citizens are well documented but not adequately addressed. Hence, there is a need to go
beyond the economic perspective, to take explicit account of the socio-cultural concerns
and focus more on poverty at the grassroots level. It is seen in a number of countries that
unequal distribution of resources, opportunities and participation across communities and
geographic regions leads to social tension and conflict, which is a threat to national and
international security and prosperity.
2.2 International level
The World Conference on Agrarian Reform and Rural Development (WCARRD) 1979
adopted a broad-based strategy for national progress based on growth with equity and
participation. It emphasised the redistribution of economic and political power, fuller
integration of rural areas into national development efforts, expanded opportunities of
employment and income for rural people and development of farmers’ associations, co-
operatives and other forms of voluntary, autonomous, democratic organisations of
primary producers and rural workers. In order to implement the broad principle, it laid
down several plan of action in the areas of access to land, water and other natural
resources; people’s participation; integration of women in rural development; access to
inputs, markets and services; development of non-farm rural activities; education,
training and extension and a number of international policies for agrarian reform and
rural development. This comprehensive package was aimed at achieving sustainable and
pro-poor rural development, especially through ensuring access to land and other natural
resources as the later provide the mainstay of the majority of the rural poor.
As a part of the strategy for implementation of WACARRD, the role of the regional level
inter-government organisation was envisioned, and accordingly, CIRDAP was
established under the aegis of FAO in 1979. The objectives of the Centre were laid down
as to assist national action, promote regional cooperation and act as a servicing institution
for its member countries for rural development through research, action research, training
and documentation and information dissemination. Presently, it has 14 member countries
namely, Afghanistan, Bangladesh, India, Indonesia, Iran, Lao PDR, Malaysia, Myanmar,
Nepal, Pakistan, Philippines, Sri Lanka, Thailand, and Vietnam.
As the second milestone of the WCARRD, the International Conference of Agrarian
Reform and Rural Development (ICARRD) was jointly organised by FAO and the
Government of Brazil with the slogan A Vision for the Future from 7 to 10 March 2006
in Porto Alegre, Brazil. FAO invited CIRDAP to present an overview paper on the
current trends and emerging issues in rural development in the Asia-Pacific region.
CIRDAP overview paper briefly reviewed the current trend of socio-economic
development of Asia-Pacific region. The paper identified emerging issues of agrarian
reform and rural development and suggested a closer linkage in order to benefit from the

10
interdependent nature of these two sectors for sustainable development. The full paper
can be downloaded from www.cirdap.org.sg.
The Conference extended unanimous support to the view that agrarian reform (AR) and
rural development (RD) are crucial and inter-dependent for meeting the World Food
Summit targets and the MDGs, relating to poverty and hunger. The members emphasised
that great disparities continue to exist in ownership and access to land and other natural
resources and that the concentration of land ownership is increasing. The members
recognised that achieving successful agrarian reforms and rural development is very
challenging, with some members reporting success and many identifying the difficulties.
This reflects the great diversity of circumstances and experiences in designing and
implementing agrarian reforms and rural development policies. It was unanimously
agreed that the agrarian reforms and rural development agenda have a great deal yet to
achieve and that new visions and political will are crucial to deliver secure and equitable
access to land and other natural resources and ensure sustainable rural development.
The Conference appreciated the role of regional organisations that were originally
established to support follow-up to WCARRD (e.g. CIRDAP, CIRD-Africa and others)
and that lessons be drawn to contribute to realistic and appropriate follow-up for
ICARRD. Members recommended that technical support should be provided to
strengthen these regional organisations, in order to enable them to generate
comprehensive knowledge at both national and regional levels, to facilitate
interministerial and intersectoral coordination on rural issues at the national level, and
facilitate policy dialogue, cooperation, exchange of good practices, and monitoring at the
regional level.
3.0 Future Challenges
As discussed above, economic development does not necessarily ensure human
development, equitable distribution of resources and sustainable natural resources
management, among others. Hence, in order to achieve broad based economic growth
and development, there has been a strong suggestion to take area-specific comparative
advantage through strong local governance structure and process, which can be
instrumental for mobilising local resources, ensuring people’s participation in the rural
development activities and enabling several institutions such as NGOs, microfinance
institutions and private sector to function at the local level. This may be one way of, what
Amartya Sen advocates, maintaining a balance between economic development and
democratisation. Moreover, in the changing context of market-led globalisation, several
South-East Asian countries have taken rural development for product development and
linkage with market for rural products. Hence, in order to make meaningful progress in
rural development and poverty alleviation, the following forward looking measures are
suggested:
3.1 Decentralised development
The decentralisation policy is advocated to deal with the heterogeneity in geographic
location, climate and area specific comparative advantage; presence of various ethnic
communities with their own language, culture and skills; and variation of the level in

11
development across territories. Traditionally, designing and implementing rural
development programmes have been carried out at the national level, deriving from the
national level periodic plans prepared by the central level planning bodies. However,
exclusive reliance on centralised planning for rural development is criticised on three
grounds viz. efficiency, equity and participation. On the efficiency ground it is said that
identification of local problems, location of projects and mobilisation of local resources
cannot be done at the central level. On the equity aspect, the decision-makers of the
central level most likely represent the interest of the dominant elite classes than the rural
poor. Finally, on the participation issue, the identification and mobilisation of local
resources and their deployment in accordance with the popular need require direct
participation of the beneficiaries. Such participation will also ensure ownership of the
project among the beneficiaries.
Therefore, involving local governing bodies in rural development process through
decentralisation is widely advocated. Several policy packages of decentralisation
measures are applied with appropriate combination, such as delegation of authority from
higher offices to their field office; deconcentration of national-level offices at provincial,
district and sub-district levels and authorities delegated to function as the executive
branches; devolution of decision-making authorities by law to lower level representative
bodies such as provincial and district councils; and privatisation of public sector
industries, product development, marketing etc. where government cannot perform
efficiently. Recently, several types of NGOs and civil societies have emerged to
complement the government especially in the areas of social mobilisation, microfinance,
awareness building and advocacy.
Another important area of decentralised development is building local capacity/
capability. Because in many developing countries, the policymakers tend to assume that
policy implementation is a simple and routine administrative and technical process.
Accordingly, there is usually a tendency to give inadequate attention to implementation
aspects while designing policies and programmes. The following are some of the gaps
that undermine the local capacity/capability in management of rural development
process:
ƒ Authority gap: in many cases the lower-level offices are given authority to
‘propose projects’ but important decisions such as allocation of resources or
deployment of manpower are retained at the higher level. Hence, local-level
offices do not take any initiative at their own risk.
ƒ Manpower gap: rural institutions are mostly understaffed with inadequate trained
manpower and insufficient incentives to retain them in the rural areas.
ƒ Resources gap: unless the political leadership is committed to rural development,
it does not attract attention of the policymakers for greater allocation of resources.
Similarly, local resources can be mobilised mainly in kind i.e. local materials and
labour contribution. Hence, there is always a resource gap for rural development
programme.
ƒ Management capacity: in the dispersed and remote geographic setting
coordination among various actors and stakeholders, sequencing of activities and

12
 time management for rural development are always a big challenge, which is in
sharp contrast with the low level of management capability of local level
institutions.
3.2 Larger projects/investments for poverty alleviation
Rural development and poverty alleviation cannot just be seen as a welfare programme.
If the rural poverty is to be addressed effectively, larger investment would be required to
create economic activities and employment opportunities. Malaysia is a good example
where rural development was accompanied by massive investment in land development,
infrastructure development, plantation and rural industrialisation. Similar analogy can be
drawn in the South Asian region. The following map shows that in South Asia, in spite of
the endowment of fertile soil, water resources and long history of civilisation, the
majority of the poor people are disproportionately concentrated in the Himalayan region
and Ganga-Brahmaputra basin: Nepal, Bhutan, north-east India and Bangladesh.
Map: Concentration of Poor People in South Asia

In order to generate economic activities, employment opportunities and income, thereby

addressing poverty in a meaningful way, development of the Himalayan water resources
holds a great promise for the entire region. There are possibilities for generating vast
amount of renewable and clean hydro-energy along with controlled water for irrigation
and drinking water from the Himalayan rivers. There is enough demand for the products
of the water sector and several projects are available with some degrees of feasibility
studies with promising results. Hence, large investments for mega size multi-purpose
water resources projects might hold the key for generating multiplier effect in the
economy of the entire region.
Similarly, several countries in the Middle-East and Africa have rich natural resources,
such as oil, natural gas, gold, diamond etc., but very poor record in human development,
food security and natural resource management. Hence, these countries can allocate
sizable resources for area specific development programme, in which rural people can

13
take economic advantage. Such priority investment would be the key for sustainable rural
development and poverty alleviation.
3.3 Knowledge management for RD policy development and implementation
The RD and PA programmes are implemented by several agencies of government,
donors, INGOs, NGOs, civil societies and private sectors. Hence, besides the
coordination gap among implementing agencies, there is a real problem of dis-
aggregating macrolevel policy into several sectoral programmes with microlevel action
plans and, again, aggregating the achievements and issues at macrolevel.

Diagram: Schematic Diagram of RD Policy Research and Development

Cabinet

Sectoral
Ministries Ministry of
Local
Government

Deconcen-
Policy Research Provincial
trated Field
and Council
Officers
Training Institutions

Field District
Extension Council
Agents

Private Microfinancial NGOs Civil

Sector Institutions Societies

14
Similarly, in order to synergise the services of NGOs, civil societies, microfinance
institutions and private sectors at the local level, there is a need for a facilitating,
coordinating and regulating mechanism at the local level. With such mechanism these
institutions can make good contribution to social mobilisation, self-employment
generation, product development and marketing of products. Therefore a research and
training institution would play a vital role in developing roles of different institutions,
monitoring the implementation process at all level, training of field-level functionaries,
and arranging periodic review meetings involving all ministries and field level
functionaries. Such arrangements may fill the knowledge gap, minimise the coordination
gap and enhance the implementation capability of the institutions, as demonstrated in the
diagram.
3.4 Role of international bodies and donors to address global issues
In the new context of globalisation, knowledge, ideas and know-how are the potential
sources of economic growth and development, along with the application of new
technologies. The global knowledge network is transforming the demands and the ways
one thinks and works. It calls for integration, in a different connotation, to integrate the
content, context, and infrastructure in order to harness the potentials of the countries, to
instil important value for success, to find a viable mechanism to raise capacity and the
productivity of people to cope with and better address the diverse demands and needs of
the countries. It requires the capabilities to address markets from local to global,
changing the technological environment, adding value to information, product and
services offered by the organisation while opening wide access of information to a very
large audience throughout the world. In order to make the globalisation useful to all, the
international bodies/donors have a vital role to play. However, some of the recent
international reports show disturbing findings.
Firstly, the FAO report on The State of Food Insecurity in the World 2006 found that the
World Food Summit 1996 had set a target to halve the hunger by 2015, on which no
progress has been achieved 10 years down the line. Compared with 1990-92, the number
of undernourished people in the developing countries has declined by a meagre 3 million
in 2005 – a number within the bounds of the statistical error.
Secondly, the Global Employment Trend for Youth 2006 of ILO found that the number of
unemployed youth (15-24 years of age) has risen sharply over the past decades, leaving
about one third of the world’s young people without jobs. This shows the inability of
economic growth to create enough productive employment opportunities for the world’s
young population.
Thirdly, the UK government-commissioned report on The Economics of Climate Change
found the looming environmental catastrophe within a foreseeable future. The author, the
former World Bank chief economist Sir Nicholas Stern said that the world must be
prepared to pay now to prevent economic fallout in the near future which could be on the
scale of the Great Depression of the 1930s.
Finally, a recent study on the Impact of Biodiversity Loss on Ocean Ecosystem Services,
found that if the global trend for over-fishing and pollution continues, the fish and sea-

15
food could disappear by 2048, which will threaten human food supplies and the
environment.
These hard facts illustrate the failures of the market-led globalisation framework to
generate sustainable results at several fronts such as food security, employment
generation, human development and sustainable natural resource management. In this
context, the role of the international organisations needs to be redefined to focus more on
mitigating the adverse effects of the market-led globalisation and help those who are left
behind by the market forces. Similarly, international bodies can play an effective role in
sustainable management of common natural resources.

References
FAO. 2006. The State of Food Insecurity in the World 2006. Rome: FAO.
ILO. 2006. Global Employment Trends for Youths. Geneva: ILO.
Paudyal D.P.2006. From the Peasant Charter to the ICARRD: An Overview of the Current Trends and
Emerging Issues in Rural Development in the Asia-Pacific Region. Asia-Pacific Journal of Rural
Development, Vol. XVI, No.1.
Stern Review. On the Economics of Climate Change. A study commissioned by the Government of UK.
Sen, A.K. 2000. Development as Freedom. Oxford University Press.
Worm, Boris et al. 2006. Impact of Biodiversity Loss on Ocean Ecosystem Services. Science 3, Vol. 314, No.
5800. Pp. 787-790.

16
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Factors Influencing Mobile Services

Adoption in Rural India
Abhay Jain* and B.S. Hundal**

Abstract
Despite several attempts over the last few years, telecom infrastructure in rural areas is
lagging behind the expected levels. There has been a phenomenal spurt in the growth of
tele-density in the country, with the evolution of new wireless technologies, but the gap
between the urban and rural tele-density has been increasing. Various policy initiatives
are on the way to reduce this widening gap, which in turn, leaves the impression of
tremendous potential for growth in the rural areas. To make the adoption and diffusion
growth possible, the companies are constantly facing certain challenges in tackling rural
market viz., understanding rural consumers, reaching products and services to remote
rural locations and communicating with vastly heterogeneous rural audience. While
mobile phone usage in rural areas is rather an unexamined genre in academic literature,
this explanatory study investigates the factors influencing the rural consumer buying
behaviour towards mobile phone and making choice of service providers prevailing in
mobile phone markets. The data have been collected from the rural regions of Punjab,
India during July to December 2005 and include 1357 respondents who have adopted
mobile phones. The study concludes that rural people extremely desire the facilities and
knowledge along with latest technology to make choice about mobile sets and service
provider.

1.0 Introduction
Over the past few years, mobile phone adoption has generated a significant amount of
hype and interest and especially in India where wireless industry is the fastest growing
market in the Asia-Pacific region with a compounded annual growth rate (CAGR) of 40-
42 per cent. But, at the same time, Indian telecom infrastructure in rural areas is lagging
behind the expected levels and consequently, the gap between the urban and rural tele-
density has been increasing. The urban tele-density has surged to 31.1 per cent while the
rural tele-density has gone up by an abysmal 1.94 per cent (Ringing Data, Financial
Express 2005). Two major reasons for this are the perceived lack of profitability of rural
telecommunication and the lack of appropriate policies and strategies to provide
Universal Access.
Experiments like Hindustan Lever’s project Shakti, ITC e-chaupal to n-Longue’s etc. are
an attempt at wiring up Rural India. Though the efficient policy development at the
*
Lecturer, Department of Commerce and Business Management, Guru Nanak Dev. University, Amritsar, Punjab,
India. E-mail: abjain22@yahoo.com
**
Reader, Department of Commerce and Business Management, Guru Nanak Dev. University, Amritsar, Punjab,
India. E-mail: bikramhundal61@yahoo.co.in

17
government/corporate level is required, at the same time we cannot neglect the actual
conditions at the ground level for the effective penetration of mobile market in rural
areas. Although it is evident that the large players in the telecommunication business
constantly conduct market research, the problem is that the results obtained are usually
kept inside company walls, and therefore consumer behaviour in mobile phone industry
is an unexamined genre in academic literature. Corporations which have understood the
psyche of rural consumers and markets have notched up successes.
2.0 Review of Studies
Pakola et al. (2004) surveyed 397 Finnish consumer-purchasing motives on one hand and
factors affecting operator choice on the other. The result indicate that while price and
properties were the most influential factors affecting the purchase of a new mobile
phone, price, audibility and friend’s operators were regarded as the most important in the
choice of the mobile phone operator. The paper concludes with a discussion of
contributions and proposes ideas for future studies in the un-researched area.
Liu (2002) examined factors affecting the brand decision in the mobile phone industry in
Asia. It was concluded that the choice of cellular phone is characterised by two distinct
attitudes to brands: attitude towards the mobile phone brand on one hand and attitude
towards the network on the other. While price and regularity of service were found to
dominate choices between network providers, choices between mobile phone brands
were affected by new features such as memory capacity and SMS-options, more than
size. The trends will actually be not towards smaller phones but towards phones with
better capability and large screens.
Riquelme (2001) conducted an experiment with 94 customers to identify the amount of
self-knowledge consumers have while choosing mobile phone brand. The study was built
upon six key attributes (telephone features, connection fee, access cost, mobile-to-mobile
phone rates, call rates and free calls) related to mobile phone purchasing. The research
shows that consumer with prior experience about a product can predict their choices
relatively well but consumers tended to overestimate the importance of features, call rates
and free calls and underestimate the importance of a monthly access fee, mobile-to-
mobile rates and connection fees.
Kesti and Ristola (2003) investigated consumer intentions to use different mobile
services. To this end mobile services had been tested in a real, interactive situation by
voluntary test users. This paper also considered the needs people see themselves having
in the mobile commerce context in the future. The field trial’s focus was on testing
mobile services and technology in an actual end user environment. The main findings of
the study indicated that the perceptions users got from testing mobile services affect their
intention to use those kinds or similar services in the future. The results also indicated
that there are significant differences while examining two kinds of groups; low-interest
users and high-interest users. The test users regarded the guidance services as the most
important, followed by mobile ads and communication services. Furthermore, there were
statistically significant differences in means between different types of users and their
evaluation of the three services groups.

18
3.0 Need of the Study
The review of the literature on marketing reveals that the studies on consumer behaviour
in rural areas should be given more consideration now. Faced with the prospect of growth
in the mobile phone user base hitting a saturation point in big cities, Indian mobile
service providers are gearing up to delve deeper into under-penetrated rural areas. The
behaviour pattern of rural consumers is starkly different from region to region (Pareek
1999). For marketers to understand, it becomes imperative for them to visit and survey
the prospective market. Visiting slums can provide the much needed experience as it has
a unique consumption pattern. The village or region from where migration has taken
place defines the identity. This is one reason a blanket air market strategy is not good but
evaluation of consumption pattern will give insights in rural habits. Often the actual
scenario in rural area is different from the brands assumption. The rural market is full of
contrasts and complexities. Actual groundwork is the only way to know the rural psyche.
3.1 Objectives of the study
This paper attempts to identify the factors affecting the choice of mobile phone on one
hand and factors affecting the choice of an operator among the rural people. A statistical
approach ‘Factor Analysis’ has been used for the study.
3.2 Methodology
The study is based on primary data collected from the adopters of mobile phones in rural
areas of Punjab State with the help of well-drafted pre-tested structured questionnaire in
Punjabi (regional language) and English. The universe of the study consists of the 1357
consumers who have mobile phones from the villages of 141 blocks of 17 districts of
Punjab State. For the collection of primary data, Rural Punjab has been concentrated.
Gupta (1979) emphasises that specific area studies have the advantage of overcoming the
regional differences in natural and geographical endowment. The respondents being the
adopters of mobile phones are selected by following the non-probabilistic convenience
sampling techniques as it is appropriate for exploratory studies. It will not be out of place
to mention here two things: firstly, in convenience sampling, respondents (who were seen
using/have possession of mobile phones) are selected because they happen to be in the
right place at the right time, and secondly, convenience-sampling technique is not
recommended for descriptive or casual research but they can be used in exploratory
research for generating ideas (Malhotra 2005). Questions inquiring the choice of a mobile
phone and the operator were implemented with 18 statements, respondents had to put
importance rate on seven-point Likert Scales. Rural consumer purchasing behaviour
concerning the choice of a mobile phone as well as operator are not well known in
theory, i.e. no commonly accepted knowledge of the factors influencing consumer’s
decision making exists. Therefore, the results obtained have some limitations and should
be considered tentative.
4.0 Data Analysis
Previous studies on mobile services as well as theories of consumer behaviour have
shown demographics to be a factor influencing the adoption of technology-based product
and services (Agarwal and Prasad 1999).

19
Table 1: Demographic Characteristics of Sampled Respondents
Number of respondents Percentage
Gender:
Male 801 59.02
Female 556 40.98
Age (years):
Under 20 92 6.78
20-40 846 62.34
40-60 403 29.70
60 and above 16 1.18
Occupation:
Farmers 652 48.05
Microentrepreneurs 348 25.64
Employed/salaried 142 10.46
Professionals 128 9.43
Students and others 87 6.42
Income:
< Rs.10,000 429 31.61
10,000 – 20,000 715 52.69
>20,000 213 15.70
Education:
Below Metric 637 46.94
Higher Secondary 256 18.87
Graduation 178 13.11
Post-Graduation 102 7.52
Any others 184 13.56

The demographic characteristics of the respondents depict that the majority of users
(62.34%) belong to 20-40 age group, followed by 40-60 age group (29.70%). This
reveals that the adopters of mobile services are relatively young. It is further revealed that
farmers comprised the maximum proportion (48.05%) followed by microentrepreneurs
(25.64%), and service class (10.46%). It reveals that mobile phone is beneficial for the
rural people, who are engaged in agricultural and small businesses. As far as the income
level of the respondents is concerned, most of the respondents (52.69%) belong to
Rs.10,000-20,000 income group, followed by less than Rs.10,000 income group
(31.61%). The Table 1 also shows that most of the respondents (46.94%) are below
metric, followed by higher secondary (18.87%) and graduates (13.11%). This signifies
that education level also plays a dominant role in the adoption and expansion of mobile
services in rural areas.
5.0 Factor Analysis
Explanatory factor analysis is used in order to identify underlying constructs and
investigate relationships among key survey interval-scaled questions regarding the
factors affecting the choice of mobile phone and mobile service provider from 1357 rural

20
respondents. To test the suitability of the data for factor analysis, the following steps
have been taken:
i. The correlation matrices are computed and examined. It reveals that there are
enough correlations to go ahead with factor analysis.
ii. Anti-image correlations were computed. These showed that partial correlations
were low, indicating that true factors existed in the data.
iii. Kaiser-Meyer-Olkin Measure of Sampling Adequacy (MSA) for individual
variables is studied from the diagonal of partial correlation matrix. It is found to
be sufficiently high for all the variables. The measure can be interpreted with the
following guidelines: 0.90 or above, marvellous; 0.80 or above, meritorious;
0.70 or above, middling; 0.60 or above, mediocre; 0.50 or above, miserable; and
below 0.50, unacceptable (Hair et al. 1995).
iv. To test the sampling adequacy, Kaiser-Meyer-Olkin Measure of sampling
adequacy is computed which is found to be 0.711 for mobile phone choice and
0.694 in choice of mobile service provider. It is indicated that the sample is
good enough for sampling.
v. The overall significance of correlation matrices is tested with Bartlett Test of
Sphericity for choice of mobile phone (approx. chi-square = 1238.151
significant at 0.0000) and for choice of mobile service provider (approximately
chi-square = 1132.164 significant at 0.000) provided as well as support for the
validity of the factor analysis of the data set.
Hence all these five standards indicate that the data are suitable for factor analysis. There
are two basic methods that the analyst can use to obtain factor solutions namely,
Common Factor Analysis and Principal Component Analysis. Common Factor Analysis
technique uses an estimate of common variance among the original variables to generate
the factor solution. Because of this, number of factors will always be less than the
number of original variables. So, choosing the number of factors to be kept for further
analysis is more problematic while using common factor analysis. However, the Principal
Component Analysis technique overcomes this limitation. This method provides a unique
solution, so that the original data can be reconstructed from the results. It looks at the
total variance among the variables and the solution so generated should include as many
factors as there are variables. So, Principal Component Analysis is always recommended
when the researcher’s primary concern is to determine the minimum number of factors
that will account for maximum variance in the data for use in the subsequent multivariate
analysis. In the present study Principal Component Analysis has been used (also referred
as R-factor analysis).
The number of factors to be extracted becomes an important issue in the absence of any
set criterion. The four possible criteria are: (i) in a priori criterion, the analyst already
knows how many factors to extract and accordingly instructs the computer; (ii) in latent
root criterion, only those factors which have latent roots greater than one are considered
significant; (iii) in percentage of variance criterion, the cumulative percentage of variance
extracted by successive factors is considered; (iv) in Scree Test criterion, at least one

21
factor more than latent root criterion is usually extracted. It is used to identify the
optimum number of factors that can be extracted before the amount of unique variance
begins to dominate the common variance structure.
In the present study exploratory efforts were made with all of the above methods.
Initially, Latent Root was used as guideline and then the Scree Test was used. In all the
attempts percentage of the variance explained was also taken into consideration. Further
an interpretation and assessment of the structure matrix was made in each case. Thus,
several factor solutions with different number of factors were examined before a
satisfactory solution was reached.
An important step in Factor Analysis is the rotation of factors. With the rotation of factor
matrix, it becomes simple and easier to interpret by making the loadings for each factor
either large or small, not in between. For rotation, either orthogonal or oblique method
can be employed. In Orthogonal Rotation Method, the axes are maintained at 90 degrees
so that the resulting factors are uncorrelated. In Oblique Rotation method, the axes are
rotated, but the 90-degree angle between them is not maintained, this makes the method
more flexible. However, analytical procedure for oblique rotations is still controversial.
Within orthogonal method, either Varimax or Quatrimax method can be employed.
Varimax method simplifies the columns in a matrix whereas Quatrimax method stresses
simplifying the rows. Further, it was assumed that no correlation exists among factors,
so orthogonal rotation along with the Varimax method of rotation of factors were used in
order to have more clarity in factor solution. Varimax rotation is probably the most
popular Orthogonal Rotation procedure. The Varimax criterion maximises the sum of the
variances of the squared loadings within each column of the loading matrix. This tends
to produce some high loadings and some loadings near zero, which is one of the aspects
of simple structure. Moreover, this is an orthogonal method of rotation that minimises the
number of variables with high loadings on a factor, thereby enhancing the interpretability
of the factors (Malhotra 2005). So, in view of the above analysis, Varimax method of
rotation of factors was used.
5.1 Criteria for the significance of factor loadings
In interpreting factors, a decision must be made regarding which factor loadings are
worth considering. A factor loading represents the correlation between an original
variable and its factor. The criterion given by J. Hair where factor loading based on
sample size is taken as basis for decision about significant factor loading is adopted. For
sample 1357 respondents, a factor loading of 0.40 has been considered significant.
After a factor solution has been obtained, in which all variables have a significant loading
on a factor, then it is attempted to assign some more meaning to the pattern of factor
loadings. Variables with higher loadings are considered more important and have greater
influence on the name or label selected to represent a factor. Hence, all the underlined
variables were examined for a particular factor and placed greater emphasis on those
variables with higher loadings to assign a name or label to a factor that accurately
reflected the variables loading on that factor. The names or label is not derived or
assigned by the factor analysis computer programme; rather, the label is intuitively
developed by the factor analyst based on its appropriateness for representing the

22
underlying dimension to a particular factor. All the factors have been given appropriate
names on the basis of variables represented in each case. The names of factors, the
statement, the label and factor loading have been summarised in Tables 3 and 5.
5.2 Factors affecting the choice of mobile phone
There were only four factors each having Eigen value exceeding one for mobile-banking
drivers. The Eigen values for 4 factors were 6.514, 4.248, 2.162 and 1.112 respectively.
The percentage of total variance is used as an index to determine how well the total factor
solutions accounts for what the variables together represent. The index for the present
solution accounts for 78.882 per cent of the total variations for the choice of mobile
phone. It is a pretty good extraction to economise on the number of choice factors (from
18 to 4 underlying factors), while losing only 21.118 per cent information content for
choice variables. The percentage of variance explained by factor 1 to 4 is 36.171 per cent,
21.911 per cent, 12.652 per cent and 8.148 per cent respectively.
Large communalities indicate that a large number of variance has been accounted for by
the factor solution. Varimax rotated factor analytic results for mobile-banking stimulators
are shown in Table 2.
Table 2: Principal Component Analysis with Varimax Rotation

Statement numbers F1 F2 F3 F4 Communalities

S1 0.545 0.815 0.447 -0.163 0.729
S2 -0.355 0.243 0.795 0.269 0.736
S3 0.684 0.383 0.471 0.266 0.825
S4 0.456 -0.177 0.512 0.874 0.846
S5 0.112 0.467 0.817 0.521 0.922
S6 0.752 0.101 -0.220 -0.428 0.728
S7 0.602 0.765 0.517 0.425 0.745
S8 0.128 0.416 -0.248 0.621 0.824
S9 -0.342 -0.659 0.419 0.116 0.689
S10 -0.085 0.406 0.345 0.745 0.752
S11 0.738 0.423 0.043 0.157 0.732
S12 0.302 -0.219 -0.684 0.224 0.861
S13 0.472 0.475 0.028 -0.710 0.754
S14 -0.075 0.737 0.293 0.156 0.816
S15 0.067 -0.091 0.672 0.302 0.750
S16 -0.215 0.194 -0.651 0.107 0.828
S17 -0.698 0.071 0.393 0.100 0.731
S18 0.110 0.343 0.046 -0.846 0.768

Eigen values 6.514 4.248 2.162 1.112

Percentage of 36.171 21.911 12.652 8.148
variation
Cumulative 36.171 58.082 70.734 78.882
percentage of
variation

23
Table 2 shows that after 4 factors are extracted and retained, the communality is 0.729
for variable 1, 0.736 for variable 2 and so on. It means that approximately 72 per cent of
the variance of variable 1 is being captured by our 4 extracted factors together. The
proportion of variance in any one of the original variables, which is being captured by the
extracted factor, is known as communality (Nargundkar 2002).
Table 3: Naming of the Factors

Factor Name of Label Statements Factor

number dimension loading
F1 Convenience S6 I prefer the mobile phone, which is easily available 0.752
and assistance in my area.
S11 I always take into account the service centres of 0.738
mobile phone in my area.
S17 Demo/knowledge of the functions of the mobile -0.698
phone is not necessary for me.
S3 I like the mobile phone with regional language. 0.684

F2 Price S1 I often find myself checking prices. 0.815

consciousness S7 I always stick to guarantee/warrantee to be 0.765
provided.
S14 I usually purchase where instalment/easy loan 0.737
facility is provided for the purchase of mobile
phone.
S9 Attracting schemes with the mobile phone don’t -0.659
change my choice of purchasing.

F3 Brand choice S5 I like the latest technologies enabled set. 0.817

S2 Image of the mobile phone is very important while 0.795
purchasing.
S12 I don’t consider the external look of the mobile -0.684
phone while purchasing.
S15 Audibility of the mobile phone. 0.672
S16 Properties in the mobile phone. -0.651

F4 Influential S4 My neighbours’ usage influences my purchase 0.874

persons decision.
S18 I always take my own decision while choosing a -0.846
mobile phone.
S10 My friends/relatives recommended me. 0.745
S13 Advertisement/media influenced me more to buy -0.710
the mobile phone.
S8 Dealer recommended me to buy a particular mobile 0.621
phone.

In the Table 3, the respondents considered convenience and assistance as the first
dominating factor. In this factor the role of manufacturer played a significant role in
affecting the choice of mobile phone. The rural respondents preferred the mobile phones
that were easily available, suit their environment and they have proper knowledge
regarding how to operate and use the mobile phone. The second significant factor was the

24
price consciousness. It indicated that the rural respondents check the price while
purchasing and stick to the guarantee and warrantee. Since most of the rural respondents
had low income, they often particularised the loan facility and the sale promotion
schemesassociated with the mobile phones. Brand Choice name had been given to the
third important factor. It specified the taste and awareness level of the rural respondents,
how they cope with the new technology and features. Rural respondents took into their
consideration the latest technology, image of the company, fascinating outlook and the
prevailing features, which helped them in choosing a particular mobile phone. The last
factor, which could be assumed as the closest choice affecting factor surrounding the
rural respondents, had been the influential persons. The rural people had been found
more influenced by the neighbours’ usage. It is pertinent to mention here that own
decision and media has been regarded here as the negligible impact on the choice of
buying a mobile phone. The rural respondents have been seen more relying upon the
external pressure than own choice.
5.3 Factors affecting the choice of service provider
The choice of service provider was affected primary by facilitating factor. Network
coverage, quality of services, easy availability of connection and bill payment centres
Table 4: Principal Component Analysis with Varimax Rotation

Statement numbers F1 F2 F3 F4 F5 Communalities

S1 0.882 0.206 0.403 0.675 0.277 0.865
S2 0.483 -0.028 0.691 0.011 -0.037 0.771
S3 0.596 0.742 0.277 0.088 0.267 0.781
S4 -0.406 -0.224 0.465 0.619 0.345 0.792
S5 -0.075 0.107 -0.124 0.162 -0.585 0.871
S6 -0.119 0.009 0.571 0.042 -0.079 0.894
S7 -0.613 -0.017 -0.428 0.325 0.178 0.799
S8 -0.188 0.092 0.594 -0.219 0.384 0.828
S9 0.347 -0.115 0.247 0.594 0.408 0.874
S10 0.727 0.032 -0.173 0.081 0.057 0.742
S11 0.294 0.066 0.624 -0.095 0.207 0.762
S12 -0.016 0.051 0.365 -0.055 -0.691 0.791
S13 -0.075 0.674 0.465 0.277 0.088 0.845
S14 0.026 0.302 0.169 -0.153 0.756 0.864
S15 -0.124 -0.178 0.413 -0.551 0.335 0.774
S16 -0.088 0.121 -0.016 -0.224 0.714 0.764
S17 0.756 0.327 0.028 -0.017 -0.417 0.878
S18 0.418 -0.705 0.189 -0.120 0.187 0.815
Eigen values 4.668 3.160 2.686 2.198 1.998
Percentage of 31.162 13.947 12.729 6.917 5.187
variation
Cumulative 31.162 45.109 57.838 64.755 69.942
percentage of
variation

25
Table 5: Naming of Factors

Factor Name of Label Statements Factor

number dimension loading
F1 Facilitating factor S1 I would like to have a service provider 0.882
who has wide network coverage.
S17 Quality of services is an important 0.756
variable while opting for mobile service
provider.
S10 I always prefer easy availability of 0.727
connection in my area.
S7 I do not bother about the convenience of -0.613
bill payment centres.
F2 Effectiveness S3 I feel proud in adopting service provider 0.742
having good Image.
S18 I don’t consider other people’s -0.705
satisfaction experience about the service
provider.
S13 I like to try the service provider who 0.674
provides more value-added services.
F3 Dexterity S2 Customer care availability helps me a lot 0.691
in adopting a mobile service provider.
S11 Proper guidance and knowledge provided 0.624
by service provider has advantage over
other service providers in the market.
S8 Dealer recommended me to have a 0.594
specific mobile service provider.
S6 I would like to have less paper 0.571
formalities.
F4 Relative advantage S4 I am attracted towards the number of free 0.619
calls/messages given by service provider.
S9 Low rental and low charges for 0.594
calls/messages motivate me to adopt
service provider.
S15 I don’t take into account the variety of -0.551
plans and attracting schemes provided by
service provider.
F5 Influential person S14 My neighbours/fellow has the connection 0.756
of service provider.
S16 Friends/family recommended me. 0.714
S12 It is my own decision to have specific -0.691
service provider.
S5 Advertisement/media influences me most -0.585
to have a specific service provider.

26
had affected their choice relatively much. The service provider has to concentrate much
on this facilitating factor in order to grab the market. Effectiveness had been regarded as
the second most important factor. The respondents had been seen enquiring from others
regarding the satisfaction level from their service providers. Besides, image of the service
provider and value-added services provided had great impact on the decision regarding
choice of service provider. Dexterity was the third significant factor. The skilfulness of
the service provider viz. customer care facility, guidance, dealer recommendation and
less paper formalities had been considered valuable for them and affected the respondent
significantly. The next factor had been named as relative advantageous. It is the general
phenomenon that people are always attracted towards the things having low price, high
quality and variety. The same thing had been perceived. Free calls/message facility, low
rental, variety of plans had attracted the respondents towards them. Influential person
factor had been regarded as the least important factor affecting the choice of service
provider. Neighbours’ choice had affected the respondent’s choice much.
6.0 Conclusion and Implication
Rural telecommunication has been a significant area on which the government has been
giving sustained emphasis to bring down the widening gap between urban and rural tele-
density. Several measures have been taken and many others are in queue to make rural
telecom more accessible. In order to understand the impact of these measures and
policies, it seems to be more appropriate to know first what actually rural consumer
perceives. This exploratory study was particularly an attempt to increase the in-depth
understanding of rural consumer regarding mobile phone market. The study was an
attempt to throw a flood of light on the much-unexamined area regarding the choice of
mobile phone and service provider in rural India.
The main results of the study indicate that consumer education is an integral part of rural
marketing strategy. In the study, most of the respondents had the education level below
metric; this might be the reason that the rural respondents were not much influenced by
media/advertising. Even, due to the lack of knowledge and education, they were unable
to make use of their own mind and have to depend upon others. So, apart from formal
media like newspaper, television, radio, cinema and direct mail, rural-specific promotion
methods like demonstrations, puppet shows, house-to-house campaigns, processions,
rural melas, dance programmes etc. would be more useful in attracting the attention of
rural consumer. While making the choice of mobile phone, rural consumer put more
emphasis on the convenience and assistance, price, features and influential person.
Though the rural consumers were having low income and knowledge, they preferred to
step into the shoes of modernity such as new technology enabled, advance features in
mobile set etc. with proper guidance. It must be taken into account that they wanted to
make the optimum utilisation of their hard-earned money. While making the choice
regarding the service provider, they stressed facilities provided, effectiveness, dexterity,
relative advantage and influential person. The rural consumer perceived that service
providers more capable which provide more quality facilities at low price along with
rural financing.
Despite the fact that the results of the study are tentative, the findings add to the existing
rare literature on consumer behaviour in Indian rural mobile market. The results endowed

27
with interesting aspects of mobile phone choice on one side by giving the fact that rural
consumer inclined towards the convenience and proper guidance/knowledge services,
and on the other side by arguing that choice of service provider was mostly affected by
the facilitating and effectiveness factor. Currently, India is one of the most assertively
competed markets, and the competition is driven by low-price rentals and free value-
added facilities. In addition to extending the understanding of consumer behaviour in
mobile services context, the research presented also has practical implications for
managers and policy makers who have to make strategies and decision in order to cater to
this hitherto unexplored new technology-based service market.

References
Agarwal, R., and J. Prasad. 1999. Are Individual Differences Germane to the Acceptance of New Information
Technologies? Decision Science, Vol.30 (2). Pp. 361-391.
Gupta, K.P. 1979. Alternatives in Developing Framework of Research. ICSSR Newsletter, 10 (April-September).
Hair, J.F., Ralph E. A., Ronald L. T., and William C.B. 1995. Multivariate Data Analysis. 4th Edition. New
Jersey: Prentice-Hall.
Kesti, M., and A. Ristola. 2003. Tracking consumer Intention to use Mobile Services: Empirical Evidence from
a field trial in Finland. Accessed on November 2005 available at http://www.rotuaari.net/
downloads/publication-10.pdf.
Liu, C.M. 2002. The effects of promotional activities on brand decision in the cellular
Telephone industry. The Journal of Product and Brand Management, 11(1). Pp. 42-51.
Malhotra, N.K.2005.Marketing Research:An Applied Orientation. P.568. Pearson Edu.(India Branch). New Delhi.
Nargundkar, R. 2002. Marketing Research: Text and Cases. New Delhi: Tata McGraw Hill Publishing Co.
Pakola, J., M. Pietila, and R. Svento. 2003. An Investigation of consumer behaviour in mobile phone markets in
Finland. Proceedings of 32nd EMAC Conference, Track: New Technologies and E-marketing. Accessed on
November 2005 available at www.oasis.oulu.fi/publications/jem-05-hk.pdf.
Pareek, V. 1999. Stop. Look. Go. Advertising and Marketing, Vol. 3. P. 58.
Riqulme, H. 2001. Do consumers know what they want? Journal of Consumer Marketing 18(5). Pp. 437-448.
Ringing Data : Dust off Rural Telecom Review Policy. Financial Express, Dec.26, 2005. Accessed on January
05, 2006 from http://www.financialexpress.com/fe_full_story.php?content_id=112558

28
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Adoption and Diffusion of Integrated Pest Management

Technology: A Case of Irrigated Rice Farm in
Jogjakarta Province, Indonesia
Joko Mariyono*

Abstract
The institutionalisation of IPM technology in Indonesia relies on farmer-to-farmer diffusion.
The study aims to analyse the adoption and diffusion of the technology at farm level. Data
come from a farm survey conducted in three consecutive seasons of 2000/2001. Two
villages in Moyudan sub-district of Jogjakarta Province, where a chronic pest infestation
exists, are chosen. Sixty rice farmers have been interviewed. The study shows IPM
technology has been adopted by both IPM-participating and non-IPM participating farmers.
There is an indication of IPM knowledge diffusion occurred in the area of study where
every farmer faces the same problems of severe pest infestations.

1.0 Introduction
Integrated pest management (IPM) technology, a package of practices that utilises natural
predators and careful timing of right doses, is one of the most important measures to cut
the use of pesticides. The technology has been introduced in Indonesia to cope with
problems resulting from the unwise use of pesticides during the 1970s and 1980s. It is not
surprising evidence that the application of pesticides during the periods has increased
substantially along with incredible amount of subsidies (Irham and Mariyono 2002). This
was considered one of the key successes in the intensification programme in Indonesia,
that is, the substantial increase in rice production because of the increase in crop yield
through intensive use of inputs including chemical pesticides. But, most of the researchers
still believe that without indiscriminate use of pesticides, increases in application of
pesticides lead to a number of consequences such as elimination of natural enemies,
pesticide-resistant pests, and frequent pest outbreaks (Barbier 1989; Matteson 2000). In
addition, pesticides also impacted on human health (Kishi et al. 1995; Pawukir and
Mariyono 2002) and the environmental contamination (Bond 1996).
1.1 IPM programme and dissemination of IPM technology
The Indonesian Government was implementing the IPM programme with support of the
UN's Food and Agriculture Organisation (FAO) since May, 1989 for dissemination of

*
Ph.D candidate in Economics at The Australian National University, Canberra, Australia. Email:
joko.mariyono@anu.edu.au

29
IPM technology among rice-based farmers through a pilot project. The programme
‘provides an ideal case to contrast extension for sustainable agriculture with that
supporting high external input agriculture. IPM is being introduced into a farming system,
irrigated rice, in which the Green Revolution has been successful during the past twenty
years’ (Rolling and van de Fliert 1994: 98).
This programme was realisation of a Presidential Decree (INPRES 3/86), three years
before, which banned 57 brands of pesticides from rice cultivation, and declared IPM the
national pest control policy. A policy measure progressively reduced the subsidy on
pesticides, which was previously 85 per cent, to zero in 1990 (Untung 1996). These policy
measures created a favourable climate for the implementation of Indonesia's National IPM
Programme. It is the first phase (1989-1992) of large-scale attempt to systematically
introduce sustainable agricultural practices as a national, public sector effort. During the
first phase, it had trained around 200,000 farmers intensively through a number of training
packages, namely the farmers’ field school (FFS). Locations of FFS were purposively
selected with criteria of easy accessibility and the presence of active farmer groups.
Farmers participating in the school were also purposively selected for the programme.
More prosperous and better informed farmers in the villages were selected and encouraged
to be participants of the field school.
The second phase (1993-1999) was sponsored by the World Bank. Since 1994, the pilot
FFS activities had been taken over by the National IPM Training Project funded by the
World Bank (World Bank 1993). The project promoted IPM and improved crop
cultivation of rice and other food and horticultural crops. More regions had been covered
and more actors had been involved. However, the target was not to reach all Indonesian
farmers. The strategy of the programme was to train a fraction of farmer community,
instead of providing training to all farmers in the community. Thus, the spread of IPM
knowledge relied on farmer-to-farmer diffusion. During implementation of second phase
of the project, villages were still subjectively selected with the same criteria by the project
management in collaboration with agricultural services officials both in provincial and
district levels. With the assistance from agricultural office at sub-district level and farmer
group leaders, farmers were also purposively selected with certain criteria, for instance:
rice farmer, literacy, and ability to actively discuss.
1.2 Farmers’ field school
The heart of IPM programme in Indonesia is FFS, a process of learning by doing.
Practically, the programme enhances human capital (Ekowarso 1997). The World Bank,
along with a number of development agencies promoted FFS since it is a more effective
method to extend science-based knowledge and practices (Feder et al. 2004a; 2004b).
Farmers’ field school applies a participatory approach to provide assistances for farmers to
develop their capability on analytical skill, critical thinking and creativity such that
farmers can make better decision. In short, the objective of FFS is to enhance human
resource development, in which farmers become experts of IPM in their paddy fields.
Farmers are expected to be able to conduct observations, to analyse agro-ecosystems, to

30
make decisions, and to implement pest control strategies based on the results of their field
observations. Farmers will obtain those capabilities from participating FFS. In reality,
IPM involves not only pest control but also other aspects of farming such as balanced and
efficient fertilising, efficient use of water, crop rotation and soil conservation. The
following IPM principles are central to the FFS: growing healthy crops; conserving and
utilising natural enemies; carrying out regular field observations; and developing farmers
as IPM experts in their own field (Untung 1996).
A unit of FFS consists of a training group of 25 farmers, selected either from one farmer
group, or across such groups within one village. To encourage woman participation, it is
expected that 30 per cent of participants are woman farmers (Kingsley and Siwi 1997).
FFS starts with a ‘ballot box’ pre-test of knowledge and ends with a post-test. Several
weeks before planting, there are pre-preparation meetings to organise identification of
communities that fulfil the criteria for establishing FFS; and identification of suitable
participants. Observation, analysis and action FFS for rice hold 12 times of weekly
meetings throughout one planting season of rice (around three months). The first meeting
begins two to three weeks after transplanting. This is to enable observations of all critical
stages of growth and development of crop. FFS uses a framework of an agro-ecosystem
analysis. The agro-ecosystem analysis is based on about 1000m2 of rice field that is
divided into two plots: IPM plot and another plot based on locally conventional
management of which the application of pesticides eliminates natural enemies of insect
pests.
The key processes and elements above have to be fulfilled to enable that FFS runs
adequately, and that FFS needs timely and sufficient material and financial supports. As
cited by Agro-Chemical Report (2002), a unit cost of FFS in 1996-97 fiscal years was, on
average, US$599. It constituted honorarium of the facilitator, preparation and coordination
expenses, facilitator’s transport, materials, refreshments, compensation of land used for
field trial, stipends for participants, and field day or ceremonial of closing FFS.
1.3 Statement of problems
During the first few years of the IPM Programme, pesticide use dropped by approximately
50 per cent and yields increased by around ten per cent (Pincus 1991). Despite this
impressive success, work of Irham (2001; 2002) shows that the programme has
diminished pesticide use and increased productivity and the household incomes of
different socio-economic groups. These confirm the fact cited by Useem et al. (1992) and
Untung (1996) that Indonesian IPM programme has been successful in reducing pesticide
use and escalating rice production. Winarto (1995) gives the impression of supporting the
success of IPM programme by showing process of transfer IPM knowledge at farmer level.
Studies by Paiman (1998), Kuswara (1998) and Susianto et al. (1998) highlight some
cases of successful IPM implementation in some sub-districts. In wider scale, Irham
(2002: 75) sums up the impact study of Indonesian IPM programme conducted by
SEARCA in 1999 that ‘at least the farmers can maintain the current yield with lower cost
of pesticide’. In the Agro-Chemical Report (2002), it is stated that Indonesia has been one

31
of the leaders in the use of IPM in Asia. Since 1989, a national IPM programme has
helped farmers in Indonesia to reduce their dependence on pesticides and increase their
harvests. It has also dramatically reduced the incidence of pesticide-related illnesses and
environmental pollution.
However, the successful implementation of IPM technology in Indonesia has been
questioned by Feder et al. (2004a) finding that the IPM training has failed to deliver IPM
technology since there is no difference between the farmers participating in the training
and the ones not participating in, in terms of growth in rice yield and reduction in pesticide
use. Another study by Feder et al. (2004b) shows that the diffusion of IPM knowledge
among farmers is also not the case, and there is no evidence that the expected
environmental and health benefits of the programme are significant. This is
understandable because non-IPM participating farmers may face different problems from
IPM-participating farmers. The expectation on adoption of the technology by Indonesian
farmers is questionable since the target of IPM programme was not to reach all Indonesian
farmers. As previously mentioned that the strategy of the programme was to train a
fraction of farmer communities, and the spread of IPM knowledge relies on farmer-to-
farmer diffusion.
This study aims to analyse the adoption and diffusion of IPM technology in rice farming.
The entry point of introducing IPM technology is plant protection. Farmers facing serious
problems of pest infestation are expected to be responsive to adopting the technology.
Once the technology has been adopted by some farmers, other farmers having the same
problems will adopt the technology. In other words, there is fast diffusion of the
technology among farmers.
2.0 Material and Methods
2.1 Source of data and location
Data come from a farm survey conducted in three consecutive seasons of 2000/2001 in
two villages. Sixty rice farmers are purposively selected in the survey, out of which thirty
have been graduated from FFS. Moyudan sub-district of Jogjakarta Province was selected
as the study area The region constitutes one of the rice production centres in Java where
IPM programme has been promoted intensively by local and national government. Every
year, a number of FFSs have been set up following the introduction of the IPM
programme. Importantly, this sub-district is considered an area of endemic pest
infestations of rice. This is because this area has a technical irrigation system allowing
farmers to cultivate rice throughout a year. Being an endemic area of pest infestations, it
is expected to stimulate the adoption of IPM technology by IPM-graduate farmers, and
enable diffusion of the technology by non-graduate farmers neighbouring the graduates.
2.2 Underlying theory
A production function explained in the microeconomic theory is used as fundamental
analysis. Related to the introduction of new technology, the production function is
mathematically expressed as:

32
 Y = F (X, L, T , S ) (1)
where Y is output, X is vector of inputs, L is land, T represents different technology
and S represents different states of nature. In Asian developing countries, it has been
pointed out by Hayami and Ruttan (1985) that agricultural production technology
exhibits constant returns to scale. This means that output will be multiplied by a
factor λ if all inputs and land are multiplied by the same factor, such that:
λY = F (λX, λL, T , S ) (2)
for any λ>0. If λ = 1 L , the production function can be expressed as yield function,
that is: y = f (x, T , S ) (3)
where y = Y L , and x = X L . Variable land, L collapses because it becomes
unity and constant.
2.3 Economic modelling
Yield function is developed in estimating the functional relationship. Pesticide input is
used as independent variable to know the effect of pesticide on rice yield. To detect the
impact of the IPM-technology on rice yield, a dummy variable is introduced into the
model. A Cobb-Douglas model is used in this study. Soekartawi et al. (1986) stated that
the Cobb-Douglas model is suitable to estimate agricultural production functions. In terms
of a log-linear functional form, the Cobb-Douglas model is formulated as:
y = α + ∑ β i ln x i + γT + ∑ δ j S j + ε
i j
(4)

where y is yield of rice (kg/ha); x is inputs for i = 1, 2, ... 8; T is dummy variable for
IPM-graduate farmers; Sj is dummy variables for j=1, 2 for first and second dry
seasons respectively; α, β, γ and δ are coefficients of regression; and ε is error terms.
Having a strong claim about the superiority of the IPM technology, it is assumed that if
the all farmers have adopted the technology, there is no difference in productivity between
IPM-graduate and non-graduate. The diffusion of IPM technology will be indicated by
insignificance of γ. Descriptive analyses on pest infestation, pesticide use and micro-
economic indicators are primarily used to justify the estimated econometric outcome.
3.0 Results and Discussion
3.1 The impact of IPM on rice yield
The impact of IPM-training on yield of rice is given in Table 1. Even though the
coefficient determination is relatively low, overall test shows high level of significance.
The low coefficient determination is a common case in estimating yield function, because
land, which is highly correlated with production, disappears in the model. It can be seen
that KCl and labour inputs significantly increase the yield of rice. On the other hand, pest
infestation significantly reduces the yield. It is logical because pesticides have no impact

33
on yield. An interesting phenomenon is that yield of rice in second dry season is the
highest. Referring to the system irrigation in this area, this case is understandable.
Table 1: Regression Results of Yield Function
Estimated results
Variables
Coefficients t-value
Intercept 7.6131 14.73***
Urea 0.0158 0.21ns
TSP -0.0046 -0.07ns
KCl 0.0037 1.51*
Seed 0.0411 0.60ns
Non-liquid pesticides -0.0042 -1.37ns
Liquid pesticides 0.0203 0.96ns
Labour 0.1001 2.11**
Pest infestation -0.0683 -2.43***
IPM graduate 0.0253 0.79ns
First dry season 0.0154 0.46ns
Second dry season 0.0497 1.45*
R2 0.1518
F-test 2.732***
Note : *** = Significant at 1% ** = Significant at 5% * = Significant at 10%

IPM graduate has no impact on yield of rice. Assuming that IPM technology can increase
productivity, there are two possible cases explaining the finding. Firstly, IPM graduates do
not adopt IPM principles, such that there is no difference in yield of rice between IPM
graduate and non-graduate. Secondly, non-graduates have adopted IPM technology
because the technology has diffused among farmers in the villages through informal
forums. It seems that the second case is acceptable because at aggregate provincial level,
IPM technology has escalated rice production (Mariyono et al. 2002). Farmers adopt IPM
technology because of the fact that pesticides are no longer useful in controlling pest
infestation in rice (Mariyono 2001). This was indicated in the yield function. As
mentioned by Lichtenberg and Zilberman (1986), pesticides will have positive impact on
yield or production if susceptible pest infestation exists. When pesticides are not effective,
farmers need to choose alternative pest controls. IPM has been introduced as an alternative.
Note that the farmers in the study area still use pesticide. Although both groups apply
pesticides for their crop protection, their philosophy of using them is different.
Table 2 shows that most of IPM-graduates use pesticides only when pest infestation exists,
and only few of them use pesticides for preventive motive. This is understandable as they
received the IPM principle during training. On the other hand, most of IPM-graduates
regarded pesticide as a preventive measure regardless the level of seriousness of pest
infestation although some of them apply pesticides only when severe pest infestation
exists.

34
Table 2: Philosophy of Using Pesticides
Types reason IPM-graduate farmer Non-graduate farmer
(%) (%)
Pest infestation exist 66.7 40.0
Preventive motive 16.7 50.0
High yield motive 10.0 16.7
Neighbours’ influence 6.7 10.0
Source: Farm Survey 2001.

3.2 Adoption and diffusion of IPM technology

It is necessary to understand the implementation of the IPM technology by the farmers
after analysing the impacts of IPM programme on pesticide use and yield of rice.
Table 3 shows that resistant variety has been adopted by both IPM-trained and non-IPM
graduates. This is a strong indication of which IPM knowledge has diffused to non-IPM
Table 3: Implementation of Components of IPM Technology
Important components of IPM technology IPM-graduate Non-graduate Gap
farmer (%) farmer (%)
Pest-resistant variety 100 100 0
Technical culture
a. Land preparation 100 00 100
b. Sanitation 100 66.7 33.3
c. Fallow 100 100 0
d. Crop rotation 0 0 0
e. Simultaneous planting 100 100 0
f. Crop density 100 100 0
g. Recommended fertilisers 83.3 10 77.3
Mechanical rat control
a. Mass killing 100 100 0
b. Trapping 56.7 20 30.7
c. Sulphured smoking 66.7 23.3 43.4
Chemical insect control
a. Pesticides 100 100 0
b. Non pesticides 53.3 0 53.3
Observation
a. Every 1-3 days 80 53.3 26.7
b. Every 4-7 days 20 33.3 -13.3
c. Every > 7 days 0 13.3 -13.3
Average 72 53 19
Source: Farm Survey 2001.

35
graduates. Pearson et al. (1991) pointed out that the development and dissemination of
new varieties of rice are significant to the success of IPM and the continued expansion of
rice output. Both different groups of farmers have applied recommended technical culture
concept for fertiliser application particularly for Potassium fertiliser. Only a few of non-
IPM graduates use this fertiliser. One of the applicable reasons is that the non-IPM
graduates still think the necessity of this fertiliser while its price is little bit higher
compared to other types of fertiliser. Both groups have fallowed land and simultaneous
planting. A long fallow period is one of appropriate measures to control pests because it
can cut the lifecycle of pests. It is stated that long fallow periods together with
simultaneously synchronised planting are important components of IPM strategy
(Matteson 2000).
It is important to note that none of groups applies crop rotation. This becomes the main
reason of high possibility of pest and disease infestations to exist in this area. The study
also found that IPM graduates applied more mechanical pest control compared to that of
non-IPM graduates particularly for rat control by using trapping tools and sulphured-
smoke. IPM graduates also apply non-pesticide material such as salt and ashes regardless
the effectiveness of such measures. One of the distinct different between both different
groups is that the IPM graduates have better observation on their crops condition
compared to that of non-IPM graduates. Overall, the gap in adopting the components of
IPM technology is not wide a lot. This is an indication that the principle of IPM has been
widely known by both groups of farmers in the study area. One important factor is that all
farmers face the same problems of massive pest infestation. The farmers get incentives in
adopting the IPM technology in coping with the risk of crop loss.
4.0 Conclusion
The Indonesian Government has institutionalised IPM technology to replace the previous
chemical intensive technology, which has been introduced during the green revolution.
However, it was doubted whether the IPM technology can be adopted by farmers directly
participating in the IPM training and diffused among non-participating farmers. This is
because farmers have been accustomed with chemical intensive technology for long
period and the number of IPM-participating farmers is very small compared with total
number of Indonesian farmers. Comparing IPM-participating and non-IPM-participating
farmers in area where serious and chronic pest infestations exist, this study shows that
there is no different productivity of rice. Both groups of farmers have applied the
important components of IPM technology. This is an adequate indication that in an area
where farmers face serious problems of pest infestation, IPM technology has been
responsively adopted by both IPM-participating and non-IPM-participating farmers. It is
the case that the technology has diffused to other neighbouring farmers.

36
References
Agro-Chemical Report. 2002. Pesticide and Policy: the Impact of Integrated Pest Management (IPM) on the
Indonesian Economy. Agro-Chemical Report, Vol. 2, No. 2. Pp. 18-20.
Barbier, E.B. 1989. Cash Crops, Food Crops, and Sustainability: The Case of Indonesia. World Development,
Vol. 17, No. 6. Pp. 879-895.
Bond, J.W. 1996. How EC and World Bank Policies Are Destroying Agriculture and the Environment.
Singapore: AgBé Publishing.
Ekowarso, J. 1997. Program Nasional Pengendalian Hama Terpadu Selaku Upaya Pemberdayaan SDM melalui
Proses Pengalaman Petani. Paper presented at Lokakarya Pemasyarakatan Pengendalian Hama Terpadu
Propinsi Daerah Istimewa Yogyakarta. 20-22 March 1997. Yogyakarta.
Feder, G., R. Murgai, and Quizon J. B. 2004b. The Acquisition and Diffusion of Knowledge: The Case of Pest
Management Training in Farmer Field Schools, Indonesia. Journal of Agricultural Economics, Vol. 55, No 2.
Pp. 221-243.
Feder, G., R. Murgai, and J. B. Quizon. 2004a. Sending Farmers Back to Schools: The Impact of Farmer Field
School in Indonesia. Review of Agricultural Economics, Vol. 26, No. 1. Pp. 45-62.
Hayami, Y., and V. W. Ruttan. 1985. Agricultural Development: An International Perspective. Baltimore: John
Hopkins University Press.
Irham. 2001. Impact of IPM Program on Rice Production and Income. In Toward the Harmonization between
Development and Environmental Conservation in Biological Production. Proceedings of the International
Workshop. Japan: The University of Tokyo. Pp. 21-23.
Irham. 2002. Integrated Pest Management Program, Pesticide Use and Rice Productivity: A Case Study in
Yogyakarta, Indonesia. In A. Herrmann. and S. Schumann (eds.), Proceedings of the Int. Workshop on
Environmental Risk Assessment of Pesticides and Integrated Pesticide Management in Developing Countries,
6-9 Nov. 2001 Kathmandu, Nepal. Pp.73-81.
Irham, and J. Mariyono. 2002. The impact of Integrated Pest Management on pesticide demand for soybean
farming in Yogyakarta, Indonesia. In A. Herrmann and S. Schumann (eds.), Proceedings of the Int. Workshop
on Environmental Risk Assessment of Pesticides and Integrated Pesticide Management in Developing
Countries, 6-9 Nov. 2001. Kathmandu, Nepal. Pp. 62-72.
Kingsley, A. M., and S.S. Siwi. 1997. Gender Study II: Gender Analysis, Agricultural Analysis, and
Strengthening Woman’s Leadership Development in the National IPM Program. Jakarta: Indonesian National
IPM Program.
Kishi, M., N. Hirschorn, M. Djayadisastra, N. L. Satterlee, S. Strowman, and R. Dilts, 1995. Relationship of
Pesticide Spraying to Signs and Symptoms in Indonesian Farmers. Scandinavian Journal of Work and
Environmental Health, Vol. 21. Pp. 124-133.
Kuswara, E. 1998. IPM in Marga Sub-District, Tabanan District, Bali. In Community IPM: Six Cases from
Indonesia. Semi-annual FAO IPM Technical Assistance Progress Report, Jakarta, Indonesia. Pp.100-137.
Lichtenberg, E., and D. Zilberman. 1986. The Econometrics of Damage Control: Why Specification Matters.
American Journal of Agricultural Economics, Vol. 68, No. 2. Pp. 261-273.
Mariyono, J. 2001. Hubungan antara Serangan Hama dan Penggunaan Pestisida: Pendekatan Analisis
Ekonometrik pada Padi dan Kedelai. Jurnal Perlindungan Tanaman Indonesia, Vol. 8, No. 1. Pp. 54-62.
Mariyono, J., Irham, and K. Suratiyah. 2002. Dampak Teknologi Pengendalian Hama Terpadu Pada Permintaan
Pestisida pada Padi di Yogyakarta. Agrosains, Vol. 15, No. 1. Pp. 141-149.
Matteson, P.C. 2000. Insect Pest Management in Tropical Asian Irrigated Rice. Annual Review of Entomology,
Vol. 45. Pp. 549–574.
Paiman. 1998. An IPM Case Study in Ngantang Sub-District, Malang District, East Java. In Community IPM: Six
Cases from Indonesia. Semi-annual FAO IPM Technical Assistance Progress Report. Jakarta, Indonesia. Pp.
168-189.

37
Pawukir, E. S., and J. Mariyono. 2002. Hubungan antara Penggunaan Pestisida dan Dampak Kesehatan: Studi
Kasus di Dataran Tinggi Alahan Panjang Sumatera Barat. Jurnal Manusia dan Lingkungan, Vol. IX, No. 3.
Pp. 126-136.
Pearson, S., R. Naylor, and W. Falcon. 1991. Recent Policy Influences on Rice Production. In Pearson, S.;
Falcon, W; Heytens, P.; Monke, E. and Naylor, R. (eds.) Rice Policy in Indonesia. Ithaca: Cornell University
Press. Pp. 8-21.
Pincus, J. 1991. Farmer Field School Survey: Impact of IPM Training on Farmer’s Pest Control Behavior. IPM
National Program: Jakarta.
Rolling, N., and E. van de Fliert. 1994. Transforming Extension for Sustainable Agriculture: The Case of
Integrated Pest Management in Rice in Indonesia. Agriculture and Human Value, Vol. 11, No. 2/3. Pp. 96-108.
Soekartawi, A. Soehardjo, J. L. Dillon, and B. J. Hardaker. 1986. Ilmu Usahatani dan Penelitian untuk
Pengembangan Petani Kecil. Jakarta: UI Press.
Susianto, A., D. Purwadi, and J. Pincus. 1998. Kaligondang Sub-District: A Case History of IPM Sub-District. In
Community IPM: Six Cases from Indonesia. Semi-annual FAO IPM Technical Assistance Progress Report.
Jakarta, Indonesia. Pp. 15-99.
Untung, K. 1996. Institutional Constraints on IPM Implementation in Indonesia. Publication of the Pesticide
Policy Project, Publication Series, 3A. Pp. 37-47.
Useem, M., L. Setti, and J. Pincus. 1992. The Science of Javanese Management: Organizational Alignment in an
Indonesian Development Programme. Public Administration and Development, Vol. 12. Pp. 447-471.
Winarto, Y.T. 1995. State Intervention and Farmer Creativity Integrated Pest Management among Rice Farmers
in Subang, West Java. Agriculture and Human Value, Vol. 12, No. 4. Pp. 47-57.
World Bank. 1993. Staff Appraisal Report. Indonesia Integrated Pest Management Training Project. Washington:
The World Bank.

38
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Exploring Predictive Strengths of Stochastic Pavement

Deterioration Models: A Case Study of
Thailand’s Highway Network
Satirasetthavee Dussadee* and Herabat Pannapa**

Abstract
Deterioration modelling is a core element in any infrastructure management system.
Higher accuracy of prediction models results in better maintenance planning in terms of
maintenance needs prediction, strategy selection and budget allocation. This research
aims at enhancing existing deterministic deterioration models used by Thailand
Department of Highways. Stochastic deterioration models, namely a time-based model
and a discrete-time state-based model, are proposed to overcome the limitations of
deterministic models by capturing the variation and uncertainty of the deterioration rate.
Predictive strengths of both models are explored in terms of their precision and
confidence of deterioration models, the integration of relevant variables, and the effect of
duration dependence. The results of deterioration rate prediction based on the developed
models demonstrate that the time-based models are suitable for the project-level or
network-level management of road networks with similar characteristics while the
discrete-time state-based model is stronger in terms of condition prediction and
comparing the effects of relevant variables for Thailand road network.

1.0 Introduction
Deterioration models are a core element in any infrastructure management system.
Precision and confidence of deterioration modelling plays an important role in model
development. Errors in prediction (Mauch and Madanat 2001; Mishalani and Madanat
2002) models should be minimised since the overall maintenance planning and its data
analysis depend on the forecast of future condition which is related to the maintenance
selection and planning. Deterministic models and probabilistic models are the two main
types of deterioration models. ‘The deterministic models, such as some of empirical-
mechanistic models, and regression models, have been developed to explain the
relationship between related factors and deterioration rate’ (Li et al. 1997). However, the
deterministic model cannot directly capture the variation or uncertainty related to the
pavement deterioration rate (Jiang and Sinha 1992; Silva et al. 2000; Mishalani and
Madanat 2002).

*
Ph.D candidate in Construction Engineering and Infrastructure Management, School of Engineering and
Technology, Asian Institute of Technology, Pathumthani, Thailand.
**
Assistant Professor, School of Engineering and Technology, Asian Institute of Technology, Pathumthani,
Thailand. Email: pannapa@ait.ac.th

39
Probabilistic deterioration models were introduced to directly explain the stochastic
nature of pavement deterioration in which several factors cannot be captured by available
data (Mishalani and Madanat 2002). The discrete-time state-based model and the time-
based model are the general forms of the probabilistic models (Mauch and Madanat
2001). The discrete-time state-based models, i.e., the Markov and the semi-Markov
processes, describe the probability that pavement sections will undergo a change in
condition state at a given discrete time. Time-based deterioration models predict the
probability density function of the time taken by each pavement section to change its
condition state, such as the time of crack initiation (Patterson and Chesher 1986). The
effects of relevant variables, such as annual daily traffic volume, geographic regions and
highway classes, can be integrated into the time-based models (Mauch and Madanat
2001). Mauch and Madanat (Li et al. 1997) further discussed that the time-based models
can be transformed into discrete-time state-based models because the relationship
between the discrete-time state-based models and the time-based model is similar to the
relationship between the Poisson process and the exponential distribution.
The main objectives of this paper are: (1) to develop the time-based model and discrete-
time state-based deterioration models; (2) to explore the predictive strengths of pavement
stochastic deterioration models in terms of their precision and confidence, the integration
of relevant variables, and the effect of duration dependence.
This research collected pavement condition data and other relevant data for model
development from Thailand Department of Highways (DoH). The DoH is responsible for
maintaining and preserving Thailand’s highway system which includes 48,626 km of
asphaltic concrete pavement, 2,559 km of concrete pavement, and 483 km of gravel road
(DoH 2000: 10). Thailand DoH implemented a PMS called ‘Thailand Pavement
Management System’, or TPMS, to help the DoH manage their highway pavement in the
most cost-effective manner (DoH 1988). The deterministic deterioration model that was
incorporated in TPMS is based on the deterioration model of the Highway Design and
Maintenance Standards Model (HDM-III). Three deterministic models used in the HDM-
III are the roughness progression model, the crack initiation model, and the crack
progression and the rut-depth progression model.
The highway condition data in the southern region of Thailand was used to calibrate the
HDM-III model to the local environment in 1992 (N.D. Lea. 1992). The developed
deterioration model based on the HDM-III model was not fully utilised because it
requires many artificial inputs that are not defined in the data collection process of
TPMS. Discrepancies between the required data and the data actually collected may lead
to a large error in predictions due to the scarcity of some inputs.
There is a need to modify and to overcome the limitation of the deterministic model
which cannot directly capture the variation and uncertainty in the pavement deterioration
process. DoH has been collecting their pavement condition data since 1990 based on the
defined inspection procedures of TPMS. Series of historical data can be used to develop
stochastic deterioration models that correspond with the defined condition assessment of

40
TPMS, thus eliminating the data discrepancy issue. Both discrete-time state-based and
time-based models are developed and their predictive strengths are further explored in
order to enhance the precision and confidence of deterioration models for better
maintenance planning.
Since their introduction, probabilistic deterioration models have received much attention
from pavement engineers and researchers. State-based and time-based models are the
concepts that are typically used to develop probabilistic deterioration models for
infrastructures (Mauch and Madanat 2001).
State-based deterioration models, also referred to as Markov Chains, have been applied
for predicting pavement deterioration since the early 1970s (Silva et al. 2000). The
transition probabilities can be developed from various sources of data such as historical
condition data and engineering judgment obtained from experts. Many techniques can be
applied to determine the transition matrix; for example, simple probability techniques
(Silva et al. 2000), minimisation of the summation of the absolute value of the difference
between the expected state predicted from the Markov process and the observed
pavement condition (Carnahan et al. 1987), and ordered probit models (Madanat et al.
1995).
Alternatively, the transition matrix can be derived from a time-based model. The time-
based model was first developed to determine the timing of periodic maintenance in
pavement management system (Patterson and Chesher 1986). The time-based model
views pavement deterioration as depending largely on the time of distress initiation.
Therefore, many probability distribution functions can be assigned for the hazard rate
function to simulate the changes in pavement deterioration rate. The hazard rate function
reflects the instantaneous rate (risk) at which a pavement section will undergo transition
out of its current state to the lower state after time t (Ebeling 1997; Mishalani and
Madanat 2002). Examples of probability distribution functions used in other research
works are the semi-parametric model (Mauch and Madanat 2001) and the Weibull
distribution (Patterson and Chesher 1986; Mishalani and Madanat 2002). The integrated
relevant variables in the exponential term of the hazard rate function are proposed by
Mishalani and Madanat (2002) to capture the effect of explanatory variable and dummy
variable.
Probabilistic models such as the time-based deterioration model and the discrete-time
state-based deterioration model are selected for development in this research. Once both
deterioration models are developed, their predictive strengths are explored in terms of
their precision and confidence, the integration of relevant variables, and the effect of
duration dependence. Chi-squared test is used to measure the precision and the
confidence of pavement deterioration models. The integration of the variables of the
time-based model is tested by measuring the capability to explain the effect of integration
variables for both pavement deterioration models. The duration dependence of the
pavement deterioration rate is also compared between two types of pavement
deterioration models.

41
2.0 Time-based Pavement Deterioration Model
The time-based pavement deterioration model is developed based on the probability
density function of the time taken for each pavement section to change its condition state.
The time-based model developed in this study is based on the research presented by
(Mishalani and Madanat 2002), which developed a time-based model to capture the
deterioration of bridge decks.
2.1 Hazard rate function and transition probability determination
The Weibull distribution is recommended to describe the hazard rate function (Patterson
and Chesher 1986; Mishalani and Madanat 2002). The Weibull probability density
function (PDF) can be used to describe for a continuous random variable t and it is
associated with the hazard rate function as presented in Eq. (1). The p value in the
equation represents the properties of the deterioration model in terms of duration
dependence (Mishalani and Madanat 2002). The hazard rate function is monotonically
decreasing for 0<p<1. The Weibull hazard rate function is reduced to the exponential
function (lack of memory), when p is equal to 1. If p is greater than 1, the hazard rate
function is monotonically increasing.
λ(t) = pλptp-1 (1)
Mishalani and Madanat (2002) recommended that the constant parameter λ should be
replaced by the exponential function, as shown in Eq. (2). The replacement function
helps integrate the effect of relevant variables into the hazard rate function. The replaced
hazard rate function is shown in Eq. (3). The parameters of the hazard rate function are
determined by maximum likelihood estimation (Mishalani and Madanat 2002).
λ = e-βX (2)
-pβX p-1
λ(t) = pe t (3)
where λ(t) = hazard rate function; p = parameter of Weibull distribution; t = duration; X =
column vector of exogenous variables; and β = row vector of parameters.
In addition, the equation for calculating the transition probability from condition state i to
condition state j was proposed by Mishalani and Madanat (2002), as shown in Eq. (4) to
Eq. (6). Condition state 2 represents the best condition and 0 represents the worst
condition.
P2,2 = 1-R2(t,∆) = exp[−λ2 (t + ∆) ]
p p 2 2
(4)
exp[−(λ 2 t ) ]
p 2

t+∆
∫
p 2 −1
P2,1 = p 2 λ 2 2 exp[( λ 2 t ) 2 ] * τ exp[ − ( λ 2 t ) p 2 − λ1p1 (t + ∆ − τ ) p1 ]d τ
p p
(5)
t

P2,0 = 1 – P2,2 – P2,1 = R2(t,∆) – P2,1 (6)

where Pi,j is the probability of the transition from state i to j over a period ∆; and Ri(t,∆)
is the probability of the transition out of state i.

42
3.0 Discrete-Time State-based Pavement Deterioration Model
This study adopts simple probability techniques to develop the transition matrix. The
historical condition data which is collected annually are classified into different condition
states depending on the scale of the condition index or its rating. The probabilities of
pavement sections that are in condition state i and will transition to condition state j (Pi,j
in the transition matrix) are calculated by dividing the total amount of highway sections
with condition changes from state i to state j (Ni,j) by the total amount of highway
sections in state i, (Σ Ni,j) as shown in Eq. (7) (Jiang et al. 1988).

N i, j (7)
pi , j = n

∑N
j =1
i, j

Once the transition matrix is developed, the state probability vector of defined period t
[PC]t , is obtained by multiplying the initial state vector [IC1 … ICN] with the transition
matrix [P] raised to the power of t, as shown in Eq. (8).

(8)

4.0 Model Development

The deterioration models in this research are developed based on the physical condition
data of all highway regions of Thailand DoH. The regions in Thailand are defined as the
north, north-eastern, central and eastern, and southern regions. The International
Roughness Index (IRI) is a measurement of a pavement roughness for evaluating the
pavement serviceability (Shahin 1994). The condition data are collected in terms of IRI,
and areas of minor- and major-deterioration. The IRI data is selected to be used in the
model development since it is more complete than the data for areas of minor- and major-
deterioration. The IRI data are available for years 1992-1994 and 1997-2000. Historical
condition data are screened out when the corrective maintenance activity is applied to
preserve the highway condition. However, the highway section may also be subjected to
routine maintenance activities (e.g. slurry seal for small cracking areas, grass cutting, and
unclogging of drainage system).
The condition data from DoH are classified into condition states. Ten condition states are
defined in the discrete-time state-based model development, and four condition states are
defined for the time-based model. The range of condition states in the time-based model
is wider than in the discrete-time state-based model as shown in Table 1 due to the one-
year inspection cycle. A higher inspection frequency would allow for a finer resolution in
defining discrete states. However, the condition data of two consecutive years can be

43
applied in the development of the discrete-time state-based model. Therefore, the
transition probabilities of the discrete-time state-based deterioration model are developed
based on five data sets which are the pavement condition data of year 1992-1993, 1993-
1994, 1997-1998, 1998-1999, and 1999-2000.
Table 1: Pavement Condition Data and Their Condition States
Discrete-time state-based deterioration model
Condition state 1 2 3 4 5
IRI values < 2.50 2.50 – 2.99 3.00 – 3.49 3.50 – 3.99 4.00 – 4.49
Condition state 6 7 8 9 10
IRI values 4.50 – 4.99 5.00 – 5.49 5.50 – 5.99 6.00 – 6.49 ≥ 6.50

AESAL group 1 2 3
Range of each group Less than 400,000 to More than
400,000 800,000 800,000

Time-based deterioration model

Condition state 1 2 3 4
IRI values < 2.00 2.00 – 2.99 3.00 – 3.99 > 4.00
AESAL Continuous variables

The inspection period censors the time spent by a pavement section in a specific
condition state because the condition data are collected over one-year period. Therefore,
the condition data, including the time that the pavement section spends in a condition
state must be approximated before developing the model. The following assumptions
are incorporated to determine the time spent in a condition state (Mishalani and
Madanat 2002):

• When a condition state drops by one between two consecutive inspection cycles,
the actual transition is approximated to occur exactly at the midpoint between
two corresponding inspection times.
• When a condition state drops by two states between two consecutive inspection
cycles, the time that the pavement section remains in the state is approximated
to be four months.
Four regions of the highway network and three groups of Annual Equivalent Single Axle
Load (AESAL) [groups 1, 2, and 3] are used to classify the pavement condition data for
developing the proposed models, as shown in Table 1. In addition, the collected data is
separately divided into two groups for model development and model validation after the
maintenance application is screened out from the historical condition data.
Approximately 20 per cent of each classified data group is randomly reserved for model
validation; for example, the condition data of a highway subsection is selected from
every five highway subsections.

44
4.1 Time-based deterioration model development
The main component of the time-based deterioration model is the definition of the hazard
rate function for all condition states. The Weibull distribution is selected to describe the
hazard rate function, as shown in Eq. (1). In addition, the effects of explanatory variable
and dummy variables, which are the continuous variable AESAL and different regions in
this study respectively, are integrated into the exponent of the exponential function as
expressed in Eq. (2). The geographical region variables take value of 1 if the highway
lies in the specified region. Eq. (3) presents a complete hazard rate function.
Maximum likelihood estimation is used to determine seven coefficients of each hazard
rate function based on the collected data from the DoH, as shown in Table 2. The p
values for hazard rate functions (from state 1 to state 4) are greater than 1. This means
that the hazard rate functions are monotonically increasing. Mishalani and Madanat
(2002:139) explained that if the function is monotonically increasing, the probability of
the transition out of state increases with an increasing time spent in that state.
Table 2: Estimated Parameters of Hazard Rate Function
Estimated parameters
Variables
State 1 State 2 State 3 State 4
North (β 1) -0.3780 -0.3422 -0.9426 -0.6641
North-East (β 2) -0.4490 -0.4078 -0.7580 -0.5898
East (β 3) -0.4380 -0.2844 -1.0172 -0.5303
South (β 4) -0.3641 -0.2406 -0.5250 -0.4375
AESAL (β 5) 1.10E-07 1.07E-07 1.07E-07 3.01E-07
Constant (β 6) 0.9987 0.9982 1.0017 1.0254
P 1.4007 2.0147 2.2593 2.7009
Note: Estimations of all parameters are statistically significant at 1% level.

As shown in Table 2, the coefficients for different regions are negative, but the
coefficients of AESAL are positive. The AESAL variable contributes to the reduction of
the hazard rate function through a positive coefficient. That is, when the AESAL
increases, the hazard rate will be reduced. One implication that may explain this
behaviour is that higher design standards and frequent maintenance applications are
applied to highways subjected to high traffic volume. The signs of the coefficients for the
different regions are negative. The negative sign reflects a contribution to an increase in
hazard rate.
The hazard rate functions for four condition states are shown in Figure 1. The slope of
the hazard rate function directly depends on the p value. The p value of the hazard rate
function of state 4 has the highest rate; therefore, the physical condition of pavement in
state 4 deteriorates at the fastest rate.

45
Figure 1: Hazard Rate Functions for All States (Region 2 and Group 1)
1.2 16.0

1.0 14.0
12.0
0.8
Hazard rate

Hazard rate
10.0
0.6 8.0
0.4 6.0
Hazard rate function 4.0
0.2 Hazard rate function
of condition state 1 2.0 of condition state 2
0.0 0.0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time-in-state (years) Time-in-state (years)

14.0 1.2
12.0 1.0
10.0
0.8
Hazard rate
Hazard rate

8.0
0.6
6.0
0.4
4.0
Hazard rate function 0.2 Hazard rate function
2.0 of condition state 4
of condition state 3
0.0 0.0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time-in-state (years) Time-in-state (years)

4.2 Transition probability of the time-based model

In the time-based deterioration model, the β coefficients presented in Table 2 are
substituted in the [β] of Eq (2). For comparison purposes, this research selects the mid-
value of AESAL for each group classified in Table 1 to calculate the hazard rate function
and further use in Eq. (4)-(6) to determine the transition probability of each AESAL
group. The transition probabilities of different condition states are then calculated based
on the four developed hazard rate functions. Eq. (3) is used to determine the transition
probabilities in the case that the future condition of the pavement section remains in the
same state. Eqs. (4) and (5) are used to determine the transition probabilities in the case
that the future condition of the pavement section changes by one step and two steps,
respectively.
Transition probabilities developed based on the initial condition states of 1, 2, and 3 are
illustrated in Figures 2, 3, and 4, respectively. The transition probabilities of a pavement
section that remains in the same condition state are equal to zero in different years. For
example, it takes six years for the transition probability of a pavement section that is in
state 1 and remains in state 1 (state 1-1) to equal to zero as illustrated in Figure 2. It takes
four years for the case of state 2-2 (Figure 3) and three years for state 3-3 (Figure 4). The
result shows that the probabilities of remaining each condition state significantly
decrease, when the initial condition state increases. This implies that the poor condition

46
highway has a higher deterioration rate than a normal or good condition highway. For
example, the probability from state 2 to state 2 decreases about 0.35 from year 1 to year 2
as shown in Figure 3, while the probability from state 3 to state 3 decreases about 0.50 as
shown in Figure 4.
Figure 2: Transition Probabilities for Pavement Condition State 1 (Region 2 and Group 1)

1.0
State 1-1
0.8 State 1-2
State 1-3
Probability

0.6

0.4

0.2

0.0
1 2 3 4 5 6 7 8
Time

Figure 3: Transition Probabilities for Pavement Condition State 2 (Region 2 and Group 1)

1.0
State 2-2
0.8 State 2-3
State 2-4
Probability

0.6

0.4

0.2

0.0
1 2 3 4 5 6 7 8
Time

47
Figure 4: Transition Probabilities for Pavement Condition State 3 (Region 2 and Group 1)

1.0

0.8
Probability

0.6

0.4

0.2 State 3-3

State 3-4
0.0
1 2 3 4 5 6 7 8
Time

Figure 5: Probability Transition Matrices for the Discrete-Time State-based

Deterioration Model
TRANSITION MATRIX (REGION_1 & GROUP_1) PROBABILISTIC DETERIORATION MODEL (REGION_1 & GROUP_1)

Before Pavement After Pavement Condition State After Pavement Condition State
Prediction Year
Condition State 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
1 0.441 0.389 0.098 0.048 0.024 0 0 0 0 0 1 1.000 0 0 0 0 0 0 0 0 0
2 0 0.380 0.434 0.135 0.036 0.014 0 0 0 0 2 0.441 0.389 0.098 0.048 0.024 0 0 0 0 0
3 0 0 0.376 0.420 0.131 0.052 0.021 0 0 0 3 0.194 0.319 0.249 0.132 0.065 0.028 0.009 0.003 0.001 0
4 0 0 0 0.348 0.411 0.168 0.054 0.019 0 0 4 0.086 0.197 0.251 0.203 0.124 0.075 0.036 0.016 0.007 0.006
5 0 0 0 0 0.319 0.401 0.178 0.076 0.026 0 5 0.038 0.108 0.188 0.207 0.165 0.123 0.073 0.043 0.023 0.033
6 0 0 0 0 0 0.310 0.344 0.184 0.077 0.086 6 0.017 0.056 0.121 0.168 0.167 0.150 0.104 0.074 0.047 0.096
7 0 0 0 0 0 0 0.242 0.338 0.205 0.215 7 0.007 0.028 0.071 0.118 0.141 0.149 0.118 0.097 0.071 0.201
8 0 0 0 0 0 0 0 0.241 0.325 0.434 8 0.003 0.013 0.040 0.075 0.104 0.126 0.113 0.104 0.085 0.338
9 0 0 0 0 0 0 0 0 0.196 0.804 9 0.001 0.006 0.021 0.045 0.070 0.096 0.094 0.096 0.086 0.486
10 0 0 0 0 0 0 0 0 0 1 10 0.001 0.003 0.011 0.025 0.044 0.066 0.071 0.079 0.076 0.625

TRANSITION MATRIX (REGION_1 & GROUP_2) PROBABILISTIC DETERIORATION MODEL (REGION_1 & GROUP_2)

Before Pavement After Pavement Condition State After Pavement Condition State
Prediction Year
Condition State 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
1 0.453 0.429 0.083 0.027 0.007 0 0 0 0 0 1 1.000 0 0 0 0 0 0 0 0 0
2 0 0.459 0.393 0.101 0.039 0.008 0 0 0 0 2 0.453 0.429 0.083 0.027 0.007 0 0 0 0 0
3 0 0 0.375 0.487 0.094 0.029 0.014 0 0 0 3 0.206 0.392 0.238 0.106 0.041 0.012 0.004 0.002 0 0
4 0 0 0 0.368 0.391 0.136 0.064 0.041 0 0 4 0.093 0.268 0.260 0.200 0.097 0.043 0.021 0.011 0.004 0.003
5 0 0 0 0 0.398 0.356 0.102 0.093 0.051 0 5 0.042 0.163 0.211 0.230 0.152 0.085 0.052 0.032 0.014 0.017
6 0 0 0 0 0 0.310 0.397 0.086 0.052 0.155 6 0.019 0.093 0.147 0.205 0.177 0.119 0.090 0.062 0.034 0.054
7 0 0 0 0 0 0 0.440 0.360 0.080 0.120 7 0.009 0.051 0.093 0.157 0.168 0.133 0.120 0.090 0.058 0.121
8 0 0 0 0 0 0 0 0.364 0.364 0.273 8 0.004 0.027 0.056 0.109 0.139 0.126 0.134 0.110 0.081 0.215
9 0 0 0 0 0 0 0 0 0.400 0.600 9 0.002 0.014 0.032 0.070 0.104 0.105 0.131 0.116 0.097 0.329
10 0 0 0 0 0 0 0 0 0 1 10 0.001 0.007 0.018 0.043 0.072 0.080 0.115 0.111 0.102 0.451

TRANSITION MATRIX (REGION_1 & GROUP_3) PROBABILISTIC DETERIORATION MODEL (REGION_1 & GROUP_3)

Before Pavement After Pavement Condition State After Pavement Condition State
Prediction Year
Condition State 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
1 0.568 0.316 0.065 0.035 0.015 0 0 0 0 0 1 1.000 0 0 0 0 0 0 0 0 0
2 0 0.447 0.400 0.126 0.015 0.012 0 0 0 0 2 0.568 0.316 0.065 0.035 0.015 0 0 0 0 0
3 0 0 0.363 0.451 0.107 0.060 0.019 0 0 0 3 0.323 0.321 0.187 0.103 0.042 0.018 0.004 0.001 0 0
4 0 0 0 0.400 0.439 0.117 0.044 0.000 0 0 4 0.184 0.245 0.218 0.178 0.093 0.050 0.023 0.006 0.004 0.001
5 0 0 0 0 0.425 0.396 0.113 0.057 0.009 0 5 0.104 0.168 0.189 0.207 0.147 0.090 0.052 0.016 0.020 0.007
6 0 0 0 0 0 0.321 0.491 0.170 0.019 0.000 6 0.059 0.108 0.143 0.193 0.177 0.125 0.086 0.030 0.051 0.030
7 0 0 0 0 0 0 0.235 0.118 0.471 0.176 7 0.034 0.067 0.099 0.157 0.178 0.143 0.113 0.041 0.091 0.078
8 0 0 0 0 0 0 0 0.000 1.000 0.000 8 0.019 0.041 0.065 0.117 0.156 0.141 0.125 0.048 0.129 0.159
9 0 0 0 0 0 0 0 0 0.333 0.667 9 0.011 0.024 0.041 0.082 0.126 0.125 0.123 0.048 0.154 0.267
10 0 0 0 0 0 0 0 0 0 1 10 0.006 0.014 0.025 0.055 0.094 0.102 0.109 0.043 0.160 0.391

48
4.3 Discrete-time state-based deterioration model
The historical condition data are classified based on four regions and three AESAL
groups as in DoH’s practice. The classified data are used to develop twelve (four regions
* three AESAL groups) transition matrices. The main assumption for developing the
transition matrix in this study is that the pavement condition state would not drop more
than four condition states in one year. Examples of transition matrices and probabilistic
deterioration models are shown in Figure 5. Once the transition matrices, [P] are
developed, the state probability vector at the defined period t, [PC]t can be calculated by
multiplying the initial state vector by the transition matrix raised to the power of t, as
shown in Eq. (8).
4.4 Conversion of the developed probabilistic deterioration models
For comparison purposes, the developed probabilistic deterioration models, as shown in
the right column of Figure 5, are converted into a performance curve which relates the
mean IRI values and time. The state probability vector of defined period t [PC]t and the
mid-value of each IRI state [IRI] are multiplied to determine the mean value of IRI of
each year IRIt, as shown in Eq. (9).

(9)

Figure 6: Converted Probabilistic Deterioration Models of All AESAL Groups in

Regions 1 and 2

7.0
Average IRI value (m/km)

6.0

5.0

4.0

3.0 R1 & G1 R1 & G2 R1 & G3

R2 & G1 R2 & G2 R2 & G3
2.0
0 2 4 6 8 10
Time (year)

49
Twelve converted probabilistic deterioration models, for different regions assuming an
initial state 1 are illustrated in Figure 6. The converted models show that the deterioration
rate of a higher AESAL group is lower than for a lower AESAL group. For example, the
deterioration rate of region 1 with AESAL Group 1 (R1&G1) is higher than the rate for
region 1 with AESAL Group 3 (R1&G3) as shown in Figure 6. This may be explained by
the fact that the weight and vehicle restriction policies are strictly implemented on the
AESAL Group 3 more than other groups. In addition, the routine maintenance is mostly
applied for a highway section that has a higher traffic volume since it yields higher
benefits. This results a sporadic maintenance application for highway that has a lower
traffic volume (DoH 1988).
5.0 Predictive Strengths of the Developed Models
Once both stochastic deterioration models are developed, the general characteristics and
the effect of variable are used to compare their predictive strengths. Table 3 presents the
general characteristics of the developed models in terms of data sources, duration
dependence, the variables and their effects, and the model descriptions.

Table 3: General Characteristics of the Discrete-Time State-Based and the Time-

Based Deterioration Models

Discrete-time state-based
Items Time-based deterioration model
deterioration model
Data sources • Time period of each state • Number of sections of each state
of two consecutive years

Duration • Duration dependence is included in • Time-independent Markov

dependence the p value. process with a constant pavement
• If p value is greater 1, the model is transition matrix throughout the
monotonically increasing and reflects programming period.
positive duration dependence.

The effect of • Replacing the constant parameter • Classified historical condition

relevant λ with an exponential function of the data based on relevant variables
variables relevant variables

Variables • Explanatory variable (AESAL) • Four regions (north, north-

• Dummy variable (highway eastern, east and south) and
geographical region) AESAL

In order to compare the predictive strengths of both developed models, the number of
condition states in the state vector of defined period (t) for the discrete-time state-based
model is first reclassified from ten states to four states as in the time-based model.
However, it should be noted that the use of more condition states in the discrete–time

50
state-based model can give a more precise deterioration model than fewer defined
condition states. Pavement sections in the north-eastern region subjected to AESAL
Group 1 during the period of 1998-2000 are selected as an example to illustrate the
predictive strengths of the developed models. The selected data is more complete and
shows the deterioration rate for three consecutive years.
5.1 Precision and confidence of developed models
The state probability vectors from both developed models are validated in this section.
The reserved data for model validation are classified into different states. The initial
condition state of the selected data set is in the condition state 2 since the condition data
of the selected pavement sections falls in the range of condition state as defined in Table
1. The Chi-squared test is used to determine the precision and confidence of the
developed model from year 1999 to 2003. However, the state probability vectors of the
developed models (dotted lines) in year 1999 and 2000 are only plotted against the
proportions of the model validation reserved data (solid line) as presented in Figure 7. By
using a Chi-squared test with 0.05 level of significance and two degrees of freedom, as
shown in Table 4 and Figure 7, the statistical results show that both developed models
provide acceptable state probability vectors since the Chi-squared values of both models
are less than the critical Chi-squared value. However, when comparing between the chi-
squared values of both models, the time-based model gives slightly higher accuracy of
the state probability vectors.
Table 4: The Chi-squared Goodness-of-Fit Test of Percentage Prediction for
Pavement Condition
Year Chi-squared Chi-squared
(Time-based) (State-based)
1999 0.014 0.327
2000 0.092 0.010
2001 0.056 0.086
2002 0.090 0.171
2003 0.160 0.210

AESAL Chi-squared Chi-squared

group (Time-based) (State-based)
1 0.092 0.010
2 0.368 0.021

Region Chi-squared Chi-squared

(Time- based) (State-based)

North-Eastern 0.092 0.010

Southern 0.204 0.048
Note: Chi-Squared Critical = 5.99 with 2 degree of freedom and 0.05 level of significance.

51
Figure 7: Transition Probabilities of the Developed Models

1.00

0.80

0.60
[PC ]i,t

0.40

0.20

0.00
State 1 State 2 State 3 State 4
Time-Based (1999) State-Based (1999) Actual (1999)
Time-Based (2000) State-Based (2000) Actual (2000)

5.2 The effect of duration dependence of developed models

The effect of duration dependence for both models is discussed in this section. The
transition probabilities and the average IRI values ranging from the initial year (year 0) to
year 5 are used to compare the effect of the duration dependence for both developed
models. Table 5 presents the comparisons of the transition probabilities and the average
IRI values for both models.
The average IRI values of the time-based model are higher than the discrete-time state-
based model during the observed period as illustrated in Table 5. This is due to the fact
that the time-based model is a duration-dependent model. The state probability vector of
the discrete-time state-based model is developed from a constant pavement transition
matrix throughout the programming period (or time-independent deterioration model).
This results a slight higher deterioration rate predicted by the time-based model of
approximately 4 to 9 per cent when compared to the discrete-time state-based model, as
shown in Table 5. Therefore, the p values of the time-based model for all condition states
are greater than one. This also implies that the time-based model is monotonically
increasing and can directly capture the duration dependence nature of deterioration via
the p value. The time-based model is calculated based on duration data, which includes
the effect of time taken for each pavement section to transition of its condition state to
the next.

52
Table 5: Comparisons of the Effect of Duration Dependence for Both Models

NE region NE region
(AESAL Group 1) Average (AESAL Group 1) Average
Percentage
Year IRI IRI
Time-based model State-based model differences
values values
2-2 2-3 2-4 2-2 2-3 2-4
0 1 0 0 2.500 1 0 0 2.500 0.00
1 0.747 0.135 0.117 2.870 0.786 0.194 0.020 2.734 4.99
2 0.308 0.427 0.265 3.457 0.448 0.437 0.115 3.167 9.16
3 0.070 0.491 0.439 3.870 0.225 0.473 0.302 3.577 8.20
4 0.009 0.217 0.774 4.266 0.106 0.380 0.514 3.908 9.15
5 0.001 0.054 0.945 4.444 0.048 0.258 0.694 4.146 7.20

6.0 Comparative Effect of the Variables

This section presents the comparative effect of the variables by varying one variable
while holding the other constant. The first comparative effect focuses on varying the
AESAL groups while holding the geographical location constant. The second
comparative effect varies the geographical locations while selected pavement sections are
subjected to the same AESAL group.
6.1 Comparative effect of the AESAL
Pavement sections in the north-eastern region are selected to demonstrate such
comparative effect by varying AESAL groups as shown in Figure 8. The solid line
represents the percentage difference of the average IRI values in year 2000 when the
AESAL groups are varied from Group 1 to Group 3. The bar chart shows the average IRI
values of each AESAL group. The average IRI values shown in Figure 8 are predicted by
both developed models and based on the actual condition data. The percentage different
of the average IRI value between AESAL Group 1 and Group 3 are 2.30 per cent (time-
based model), 9.55 per cent (discrete-time state-based model), 11.04 per cent (actual
data), as shown in Figure 8. According to the signs and the values of the corresponding
estimated coefficient in Table 2, the time-based model can correctly reflect a contribution
to a decrease in the deterioration rate. However, the discrete-time state-based model can
better capture the effect of the deterioration when the AESAL groups are different since
the discrete-time state-based model classifies the historical condition data in terms of
AESAL groups when the time-based model integrates all the variables in the hazard rate
function (see Eqs. 2 and 3).

53
Figure 8: Percentage Differences of the Converted Probabilistic Models for
Selected AESAL Groups

3.50 15.00

3.00

2.50

Difference (%)
10.00
IRI (m./km.)

2.00

1.50
5.00
1.00

0.50

0.00 0.00
Time-Based State-Based Actual

S AESAL Group 1 NE AESAL Group 1 Difference

6.2 Comparative effect of the highway geographical locations

This research selects pavement sections located in the southern region (S AESAL Group
1) and the north-eastern region (NE AESAL Group 1) to illustrate the effect of
deterioration. The average IRI value and the percentage difference of the average IRI
value are presented by the bar chart and the solid line, respectively, as shown in Figure 9.
Figure 9: Percentage Differences of the Converted Probabilistic Models for
Selected Geographical Locations

3.50 15.00

3.00

2.50
Difference (%)

10.00
IRI (m./km.)

2.00

1.50
5.00
1.00

0.50

0.00 0.00
Time-Based State-Based Actual

S AESAL Group 1 NE AESAL Group 1 Difference

The percentage differences of the average IRI value between the southern region and the
northern region are 4.54 per cent (time-based model), 10.22 per cent (discrete-time state-
based model), and 8.92 per cent (actual data). Both deterioration models can correctly
capture the sensitivity of the changes in condition prediction when the highway

54
geographical locations vary. However, the discrete-time state-based model can better
capture the effect of variable when comparing to the time-based model since the
percentage difference of the average IRI value of the discrete-time state-based model is
closer to the one in the actual data. Such demonstration indicates that the discrete-time
state-based model may be more applicable in the budget allocation when considering
different geographical locations.
7.0 Concluding Remarks
Many deterioration models can be applied for forecasting the future condition of
pavement. Stochastic deterioration models, the discrete-time state-based deterioration
model and the time-based deterioration model, are developed based on the collected
physical pavement condition of Thailand DoH. Predictive strengths of both models are
further explored in terms of their level of deterioration rate predictions, and the effects of
relevant variables.
The results confirm that the deterioration process of the pavement in the time-based
model does not have the lack-of-memory. The levels of deterioration rate prediction of
the time-based transition probabilities are slightly higher than the discrete-time state-
based model (Carnahan et al. 1987: 560). Comparative effects of relevant variables are
presented to describe the changes in condition prediction. The discrete-time state-based
model can capture the sensitivity of the changes in condition prediction when relevant
variables vary.
The time-based model is suitable in the project-level or the network-level management of
road networks that share similar characteristics, such as in the same region or subjected
to the same AESAL group because the time-based model incorporates the effect of
duration dependence. The p value in the time-based model dominates the effects of
changes in the relevant variables. Based on its predictive strength and its ability to
compare the effects of relevant variables, the discrete-time state-based model is more
appropriate than the time-based model in the network-level management for Thailand
DoH. In addition, risk and uncertainty of relevant variables can be included in the
discrete-time state-based model for future research.

55
References
Carnahan, J. V., W. J. Davis, M. Y. Shahin, P. L. Keane, and M. I. Wu. 1987. Optimal maintenance decisions
for pavement management. Journal of Transportation Engineering, 113, 5. P. 560. American Society of
Civil Engineers.
Department of Highways.1988. Thailand pavement management system: system BSM manual. M. Division,
Department of Highways, Thailand. 5.
Department of Highways. 2000. Annual report. Department of Highways, Thailand. 10.
Ebeling, C. E. 1997. Reliability and maintainability engineering. International Edition, Singapore: McGrawhill.
Jiang, Y., M. Saito, and K. C. Sinha. 1988. Bridge performance prediction model using the Markov chain.
Transportation Research Record: Journal of Transportation Research Board, 1180. Pp. 25-32. Washington
D.C.: Transportation Research Board.
Jiang, Y., and K. C. Sinha. 1992. Simulation approach to prediction of highway structure conditions.
Transportation Research Record: Journal of Transportation Research Board, 1347. P. 13. Washington D.C.:
Transportation Research Board.
Li, N., M. Huot, and R. Haas. 1997. Cost-effectiveness-based priority programming of standardized pavement
maintenance. Transportation Research Record: Journal of Transportation Research Board. 1592. P. 9.
Washington D.C.: Transportation Research Board.
Madanat, S., R. Mishalani, and W. H. Wan Ibrahim. 1995. Estimation of infrastructure transition probabilities
from condition rating data. Journal of Infrastructure System, 1, 2. P. 122. American Society of Civil
Engineers.
Mauch, M., and S. Madanat. 2001. Semiparametric hazard rate models of reinforced concrete deck
deterioration. Journal of Infrastructure System, 7, 2. P. 49. American Society of Civil Engineers.
Mishalani, R., and S. Madanat. 2002. Computation of infrastructure transition probabilities using stochastic
duration models. Journal of Infrastructure System, 8, 4. P. 139. American Society of Civil Engineers.
N.D. Lea International LTD. 1992. Thailand road maintenance project. Final Report. Vancouver, Canada, N.D.
Lea International LTD.: 25.
Patterson, W. D. O., and A. D. Chesher. 1986. On predicting pavement surface distress with empirical models
of failure times. Transportation Research Record: Journal of Transportation Research Board, 1095. P. 45.
Washington D.C.: Transportation Research Board.
Shahin, M. Y. 1994. Pavement management for airports, roads, and parking lots. First Edition, New York:
Chapman & Hall.
Silva, F. M., T. J. Van Dam, W. M. Bulleit, and R. Ylitalo. 2000. Proposed pavement performance models for
local government agencies in Michigan. Transportation Research Record: Journal of Transportation
Research Board, 1699. P. 82. Washington D.C.: Transportation Research Board.

56
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Socio-Economic Study of Gender Role in Farm

Production in Nasarawa State of Nigeria
S.A. Rahman*, H. Ibrahim* and H. Ibrahim*

Abstract
This study examines the gender role in the farm sector of Nasarawa State. Double-stage
random sampling techniques were used to select a sample of 90 farm households during
the cropping season of 2005. Primary data were obtained through the use of interview
schedules administered to farm families. Data analysis was done using descriptive
statistics, ordinary least squares regression model, productivity model and logit
regression model. It was observed that women had extensive work loads of 13 hours/day
and lower labour productivity compared to the men. The women seem to have lower rate
of involvement in farm decisions (42.8%) and less accessible to farm resources (11.5%).

1.0 Introduction
Over the past several decades, awareness of gender issues in development has steadily
increased. Analysis of gender disparity in resource accessibility and labour productivity
suggests the appropriate types of intervention require to achieve the sectoral goals and
priorities in a given country context (Rahman and Ajayi 2005). Genuine and balanced
development and growth will be achieved only when gender inequalities have been
redressed.
There is a lack of quantitative and qualitative data on gender disparity in resource
accessibility, labour contribution and productivity; especially the extent of women’s role
in agricultural production. Women position in meeting challenges of agricultural
development cannot be over-emphasised (Rahman and Usman 2004). In Nigeria, women
provide some 60 to 80 per cent of agricultural labour and are responsible for 80 per cent
of food production (Ingawa 1999; Mgbada 2000). Women spend more hours per day than
men in both productive and reproductive activities (Saito 1995).
Despite the important roles that women play in farm and household production, they have
not been given due recognition in the agricultural sector. There has been a great disparity
between women and men in the size of landholdings, as well as overall trends of
increasing landlessness (Quisumbing 1994). There are also constraints on women labour
time as they cannot call on the labour of other household members in the way men can
(Malena 1994). Women interest and involvement in farm decisions and production are

*
Department of Agricultural Economics and Extension, Faculty of Agriculture, Nasarawa State
University, Nasarawa State, Nigeria. Corresponding e-mail: shehurahmandr@yahoo.com

57
limited. The limitation has a lot of implications for women access to and control over
resources of their own (Morna 1990; Rahman and Alamu 2003).
The contribution of women to agricultural development should be maximised to improve
efficiency and sustainability. The recognition and promotion of women’s role in
agriculture requires the assessment of their accessibility to resources and the productivity
of their labour on farms to clarify the benefits of their participation. Women participation
in management and production decisions varies according to geographical location,
culture and religion. In northern Nigeria, women predominate crop processing and
storage. Fulani women are famous for the rearing of sheep, fowls, cattle and processing
of milk. Middle-belt women are generally involved in land clearing and preparation to
the marketing of farm produce. In south eastern Nigeria, women are engaged in land
preparation down to weeding and harvesting. In the riverine areas, women gather
shellfish and processed fish caught by men (Audu and Onoja 2005). Women in Nasarawa
State also play an active and significant role in farm activities. This study therefore,
assessed time budget in farm households, gender-labour differentials and productivity on
farms; and gender-based resource accessibility in farm households in Nasarawa State of
Nigeria.
2.0 Methodology
Primary data were obtained through a survey of farm households using interview
schedules administered to farm families by six enumerators, three of each sex. The
female enumerators interviewed the women while the male enumerators interviewed the
male household heads. Double-stage random sampling techniques were used to select a
sample of 90 farm households during the cropping season of 2005.
There are three geo-political zones in Nasarawa State. Two Local Government Areas
(LGAs) were randomly selected in each zone and then fifteen farm households were
randomly selected in each of the six LGAs. Data analysis was done using descriptive
statistics and ordinary least squares regression model (Mirer 1990). The regression model
was used to examine the gender-labour productivity differentials and identify factors
determining the accessibility of women to farm resources. Logit model (Pindyck and
Rubinfeld 1976) was employed to examine responsiveness of women satisfaction in the
farm sector.
The productivity model was specified as:
Log Yv = log bo + b1 log Sc + b2 log Ft + b3 log Lm + b4 log Lf + log µ
where,
Yv = value of crop output (N/ha)
Sc = seed cost (N/ha)
Ft = Fertiliser used (kg/ha)
Lm = male labour (man-hours/ha)
Lf = female labour (man-hours/ha)

58
 bo = constant term
b1, b2, b3, b4 = Regression coefficients
µ = Error term
Note: Output and seed were measured in monetary value because different crops
grown per household were considered together.
Marginal productivity of male labour (MPML) was expressed as:
MPML = b3.yv - - - - - - -- - (ii)
Lm
Where,
b3 = Regression coefficient for the male labour
Lm = Male labour
Yv = value of crops output
Marginal productivity of female labour (MPFL) was expressed as:

MPFL = b4.yv - - - - - - -- - (iii)

Lf
Where,
b4 = Regression coefficient for the female labour
Lf = Female labour
Yv = value of crops output
Gender-labour productivity differentials (GLPD) were estimated as follows:
GLPD = b3.Yv - b4. Yv
Lm Lf
= Yv (b3/Lm – b4/Lf) ----------------- - - - - - - (iv)
The regression model used in identifying the factors determining the accessibility of
women to farm resources was expressed in the form of double logarithmic function as
follows:
Log RA = logβ0 + β1log FD + β2log VC + β3 log CP+ β4log NC+ β5log ES +
β6log AG + β7log IL + logµ - - - - -- - - - (v)
where,
RA = Rate of Accessibility of Woman to Resources (%)
FD = Rate of Involvement in Farm Decision (%)
VC = Value of contribution to Farm Production (N)
CP = Cooperative Participation (years)
NC = Number of Children

59
 ES = Educational Status (years)
AG = Age of Woman (years)
IL = Income Level of Woman (N)
β0 = Constant term
β1 – β7 = Regression Coefficients
µ = Error term
Logit regression model is one of the binary choice models in which a dichotomous
regression variable is considered as the dependent variable (Rahman and Alamu 2003).
The model is based on the cumulative logistic probability function. The probability:
(yi = 1, Xij, j = 1-6)
ezi ; i = 1-90 - - - - -- - - (vi)
= ezi+1

where,
Zi = α + β1 Xi1 + β2 Xi2 + - - - - - - - - - - - - - - (vii)
Zi = cumulative logistic distribution.
Zi is a theoretical or unobservable variable. In order to obtain the value of Zi, the
likelihood of observing the sample needs to be formed by introducing a dichotomous
variable yi such that:
1 if ith woman was satisfied with her participation in farm production
Y= O if ith woman was not satisfied with her participation in farm production

The six factors influencing the satisfaction of the ith woman in farm production are
defined as:
X1 = Household farm size (ha)
X2 = Household size
X3 = Woman share of farm income (N)
X4 = Educational status (years)
X5 = Credit (N)
X6 = Cooperative participation (years)
β1- β6 = Regression coefficients
α = Constant term

60
Marginal probability was considered as a proxy for marginal satisfaction. This was
obtained based on a derived expression from the logit model as:
dp = bj [P(I-P)] - - - - - - - - - - - - - -- - - - - -- - - (viii)
dxj
In order to determine the responsiveness of women satisfaction in farm production with
respect to jth factor, the elasticity of satisfaction; which can be defined as the ratio of
percentage change in satisfaction (in probability terms) to percentage change in
magnitude of the factor in question is expressed as:
Es = bj Xj (I-P) - - - - - - - - -- - - -- -- - - - (ix)
where,
bj = Estimated logit regression coefficient
p = Estimated probability of satisfaction
Es = Elasticity of satisfaction
Xj = women jth attribute (factor).
3.0 Results and Discussion
3.1 Gender-labour in farm household activities
The time budget analysis (Table 1) indicates that women carry the major responsibility
for both farm production and domestic work. Women spend more hours per day than men
in both productive and reproductive activities. They are mainly responsible for cooking,
cleaning house and child care. They also play significant role in farm works. Generally,
women provided most of the labour for the farm household activities. However, the
decision making at the household level continues to be male-dominated.
Table 1: Time Budget Analysis per Household (hrs/day)
Chore Men Women
Farm work 4.8 2.4
Cooking 0.3 3.3
Fetching water 0.5 1.0
Fetching firewood 0.9 0.8
Cleaning house 0.4 2.2
Child care 0.4 2.0
Processing of farm produce 0.4 1.3
Total 7.7 13.0

61
Gender involvement in the various farming operation (Table 2) varies considerably. Men
do much of the land clearing, ridging and weeding. Women do about 80 per cent of the
threshing and participated fairly equally in harvesting. Women’s work is getting harder
and more time-consuming due to ecological degradation and changing agricultural
technologies, which are aggravated by their lack of access to extension services. Rahman,
Gabriel and Marcus (2004) reported a similar finding among farming households in
southern Kaduna State, Nigeria.
Table 2: Gender-Labour Contribution to Specific Farm Operation (hrs/ha)
Farm operation Men Women** t-value*
Land clearing 21.1 8.2 2.842
Ridging 27.3 7.6 3.111
Planting 20.0 10.5 2.250
Fertiliser application 12.5 6.3 2.048
Weeding 63.1 35.0 3.062
Earthing-up 23.4 11.8 2.971
Harvesting 45.3 39.7 3.193
Transporting 9.2 17.0 1.972
Threshing 8.1 30.0 3.942
* In every case, the gender-labour difference is statistically significant at the 5 per cent levels.
** There is no statistical difference in labour contribution for farm operations by women in the 3 zones.

Table 3: Results of Regression Analysis Showing the Gender-Labour Effects on

Farm Production
Variable Regression t-value* Level of significance
coefficient
Constant 3.612 4.205 0.05
Seed 0.085 1.011 NS
Fertiliser 0.573 1.204 NS
Men labour 0.854 3.522 0.05
Women labour 0.610 2.823 0.05
R2 0.63
NS = Not significant.
* In every case, the gender-labour difference is not statistically significant at the 5 per cent levels.

3.2 Gender-labour productivity in farm production

The regression results in Table 3 reveals that the percentage variation in farm output as
explained by the variables in the productivity model was 63 per cent. The regression
coefficients for both men and women labour were significant at 5 per cent level.

62
Agricultural productivity referred to the ratio of farm outputs to farm inputs used.
Marginal productivity measures the extra output produced as a result of a unit increase in
the farm input. Productivity could be assessed in physical or monetary terms. In the
monetary terms as used in this study, marginal value productivity of men labour on farms
in Nasarawa State was greater than that of women. The gender-labour differential in
productivity was estimated as N7.29k per man-hour (Table 4). This implies that every
one hour spent by men on farm works produced more extra output compared to the case
of women.
Table 4: Gender-Labour Marginal Productivity Differentials (N/Man-Hours) in
Farm Production
Estimate Men Women*
Minimum 18.34 11.75
Maximum 52.61 48.12
Average 26.73 19.44
St. deviation 5.64 8.32
CV (%) 21.10 42.80
Differential 7.29
CV = Coefficient of variation.
* There is no statistical difference between the productivity of women in the different zones/regions.

This phenomenon could be attributed to the fact that men have more access to productive
resources compared to the women. Also, it has been reported in several studies (Gabriel
1991; Rahman and Usman 2004) that women’s working day is, on the average, longer
than that of men because they carry major responsibility for both farm production and
domestic works. This could negatively affect their labour productivity in farm
production.
3.3 Gender-based accessibility to farm resources
The fact is that in the study area, women generally do not control a lot of finance in the
households, and their access to and ownership of productive resources are usually
limited. Most households (92%) are headed by men and with their superior financial
status, they own and control the means of production. This study observed very low
accessibility rate (11.5%) of women to productive resources in Nasarawa State (Table 5).
The women poor financial status and poor access to credit facilities do not allow them to
acquire farm inputs.
The socio-economic factors that influenced women accessibility to farm resources are
shown in Table 6. The variables included in the regression model explained 65 per cent
of variation in the rate of women accessibility to farm resources in Nasarawa State.
Women involvement in farm decision making, the value of their contribution to farm
production, their income level and number of children were the significant factors that
positively determine the rate of their accessibility to farm resources.

63
Table 5: Gender-based Differences in Accessibility to Farm Resources
Rate of Accessibility (%)
Farm Resource Women Men Disparity
Farm land 17.25 82.75 65.50
Labour 13.36 86.64 73.28
Fertiliser 9.14 90.86 81.72
Credit 6.18 93.82 87.64
Other farm inputs 11.55 88.45 76.90
Average 11.50 88.50 77.00
Note: Other farm inputs include seeds (local or improved) and chemicals (herbicides or/and insecticides).

Table 6: Factors Determining Women Accessibility to Farm Resources

Regression t-value Level of
Factors coefficient significance
Constant 0.412 2.940 0.05
Farm decisions 0.305 3.017 0.05
Value of contribution 0.249 2.876 0.05
Cooperative participation 0.062 1.162 NS
Number of children 0.193 3.604 0.05
Educational status 0.029 1.007 NS
Age 0.056 1.249 NS
Income level 0.118 3.062 0.05
R2 0.65
NS = Not significant at 5 per cent level.

3.4 Gender-based involvement in farm decision making

As women are the key resources in farm production, their productivity depends on the
rate of their involvement in farm decision making. This study observed that women
enjoyed about 43 per cent rate of involvement in farm household decisions
Table 7: Gender-based Differences in Involvement in Farm Decision Making
Rate of accessibility (%)
Decisions with regards to: Women Men Disparity
Selection of enterprises 25.75 74.25 48.50
Inputs procurements 27.15 72.85 45.70
Inputs allocations 32.03 67.97 35.94
Selling of produce 48.12 51.88 03.76
Consumption of produce 51.50 48.50 03.00
Processing of produce 64.82 35.18 29.64
Storing of produce 50.15 49.85 00.30
Average 42.79 57.21 14.42

64
(Table 7). Women had above 50 per cent involvement rate in farm decision with regards
to consumption of produce, and below 50 per cent for selection of enterprises, inputs
procurements, inputs allocation and selling of produce. This finding compares favourably
with that of Mbatia (1998) as well as Hanger and Morris (1973).The two studies have
shown that women contribution to farm decision making is less than 50 per cent. It is
significant to note that decisions and responsibilities taken by women are related to the
farm tasks they perform.
3.5 Responsiveness of women satisfaction in farm production
The study identified farm size, household size and share of farm income as the significant
factors determining women satisfaction in farm production. These factors have positive
coefficients that are significant at 5 per cent level (Table 8). The marginal satisfaction of
the factors was estimated to understand the extent of influence of each of the factors on
women satisfaction. The rate of change in women satisfaction as induced by unit change
in a factor was highest for the share of farm income gained by the women, followed by
farm size and household size. In terms of elasticity measurement, women satisfaction
seems to be highly responsive to farm size, followed by household and cooperative
participation.
Table 8: Logit Estimates and Women Satisfaction in Farm Production
Regression t-value Marginal Elasticity of
Factors coefficient satisfaction satisfaction
Farm size (X1) 0.0724 1.985* 0.018 0.094
Household size (X2) 0.0361 2.116* 0.009 0.056
Share of farm income (X3) 0.0945 2.574* 0.024 0.003
Educational status (X4) 0.0075 1.022 0.002 0.016
Credit (X5) 0.0214 1.311 0.005 0.001
Cooperative participation (X6) 0.0322 1.068 0.008 0.050
Constant (a) 0.0577 1.426 NA NA
- 2log likelihood = 53.629
probability = 0.48
* = significant at 5 per cent level.
NA = Not applicable.

4.0 Conclusion
Gender analysis usually suggests appropriate type of intervention required to improve a
given system. The gender-based differences in agricultural activities, resources and
benefits being revealed in studies of this kind are supposed to alert the policymakers to
potential gender-differentiated outcomes for the purpose of reducing gender inequality in
the societies. This study revealed that women have extensive work loads with dual
responsibility for farm household production. As their work is getting harder and more
time-consuming due to ecological degradation and changing agricultural technologies,
they have not been given the opportunity to realise their full potential in terms of their

65
labour productivity. Women are less involved in farm decision making, less accessible to
productive resources and less satisfied with their participation in farm production. For
sustainable agricultural development, women are to be perceived as the subjects of
development and their contribution to agricultural production should be given adequate
recognition.

References
Audu, S.I., and A.O. Onoja. 2005. Economic Analysis of Activities of Women in Agricultural Production in
Ankpa local Government Area, Kogi State Nigeria.Production Agriculture and Technology 1(1): 9-16.
Gabriel, T. 1991. The Human Factor in Rural Development. London: Belhaven Press.
Hanger, J. and J. Morrris. 1973. Mwea: An irrigated Rice Settlement in Kenya. In Chambers, R. and J. Morris
(eds):Africa Studies No. 83 IFO.Munchen.
Ingawa, S.A. 1999. Welcome Address at the National Workshop for Women in Agriculture, held in FACU
headquarters, Sheda, Abuja, Nigeria. 31August-2 September.
Malena, C. 1994. Gender Issues in Integrated Pest Management in African Agriculture. NRI Socio-Economic
Series. National Resources Institute. Chatham.
Mbatia, O.L.E. 1998. Economic Analysis of Irrigated Horticultural Production: A Case Study of Gender Issues
in Kenya. In Laubseher, J.(ed) Farm and Farmer Organization for Sustainable Agriculture in Africa.
Proceedings of the 4th Biennial Conference of the African Farm Management Association. 26-30 January,
1998. University of Stellenbosch, South Africa.
Mgbada, J. K. 2000. Production of Staple Crops by Rural Women in Enugu and Ebonyi States: Lessons for
enhancing poverty alleviation programmes. In T.A. Olowu edited Agricultural Extension and Poverty
Alleviation in Nigeria. Proceedings of the 6th Annual National Conference of the Agricultural Extension
Society of Nigeria. Pp. 10-12.
Mirer, T.W. 1990. Economic Statistics and Econometrics. Second Edition. New York: Macmillan Publishing
Company.
Morna, C.L. 1990. Women Farmers emerge from the shadows. African Farmers, No. 3. April.
Pindyck, R.S., and D. L. Rubinfeld. 1976. Econometric Models and Economic Forecasts. New York: McGraw
Hill Book Company.
Quisumbing, A. 1994. Gender Differences in Agricultural Productivity: A survey of empirical evidence.
Discussion Paper,Series No. 36.Washington DC: Education and Social PolicyDepartment, World Bank.
Rahman, S.A., and J. F. Alamu. 2003. Estimating the Level of Women Interest in Agriculture: An Application
of Logit Regression Model. The Nigerian Journal of Scientific Research, 4 (1): 45-49.
Rahman,S.A., J. Gabriel, and N. D. Marcus. 2004. Gender Differentials in Labour Contribution and
Productivity in Farm Production in Kaduna State of Nigeria. The Journal of Family Development, 1(2):12-
21.
Rahman, S.A., and J. I. Usman. 2004. Corparative Analysis of Women Participation in Agricultural Production
in Northern and Southern Kaduna State, Nigeria. In O. O. Agbede, M. B. Idris, S. A. Rahman, M. A. Ari; I.
M. Ogara, and G. N. Asumugha edited Mobilizing Investors for Sustainable Agricultural Research,
Deveopment and Production in Nigeria. Proceedings of the 38th Annual Conference of the Agricultural
Society of Nigeria. Pp.103-113.
Rahman, S.A., and F. A. Ajayi. 2005. Gender-based Analysis of Accessibility to Resources in the Farming
Communities of Kaduna State, Nigeria. Paper presented at the Maiden Academic Seminar organised by the
Academic Staff Union of Polytechnic, held at Nasarawa State Polytechnic, Lafia. 26-28 July, 2005.
Saito, K. A. 1995. Raising the productivity of Women Farmers in Sub-Saharan Africa. In J. P. Srivastava and H.
Alderma edited Agricultural and Environmental Challenges. Proceedings of the 13th Agricultural Sector
Symposium. World Bank. Pp. 147-152.

66
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Economic Evaluation of Rice-Prawn Gher Farming

System on Soil Fertility for Modern Variety of
Rice Production in Bangladesh
Basanta Kumar Barmon*, Takumi Kondo* and Fumio Osanami*

Abstract
The present study attempts to examine the economic evaluation of rice-prawn gher
farming system on soil fertility for modern variety (MV) rice production in Bangladesh.
Laboratory based experimental data and field survey data were used in the present
study. To test the change in soil fertility, soils were collected before (after prawn
production) and after rice production (before prawn production) and analysed in the
Soil Resources Development Institute (SRDI) laboratory in Khulna, Bangladesh. The
findings of the study indicated that the farmers applied comparatively less chemical
fertilisers in MV rice production under the rice-prawn gher farming system compared to
usual year-round MV rice production. The cost of fertilisers for MV rice production was
about seven times higher than MV rice production under the rice-prawn gher farming
system. However, per hectare MV rice yield under farm condition was almost similar to
MV rice yield under the rice-prawn gher farming system. The leftover feeds for prawn
production and faeces of prawn and fishes have significant impacts on soil fertility for
MV rice production. The rice-prawn gher farming is a cost-saving technology for MV
rice production.

1.0 Introduction
Rapid population growth, increased food demand, and urbanisation are the main causes
that have created tremendous pressure on agricultural land, making it an increasingly
scarce resource (Bhuiyan 2001). As a result, agricultural land per capita is decreasing
over the years in Bangladesh (Alexander et al. 1998; Akteruzzaman 1998). Food
insecurity is a critical concern since the independence of Bangladesh. Despite a
noticeable improvement in rice production in recent years still it falls short of attainable
levels. The government of Bangladesh imports a large amount of rice from aboard to
meet up the domestic demand (Zaman 2001). Proper soil and crop management with high
yielding modern varieties (MV) of rice is the logical way to raise the gross production to
meet up the domestic demand for food using limited land resources (Islam et al. 2002).
The rice-prawn gher farming system that locally known as gher revolution, (Kendrick
1994) diffused rapidly compared to ‘Green Revolution’ in Bangladesh, mainly because
of indigenous natural resources used at the early stage (Barmon et al. 2005). This farming
has positive impacts on labour market and increased daily labour wage compared to MV
boro and local aman rice (Barmon et al. 2004c). The agricultural and household income
of rice-prawn gher farmers are more than seventeen and two times higher compared to
MV and local aman rice farmers (Barmon et al. 2004a, 2004b). The shrimp gher farming

*
Laboratory of Development Economics, Department of Agricultural Economics, Graduate School of
Agriculture, Hokkaido University, Japan. Corresponding e-mail: bkbarmon@yahoo.com

67
has negative impacts on environments and ecology in the coastal region in Bangladesh
(Asaduzamman et al. 1998; Nijera Kori 1996; and Sobhan 1997; Bhattacharya et al.
1999; Datta 2001), whereas the impact of rice-prawn gher farming on the environment is
ambiguous. Ali (2004) conducted a study on the impact of the cultivation of High
Yielding Variety (HYV) of rice using modern inputs such as power tillers, low-left
irrigation pumps, chemical fertilisers and pesticides during the period from 1985 to 2000.
Ali (2004) concluded that the modern inputs, especially the chemical fertilisers and
pesticides, have negative impacts on soil quality and land degradation. Information on
the impacts of rice-prawn gher farming system on soil fertility of MV rice yield in the
southwest Bangladesh is quite insufficient. Therefore, the present study has been
conducted to the economic impacts of rice-prawn gher farming system on soil fertility of
MV rice field.
2.0 Material and Methods
2.1 Soil sampling
The present research was conducted at Bilpabla and Chanchra villages, which are under
Khulna and Jessore districts, respectively, of Bangladesh in order to assess the impact of
rice-prawn gher farming system on soil fertile of MV rice field (Figure 1). Bilpabla
Village was selected purposively because it is one of the typical villages in rice-prawn
production. Chanchra Village was also selected purposely because the cropping pattern
of Chanchra Village was similar to Bilpabla Village prior to the gher farming system.
Moreover, Chanchra Village is only 60 km far from Bilpabla Village. The altitude level
of rice fields in Chanchra Village is slightly higher than the land for gher farming in
Bilpabla Village. As a result, the farmers in Chanchra Village cannot convert their
rice field into gher farms. The farmers in Chanchra Village mainly grow MV rice
twice a year.
A total of sixty-five farmers (32 from rice-prawn gher farming and 33 from year-round
MV rice farming) were randomly selected for this study. Laboratory-based experiment
and farm surveys were conducted in this study. Soil samples were colleted only from
gher farming. The soil samples used for the research were collected from 50 gher plots of
32 farmers from different locations before (after prawn production) and after MV rice
cultivation. Soil sample collection procedures were conducted in two phases. First phase
was conducted in the end of December (after prawn production) in 2004 and the second
phase was in the mid of May in 2005 (after paddy production). Each sample of soil was a
well mixture of nine sub-samples those were collected from nine different places of a
particular gher farm. The soils were taken from 0-15 cm depth, which represented the
cultivated topsoil (BARC 1997). After collecting soils, the sample soils were placed in
polythene bags and well dried by natural sunshine. After well dried the soils were again
placed in polythene and levelled numerically and sealed for transportation to the
laboratory for testing. As the present study is conducted to explore the impact of rice-
prawn gher farming on soil fertility, therefore, the samples as well as plots number were
identified with same numerical value before (after prawn production) and after the
production of paddy (before prawn production).

68
Figure 1: Rice-Prawn Gher Farming Area (Study Area)

Map of Bangladesh

Rayermahal village

Bilpabla village •Dhaka

•Chuadanga
MV paddy farming
area (soil nutrient)
•Khulna

Rice-prawn gher Bay of Bengal

farming area

Note: ・indicates the soil sampling rice-prawn gher farms

River:

Figure 1. Rice-prawn gher farming area (study area )

Before testing the nature of plant nutrients, the sample dried soils passed through a 2-mm
mesh sieve to remove stones and plant roots and prepared for saturated extracts. The
National Soil Resources Development Institute (SRDI) of Bangladesh developed
standard laboratory to determine the major plant nutrients such as N, P, K, Ca, Zn, S, Mn,
B and Cu.
2.2 Soil analysing methods
Total nitrogen (T-N) was determined by dry combustion method (Sumigraph NC-1000,
Chromatograph machine). The pH of the soil was measured by a glass electrode pH
meter using soil: water suspension ratio of 1:2.5 as described by Jackson (1962).
Potassium (K) and phosphorus (P) were analysed by ammonium acetate extraction
method (Barker and Surh, 1982) and Olsen’s method (Olsen and Somners 1982),
respectively. Sulphur (S) and Zinc (Zn) Cupper (Cu) and Manganese (Mn) were analysed
by Turbidimetric and DTPA extraction method (Baker and Amacher 1982). Calcium
(Ca) and Boron (B) were analysed by extractant composition (Haby et al. 1990) and
Truog method (Berger and Truog 1940).
2.3 Farm survey
The present study collected various combinations of prawn feeds that the selected gher
farmers used in prawn production cycle in 2004 through a comprehensive cross-sectional
field survey, and these were also analysed in the laboratory. Along with soil samples, the

69
information on applied chemical fertilisers of MV rice field and MV rice yield was
collected both from rice-prawn gher and year-round MV rice farming villages.
3.0 Rice-Prawn Gher Farming System in Bangladesh
In Bangladesh, two types of gher farming are operated: one is brackish water based
shrimp culture, and another is fresh water based rice-prawn culture. Shrimp gher farming
is large in size and scale, and needs saline water, whereas prawn gher farming is
comparatively small in size and scale, and needs fresh water.
The term rice-prawn gher refers to a modification of rice field that has been used for
prawn and rice cultivation. The main land (locally known as Chatal) of gher is
surrounded by high wide dikes and canals that lie in the periphery of the dikes (Figure 2).
The whole area of gher remains under rain-water from June to December, when the
farmers cultivate prawn (Macrobrachium rosenbergii) and carp fishes. During the period
from January to April, the land becomes dry but sufficient water still remains in the
canal. As a result, farmers can use this water to grow MV boro rice on chatal. Moreover,
farmers also grow vegetables on the dikes throughout the year.
Figure 2: Diagram of Rice-Prawn Gher Farming System and Soil Fertility
Process for MV Paddy Production
S u p p lie d A v a i la b i l it y o f
p ra w n fe e d R id g e n u t r ie n t s fo r M V
p a d d y p r o d u c t io n

D ik e D iikk ee
D
C aa nn aa ll
C

P P oPno d
n od
Dike
Dike
Dike

Canal
Canal

Dike
Canal
Canal

Dike
Dike

PP aa dd dd yy ffiiee lldd
Dike

P P ondo
nd

C aa nn aa ll
C
D ik e D iikk ee
D

L e ft o v e r fe e d L e ft o v e r fe e d c o n v e r t s o il
n u t r ie n t s fo r M V p a d d y
p r o d u c t io n

a ) G h e r fa r m in g in r a in y s e a s o n b ) G h e r f a r m in g in w in t e r s e a s o n

F ig u r e 2 . D ia g r a m o f r ic e - p r a w n g h e r f a r m in g s y s t e m a n d s o il f e r t il it y p r o c e s s f o r M V
p a d d y p r o d u c t io n .

70
3.1 Types of feeds and feeding system
At the early stage of rice-prawn gher farming (1985-1995), farmers mainly applied meat
of mud snails as feed for prawn. When the whole wetlands (swamplands) converted into
gher farming, mud snails were disappeared gradually from these areas. Farmers are now
importing mud snails from neighbouring districts as well as India. As a result, the price
of mud snails has increased more than five times from within 15 years. At present, the
farmers are applying various combinations of feed to prawn from their own experiences.
Farmers apply two categories of feeds during the whole production cycle of prawn.
Category-I includes eggs, semai (made from fine flour of wheat), stater-1 and sugi (made
from fine flour of wheat). These feeds are applied to prawn until 45 days from releasing
to gher pond (table 1). Category-II of prawn feed includes meat of mud snail, fishmeal,
soya bean oilcake, mustard oilcake, coconut oilcake, broken maize and rice, husk of
wheat, polish of rice, chira, and pulses etc. and provides to gher as prawn feed from 45
days to harvesting (December).
Table 1: Types, Application Times and Retail Market Price of Various Feeds for
Prawn Production in 2005
Types of feeds Retail market price Types of feeds Retail market price
Category-I (0 to 45 days) Category-II (After 45 days to harvesting)
Eggs 4 Tk./Egg Meat of mud snail 9 Tk./kg
Semai 25 Tk./kg Fishmeal 20 Tk./kg
Starter-1 45 Tk./kg Soya been oilcake 22 Tk./kg
Sugi 25 Tk./kg Mustard oilcake 15 Tk./kg
Coconut oilcake 12 Tk./kg
Broken maize 14 Tk./kg
Broken rice 12 Tk./kg
Husk of wheat 13 Tk./kg
Polish of rice 7 Tk./kg
Chira* 20 Tk./kg
Pulses 25 Tk./kg
Notes: 1) 1US$ = 68.85 Taka, May, 2006.
2) * made from rice.
Source: Field survey 2005.

The farmers generally apply feed twice a day- morning and evening. Prawn farmers use
three types of supplementary feed: processed feed, homemade feed and snail. The
ingredients of homemade feed include oil cakes, polished rice, molasses, wheat bran,
wheat, fishmeal and other feeds in fixed proportions from their own experiences. Some
farmers also use oil cake, wheat bran, and boiled rice directly without mixing. The meat
of mud snails is commonly used as prawn feed. The farmers use these feeds during the
whole cycle of prawn production.
The farmers apply feed as semai (made from wheat bran), starter-1, mixed feed, broken
rice, chira (made from rice), wheat bran, mastered and soya bean oil cake, snail, fishmeal

71
and pulses at the different growth stages of the prawn (Table 2). The farmers do not
supply any additional feed for carp fish because the carp fish uses leftover prawn feed.
Table 2: Descriptive Statistics of per Hectare Feeds, Fingerlings of Prawn and Carp
Fishes under Rice-Prawn Gher Farming, 2004
Minimum Maximum Mean Standard Used numbers
deviation of gher plots
Fingerlings
Prawn fingerlings (No.) 7,410 39,520 21,905 7,245 50
Fish fingerlings (kg) 0 123.5 33.5 28.8 49
Feeds
Mud Snail (kg) 0 5,928.0 1,733.0 1,515.0 45
Fish meal (kg) 0 2,223.0 573.8 507.1 42
Pulses (kg) 0 1,111.5 277.7 315.4 31
Cooked rice (kg) 0 988.0 71.6 177.2 15
Oilcake (kg) 0 617.5 118.2 173.7 30
Chira (kg) 0 889.2 80.7 158.3 38
Wheat bran (kg) 0 2,074.8 544.1 585.8 38
Others (kg) 5 6,527.0 302.0 937.0 50
Source: Field survey 2005.

3.2 Ploughing system of rice field of rice-prawn gher farming system

Generally, rice fields are ploughed several times before transplanting of seedlings, using
power-tiller or bullock or tractor. But in case of rice-prawn gher farming system, lands
are ploughed after harvesting of rice. The main reasons are that after prawn harvesting,
the rice fields (mid-field of gher farming) are not dry enough for ploughing or sometimes
the field retain small amount of water which is also suitable for plantation. Moreover,
soils of the mid-fields become muddy for transplanting because during prawn harvesting,
these soils are well mixed. However, a small number of rice fields are also cultivated
before transplanting which are located in comparatively high altitude level and near to
roadside.
3.3 Irrigation system for MV boro rice production
The canals of rice-prawn gher farming system retain sufficient water for irrigation of MV
paddy production. The rice fields of rice-prawn gher farming are irrigated from canal
using indigenous manmade tools such as doon and sewing basket. In some areas, shallow
tube-wells are also used to irrigate the large area gher farm. At present, the farmers dig
new canals inside the gher plots and filled up the old canals using the soil of new canals.
This practice of management makes the soil fertility for the good MV rice as well as
prawn production. It is farmer’s belief, if the position of canals shifts after every 3-4
years, the soils become more fertile for crops production.

72
3.4 Prawn harvesting and soil fertilisation
Usually farmers harvest prawn from gher using cast-net or draw-net five to six times
from October to end of December in a production cycle. Draw-net is used for the larger
gher farms and cast-net for comparatively small gher farms. At least six labours are
needed for cast-net and more that eight labours are needed for draw-net depending on the
length of the draw-net.
Prawns try to hide in mud during harvesting by nature. To observe the hiding nature of
prawn in mud, two labours pull the rope of net from both sides, and remaining labours
harvest prawn by hand from clay behind the nets. When the labours search prawn in mud,
the soils becomes upside down smoothly. Moreover, some aquatic habitats and various
varieties of algae are grown on the bottom of gher farms during prawn production, and
these aquatic habitats and algae are mixed with soil during harvest of prawn. These
aquatic habitats and algae are rotten naturally and make the land fertile for MV
rice production.
4.0 Cropping Patterns
Currently three types of rice are being produced in Bangladesh in three distinct seasons:
aus (April to August), transplanting aman (T. aman) (August to December), and boro
(January to April). Among them, aus and T. aman rice are produced in rain-fed water,
and MV boro rice is produced in irrigated water (ground water or rivers and canals).
Irrigation and chemical fertilisers are not used in local aus and T. aman production
because the rice fields are submerged. Farmers transplant MV boro rice from mid-
January to mid-February, and harvest from mid-April to mid-May. Farmers usually use
chemical fertilisers, pesticides, and irrigation for boro rice production. Along with rice
crops, farmers also cultivate oil seeds, potatoes, and vegetables in the comparatively high
land during the winter season.
The cropping pattern of the study villages is presented in Figure 3. Before the advent of
gher farming in the study village, farmers cultivated broadcasting aus and aman rice in
swampland, and transplanting aman rice in the upper lands. The familiar broadcasting
aus and aman rice have almost disappeared mainly because of siltation of inland rivers
and canals, embankment of rivers, and environmental changes. Oil seed crops such as
rape, mustard and sesame were also produced in the comparatively high-altitude land
located on the riverside. The life cycle of broadcasting aman was longer than that of
broadcasting aus rice though the sowing time is the same for both types. The sowing time
of aus and aman rice is in April/May; harvesting of broadcasting aus takes place in
August while harvesting of broadcasting aman takes place in December. The farmer
sows aus and aman seeds together in April/May because after June/July the whole area is
submerged under water due to heavy rain, and at that time it is not possible to transplant
aman (T. aman). This system of producing local aus and floating aman rice together is
known locally as Domuti.

73
Figure 3: Changes in Cropping Patterns after Introduction of
Rice-Prawn Gher Farming
Crops Months
Jan Feb Mar April May June July Aug Sept Oct Nov Dec
Before Gher Farming
Swampland:
Aus paddy
Aman paddy
High Land:
Aus paddy
Aman paddy
Rape/Mustard/Sesame
After Gher Farming
Prawn
Carp fish
Boro paddy
Vegetables
High Land:
Boro paddy
Aman paddy
Notes: indicates the period up until the sowing paddy, and releasing of prawn and fish is carried out.
indicates harvesting time starts.
Source: Field survey 2006.

The gher farming system has dramatically changed the cropping patterns in the study
area (Figure 3). The construction of rice-prawn gher farming has created opportunities
for crop diversification. Along with prawn and fish, farmers can now cultivate MV boro
rice in the central field and vegetables on the dikes of the gher, mainly for domestic use.
Prior to the gher farming, the farmers used to cultivate oil seeds such as rape, mustard
and sesame after the harvest of local broadcasting aman rice. However, the gher farmers
are not able to cultivate oil seeds due to the physical construction of gher farming. The
gher farming system has increased vegetable production compared with the past. The
farmers have also planted fruit trees (coconut, mango, guava, jackfruit, banana, papaya,
etc.) on the dikes. The production period of prawn and fish is from May/June to
December/January; the MV boro rice is from the end of January to the end of April, and
vegetables are produced throughout the year.
5.0 Effects on Prawn Production of Soil Fertility
The effects of rice-prawn gher farming on soil nutrients are presented in Table 3. It
appears from the table that all micro- and macro-nutrients of soil have decreased after
rice production. However, the concentration of all soil nutrients was higher before rice
production (after prawn production) than after rice production (before prawn production).
The same soil nutrients content varied widely before and after rice production, mainly
because of the location of gher plots and soil textures (Table 3). The availability of all

74
nutrients also significantly varied among the samples before and after prawn production
because the farmers apply various combinations of feeds during prawn production. The
changes in different soil nutrients content are discussed below:
Soil pH
The term pH refers to the alkalinity or acidity of a growing media water solution. This
solution consists of mineral elements dissolved in ionic form in water. Availability of
essential plant nutrients depends on the soil pH (Rahman and Ranamukhaarachchi 2003).
Soil pH and base saturation are the important chemical properties that influence soil
nutrient availability, plant growth (Fageria and Baligar 1998) and the activities of soil
micro-organisms and organic matter decomposition (Mengel and Kirkby 1982). All
plants are not able to tolerate acidic or alkaline soils. Most of the field crops prefer
neutral or slightly acidic soils because it promotes the solubility of micro- and macro-
nutrients for the plant growth and development. Rice crops usually prefer slightly low-
acidic soils compared to other crops. The mean soil pH rice field before and after rice
production was 7.69 and 7.05, respectively. This implies a good indicator of suitable MV
rice production.
Soil organic matter and C:N ratio
Soil organic carbon, being the biggest part of the soil organic matter (SOM), is
considered perhaps the most important indicator of soil quality and productivity. SOM
affects soil structure, water storage capacity and nutrient supply. It helps in maintaining
the soils’ tilth, aiding infiltration of water and air and reducing erosion. However, it
should be noted that the importance of these benefits differs by region and soil type. The
organic carbon content does not alone determine soil fertility but becomes important in
the context of the carbon/nitrogen (C/N) ratio of a soil. The present study has analysed
only the carbon/nitrogen ratio to represent the proxy of soil organic matter (SOM). The
C:N ratio of soils were almost the same before and after rice production. On an average,
high concentration of C (before rice production: 71g/kg and after rice production: 72g/kg,
table 4) in the soil of rice fields under gher farming is higher than other rice farming
(year-round rice production: 61g/kg: experimental result). This indicates that the gher
farming system has significant impacts on the soil fertility for MV rice production.
Nitrogen
Nitrogen is a major component of proteins, hormones, chlorophyll, vitamins and
enzymes essential for plant life. It is the most important soil element for crop production.
The availability of optimal nitrogen for crop production influences the crop yields, and
deficiencies reduce yields. Nitrogen losses from soils have negative impact on water and
air quality (Tisdale et al. 1993; Pierzynski et al. 1994; Foth and Ellis 1997). Leaching,
surface runoff, dentrification and ammonia volatilisation are the main ways to losses
nitrogen (Cai et al. 2002; Kyuma 2004). The nitrogen content of soils was varied widely
among the rice fields before and after rice production of gher farming in the same study
village. The nitrogen content of soil has decreased after rice production (Table 3).
Moreover, a part of the nitrogen content was uptaked by rice during growth
stage (Table 3).

75
Table 3: Descriptive Statistics of Soil Fertility Status (0-15cm) in Rice-Prawn Gher
Farming System in Khulna District, Bangladesh
Rice-Prawn Gher Farming
Particulars Before rice production After rice production
Mean SD Min Max Mean SD Min Max
pH 7.69 0.746 5.0 9.1 7.05 0.424 6.2 7.9
Nitrogen (N%) 0.46 0.118 0.29 0.75 0.37 0.102 0.19 0.61
mg/100g soil
Potash (K) 1.25 0.403 0.62 2.89 0.97 0.23 0.61 1.60
Calcium (Ca) 29.9 6.41 16.25 49.05 23.23 4.33 14.75 32.75
µg/g soil
Phosphorus (P) 12.23 7.73 2.26 37.00 9.56 5.89 2.16 34.00
Zinc (Zn) 2.17 1.09 0.80 6.8 1.59 0.937 0.60 5.82
Sulpher (S) 232.5 95.1 49.1 410.6 176.2 80.4 49.1 381.6
Manganese (Mn) 108.48 38.2 39.4 212.0 88.24 22.5 50.0 173.6
Boron (B) 2.87 0.655 1.75 4.23 2.05 0.549 1.12 3.51
Copper (Cu) 12.54 3.71 5.42 21.81 10.64 2.11 4.42 14.42
g/kg
Total Nitrogen (N) 5.566 1.902 2.4 9.5 5.437 2.046 2 9.5
Total Carbon (C) 71.1 30.25 31 144 72.27 30.73 22 144
Ratio
C/N ratio 12.517 1.214 11 15 13.067 1.015 11 15
Notes: 1) Soil sample size was 50 from 32 gher farmers.
2) SD indicates standard deviation.
Source: Laboratory experiment 2005.

Phosphorous
Phosphorus is an essential nutrient for plant growth. It is also necessary for seed
germination, photosynthesis, protein formation and metabolism in plants. It is essential
for flower and fruit formation. Deficiency symptoms are purple stems and leaves;
maturity and growth are retarded. Yields of fruit and flowers are poor. Premature drop of
fruits and flowers may often occur. In general total P content in soil varies the range
between 200-1500mg/kg (Mengek and Kirkby 1982). In the study area, the P content is
higher than the general P content level in the soil. The P content has decreased significant
(significant at 1% level) after rice production.
Potassium
Potassium is necessary for formation of sugars, starches, carbohydrates, protein synthesis
and cell division in roots and other parts of the plant. It helps to adjust water balance,
improve stem rigidity and cold hardiness, enhance flavour and colour on fruit and
vegetable crops, increase the oil content of fruits, and is important for leafy crops.
Deficiencies result in low yields, mottled, spotted or curled leaves, scorched or burned
look to leaves. The K content has varied widely among the soil sampling rice fields like

76
N and P. The K content has decreased significantly (significant at 1% level) after rice
production like N and P.
Other soil nutrients
Other soil nutrients such as Ca, Zn, S, Mn, B and Cu have varied widely among the rice
fields and decreased significantly (significant at 1% level) like NPK after rice production.
5.1 Analysis of applied feeds to prawn production
The present study has only analysed N, C contents and C:N ratio of applied feeds for
prawn production because N and C contents are the main factors for soil fertility among
the macro- and micro-nutrients and these are presented in Table 4. The table shows that
the contents of C, N and C:N ratio of applied feeds to prawn production were higher than
the contents of soil samples before and after rice production. The contents C, N and C:N
ratio were higher than the soil samples. As a result, the leftover feeds and faeces of
prawn and fishes enhance the soil fertility of rice fields.
Therefore, it can be concluded from the Table 3 and 4 that the rice-prawn gher farming
system has significant impact on soil fertility for MV rice production.
Table 4: Nitrogen (N) and Carbon (C) Contents Applied Feeds of Prawn Production
Name of the feeds N (g/kg) C (g/kg) (C:N) ratio
Starter-1 50.7 394.5 7.8
Wheat noodles 19.6 445.3 22.7
Fish meal 42.7 415.4 9.7
Wheat flour 20.4 442.3 21.7
Beaten rice 13.5 443.0 32.7
Maize grain 15.0 443.9 29.6
Legume grain 48.5 445.6 9.2
Broken rice 19.3 418.9 21.7
Oil cake (mustard) 57.3 459.7 8.0
Oil cake (soya bean) 75.5 448.3 5.9
Wheat husk 29.3 447.1 15.3
Wheat bran 25.4 465.3 18.3
Meat of mud snail 95.9 400.5 4.2
Source: Experimental results 2006.

5.2 Application of chemical fertiliser for MV paddy production

Chemical fertilisers such as urea, triple super phosphate (TSP), muriate of potash (MP),
gypsum, and zinc sulphate are commonly used in MV of paddy production in Bangladesh
(BRRI 1997). Usually the farmers do not use any chemical fertilisers except homestead
manure and cow dung for local aus and aman paddy production. However, the farmers
apply various types of chemical fertiliser for MV aman and boro paddy production.
Application of chemical fertilisers for MV production and MV rice production under
rice-prawn gher farming system are exhibited in Table 5.

77
Table 5: Per Hectare Chemical Fertiliser Use in MV Rice Field and MV Rice
Yield between Two Farming Systems
Application of chemical fertilisers
Farming system Urea TSP MP Gypsum Rice yield
(kg) (kg) (kg) (kg) (kg)
Paddy production under gher farming (GF) 51.3 35.3 6.2 12.8 4,252
MV Paddy farming (PF) 250.9 182.3 79.7 214.2 5,014
Ratio (PF/GF) 4.89 5.16 12.85 16.73 1.17
t-statistics 16.28* 11.16* 15.64* 10.44* 4.48*
Notes: 1) TSP and MP indicate Triple super phosphate and Muriate of potash, respectively.
2) One mound equivalent to about 40 kg in the locality
3) * 1% level of significance
4) Sample size of gher, and MV paddy farming were 32 and 33, respectively.
Source: Field survey 2005.

The figures in the table shows that the farmers apply, on an average, 250.9 kg of urea,
182.3 kg of TSP, 79.7 kg of MP, and 214.2 kg of gypsum per hectare year-round MV
rice production, and only 51.3 kg of urea, 35.3 kg of TSP, 6.2 kg of MP, and 12.8 kg of
gypsum per hectare MV rice production under rice-prawn gher farming system. In other
words, the farmers use more chemical fertilisers in MV rice production compared to MV
rice production under rice-prawn gher farming system, and the application rates of all
chemical fertilisers in per hectare rice production are significantly differ from farmers to
each other. The main reason is that the farmers apply various combination of feed into
gher plots during prawn production (see Table 2). However, the prawns do not eat the
supplied feeds properly. The leftover feed during prawn production and faeces of prawn
and fish make the paddy field fertile and the rice crop takes necessary nutrients from the
fertile field. In addition to this, some aquatic habitats are grown during the prawn
production, and these aquatic habitats are used as composed manure to rice production
under gher farming system. Therefore, it can be assumed that the leftover feeds, faeces of
prawn and fish, and the aquatic habitats make the soil fertile that reduces the application
of chemical fertilisers in MV rice production under rice-prawn gher farming system.
Rice yield per hectare of two agricultural systems is also presented in Table 5. The
figures in the table indicate that the yield rice is higher in year-round rice farm than rice-
prawn gher farming system. It was observed during the farm survey that per hectare rice
yield of some rice plots under rice-prawn gher farming system was higher than year-
round MV rice farm. However, on an average, the per hectare rice yield was lower in
rice-prawn gher farming system than year-round MV rice farming. One of the main
reason is that the gher farmers are not serious about rice production like prawn
production because rice is not profitable crop like prawn production. The farmers mainly
produce rice for home consumption.
5.3 Comparison of fertilisers’ cost between two agricultural systems
Input cost per hectare of chemical fertilisers for MV rice production under rice-prawn
gher farming system and year-round MV rice farming is presented in Table 6. It is
revealed that the gross costs for fertilisers are Taka 5,488 and Taka 819 for MV rice

78
production under the gher farming system and year-round rice farming. This implies that
the gross input cost of fertilisers for year-round MV rice farming is about seven times
higher compared to MV rice production under rice-prawn gher farming system.
Therefore, it may be concluded that the locally adopted rice-prawn gher farming
technology has significant impacts on soil nutrients for MV rice production. In other
words, this farming system has reduced the production cost system for MV rice
production compared to MV rice production in Bangladesh.
Table 6: Comparison of per Hectare Cost of Chemical Fertilisers between
Rice- Prawn Gher and MV Paddy Farming
Application of chemical fertilisers
Farming system Urea TSP MP Gypsum Total cost Ratio
(Taka) (Taka) (Taka) (Taka) (Taka) (PF/GF)
Paddy production under 294 391 55 79 819
gher farming (GF)
6.7
MV paddy farming (PF) 1,438 2,019 703 1,328 5,488
Notes: 1) TSP and MP indicate Triple super phosphate and Muriate of potash, respectively.
2) Retail price of per kg urea, TSP, MP and gypsum is Tk 5.73, Tk 11.08, Tk 8.82 and Tk 6.2, respec.
3) Sample size of gher, and MV paddy farming were 32 and 33, respectively.
4) 1$US= Taka 68.85, March, 2006.
Source: Gateway to Fertiliser Information in Bangladesh 2005.

6.0 Conclusion
The rice-prawn gher farming system is an indigenous agricultural system solely
developed by farmers since mid-1980s. The rice-prawn gher farming system has
significant impacts on soil fertility for MV rice production. The farmers apply
comparatively less chemical fertilisers in MV rice production under the rice-prawn gher
farming system compared to MV rice production in Bangladesh and are statistically
significant between the two agricultural systems. Input cost per hectare of chemical
fertilisers for MV rice production is about seven times higher than MV rice production
under rice-prawn gher farming technology. However, production per hectare of
traditional year-round MV rice farming is more than one times (1.17) higher than MV
rice production under the rice-prawn gher farming system. Therefore, it can be concluded
that rice-prawn gher farming system is a cost-saving technology for MV rice production.

79
References
Akteruzzaman, M. 1998. The Impact of Privatization and Deregulation on New Technology Diffusion and
Income Distribution in Rural Bangladesh. Ph.D. dissertation. Laboratory of Development Economics,
Department of Agricultural Economics, Hokkaido University, Japan.
Ali, A. M. S. 2004. Technological Change in Agriculture and Land Degradation in Bangladesh: A Case Study.
Land Degradation and Development, Vol. 15, Issue 3. Pp. 283-298.
Asaduzamman, M., and K. A. Toufique. 1998. Rice and Fish: ‘Environmental Dilemmas of Development in
Bangladesh’. In Growth or Stagnation? A Review of Bangladesh’s Development 1996, Center for Policy
Dialogue. University presses Ltd. Dhaka (mimeo).
Baker, D. E., and M. C. Amacher. 1982. Nickel, copper, zinc and cadmium. In A. L. Page, R. H. Miller, and D.
R. Keeney edited Methods of soil analysis, part 2. American Society of Agronomy, Madison, Wisconsin. Pp.
323-336.
BARC, 1997. Fertilizer Recommendation Guide. Bangladesh Agricultural Research Council. Dhaka.
Barker, D. E., and N. H. Surh. 1982. Atomic absorption and flame emission spectroscopy. In A. L. Page, R. H.
Miller, and D. R. Keeney edited Methods of soil analysis, part 2. American Society of Agronomy. Madison,
Wisconsin.
Barmon, B. K., T. Kondo, and F. Osanami. 2004a. Impact of Rice-Prawn Gher Farming on Agricultural and
Household Income in Bangladesh: A Case Study of Khulna District. Journal of Bangladesh Studies, Vol. 6,
No. 1 & 2, Pp. 51-61. The Pennsylvania State University. USA.
Barmon, B. K., T. Kondo, and F. Osanami. 2004b. Impacts of Rice-Prawn Gher Farming on Cropping Patterns,
Land Tenant System, and Household Income in Bangladesh: A Case Study of Khulna District. Asia-Pacific
Journal of Rural Development, Vol. 14, No.1, Pp.10-28.
Barmon, B. K., T. Kondo, and F. Osanami. 2004c. Labor Demand for Rice Prawn Gher Farming in Bangladesh:
A Case Study of Khulna District. The Review of Agricultural Economics, Vol. 60, Pp. 273-287.
Barmon, B. K., T. Kondo, and F. Osanami. 2005. Historical Background and Diffusion Process of Rice-prawn
Gher Farming in Bangladesh: A Case Study of Khulna District. The Review of Agricultural Economics, Vol.
61, Pp. 85-100.
Berger, K. C., and E. Truog. 1940. Boron deficiency as revealed by plant and soil tests. J. Am. Soc. Agron. 32:
297-301.
Bhattacharya, D., M. Rahman, and F. Khatun. 1999. Environmental Impact of Structural Adjustment Policies:
The Case Export Oriented Shrimp Culture in Bangladesh. Dhaka: Centre for Policy Dialogue (CPD).
Bhuiyan, N.I. 2001. Application of Integrated Plant Nutrition System (IPNS) in Agriculture- Bangladesh
Experiences. Country paper presented in ESCAP Organized Regional Workshop on ‘Integrated Plant
Nutrition System (IPNS) Development and Rural Poverty Alleviation’ held on 18-20 September. Bangkok,
Thailand.
BRRI. 1997. BRRI Annual Report for 1997. Bangladesh Rice Research Institute. Gazipur, Bangladesh.
Cai, G.X., D.L. Chen, H. Ding, A. Pacholski, X. H. Fan, and Z.L. Zhu. 2002. Nitrogen losses from fertilizers
applied to maize, wheat and rice in the North China Plain. Nutrient Cycling in Agroecosystems. Pp. 63, 187-
195.
Datta, D.K. 2001. Ecological Role of Fresh Water Apple Snail Pila Globosa and the Consequences of its Over-
harvesting from Beel Ecosystem of Bagerhat and Gopalgonj District. A Study Report. Study carried out
jointly by Khulna University and GOLDA Project of CARE Bangladesh. Funded through Department for
International Development.
Duxbury, J.M. 2001. Sustainable of Post-Green Revolution Agriculture: The Rice-Wheat Cropping System of
South Asia, Annual Report, Submitted to the Soil Management CRSP Management Entity, University of
Hawaii by Cornell University, Department of Crop and Soil Science. New York.
Fageria, N.K., and V.A.C Baligar. 1998. Growth and nutrient uptake by common bean, lowland rice, corn,
soybean, and wheat at different soil pH and base saturation on an inceptisol, TEKTRAN. United States
Department of Agriculture. Agriculture Research Service.
Foth, H.D. and B.G. Ellis. 1997. Soil Fertility. 2nd Ed. CRC Press. Boca Raton, FL.

80
Haby, V.A., M.P. Russelle, and E.O. Skogley. 1990. Testing soils for potassium, calcium and magnesium. In
R.L. Westerman edited Soil testing and plant analysis. 3rd Ed. Pp. 181-227. Soil Science of America.
Madison, WI.
Islam, M.S., M.A. Sattar, M.A. Rahman. 2002. Integrated Nutrient Management for Different Cropping
Patterns in Bangladesh. 17th World Congress of Soil Science (WCSS), Symposium 14, Paper No. 1284, 14-
21 August. Bangkok, Thailand.
Kendrick, A., 1994. The Gher Revolution: The Social Impacts of Technological Change in Freshwater Prawn
Cultivation in Southern Bangladesh. The Report of a Social Impact Assessment Prepared for CARE
International in Bangladesh with Support from the Bangladesh Aquaculture and Fisheries Resource Unit
(BAFRU).
Kyuma, K. 2004. Paddy Soil Science. Kyoto University Press. Kyoto, Japan
Market Monitoring & Information System (MMIS). 2005. Gateway to Fertilizer Information in Bangladesh.
Ministry of Agriculture, Bangladesh Secretariat. Dhaka. http://www.fadinap.org/Bangladesh/ documents/
McLean, E.O. 1982. Soil pH and lime requirement, Methods of soil analysis, part 2, chemical and
microbiological properties. 2nd edition, Number 9 (Part 2). Pp. 199-224. American Society of Agronomy.
Madison, WI.
Mengel, K., and E.A. Kirbky. 1982. Principles of Plant Nutrition (3rd ed.). P. 59. International Potash Institute.
Nijera Kori. 1996. The Impact of Shrimp Cultivation on Soils and Environment in Paikgacha Region. Khulna
(Limited to polders 20, 21, 22, 23, and 24).
Olsen, S.R., and L.E. Sommers. 1982. Phosphorus. In A.L. Page, R.H. Miller, and D.R. Keeny edited Methods
of soil analysis, Part 2.Chemical and microbiological properties. American Society of Agronomy. Madison,
WI.
Pierzynski, G. M., J.T. Sims, and G..F. Vance. 1994. Soils and environmental quality. Fl: Lewis Publishers.
Rahman M.M., and S.L. Ranamukhaarachchi. 2003. Fertility status and possible environmental consequences
of Tista Floodplain soils in Bangladesh, Thammasat International Journal of Science and Technology, 8(3).
P. 11-19.
Sobhan, A.S. 1997. Bangladesh Country Paper on Shrimp Sector. Paper Presented at the International Training
Course on Pollution Control and Minimization in Small and Medium Sized Marine Food Processing
Industries in the Developing Countries of South Asia, 28-30 October, Bangkok.
Tisdale, S.L., W.L. Nelson, J.D. Beaton and J.L. Havlin. 1993. Soil fertility and fertilizers. 5th Ed. New York:
MacMillan Publishers.
Zaman, Z.U. 2001. A study on Rice Marketing System and Price Policy in Bangladesh. Ph.D. dissertation.
Laboratory of Agricultural Marketing, Department of Agricultural Economics, Hokkaido University. Japan.

81
 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS
(Organ of the Indian Society of Agricultural Economics)

Vol. 62 JANUARY-MARCH 2007 No.1

CONTENTS
SUPPLEMENT TO THE CONFERENCE NUMBER:
JULY-SEPTEMBER 2006
Presidential Address:
Conference Keynote Papers Abhijit Sen
Strategies for Agricultural Development in North-East India: B.C. Barah
Challenges and Emerging Opportunities
Rural Infrastructure and Growth: An Overview P. Satish
Summaries of Group Discussion:
Agricultural Development in the North-East: Status, M. P. Bezbaruah
Assessment and Prospects
Trends in Food Consumption and Nutrition – Food Security Concerns Ramesh Chand
Rural Infrastructure and Growth Vasant P. Gandhi
ARTICLES
Energy Costs and Groundwater Withdrawals: Results from an F.A. Shaheen and
Optimal Control Model for North Gujarat R.L. Shiyani
Efficiency of Market Behaviour of NTFPs for Households Debnarayan Sarker and
under JFMP: A Case Study in West Bengal Nimai Das
RESEARCH NOTES
Economics Cardamom Cultivation in Kerala P.K. Varghese
Economics and Determinants of Fish Production and Its Effects on Kehar Singh
Family Income Inequality in West Tripura District of Tripura
Factors Affecting the Adoption of Resource Conservation Technology: Vijaya Lasmi Pandey and
Case of Zero Tillage in Rice-Wheat Farming Systems Vinod Mishra
Changing Pattern of Agricultural Productivity in Brahmaputra Valley Surendra Singh and
Bimal Sharma

BOOK REVIEWS* PUBLICATIONS RECEIVED*

Annual Subscription Rates
Individual Membership Fee : Rs.300.00 £35.00 $75.00
Life Membership Fee : Rs.3,000.00 £350.00 $75.00
Institutional Subscription : Rs.600.00 £54.00 $125.00

Please address correspondence to the Hon. Secretary & Treasurer, The Indian Society of Agricultural
Economics, 46-48, Esplanade Mansions, M.G. Road, Fort. Mumbai-400 001 (India)
Telephone: 022-22842542; Fax: 091-022-22838790; E-mail: isae@bom7.vsnl.net.in; website: www.isaeindia.org

82
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Soil Hydraulic Characterisation under Different Cropping

Patterns in Sloping Agricultural Lands in Sri Lanka
A.A. Rivas*, R.S. Clemente*, S.L. Ranamukhaarachchi**, A. Das Gupta*,
M.A. Zoebisch**, S. Thevachandran** and M.S.D.L. De Silva***

Abstract
The highlands of Sri Lanka are generally considered as agriculturally productive, and
yield potentials are high. However, in recent years, most agricultural lands in the hilly
areas are being over-exploited, ignoring long-term negative impacts on the soils and the
environment. Different forms of cropping can also modify soil conditions which can have
positive or negative consequences on the soil quality. Thus, the objectives of this study
were to evaluate the effects of different cropping patterns under different terrains in the
Welimada Division of the upper watershed of Sri Lanka on soil quality using selected
soil hydraulic characteristic indicators, and to identify which among them provides more
positive effects on soil hydraulic quality for crop growth and production. The soil
hydraulic characterisation conducted in the Yala (dry) and Maha (rainy) seasons in year
2004 involved measurements of infiltration, hydraulic conductivity, moisture retention
characteristics, texture, and organic matter in selected zones of the Welimada watershed.
Typical transects or hillslopes were selected based on varying cropping pattern and
slope, and were divided into three or two field sites along the slope. A more integrated
approach of assessing soil hydraulic quality based on soil hydraulic quality index was
adopted to compare the hydraulic quality of soils under different cropping patterns.

Analysis of the data gathered indicated that the effects of different cropping patterns are
more pronounced in the rooted subsoil layer. In general, the potato-fallow-vegetable-
fallow cropping rotation was identified as the pattern which provided more positive
effects on the soil hydraulic quality in the context of enhancing crop growth and
production. The more agriculturally intensive irrigated lowlands, particularly the paddy-
potato-vegetable rotation, showed the potential of increasing the income of farmers
without adversely affecting soil hydraulic quality.

1.0 Introduction
The increase in the demand for food and fibre has resulted in the utilisation of sloping
lands for crop production (Jayakody et al. 2004). This is particularly true in the
highlands of Sri Lanka where sloping lands are being used under intensive cultivation.
Various cropping patterns are practised depending on different forces (economy, climate,

*
Water Engineering and Management Programme, Asian Institute of Technology, Pathumthani, Thailand.
**
Agricultural Systems and Engineering Programme, Asian Institute of Technology, Pathumthani, Thailand.
Corresponding e-mail: ranamuka@ait.ac.th
***
Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka.

83
etc.) influencing farmers’ selection of crops. The use of sloping lands for food production
may temporarily address the demands of growing population, but its sustainability is
under question as evidences of soil erosion and land degradation have been widely
reported. Every type of land management, such as cropping pattern, has its own effects
on the soil (Rachman et al. 2003), which in turn determines whether the practice is
sustainable. Among the soil properties, which are affected by land management as well
as influence the quality of soil and consequently its productivity, are the soil hydraulic
properties (Lekamalage 2003). It is therefore necessary to evaluate the effects of different
cropping patterns under different terrains in the Welimada Division of the upper
watershed on soil quality using selected soil hydraulic characteristic indicators, and to
identify which among them provides more positive effects on soil hydraulic quality for
crop growth and production.
2.0 Materials and Methods
2.1 The study area
The area selected for this study is the Welimada Divisional Secretary division since a
long history of cropping pattern is being practised in its sloping lands. Located at
Latitude 6°54’4’ N and Longitude 80°55’22’, Welimada has a total land area of 192.61
km2. It is located at about 204 km east of Colombo, the country’s capital, and is part of
the Uma Oya catchment, which is tributary to the largest river in the country, the
Mahaweli. The average altitude in the area is at 1017m AMSL, hence considered as an
upcountry (elevation > 900m AMSL). It is situated along the boundary of two climatic
zones, namely, intermediate (east portion) and wet zones (west portion). Most of it,
however, belongs to the intermediate zone, including the Grama Niladhari (GN- i.e.
Government Officer responsible at the village level) Divisions of Palugama Ella,
Idamegama and Kebillegama, where the specific sites were located.
The topography in the area is complex. Slope varies from undulating in lands adjacent to
stream or rivers to steeply dissected ones, with inclination of more than 60 per cent in
upper portions of the slope. The soil in most parts is Red Yellow Podzolic with small
parts of Lithosols and rock outcrops. The typical hillslope in the area is divided into two,
namely, upland (rain-fed) and lowland (irrigated). The irrigation canals, constructed of
concrete, stones or earth some 50 years ago, serve as the division between them.
As with the other parts in the country, Welimada has two seasons: the Maha (rainy) and
Yala (dry) seasons. Maha usually starts in late September or early October and ends in
early January, and Yala season starts from March and ends in August. The average
rainfall for the last five years, i.e. from 2000 to 2004, is 1320 mm, and the temperature
ranges from 12°C in December to 32°C in August 2004. This is the typical range of
values of temperature in areas within the intermediate zone.
Most parts of Welimada Division are utilised for cereal or vegetable cultivation (52%),
and only small portion is planted with tea (13%). The major crops are rice, potato, beans,
tomato, and cabbage, among others, which are cultivated as dictated by climatological

84
conditions, socio-economic forces, and availability of resources. The upland is usually
cultivated only once during Maha as rain-fed, while the lowland, which has irrigation
water, is usually cultivated three times a year. Only vegetables or root crops are
cultivated in the upland, while on the other hand, paddy is usually cultivated first in the
lowland followed by either potato, beans, tomato, cabbage, or other crops. The cropping
pattern indicates that the lowland is under intensive agricultural practices and hence more
vulnerable to degradation.
2.2 Data collection
The selection of sampling sites was based on transects. Typical transects in the area were
identified with due consideration to the category of land (upland or lowland), slopes and
cropping patterns. Each transect was divided into three sites along the hillslope,
depending on the number of significant portions along the slope. Hence, three types of
transects were identified, namely; upland-upland-lowland, upland-lowland-lowland, and
upland-lowland. With respect to slopes, three slope ranges per upland and lowland
category were considered: <30%, 30-60%, >60%. The cropping patterns common in the
study area and investigated were; (1) vegetables only–fallow (VF) [upland], (2) potato-
fallow-vegetables–fallow two-year rotation (PVF) [upland], (3) paddy-potato-vegetables
(PPV) [lowland], and (4) vegetables-vegetables-fallow (VVF) [lowland], plus (5)
uncultivated lands (UC) [upland], which were either secondary forest or bush or grass
lands.
Figure 1: Sampling Scheme for the Study

Transect/hillslope

Three (field) sites per transect/hillslope

Top portion Middle portion Bottom portion

Layers per field site

Top layer (Ap horizon) Second (rooted subsoil) layer

- Infiltration Char. (5 replications) - Kfs (3 replications)

- Kfs (3 replications)
- WRC (3 replications)
- BD (3 replications)
- Texture (1 sample)
- OM (1 sample)
- WRC (3 replications)
- BD (3 replications)
- Texture (1 sample)
- OM (1 sample)

85
In every field site, the soil properties of the top two layers, i.e. Ap horizon and rooted
subsoil layer, were considered. The sampling scheme for soil hydraulic characterisation
is shown in Figure 1. The numbers of representative samples or replications taken are
indicated in parenthesis. The time of sampling and testing was chosen when the field was
least disturbed, i.e. before harvesting or uprooting for lowland, and before land
preparation for upland fields. A soil pit was also dug in each field to determine soil
profile and structure before sampling and testing were conducted.
Cropping patterns and management practices of farmers were determined through
interviews and group discussions, while gathering data for the soil properties was done in
two ways: in-situ and laboratory measurements. Where data on other soil properties
could not be obtained primarily due to technical constraints, existing estimation methods
utilising the measured data were applied. The details on the methods used in this study
are given in Table 1.
Table 1: Methods Used for the Determination of Selected Soil Properties

Parameter Method References

Soil profile and structure In-situ Inspection FAO 1977
Soil texture Hydrometer Method Gee and Bauder 1986
Bulk density (BD) Core Method Blake and Hartge 1986
Soil organic matter (OM) Modified Walkley & Black Sims and Haby 1971
Infiltration characteristics Single Ring Cylinder Method Baragello et al. 2004
Field saturated hydraulic Guelph Permeameter SMEC 1991
conductivity (Kfs)
Water retention characteristics Hanging Water Column Klute 1986
(WRC)
Field capacity (FC) Hanging Water Column Klute 1986; Lal 1981
Permanent wilting point (PWP) Soil-Moisture Characteristic Ghosh 1976
Model

2.3 Identification of soil hydraulic quality index

Soil quality is a general characteristic of soil that is a function of many soil properties.
Among the components that describe soil quality are the soil hydraulic properties. In this
study, the assessment of soil hydraulic quality among different cropping patterns was
based on the procedures described by Andrews et al. (2002). Soil hydraulic quality
indices (SHQIs) specific for the locality were developed to describe, as a single
parameter, the qualities that relate to the soil hydraulic properties. These SHQIs
contain only those soil properties that are responsible for the observed differences
between different cropping patterns. The procedure to determine SHQI consists of the
following steps:

86
• Selection of initial dataset- This dataset is composed of soil properties that are
assumed to be related, directly or indirectly, to soil hydraulic quality. The selection
of these soil properties was based on previous studies and published documents
(Azam 2004; Cho 2003).

• Determination of the minimum dataset indicators- These indicators derived from

the initial dataset are those that best represent the soil hydraulic functions in the
context of land use in the study area. Principal components analysis (PCA) was
applied to identify the minimum dataset (MDS). The procedure described by Brejda
et al. (2000) was used to derive principal components (PCs) based on the Eigen
values, which are the amount of variance explained by the factor. Only PCs with
Eigen values ≥ 1 were retained for further analysis; and only the highly weighted
variables (variables within 10 per cent of the highest factor loading, in absolute
values) were retained from each PC for the MDS. When more than one variable was
retained in a certain PC, a cross checking was done to avoid involving redundant
variables by looking at the correlation matrix. If the highly weighted variables were
correlated (r ≥ 0.60), then the variable with smaller factor loading was considered
redundant and hence eliminated in the MDS, otherwise, each variable was
considered important and therefore included in the MDS.

• Scoring of the minimum dataset indicators- This was based on the performance of
the indicators within the dataset and involved conversion of the selected data into a 0
to 1 scale. This study used the equations proposed by Diack and Stott (2001). These
equations, which were based on the direction - positive or negative - of influence of
the soil properties on the overall soil hydraulic quality, are as follows:
y = (x – s) / (1.1 t – s) for ‘more is better’ (1)
and,
y = 1 – {(x – s) / (1.1 t – s)} for ‘less is better’ (2)
where y is the score of the individual MDS indicator, x is the value of the soil
property to be converted into a 0 to 1 scale value, s is the lowest possible
value of the soil property (s=0 is considered in this study), and t is the
highest value for the same soil property in the dataset.

• Development of comparative SHQIs- This was done by assigning weights on the

MDS indicators and summing the weighted MDS indicators for each site to obtain
the SHQI specific to the site determined by the following formula:
n
SHQI = ∑ [(Wi )( Si )] (3)
i =1

where Wi is the PC weight (variance of PC / cumulative variance of all the

selected PC), and Si is the indicator score (y in equations 1 and 2).

87
2.4 Statistical analysis
Descriptive statistics were used to analyse the gathered data. Student’s t-test tools and
analysis of variance (ANOVA) were applied to test the significant differences in
indicators of soil hydraulic properties between different soil layers, field categories,
locations along the slope, cropping patterns and slopes. The calculated means of SHQI
for each land use pattern were also compared using t-test. It was assumed that higher
index scores meant better soil hydraulic quality. Statistical package software SPSS
version 11.5, in particular, was used for the data analysis.
3.0 Results
3.1 Initial dataset of soil hydraulic quality parameters
The selected initial set of soil properties for the development of SHQIs based on previous
studies and published documents are presented in Table 2. The rating scales for ranking
soil structural properties are based on FAO (1977).
Table 2: Selected Soil Parameters for Developing Soil Hydraulic Quality Indices

Soil hydraulic quality parameters

Methods
Ap horizon Rooted subsoil layer

• Profile description • Profile description

- Soil structure (pore size and
In situ distribution, aggregate size and
grade, and packing density)
- Root density
• Field saturated hydraulic
conductivity
• Infiltration characteristics
- Soil structure (pore size and
distribution, aggregate size and
grade, and packing density)
- Root density
• Field saturated hydraulic
conductivity

• Soil texture (per cent sand, silt and
clay)
• Bulk density
Laboratory
• Soil water content at field capacity,
permanent wilting point and
available water-holding capacity
• Soil organic matter
• Soil texture (per cent sand, silt and
clay)
• Bulk density
• Soil water content at field capacity,
permanent wilting point and
available water-holding capacity
• Soil organic matter

3.2 Soil hydraulic quality indices

Individual comparisons of hydraulic properties have been found not conclusive for
identification of land use or cropping patterns that have more favourable effects on soil
hydraulic quality in the context of agricultural production, i.e. the soil hydraulic quality
status that helps make more water available to plants for a longer time period without
negatively affecting the other processes required for plant or crop growth. In fact,

88
Reganold and Palmer (1995) pointed out that the comparative assessment of different soil
properties using single parameters has limitations, hence not sufficient in evaluating and
comparing the soil hydraulic quality of different systems. According to Cho (2003), this
approach does not also provide comprehensive quality assessment since the values of
selected parameters may provide no correlation with each other. Soil hydraulic quality
needs to be described by highly correlated attributes or properties (Azam 2004). Thus, a
more reliable and comprehensive assessment of soil hydraulic quality that takes into
account the interactions among soil properties (e.g. SHQI), was adopted and used for this
purpose.
3.2.1 Ap horizon
Derived from the initial dataset of soil properties (Table 2), six soil properties (Table 3)
in six principal components (PC) were identified by the PCA to be included in the MDS
for the Ap horizon. These properties are important indicators and can be used for the
development of the SHQI for the study area. It is evident from the table that about 90 per
cent of the total variance of the entire dataset can be explained by the MDS indicators
listed in the table. It can be observed that the MDS list is a combination of different soil
properties responsible for the integrated hydraulic function of the soil. In the list, two are
direct measurements of hydraulic properties (PWP and Kfs), two are indirect
measurements of soil structure (root density and pore size distribution), and one each for
physical (packing density) and biological (OM) properties. Pore size distribution is an
important indicator of soil hydraulic quality, especially in terms of soil water flow and
soil water content (Alegre et al. 1986). Root density as well as other soil structure related
properties directly govern the soil water regime (Cho 2003). Packing density is related to
the soil compaction and soil structure, thus affecting the hydraulic quality of the soil. The
inclusion of OM as a key indicator is very much needed as OM is considered the central
property of soil (Brady 1984) affecting soil functions including the hydraulic aspect.
Table 3: Minimum Dataset Indicators for the Ap Horizon

Variance Cumulative Indicators

PC Weight
(%) variance (%) (soil properties)

1 31.45 31.45 0.350 PWP

2 18.07 49.52 0.201 Root density
3 13.72 63.24 0.153 Pore size distribution
4 11.70 74.94 0.130 OM
5 8.49 83.43 0.094 Packing density
6 6.50 89.93 0.072 Kfs

Analysis of the SHQIs of different land use or cropping patterns in the Ap horizon
revealed no significant difference on their soil hydraulic quality status (Table 4).
Uncultivated fields had the highest SHQI value while the VF system got the lowest. The
higher value of SHQI in uncultivated fields could be due to the absence of intense tillage,

89
thus conserving the soil hydraulic property compared to others. The similarities of the
SHQIs in the cultivated systems could verify the short-term effects of tillage or
cultivation in the top layer on soil hydraulic properties since the sampling and
measurements were done during the least disturbed state.

Table 4: Soil Hydraulic Quality Indices (SHQIs) for Ap Horizon under

Different Cropping Patterns

SHQI*
Cropping pattern Rank
(Mean ± Std. deviation)
Uncultivated (sec. forest or grassland) 0.565±0.163 Highest
Potato-fallow-vegetable-fallow rotation 0.520±0.156
Paddy-Potato-vegetable rotation 0.507±0.087
Vegetable-vegetable-fallow rotation 0.480±0.123
Vegetable-fallow rotation 0.415±0.106 Lowest
* Not significant at the level of p ≤ 0.05.

3.2.2 Rooted subsoil layer

The PCA identified five soil properties namely, per cent sand, Kfs, aggregate size, OM,
and pore size from the initial dataset to compose the MDS in four PCs (Table 5). As with
the Ap horizon, Kfs and OM appeared again among the important factors affecting the
soil hydraulic quality. In this case, sand content had the highest contribution. Sand and
other textural properties are inherent properties of the soil and are not much affected by
external factors. They, however, still affect the hydraulic behaviour of soil with the fact
that sand facilitates movement of water within the soil. It is also reasonable that sand
appeared as the most significant contributor in the subsoil layer considering that this
layer is less influenced by external factors. Also, aggregate size is a structural property;
hence it has a dominating influence in overall hydraulic quality (Cho 2003). The last
parameter, pore size, is considered a direct indicator of soil hydraulic quality (Azam
2004).
Table 5: Minimum Dataset Indicators for the Rooted Subsoil Layer

Variance Cumulative variance Indicators

PC Weight
(%) (%) (soil properties)

1 39.51 39.51 0.465 Per cent sand

2 18.60 58.11 0.219 Kfs
3 17.91 76.02 0.211 Aggregate size
4 9.03 85.04 0.106 OM and pore size

Among the five cropping patterns considered, significant differences between the five
SHQIs were observed at a highly significance level of p ≤ 0.01 (Table 6), and

90
correspondingly five ranks were also identified (Figure 2). The PVF system scored the
highest SHQI while the VF system had the lowest value again. UC system was second
highest, followed by two lowland cropping patterns, PPV and VVF. These significant
differences could indicate that the effects of different cropping patterns on soil hydraulic
quality are long term and more stable in the subsoil layer, which is not much influenced
by external factors compared to the surface layer.
Table 6: Soil Hydraulic Quality Indices (SHQIs) for Rooted Subsoil Layer under
Different Cropping Patterns

SHQI*
Cropping pattern Rank
(Mean ± Std. deviation)
Highest
Potato-fallow-vegetable-fallow rotation 0.730±0.071a
Uncultivated (sec. forest or grassland) 0.675±0.064ab
Paddy-potato-vegetable rotation 0.600±0.017abc
Vegetable-vegetable-fallow rotation 0.553±0.055bc
Lowest
Vegetable-fallow rotation 0.485±0.050c

* Significant differences at p ≤ 0.01 are denoted by different letters according to least significant difference
(LSD) method.

Figure 2: Soil Hydraulic Quality Index (SHQI) Ranks for the Rooted Subsoil
Horizon among Different Cropping Patterns

0.85

0.8

0.75
Soil Hydraulic Quality Index

0.7

0.65

Rank 1
0.6

Rank 2
0.55

Rank 3
0.5

Rank 4
0.45

Rank 5
0.4
PVF UC PPV VVF VF

Cropping Pattern

91
4.0 Discussion
It can be observed from the SHQIs of the different cropping pattern in the top two soil
layers that the differences in the rooted subsoil layer is more pronounced. It should be
noted, however, that though there were no significant differences among the SHQIs in
the Ap horizon, the ranking of the different cropping patterns with respect to the SHQI
values appear to be consistent with that in the subsoil layer. This could only indicate the
consistent influence of the different cropping pattern to the soil hydraulic quality. The
ranking of the selected cropping patterns could probably be explained by the OM
content. OM is the central property of soil that could influence other soil hydraulic
properties. It plays a vital role in the soil hydraulic quality by building stable aggregates,
encouraging the activities of microflora (plant roots) and fauna (earthworms, burrowing
insects, etc.) and resisting compaction, which consequently govern the infiltration
potential, storage, redistribution and drainage characteristics of soil and soil water
(Kleinhenz and Bierman 2001). In this case, PVF system has the highest while VF
system has the lowest OM content in both soil layers (Rivas 2005). The VVF system
also has high OM content, which could be due to high application of organic manure in
the field, but its effects might have been buffered by its inferior values for other soil
structure properties like root density, pore size, and aggregate size (Rivas 2005).
It can also be observed that, in general, the SHQI values in the Ap horizon are much
lower than those in the subsoil layer. This may indicate deterioration of the quality of soil
in the Ap horizon.
However, what is more important behind the SHQI values is its effects on soil water
content and movement that are beneficial for crop production. It should also be noted that
the soil water movement in one layer of soil is not only governed by its hydraulic quality
but also influenced by the boundary conditions – these include the underlying layer
(Linsley et al. 1988; Ward and Robinson 2000). Thus, in this case the soil water content
and movement in the Ap horizon is further controlled by the hydraulic quality of soil in
the underlying layer, that is, the rooted subsoil layer. Hence, though the Ap horizon has
no significant differences in SHQI, its hydraulic behaviour may still differ due to the
differing hydraulic quality in the subsoil layer.
It can therefore be inferred that for the upland fields, the soils in PVF system have more
positive hydraulic behaviour for crop growth and production, while for lowland fields,
the soils in PPV system functions better. This deduction verifies the farmers’ indigenous
knowledge and belief that changing crops within a year, along with incorporating organic
matter or residue, more particularly the PPV rotation system has positive effects on the
quality of soil as evident in the increase in the yield of their crops.
In general, it was found that the different cropping patterns significantly affect soil
hydraulic properties, and these differences are much more pronounced in the rooted
subsoil layer. Using SHQI, an integrated approach which takes into account the
combined effect of different soil properties on the hydraulic quality of soil, it was found

92
out that the PVF system showed the highest soil hydraulic quality. Thus, if a cropping
pattern is to be selected from each field category in the context of soil hydraulic quality,
PVF system is recommended for upland fields and the PPV system for the lowland fields.
The importance of knowing that the hydraulic quality of soil under the PPV system is
higher than the VVF system, is essential for the farmers. This is not just because it
verifies their belief that PPV system could provide positive contribution to soil quality
but because of its contribution to the livelihood of farmers. The fallow component in the
VVF system is among the major factors, and this system is mainly practised by the
farmers in the study area. It is to minimise the negative effects of intensive cultivation on
soil quality, thus urging them to sacrifice by not planting during one growing period.
However, if circumstances allow, they may still grow three crops per year in a rotation
without adversely affecting the soil hydraulic quality by adopting the PPV system which
could result in enhanced income per year.

References
Alegre, C., D. Cassel, and D. Bandy. 1986. Effects of land clearing and subsequent management on soil
physical properties. Soil Science Society of America Journal. Madison, Wisconsin, USA 50, 1379-1384.
Andrews, S., J. Mitchell, R., Mancinelli, D. Karlen, T. Hartz, W. Horwatth, G. Pettygrove, M. Scow, and S.
Munk. 2002. On-farm assessment of soil quality in California’s Central Valley. Agronomy Journal.
Madison, Wisconsin, USA 94, 12-23.
Azam, M. 2004. Determination of a Location-Specific Soil Hydraulic Quality (SHQ) Index: A Case Study from
Northeast Thailand. Master Thesis. Asian Institute of Technology. Bangkok, Thailand.
Baragello, V., M. Iovino, and D. Elrick. 2004. A simplified falling-head technique for rapid determination of
field-saturated hydraulic conductivity. Soil Science Society of America Journal. Madison, Wisconsin, USA
68, 66-73.
Blake, G. and K. Hartge. 1986. Bulk density. In A. Klute edited Methods of Soil Analysis: Part I. Physical and
Mineralogical Methods, Second Edition. American Society of Agronomy, Inc. and Soil Science Society of
America, Inc. Madison, Wisconsin, USA.
Brady, N. 1984. The Nature and Properties of Soils. 9th Edition. New York: Macmillan.
Brejda, J., T. Moorman, D. Karlen, and T. Dao. 2000. Identification of regional soil quality factors and
indicators: Central and Southern High Plains. Soil Science Society of America Journal. Madison, Wisconsin,
USA 64, 2115-2124.
Cho, K. 2003. Land-use and Soil Quality Dynamics in an Agricultural Area of Nakhon Ratchasima Province,
Northeast Thailand. Doctoral Dissertation. Asian Institute of Technology. Bangkok, Thailand.
Diack, M., and D. Stott. 2001. Development of soil quality index for the Chalmers silty clay loam from the
Midwest USA. In D. Stott, R. Mohtar, and G. Steinhardt edited Sustaining the Global Farm. Selected papers
from the 10th International Soil Conservation Organisation meeting held on 24-29 May 1999 at Purdue
University, and the USDA-ARS National Soil Erosion Laboratory. Pp. 550-555.
FAO. 1977. Guidelines for Soil Profile Description. Second Edition. Food and Agriculture Organization of the
United Nations. Rome, Italy.
Gee, B. and J. Bauder. 1986. Particle size analysis. In A. Klute edited Methods of Soil Analysis: Part I. Physical
and Mineralogical Methods. Second Edition. American Society of Agronomy, Inc. and Soil Science Society
of America, Inc. Madison, Wisconsin, USA.

93
Ghosh, R. 1976. Model of the soil-moisture characteristic. Journal of Indian Society of Soil Science. New Delhi,
India 24, 353-355.
Jayakody, P., F. Molle, and C. Gamage. 2004. Growing pressure over land and water resources:
Transformations in the Belihuloya Catchment. Water Resources Research in Sri Lanka: Symposium
Proceedings of the Water Professionals’ Day. University of Peradeniya, Sri Lanka.
Kleinhenz, M. and P. Bierman. 2001. Soil quality in vegetable and small fruit production. In J.A. Fischer edited
Bulletin 898. The Ohio State University. Ohio.
Klute, A. 1986. Water retention: Laboratory methods. In A. Klute edited Methods of Soil Analysis: Part I.
Physical and Mineralogical Methods. Second Edition. American Society of Agronomy, Inc. and Soil
Science Society of America, Inc. Madison, Wisconsin, USA.
Lal, R. 1981. Physical properties. In D. Greenland edited Characterization of soils in relation to their
classification and management for crop production: examples from some areas of the humid tropics.
Oxford: Clarendon Press.
Lekamalage, B. 2003. Characterization of Surface Soil Hydraulic Properties in Sloping Landscapes. Master
Thesis, University of Saskatchewan. Saskatoon, Canada.
Linsley, R., M. Kohler, and J. Paulhus. 1988. Hydrology for Engineers. International Edition. London: Mc-
Graw Hill.
Rachman, A., S. Anderson, C. Gantzer, and A. Thompson. 2003. Influence of long-term cropping system on
soil physical properties related to soil erodibility. Soil Science Society of America Journal. Madison,
Wisconsin, USA 67, 637-644.
Reganold, J. and A. Palmer. 1995. Significance of gravimetric versus volumetric measurements of soil quality
under biodynamic, conventional, and continuous grass management. Journal of Soil and Water
Conservation. Ankeny, Iowa, USA 39, 131-136.
Rivas, A. 2005. Soil Hydraulic Characterization and Hydrologic Modeling of Sloping Agricultural Lands in
Uma Oya Watershed, Sri Lanka. Master Thesis. Asian Institute of Technology. Bangkok, Thailand.
Sims, J., and V. Haby. 1971. Simplified colorimetric determination of soil organic matter. Soil Science.
Hagerstown, Maryland, USA 112, 137-141.
SMEC. 1991. 2800KI Operating Instructions: Guelph Permeameter. Soil Moisture Equipment Corporation.
Santa Barbara, California, USA.
Ward, R., and M. Robinson. 2000. Principles of Hydrology. Fourth Edition. London: Mc-Graw Hill.

94
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Practitioners’ Papers
Good Governance and HRD: Case Studies of User
Managed Safe Drinking Water and
Health Projects in India
P. Durgaprasad* and P. Sivaram**

Abstract
The paper underscores the criticality of ‘Good Governance for Human Resource
Development (HRD)’, even while delineating the key aspects of people centred and
people-managed decentralised safe drinking water and sanitation (DWS) projects in
India– a major investment in Macro-HRD interventions. Capacity building of the poor
(rural) communities for enhanced absorption of development goods, services and
competencies is viewed as a necessary condition. It highlights the results of three self-
reliance oriented Public-Private and Civil Society partnership-based DWS projects,
which hold promise for `community managed good governance-based microsystems at
the grassroots level in India’. Partnership initiatives like Bone Char-based Community
and Domestic Defluoridation Systems in Nalgonda, AP; Integrated Rural Water Supply
and Environmental Sanitation Project of Mysore, Karnataka; and Bal Panchayat
Programme of the Institute of Health Management, Pachod, Aurangabad, Maharashtra,
are shown as replicable examples and sustainable projects of solid community action in
partnership with the local (Panchayat) government institutions and NGOs. The paper
concludes that `putting the instruments of change into the hands of the development
consumers’ is the most effective and proactive strategy of user-managed and
empowerment-oriented sustainable development. The lessons learnt are outlined, and the
distance to travel towards the goal of `improved overall quality of life of (rural) people’
is emphasised.

1.0 Introduction
Good governance is the key to the window of the new millennium of opportunities; the
door to human resource development (HRD) and the Gateway to equitable social and
human development. The focus of national and international development agencies is
gravitating from technical support and financial assistance towards policy processes,
community initiatives and institutional structures as essential preconditions for good
governance. The introduction of the Millennium Development Goals (MDGs),

*
Professor & Head, Centre for HRD, National Institute of Rural Development, Hyderabad, AP, India.
**
Associate Professor, Centre for HRD, National Institute of Rural Development, Hyderabad, AP, India.

95
governmental actions, public-private partnerships, civil society and community initiatives
for improved quality of life of the (rural) people confirm a renewed commitment to good
governance. Effective and efficient people-managed development, prudent public
expenditure management, and participatory monitoring and evaluation of service delivery
are considered as non-negotiable development instruments. Similarly, decentralisation of
political, administrative and financial authority to district and sub-district levels is sought
to be ensured across the country to improve public sector governance including
implementation of the key principles of Simple, Measurable, Accountable, Responsible
and Transparent (SMART) governance. Rights-based perspectives and approaches to
development are also increasingly recognised as important aspects of good governance
and sustainable development by national, multilateral and bilateral agencies, and non-
government organisations (NGOs).
In sum, Good Governance is one of the innumerable responses of society to the various
upheavals anticipated during the new millennium. The concept of good governance
envisages excellence in public administration, capacity building, generating human
capital, empowerment, self-reliance and partnerships among stakeholders in the society.
Reforms in the public sector should be formulated and analysed through critical thinking
and the goals should be to arrest corruption as well as eliminate inefficiency of civil
servants. Public officials should be reoriented to the vital values of ownership,
responsibility and respect for human dignity.
The paper highlights here the results of three case studies of self-reliance oriented public-
private and civil society partnership-based DWS projects, which hold promise for
`community managed good governance-oriented microsystems at the grassroots level in
India’. Partnership initiatives– Bone Char-based Community and Domestic
Defluoridation Systems in Nalgonda, Andhra Pradesh; Integrated Rural Water Supply
and Environmental Sanitation Project of Mysore, Karnataka; and Bal Panchayat
Programme of the Institute of Health Management, Pachod, Aurangabad, Maharashtra
are shown as replicable examples and sustainable projects of solid community action in
partnership with the local (government) institutions, NGOs and community-based
organisations (CBOs).
2.0 Bone Char-based Community and Domestic Defluoridation
Systems in Nalgonda
Health for all and the Millennium Development Goals are interdependent. Without safe
and improved access to drinking water, the overarching goal of improved health and
poverty reduction cannot be achieved. Several methods and techniques for defluoridation
of water have been tried out in various parts of the country with varying degrees of
success. Removal of fluoride from water is difficult and expensive. Current
defluoridation methods include: Reverse Osmosis, Contact Precipitation, Nalgonda
Technique of Lime and Alum filter, Activated Alumina filter and the Magnesium Oxide
filter. The bone char filter is the latest but less popular method.

96
Nalgonda, the study area, is the worst affected of all the districts in Andhra Pradesh of
India, with high fluoride levels ranging from 3 mg/L–9 mg/L. More than five hundred
villages in the district are gripped by fluorosis. It is more prevalent in areas where the
underground water resources are tapped and pumped out through bore wells, which are
often also the source of the so-called protected water supply schemes. Most people
suffering from various degrees of dental (Photo 1) and skeletal fluorosis (Photo 2)
generally faced one or more of the following health disorders: Black and brown stained
teeth; Bowlegs; Calcification of tendon muscle sheaths; Joint pains, and Indigestion.
Children are the worst affected victims of dental fluorosis.

Photo 1: Dental Fluorosis.

Photo 2: Skeletal Fluorosis.

97
2.1 Objectives and study area
• To study the initiatives of Sri Sai Oral Health Foundation (SSOHF) in
preventing and curing dental and skeletal fluorosis
• To understand the effects of bone char-based community and domestic
defluoridators on people’s health, their responses and type of participation
Nalgonda District of Andhra Pradesh State was selected for the study, where the
incidence of dental and skeletal fluorosis is the highest in the state. Six villages out of ten
where the SSOHF, an NGO, had introduced its programme of defluoridation of drinking
water, were selected on the basis of their proximity to the Mandal, the intensity of the
(fluorosis) problem and the level of people’s participation. The names of sample mandals
and villages are presented in the table here.
Table: Sample Villages and Mandals
No. Name of Village Name of Mandal
1. Madava Yadavalli Narketpally
2. Naibai Narketpally
3. Pothanampet Narketpally
4. Choutabai Narketpally
5. Vailapalle Narayanapur
6. Anthampet Marriguda

2.2 Methodology
Participatory Rural Appraisal (PRA) tools such as Focus Group Discussions (FGDs),
Semi-structured interviews, Timelines, Transects, Seasonal calendar, Semi-participant
observation etc. were used. About 100 men, women, and children were contacted for
gathering information and insights into the problems of fluorosis besides discussions with
district administrators, engineers, doctors and chemists.
Over twelve hundred Bone Char-based Household (Domestic) Defluoridators (BCDDFs)
are being used for drinking water and cooking purposes. Most of these defluoridators
are made of plastic while some are made of steel (Photo 3) whereas the Bone Char-based
Community Defluoridators (BCCDFs) are mainly based in communities/institutions. This
project is a partnership-based initiative between the local user committees and the
Panchayat. SSOHF, Nalgonda is the guiding partner.

98
 Photo 3: (i) Common Plastic Domestic Bone Char-based Defluoridator;
(ii) Domestic Bone Char-based Steel Defluoridator.

2.3 Major features of defluoridation systems and study findings

The major features of Bone char-based defluoridation systems including the technical
factors and costs are shown in Boxes 1 and 2. These systems are cost-effective and
efficient. The technical process is simple to understand and easy to implement. The cost
per litre of defluoridated water is five paise whereas the other systems cost about 20 paise
to 90 paise per litre. Of course, packaged RO-based drinking water bottles cost up to
Rupee 10 per litre.
ƒ Majority of the sample respondents belonged to the age group of 40-50 years
and were suffering from dental and skeletal fluorosis induced problems, as
outlined.
ƒ Safe water is available and accessible (defluoridated at 0.5-1 mg/L) in all the
sample villages.
ƒ BCCDFs (see Box I for its major features) could be fitted to any drinking water
source (bore well based cistern here). It processes 1,25,000 litres per 50 kg of
bone char (Photo 4).
ƒ The quantity of fluoride water processed is indicated by a meter, which also
indicates the time for bone char replacement. The raw water collection tank has
200 litres (physical) capacity for processing the fluoride water. Two litres of
fluoride water is processed per minute (3000 litres in 24 hours). Therefore, a
BCCDF runs round the clock. It does not need any electricity for operating the
system. However, bone char efficiency comes down as the fluoride levels
increase. Therefore, appropriate calculations need to be made regarding the
efficiency of defluoridation and the bone char replacement time. The fluoride
levels are monitored by lay members of the community (see Chart).

99
 Photo 4: Community Defluoridator with Bone Char.

Chart: Fluoride Monitoring

ƒ Every day, an average of 200 households fetch about 15 litres of treated water
for drinking and cooking. Rest of the drinking water and cooking water
requirements of households, i.e., 10 L/household/day is met by Harvested Rain
Water or DDFLs or both. Thus, the total requirement of 25 litres (restricted
though) of safe water/household/day is met by a combination of BCDDFs (see
Box II for its major features), rainwater harvesting and BCCDFs.
ƒ Unlike Reverse Osmosis (RO) systems of drinking water treatment, the
BCDDFs and BCCDFs do not eliminate essential minerals in treated water.
Most low-income families, especially the poor, derive significant amounts of
minerals through drinking water.

100
Box I: Key Features of Bone Char-based Community Defluoridator (BCCDF)

1. Name of BCCDF system : EnDeCo Defluoridator

2. Technology
3. BCCDF unit cost
4. BCCDF container material
5. Raw water tank capacity
6. Source to be fitted with
7. BCCDF operating energy
8. Overall processing capacity
9. Basic processing facility
10. Filtration medium
11. Filtration medium
replacement frequency
12. Processing cost
13. Fluoride content testing
14. Monitoring chart
15. Monitoring personnel
requirement
16. Comparison with RO systems
of defluoridation etc.
17. Who can manufacture
: Swedish
: Rs.11,000
: Approved PVC drums & stainless steel taps
: 200 litres
: Any ground or surface water source
: Gravitation (no electricity requirement)
: 1.25 lakh litres per 50 kg of bone char
: 2 litres of fluoride water per minute
: Bone char
: 45–90 days
: 5 paise per litre
: Fluoride testing kit includes reactive agent,
test tubes and monitoring chart
: Fluoride content is indicated by
monitoring chart showing less than
1 mg/litre (pink colour-deficit
fluoride); 1 mg/litre (Brick
colour-safe fluoride level),
and more than 2 mg/litre (lemon
yellow colour-toxic)
: Any lay person with a few hours of
training can monitor the fluoride
concentrations as shown above
: Comparing very favourably as its
operating cost is lower and essential
minerals retention capacity is higher
: Any small-scale unit dealing with PVC
drums and steel water dispensers
anywhere on the basis of the EnDeCo
technology and manufacturing
specifications available with any bone
char-based BCCDF and BCDDF
promoters

Note: The BCCDF & BCDDF designs and diagrams can be obtained from Prof. Eli Dahi,
Environmental Development Cooperation, EnDeCo, Denmark. For further reading and knowledge
can be consulted the Wiley Encyclopaedia, and Proceedings of the International Workshops on
Fluorosis Prevention and Defluoridation of Water, edited by Eli Dahi et al. (1995, 97, 2000, 2004),
International Society of Fluoride Research.

101
Box II: Key Features of Bone Char-based Domestic Defluoridators (BCDDF)

1. Name of BCDDF system : EnDeCo Defluoridator

2. Technology
3. BCDDF unit cost
4. BCDDF container material
5. Raw water tank capacity
6. Source to be fitted with
7. BCDDF operating energy
8. Overall processing capacity
9. Basic processing facility
10. Filtration medium
11. Filtration medium
replacement frequency
12. Processing cost
13. Fluoride content testing
14. Monitoring chart
15. Monitoring personnel
requirement
16. Comparison with RO systems
of defluoridation etc.
17. Who can manufacture
: Swedish
: Rs.1,800
: Stainless steel containers with approved
steel/PVC taps
: 17 litres
: Any ground or surface water source
: Gravitation (no electricity requirement)
: 15,000 litres for 6 kg of bone char
: 100 ml/minute
: Bone char
: 45–90 days
: 5 paise per litre
: Fluoride testing kit includes reactive
agent, test tubes and
monitoring chart
: Fluoride content is indicated by
monitoring chart showing less than
1 mg/litre (pink colour-deficit fluoride);
1 mg/litre (brick colour- safe fluoride
level), and more than 2 mg/litre (lemon
yellow colour-toxic)
Any lay person with a few hours of
: training can monitor the fluoride
concentrations as shown above
Comparing very favourably as its
operating cost is lower and essential
minerals retention capacity is higher
: Any small-scale unit dealing with PVC
drums with steel water dispensers
anywhere on the basis of the EnDeCo
technology and manufacturing
specifications available with any bone
char-based BCCDF and BCDDF
promoters

ƒ For benchmarking the impact of BCCDDs and BCCDFs, blood samples were
collected before launching the defluoridation programme even while exposing
the fluorosis affected persons to a variety of awareness and communication
campaigns The SSOHF intervention results indicate substantial benefits like
relief from pains, ability to do domestic works, improved digestion and return to
wage/self-employment. Positive results have been achieved. This motivated
other villagers for demanding more BCDDFs and BCCDFs.
ƒ Village-level committees manage the entire defluoridation process and outputs.

102
2.4 Sustainability and viability
The Foundation makes regular visits every month and collects raw water and urine
samples for analysis by the Institute of Preventive Medicine, Hyderabad. It was found
that the fluoride content in the defluoridated water was low and acceptable.
Investigations have been made for one year and phenomenal improvement was observed
in the health of the families that were short-listed for direct observation as well as
household interviews. The positive effects included:
50 per cent reduction in Gastro-intestinal problems; 25 per cent reduction in Neuro-
intestinal manifestation; 35 per cent reduction in Muscular manifestation; 20 per cent
reduction in Allergic manifestation; 50 per cent reduction in headache, and pain while
standing and walking reduced considerably. Compared to the alum-lime technique
adopted for defluoridation earlier, the results were better as the water was safe, clean,
colourless, odourless and tasty.
2.5 Challenges
Waste disposal (filtered fluorides) is still a problem in general (the current practice is to
store the spent [sludge] bone char in gunny bags in village corners); Bone char recycling
method is yet to be satisfactorily field-tested; Establishing bone char production and
supply points locally; Building effective linkages with abattoirs for consistent supply of
bones; Control of private markets and their exploitative practices; Ensuring realistic
water pricing and user charges. As regards safe disposal of the bone char sludge, the
Indian Institute of Science, Bangalore, India has embarked on a project for using
different types of sludge including the bone char sludge for making low-cost bricks for
housing of the rural poor. The Nalgonda sludge has also been sent there. If and when
these bricks become popular in this region, the bone char sludge disposal problem can
thus be easily solved for good.
2.6 Comments
The Bone Char-based Community and Domestic Defluoridation Systems have evidently
come to stay. The acceptance of the community indicates that these systems are
technically simple, cost-effective and easy to maintain. A major lesson that development
administrators and (public health) workers should note that microinterventions should be
preferred to macrointerventions so as to avoid wastages, even while seeking to involve
the community as user-managers. Good governance systems, as highlighted, should be
the cornerstones of decentralised development. SMART governance is bound to produce
many cascading effects in the form of diffusion of innovations and best management
practices at the grassroots levels. Development partnerships between the user
committees, community and NGOs will go a long way in replicating and sustaining small
but strong local initiatives. The state should (technically) support such initiatives as one
of the partners.

103
3.0 Partners in Development– A Case Study of Integrated Rural Water
Supply and Environmental Sanitation Project, Mysore
The Integrated Rural Water Supply and Environmental Sanitation (IRWS&ES) project,
Mysore, Karnataka, India, is a unique and massive project in Karnataka that receives its
strength from partnership-based endeavours involving the Drinking water and Sanitation
User Groups, Government of Karnataka, World Bank, NGOs, Panchayats and private/
corporate consultants.
3.1 Study objectives
• To analyse the partnership based endeavours of user committees in IRWS&ES
project
• To critically evaluate and highlight the planning and management components and
(user) management lessons of IRWS&ES
3.2 Study area
Mysore, one of the 12 IRWS&ES project districts in Karnataka, was selected for study as
all the project components were implemented and that the process of handing-over of the
village drinking water systems to the user groups had begun. Fieldwork was done in
seven project villages, namely, Kadakola, Sindhuvalli, Antharasanthe, Masahally,
Bachegowadanahally, Sathekala and Doddakavalande.
3.3 Methodology
Primarily, the Participatory Rural Appraisal (PRA) tools such as Focus Group
Discussions, Transects, and Semi-structured interviews were used to gather information
and insights on partnership-based arrangements between the government, drinking water
user groups, NGOs and Panchayat Raj Institutions (PRIs) for managing safe drinking
water systems. Discussions were also held with the key officials of the project to collect
information on joint decision-making processes and their perceptions of partnerships. A
sample of 140 project users having piped water supply household connections and 18
Village Water and Sanitation Committee members was interviewed to gather their views
on partnership endeavours in the management of the project.
3.4 IRWS&ES in Karnataka
The service levels for rural water supply in Karnataka were higher than the national
average in terms of coverage. However, water consumption was still quite low,
particularly during the dry seasons, and there were strong regional differences within the
state. Karnataka was one of the states to provide at least one safe drinking water source
for all of its 27,000 villages. Yet, the Rural Development and Panchayat Raj Department
(RD&PR) had confronted many drinking water problem villages in the state through a
survey in 1993. Moreover, sanitation levels were below the national average, with less
than two per cent of the population using sanitary latrines. Diarrhoeal diseases and

104
Malaria were considered to be among the main health problems. Some villages had also
witnessed problems of dental fluorosis, especially in north Karnataka districts.
Chlorination and deflouridation of the piped water supply system was negligible in the
villages. Also, the Water quality monitoring facilities were inadequate in terms of
staffing and equipment.
In this background, the Government of Karnataka has taken up the World Bank assisted
(IRWS&ES) project through the Government of India. The prime objective of the project
was to raise the standard of living in rural areas through improved health and
productivity by expanding access to potable rural water supply systems and
environmental sanitation facilities.
The total cost of the project in the state is Rs.447.20 crore. Of this, Rs.349.20 crore
(78%) is being met through the World Bank assistance, Rs.67 crore (15%) by the
Government of Karnataka and remaining Rs.31 crore (7%) through contributions by the
village community. The project is being implemented to benefit 48 lakh people in 1111
villages of 12 districts. The project is being jointly managed by the Village Water Supply
and Sanitation Committees (VWSCs), Project Planning and Management Unit (PPMU),
and the District Project Unit (DPU). The community plays an active role throughout the
project-cycle, i.e., in planning and implementation of project activities, in operation and
maintenance of water systems, environmental sanitation and habitat development
facilities.
3.5 NGO partnerships in software activities
The following specialised NGO/consultant partners were involved in supplementing and
complimenting the software and hardware project activities:
• Mysore Resettlement and Development agency (MYRADA) was commissioned to
facilitate social mobilisation, community participation and awareness of the linkages
between safe water, health, sanitation and well-being.
• Symbiosis of Technological Environmental Management (STEM) was made a
partner in preparation of Information, Education and Communication (IEC)
materials, awareness creation among the villagers towards health education,
sanitation and to impart training to the health facilitators.
• Jagadguru Shri Sivaratreswara Swamy (JSS) was given the job of imparting training
to technicians, masons and secretaries of Gram Panchayats (GPs) for Operation and
Maintenance (O&M).
3.6 NGO partnerships in hardware activities
• Globe Consultants were responsible for identification of water sources, construction
of overhead tanks, laying of pipelines, erection of stand-posts and drainage systems
in the villages.

105
• National Industrial Development Corporation (NIDC) and Inter-continental
Consultants and Technocrats Private Limited (ICPL) were asked to oversee the
implementation of all the schemes in the project.
3.7 VWSC partnership in management
In the pre-project phase, all the villagers were involved in the planning of the water
schemes through a series of PRA exercises organised by PPMU in collaboration with the
PRIs and NGOs. Action Plans for implementing the project were prepared in
collaboration with the users and project staff. The GPs in each project village constituted
the VWSCs. These VWSCs, registered with the PPMU, are comprised of the
representatives of the drinking water users in the villages. The VWSCs are responsible
for O&M of drinking water systems in the project villages. At present, they are meeting
the O&M expenses of drinking water systems through water charges collected from the
water users. The monthly water charge was being fixed by the VWSC keeping in view
the varying O&M expenses. The VWSCs are collecting water charges at the rate of Rs.50
per month for individual private household piped water supply connections in the sample
project villages. Also, nominal charges of Rs.10 are being collected from those
households who use the facility of public tap. Evidently, the VWSCs, standpost
committees, local youth and water users are intensively involved as partners. Further, the
stakeholders are informed of various aspects of the project in the meetings of the VWSCs
to maintain transparency and accountability.
3.8 Results and discussions
The partnership endeavours of the project are analysed through a SWOT framework that
is based on the research methods, as outlined:
3.8.1 Strengths

• Water supply, environmental sanitation and health communication have been

integrated in one investment package for the first time in the State of Karnataka.
• Community participation was one of the in-built systems in the project. The local
people’s role starts from the initial stage of the project i.e. preparation of
Engineering Study Report (ESR) by adopting PRA techniques of social mapping,
resource mapping and Venn diagram. The ESR was prepared for every project
village in the district through a private agency. The joint ESR exercise has enhanced
the project appropriateness and acceptability.
• It was made mandatory that the local water users should contribute 30 per cent of the
cost towards drainage and culvert constructions. This share is shown in a bank
account by the VWSC to the DPM. Then alone, the cited water and sanitation
schemes are implemented in the selected village under the project. The PPMU,
NGOs like MYRADA, JSS, STEM, the local panchayats and the elderly persons of
the village were involved in IEC campaign and mobilisation of 30 per cent people’s
contribution towards drainage and culvert construction. The local people have

106
 responded overwhelmingly and voluntarily in evolving and availing the project
benefits. The 30 per cent contribution of the people to the environmental sanitation
component of the project was essentially aimed as much at making the people
identifying with the project as being partners in a welfare scheme. The capital cost
for facilities like water storage in ground level and over head tanks, the cattle
troughs, dustbins, washing platforms had been met by the Project Authority. Due to
low rural income levels, there was no capital-cost recovery for water supply
facilities. However, user charges were levied by the VWSCs at the rate of Rs.50 per
month for private household piped water supply connections. Nominal charges of
around Rs.10 were being collected from those households which had utilised the
facility of public tap.
• Women’s participation was greater in the project villages when compared with the
non-project villages. This is one of the project strengths, which is stressed by the
World Bank on gender-oriented approach for effective women’s participation.
Standpost committees were formed in the project villages. Most of the committee
members are women from nearby houses who were utilising the standpost. They
were not only responsible for better upkeep of the standpost but also for economic
use of water.
• Street-level committees were also formed in the project villages for monitoring and
upkeep/cleaning of drainage system on a regular basis. The selection of committee
members was done by the VWSC.
• The IEC components are strong in the project villages. Due to the effective
functioning of extension staff, VWSCs and NGOs inasmuch as the enthusiasm of the
VWSCs, the quality of programme implementation was high. As part of IEC,
pamphlets and folders were distributed and messages were written on the walls
regarding safe water, sanitation and hygiene practices in the regional language
(Kannada). Regular house visits and street corner group meetings were also held by
the grassroots village level health facilitators. Folk media programmes were
organised alongside screening video film shows through cable operators and also by
Information and Publicity Department, Government of Karnataka. School children
were also covered under the IEC activities by the extension workers with a view to
conveying the messages to their parents and to the neighbouring families on safe
water, environment and sanitation aspects. Because of these interventions, diseases
like Malaria and Diarrhoea were negligible. The villagers mentioned that there is a
gradual increase in the overall health status of the villagers. Visiting the health
centre/doctor was a common feature even for common diseases like cold and cough,
among other ailments, owing to high levels of health consciousness. It also saved a
lot of family mandays for women and children in terms of fetching water from
distance places. Thus, the project helped in saving of many family mandays (which
otherwise would have gone waste in terms of sick days) to help augment
employment opportunities and family incomes in the project villages.

107
• Rainwater harvesting was one of the components of IRWS&ES project. NGOs,
PRIs, and government encouraged the community to adopt rain water harvesting
practices for maintaining constant ground water recharge at source for sustainable
water supply.
• Training of local youth for O&M was another in-built component of the project. The
O&M activities not only provided employment opportunities for local youth but also
helped run the drinking water systems better. People’s participation through O&M
and training of local youth was ensured for better management of the project.
• Initially, the stakeholders had to invest their own money for constructing the sanitary
latrines. Later, the health facilitator collects the filled-in subsidy application forms
from the stakeholders and submits to the VWSC. The VWSC physically checks the
constructed toilets and forwards the application to the DPM for sanction of subsidy
(Rs.1200 for general category and Rs.2000 for BPL category).
3.8.2 Weaknesses

• A negative tendency towards 30 per cent contribution was found in a few project
villages such as Antharasanthe and Sathekala. Community contribution (30%) is not
encouraging in such villages because of availability of alternative drinking water
sources like open wells, river streams nearby and piped water supply provided under
the Community Development Programme. Perhaps, the extension workers too have
failed to convince the villagers regarding the benefits of the project. Therefore, the
awareness and contribution campaigns should be revised and intensive persuasion
approaches adopted.
• The people’s participation was inadequate in some villages like Anantharsanthe and
Sathekala. This had happened because of various socio-economic and management
factors like ineffective functioning of VWSCs, non-coverage of the entire village
under the drinking water and sanitation programme and availability of (pre-project)
old/alternative drinking water systems in the villages. Improper selection of
beneficiaries under the individual sanitary scheme is another factor for low people’s
participation in these villages. It was also observed that a few representatives from
BPL and SC&ST families in these villages were not actively participating in the
programme because of their low socio-economic status or inadequate efforts of the
project authorities or both.
• In the focus group discussions, it was noted that only female and old members of the
family were using sanitary latrines. Open defecation is still prevailing in the project
villages, despite intensive IEC efforts. Consequently, the public health and hygiene
systems are under great pressure.
• The concerned NGOs have failed to create awareness for 30 per cent contribution of
drainage and culvert constructions in Sathekala and Antharasanthe villages.
Consequently, the implementation of the project was delayed in these villages. The
awareness building exercise on project modalities did not get the expected support of

108
 some political parties. Some vested interests had created an impression amongst
people that providing drinking water is the responsibility of the government and does
not need any people’s participation, particularly the contribution aspect.
• In spite of interest, some of the families in Doddakavalanede Village were unable to
accept the sanitary units because of lack of space for construction. As the
beneficiaries had to bear the cost of sanitary units initially, some of them could not
avail the subsidy benefit due to inability to invest.
• There is a belief among the local people that having a sanitary unit in the residence
would emit fowl smell. Therefore, some families did not accept them. The traditional
house plans had no provision for sanitary latrines. This has resulted in non-
acceptance of sanitary latrines by some families. Therefore, stepping-up of
programmes for house upgradation is needed so that sanitary units can be
accommodated in the houses or in the backyards. In addition, greater IEC efforts for
adoption of sanitary latrines are required.
• NGO partnerships (software) is very limited and confined to only the initial stages of
the project. They should have been allowed to continue till the end of the project. It
would have been still better had there been provision to retain them for some more
time even after handing over the project to the VWSCs for improved sustainability
of the projects.
• Irregular supply of power is the biggest problem in the project villages. Therefore,
the distribution of water with sufficient pressure in the pipelines was frequently
affected in some villages resulting in dependence of people on unsafe traditional
water sources like open wells and rivulets.
3.8.3 Opportunities

• The success of this partnership project would help replicate the project in other
uncovered problem villages/districts within and outside the state.
• Usually, girl children in the villages are assigned to fetch drinking water causing
increase in absence/school dropout rate amongst the poorer families in particular.
This project and its attendant benefits (e.g. time saving through piped water supply)
would help reduce the absence/school dropout rate and thereby increase the literacy
levels in the long run, especially of girl children.
• As safe water is provided inside or nearer the households, the distance problem has
been solved. This is saving family labour and enhancing work and family incomes.
3.8.4 Threats

• There is a lurking question of sustainability of the project after it is completely

handed over to the VWSCs/panchayats in all the project villages in the state. If the
VWSCs fail to perform the duty of O&M efficiently and consistently in all the
project villages, then, the Government of Karnataka may hand over the project to a
private agency for O&M, just as they did in the case of the Bangalore Urban Water

109
 Supply, which was handed over to a multi-national private agency for O&M in
Bangalore city.
• The present water charges and collections would only suffice to meet the costs of
power, minor repairs and salaries of maintenance staff. No financial
provisions/reserve fund exists to meet unforeseen major repair costs of motors and
pipeline systems. In case major repairs occur, the VWSCs may be forced to approach
the government.
4.0 Where Child is Mother of Development: Bal Panchayat Programme
The Bal Panchayat Programme (BPP) of the Institute of Health Management, Pachod
(IHMP), Aurangabad, Maharashtra, India is a participatory initiative for holistic
development of children as well as the disadvantaged with focus on the poor, especially
illiterate women. Children aged 6-14 years (Bal Sevaks- boys and Bal Sevikas- girls) are
the potential resource of this child-centred democratic decentralised development
programme. They provide voluntary services for the development of children and
community in their villages through social mobilisation, participation and empowerment
strategies. The broad objective is to develop skills and leadership in children through
their involvement in rural development initiatives such as Health, Drinking water and
Sanitation. The five essential components of the BPP are: Organisation of children and
leadership development through Bal Panchayats; Non-formal education of illiterate
children through a child to child strategy; Health education (child-to-adult strategy) for
changing community behaviours, and Voluntary action for village development (child-to-
community approach).
The Bal Panchayats (BPs), in session in Karanjkeda Village, have thrown up rich insights
and lessons for community-based women and child-centred development interventions.
The children were very enthusiastic in demonstrating their skills in conducting literacy
and education classes for school dropouts as well as illiterate adults, especially women.
While there was meaningful interaction between the children and adults in the
demonstration event, a significant number of onlookers comprising children and adults of
the village commended the contributions of BPs, even while posing mobilisation and
empowerment oriented questions. What was even more interesting was the demonstration
exercise presented by the children to enlighten the visitors about the modus operandi of
the BPs. The President, the Secretary and the members of the BPs played their roles as
they do in the regular BP meetings– often with focus on water conservation, rainwater
harvesting and sanitation. The confidence with which they conducted their 'BP business'
was remarkable in that there were even some animated discussions between the President
and the Members and the villagers on key aspects relating to access to water, water
quality monitoring and hygiene education. Immediately after the BP meeting was
completed, the Secretary, an 11-year-old girl, read out the minutes of the meeting, written
in the 'impromptu' meeting, with great confidence and ease in communication.
One can also see how health, water and sanitation literacy is imparted to the adults,
especially women by the children both by way of Flip charts, Role plays and

110
Demonstration. In fact, in many villages of Pachod, one gets to see a very unusual but a
critical development event of children and adults worshipping the 'Sanitation Baba',
namely, Sant Gadge Baba in the villages. This great social worker had inspired the
villagers in sanitary upkeep of both households and villages so much so there are some
total sanitation villages in the vicinity of Karanjkeda, where all the households are
connected to a semi-modern/modern liquid waste disposal and solid waste disposal
system. And unusually so, many of the total sanitation villages are comprised of
households with common outer colours (cream or white). The common colour, apart
from depicting some measure of cleanliness and equality among households, also serves
the purpose of ‘heritage building’ by way of common house designs and colours.
5.0 Conclusion
Good governance is the key to the window of the new millennium of opportunities.
Effective and efficient people-managed development, prudent public expenditure
management, and participatory monitoring and evaluation of service delivery are seen as
non-negotiable development instruments. A major lesson that development
administrators and workers should note is that microinterventions should be preferred to
macrointerventions so as to avoid wastage, even while seeking to involve the community
as user-managers. Good governance systems should be the cornerstones of decentralised
development. SMART governance is bound to produce many cascading effects in the
form of diffusion of innovations and adoption of best management practices at the
grassroots levels. Gender sensitive approaches should be built into the programme
designs, even while recognising the effectiveness of child-to-child and child-to-adult
communication strategies and action-oriented interventions. Participatory planning,
monitoring and evaluation should be integral in all microprojects. Towards capacity
building of the (user) communities, regular IEC and skill oriented training programmes
should be organised. Where necessary, external support agencies must follow a staggered
exit strategy so as to enhance the self-reliance of communities, development partnerships
between the user groups, community, NGOs and CBOs will go a long way in replicating
and sustaining small but strong local initiatives. Networks of user groups and
communities will further the process of consolidation of good governance endeavours,
knowledge generation, resource sharing and capacity building. The state should support
such initiatives as one of the partners.

111
References
Panchayati Raj Engineering Department. 2005. Action Plans of Rural Water Supply Wing. PRED, GOAP.
Brochures and Pamphlets of IRWS & E Project, Bangalore, 1999-2002.
National Institute of Rural Development. 2005. Background Material of the Workshop on Rural Drinking
Water Supply: Reforms and Bharat Nirman.10-11 August. NIRD, Hyderabad.
Choudhary, R.C., and P. Durgaprasad. 1999. Basic Rural Infrastructure and Services for Improved Quality of
Life, Vol. I & II. National Institute of Rural Development (NIRD), Hyderabad.
Durgaprasad, P. 2000. Case Study of Sri Satya Sai Safe Drinking Water Project, Ananthapur. NIRD,
Hyderabad.
Durgaprasad, P. 2001. Mahsana Report of Partnerships and Networking in Drinking Water Projects: User
Group, NGO and Training Institute Initiatives in Gujarat. NIRD, Hyderabad.
Nageswar, Rao, M. 1990. Evaluation of Hand Pump Programme in Karnataka. Institute for Social and
Economic Change, Banglore.
Planning Commission Documents on Basic Minimum Services 2003, Gol.
Report of the Bal Panchayat Programme of the Institute of Health Management, Pachod. Aurangabad.
Programme Brochures/Newsletters. 2004.
Sivarm, P. 2000. The Problems of Rural Drinking Water Supply and Sanitation. Journal of Community
Guidance and Research, Vol. 17, No. 2.
Sudhakar Rao, M. 2005. Reports on Water Quality by the Department of Mechanical Engineering. Indian
Institute of Science (IISc), Bangalore.
Water and Sanitation Research Papers and Booklets on IEC. 200-06. NIRD, Hyderabad.

112
 Asia-Pacific Journal of Rural Development
Vol. XVII, No. 1, July 2007

Land-use Suitability Evaluation Criteria for Precision

Agriculture Adoption in a Moderately Yielding
Soya bean Cropping Area in Thailand
Kishore C. Swain* and H.P.W. Jayasuriya∗

Abstract
A land-use suitability evaluation criteria was developed for Precision Agriculture
technology adoption in developing countries using a case study conducted in moderately
yielding soya bean cropping area in Thailand. Suitability maps were developed for geo-
referenced crop and soil data using GIS. The suitability criteria, comprising 12 physical
and 3 economic factors revealed that 77 per cent of the total study area was suitable for
soya bean crop cultivation. By assigning equally weighted factors, 86 per cent of the area
was found highly potential for PA adoption with higher yield goals. The yield comparison
justified the suitability classification used in selecting lands for PA adoption. Suitability
maps can assist farmers in selecting areas for step-by-step adoption depending on their
financial capabilities or in choosing other crop options for areas not suitable for soya
bean.

1.0 Introduction
Precision Agriculture (PA) helps properly in managing crop production inputs in an
environmentally friendly way while increasing profits (Jin and Jiang 2002). By using
site-specific knowledge, PA can target correct inputs; rates of fertiliser, seed and
chemicals for soil and other conditions (Bongiovanni and Lowenberg-Deboer 2004). It is
possible to point out the areas that will always have low yield. If the maximum capability
of an area is 3.40 tons an hectare, there is no need to use fertiliser for 8.15 tons, and it is
not only a waste, but also it may create negative effects (Mask 2000). Such over or under
applications are very common in developing country situation (Cook et al. 2003).
Perhaps proper crop selections and crop rotations could mitigate this problem
significantly. PA can be suitably defined as doing the right thing, at the right place and at
the right time (Colvin and Kerkman 1997) and meant to use up-to-date technology in
agriculture for obtaining maximum profit. Therefore, there is a strong possibility that
appropriate adoption of PA could be helpful in addressing the poverty alleviation, food
security, improved livelihood as well as sustainability in agricultural production under
developing country context (Maohua 2001).

∗
Agricultural Systems and Engineering Programme, School of Environment, Resources and Development,
Asian Institute of Technology, Pathumthani, Thailand. Corresponding e-mail: hpwj@ait.ac.th

113
The conceptual PA has been practised in its traditional form since ages in agricultural
communities. In the developed countries, PA with cutting-edge technology has been
practised for past decade with tremendous success (Srivastava 2001). The access to the
affordable technology and sincere government support assisted with intensive research
work has created a healthy environment for PA technology adoption in developed
countries. The meaningful adoption of PA technology could be promising in enhancing
the productivity as it has already been proved in some applications in developing
countries (Cook et al. 2003; Zhang et al. 2002). Therefore, in this study, the current
standard of the PA technology was slightly modified to suit the conditions of developing
countries.
For primary crop analysis, soya bean crop was selected for the study. Soya bean is one of
the demanding secondary crops in Thailand, which contributes around 30 per cent of the
annual demands. Soya bean is cultivated in many areas and regions showing countrywide
inconsistency in yield. In order to evaluate the land-use suitability for such cropping
areas, a land suitability evaluation criterion was developed using geographic information
system (GIS) for the selection of lands for PA technology adoption. The research was
conducted for moderately yielding soya bean grown in Khon Kaen Province of Thailand.
2.0 Basic Considerations
2.1 Precision agriculture technology for developing countries
The application of GIS techniques in PA is multipurpose, such as conservation of
important plant species in land-use planning, land-use suitability evaluations, crop
selections and rotations, irrigation and mechanisation planning etc. (Setojima et al. 1988;
Yanamoto 1988; Cruz 1992; Gouvenian 1995; Swain 2001; Runquist et al. 2001).
In developed countries large land areas are divided into square grids for analysing the
special variability of management properties. The average farm size in the developed
countries like USA is 182 hectares (Statistics on USA 2002) and that of Europe is 20
hectares (Hennessy 2003). The smaller average farm-size in the developing countries
restricts its further sub-division into smaller grids, which cannot be economically
justifiable in any means. However, in GIS, an individual small farm plot could be
considered as a single grid unit (Aronoff 1989).
The ‘between-farm’ variability was accounted for technology adoption taking each small
farm as a single grid unit considering twelve major soil nutrients and crop input
parameters. The purposes of applying appropriate rate of inputs at specified place are
achieved by estimating the management practice (viz. application rate) and the
requirement rates of the individual farmlands. The farmland size in developing countries
generally varies between 0.5 ha to 1.5 ha, was considered as an individual grid unit,
while the farmlands larger then 1.5 ha were treated by further dividing into smaller grid
units, where precision agriculture could be practised.

114
2.2 Suitability classification
Vien (1996) found the suitable land for the crop by dividing the whole land into four
classes such as highly suitable, moderately suitable, marginal suitable and not suitable
area. In this study, the crop nutrient suitability was categorised under five major classes
as previously practised by many researchers. The nomenclature followed was described
by FAO (1993) for irrigation suitability evaluation, namely, S1: Highly suitable, S2:
Suitable, S3: Marginally suitable, N1: Marginally unsuitable and N2: Permanently
unsuitable.
2.3 Suitability evaluation criteria
The land suitability evaluation study carried out by Rai (1996) in Lopburi Province,
Thailand used GIS for finding the suitable lands for crops such as rice, maize, and
cassava etc. In the study equal importance was given to all yield limiting parameters such
as soil texture, drainage, rainfall, soil depth, slope, soil reaction and fertility class with
the available secondary information. Wong et al. (2003) applied weight-in-evidence
modelling to find greater suitability area for wheat cropping in Australia during their
profitable land-use change study through precision agriculture. The inputs from the
farmer and agronomist were used to decide on the relationship between the evidence and
its postulated effects for modelling.
In the suitability evaluation, some selected crop, land and operational based yield
limiting parameters were considered as crop influencing factors in estimating the final
suitability. The parameters represented the condition for the better crop growth with
maximum crop productivity. The difficulty in comparing importance of any individual
parameter having inter-relation with others, led all the parameters to be treated equally
important. In some cases weighting factors can be taken into account considering
dominating behaviour of some parameters, especially when the application determination
process is concerned. This gives the overall view of land suitability for the crops. Final
suitability was determined by simple multi-criteria analysis as shown below:
x=n
F = (wx) ∑f
x =1
x (1)

Where, F = Final suitability value

fx = Individual suitability value
n = Number of parameters
wx = Ranking and weight factor for parameters
3.0 Methodology
The study area selected was one of the major soya bean producing areas with moderate
yield in Thailand. Five villages scattered over nearly 20 sq km area and farmers growing
soya bean crop with an average farm size in the range of 0.5 to 2 ha were selected for the

115
study. The owners of the farms were included in the study to gather farm practice
information through personal interviews. Typical rural farming patterns and practices
were observed in the selected area. Each farmland was treated as a single grid unit with
the size varied from 0.5 ha to 2 ha. Farmlands with the area greater than 2 ha were
converted into two grids, and both were treated as individual grids. This assisted in
maintaining uniform size among the grids. Soil samples were collected from all
individual grids with GPS receiver readings to determine the universal location of the
grid. Analyses of soil nutrients were carried out using the standard procedures.
3.1 Selection of parameters
Attempts were made to cover all the major parameters affecting crop growth and crop
yield under the available farming practices. All the influential factors were combined into
two classes: i) Physical factors, and ii) Economic factors.
3.1.1 Physical factors (crop and soil parameters)
The physical factors were further classified into four major groups. Each group was the
combination of two or more individual parameters. Twelve major influential soil and
crop input parameters were considered for the precision agriculture studies under
following classification.
Group A - Primary soil nutrients (nitrogen, phosphorous, potassium, calcium): Vonlaya
(1983) reported that the application of potassium and phosphorous fertiliser helps in
increasing the yield of soya bean. The soil nutrients like nitrogen, phosphorous and
potassium play vital roles in crop growth and also their available level can easily be
manipulated. Chances of over-application of inputs could not be neglected, whereas PA
discourages application of such higher levels than the required level for a better profit
margin. So the suitability range would include both positive and negative
recommendations (Table 1). The term ‘deficiency level’ (Deficiency level = Nutrient
required - Nutrient available) was used for deciding the range of suitability.
Group B - Secondary soil nutrients (pH, organic matter, soil texture): Soil organic matter
influences many of the physical, chemical and biological properties of the soil. Some of
the properties influenced by the organic matters include soil structure; water holding
capacity, nutrient contribution, biological activities, water and air infiltration rate and
pesticide activities. Soil pH and soil texture influences the organic matter decomposition,
whereas pH also has direct impact on nutrient availability and plant growth (Weaver et
al. 2004). Soil texture determines the soil pore space. Soil texture was determined from
the analysis of clay, silt and sand content using the universal triangle. The available level
was compared with the requirement to calculate the suitability range (Table 1).
Group C - Management factors (mechanisation level, irrigation frequency, drainage
level): The study by Hardjono et al. (2003) revealed that 73.1 per cent of the smallholders
engaged in oil palm production mostly adopt low to very low level of management
practices in Indonesia. Nearly similar situation exists for soya bean cropping area in

116
Thailand. Status of the mechanisation has a potential influence on the yield as well as
profit margins. Application of modern technology was determined by quantifying
mechanisation level of the farmland in terms of total power input (Singh 1994). The
availability of water and its use has a direct impact on the agricultural productivity (Zhen
and Routray 2002). Irrigation was mostly performed by the flooding method in the study
area. It was a tough task to quantify amount of water used for irrigation with the flooding
method, the irrigation frequency was taken into consideration therefore. The general
recommendation of Agricultural Department for wheat crop as reported by Hussain et al.
(2002) was to apply four irrigations in Lower Indus region. As there was availability of
water through the irrigation canal, no water scarcity was assumed for the crop. Lack of
literature on local drainage facilities for agricultural crops, the drainage level was
quantified through the farmer’s perception.
Group D - Non-fertiliser input factors (seed rate, pesticide rate): Regulation on seed and
pesticide rate should be maintained at proper level to get the maximum crop output.
Application rate crossing the range in either way could affect the crop as well as the final
profit margin.
3.1.2 Economic factors (input, return, profit)
The total household income of smallholders comprised monetary return and benefits
derived from different sources in the farm and off-farm activities (Hardjono et al. 2003).
Economic factors were determined by calculating the net farm income. The net farm
income can also be used in order to find out the area for maximum profit potential. Profit
was determined as the difference in return cost to total input cost.
3.2 Total suitability for precision agriculture
Both the physical and economic factors were assumed to have equal effect on the
suitability evaluation for the crop. After developing the suitability class for physical
factors, it was compared with the economic factors for selecting the area suitable for PA
adoption.
g
gg
Suitability for PA = gg (Value for Physical factor + Value for Economic factor) (2)
Arc View 3.2a GIS application software was used for creating GIS land-use suitability
maps. Profit maps were also developed in order to compare the current profit with the
expected profit levels. For obtaining profit maps, the following parameters were taken
into consideration:
ƒ Additional investment for PA adoptions (in terms of inputs, technologies, and
equipment)
ƒ Current profit level
ƒ Expected profit after PA adoption

117
 The profit margin for individual farmland was determined by taking the difference in the
expected profit margin with precision agriculture to the current profit margin of the
farmlands.
Table 1: Suitability Range Distribution of Physical Factors for
Precision Agriculture
Physical
factors Parameters S1 S2 S3 N1 N2
0- 5 5-8 8-11 11-14 14-16
P, mg/kg -5-0 -10-(-5) -15-(-10) -100-(-15)

K, mg/kg 0-20 20-40 40-60 (60-80) 80-200

Primary soil -20-0 -40-(-20) -60-(-40) -200-(-60)
nutrients
0-2 2-4 4-6 6-8 8-100
N, mg/kg -2-0 -4-(-2) -6-(-4) -100-(-6)
Ca, mg/kg 200-400 150-200 125-150 100-125 50-100
pH, value 6.5-7 6-6.5 5.5-6 5-5.5 3-5
OM, g/kg 20-50 16-20 13-16 10-13 6-10
Secondary
soil nutrients Loam Clay Sandy
Soil texture & Clay Loam Sandy Loamy
Clay Loam Loam Loam Sand

Seed rate, 75-95 95-115 115-130 139-220

Non- kg/ha 65-75 50-65 40-50 15-40
fertiliser 0.3-0.62 0.62-0.95 0.95-3.125
inputs Pesticides,
0.125-0.3 0.095- 0.063- 0.032- 0-0.032
kg/ha
0.125 0.095 0.063
Mechanisation 235-470 187.5-235 140-187.5 95-140 50-95
level kW/ha
Management Irrigation
6-10 5-6 4-5 3-4 0-3
status frequency/crop
Drainage Very
Level Excellent Good Fair Poor
Good
Source: Ratanmart and Kiratikasikorn 1990; Field Crop Research Institute, Bangkok, Thailand;
Discussion with expertise from FAO.

3.3 Zone selection for site-specific management

The selection of zones for site-specific management takes an important role for the
success of precision agriculture application. The management zones can be selected and
categorised for the areas having similar yield limiting factors. The zones can be selected
on the basis of fertiliser variability, pesticide variability, mechanisation levels and
irrigation etc. and can also be shown in the maps, created using the GIS software.

118
The working range of any parameter for the zone selection was determined based on the
total available rate of nutrients for the current crop. The range for parameters like N, P
and K was selected on the basis of recommended value for suitability class. The range for
seed rate and pesticide rate were selected to show the application rate from lower to
excess rate with the recommended value as middle zone. The range for mechanisation
level was decided in terms of total power input to individual farmlands.
4.0 Results and Discussion
Analysing the variation in the yield of soya bean might call for considering all possible
factors directly or indirectly influencing crop growth and productivity. The soya bean
yield map showed a distinct variation of the yield through the area, classifying the study
area under medium soya bean yield zone with high potential (Figure 1). When considered
the physical factors, only 7 per cent of the area was found to be under permanently
unsuitable class, and 86 per cent of the area was suitable for soya bean growth (Table 2,
Figures 2-6). The results of economic factor suitability study (Figure 7) showed 21 per
cent of the total area pertaining to permanently unsuitable (N2) class showing the current
low soya bean yield level. Fifty-seven per cent (57%) of the land area was fallen under
marginally unsuitable (N1) class indicating poor investment and input supply in the area.
Table 2: Land-use Suitability Area Distribution
S1, ha S2, ha S3, ha N1, ha N2 , ha Total area,
Class
(%) (%) (%) (%) (%) ha (%)
Physical factor 350.44 718.05 443.67 313.28 136.07 1961.51
suitability (17.87) (36.61) (22.62) (15.97) (6.94) (100)
Economic factor 0.95 22.72 401.68 1115.56 420.6 1961.51
suitability (0.05) (1.16) (20.48) (56.87) (21.44) (100)
Precision agriculture 90.30 525.33 626.63 534.94 184.31 1961.51
suitability (4.63) (26.78) (31.95) (27.27) (9.47) (100)
Precision agriculture 525.33 626.63 534.94 1539.96
- -
application (26.78) (31.95) (27.27) (86.00)
Source: Field study by author.

Five per cent (5%) and 9 per cent areas were under categories of highly suitable and
permanently unsuitable classes respectively (Figure 8) in the PA suitability distribution.
The remaining 86 per cent of the land was marked for precision agriculture application
(Table 2). Forty per cent (40%) area was fallen under marginally unsuitable (N2) class
indicating the need for alarming attention, which still has potential to come up as
profitable soya bean growing lands. The areas returning lowest profit, classified as
unsuitable for PA could be tried with other crop combinations.
The criteria can be applied in any region before going for PA adoption involving huge
investments. It will also enable the farmers for selecting the area for step-by-step
adoption of PA without going for full-scale adoption depending on their financial

119
capabilities. The information collected during the analysis could be employed in
developing PA technology recommendations.
The profit map (Figure 9) showed an upper side of the expected profit margin over the
current profit level. Around 90 per cent of the farmland, selected for precision agriculture
adoption, was found to give appreciably higher profit margins over the farmlands
practising traditional agriculture. The site-specific zones created will allow the farmers
for proper distribution of the individual nutrients. Figure 10 shows the site-specific zones
created for potassium application in the study area as an example.
The final suitability area was further verified with the ultimate source of return from
agriculture i.e. crop yield. The mean yield summarised with the suitability zone produced
nearly expected results (Table 3). The yield of the first two suitable classes showed
nearly similar average yield and the yield decrease with the descending order of
suitability classes. The relatively lower yield for highly suitable zone (S1) may have
appeared due to the less area under its belt. The average yield of marginally unsuitable
area (N1) is lowered in values containing a greater region underneath (27%) and should
be considered for special attention since it has potential to come up as profitable soya
bean growing lands. The minimum and maximum yield variation has also shown a
similar trend as the mean yield, with decreasing trend from highly suitable to
permanently unsuitable class, further justifying the criteria (Table 3).
Table 3: Soya bean Yield Analysis

Suitability class Area (ha) Mean yield (t/ha) Min. yield (t/ha) Max. yield (t/ha)
a a
S1 90.300 2.4697 1.8755 3.1275a
S2 525.330 2.5101a 1.6399ab 3.6926b
S3 626.630 2.3106b 1.4634bc 3.5768bc
N1 534.940 2.0269c 1.4398c 3.3743abc
N2 184.310 2.0211c 1.4191c 3.3279ac
Note: Level of significance is 0.05.

120
Figure 1: The map of soya bean yield Figure 2: The map of land-use suitability
distribution in the study area distribution based on primary
soil nutrient status

Figure 3: The map of land-use suitability Figure 4: The map of land-use suitability
distribution based on secondary distribution based non-
soil nutrients status fertiliser input factors

121
Figure 5: The map of land-use suitability Figure 6: The map of land-use suitability
distribution based on distribution based on physical
management status factors factors

Figure 7: The map of land-use suitability Figure 8: The map of land suitability
distribution based on economic evaluation for precision
factors agriculture application

122
Figure 9: The map of land suitability for Figure 10: The map of site-specific zones
expected profit margin over for potassium application.
traditional practices

5.0 Conclusion
The land-use suitability evaluation and fine-tuning of the crop selections and rotations
were found very important in optimising the land and labour productivities leading to
higher cropping intensities and yield goals in agricultural production. Some of the new
technologies and concepts used in PA seem to be appropriate for eradicating the
prevailing inefficiencies in crop production practices in the developing countries. The
study developed a methodology and guidelines for land-use suitability evaluation using
multifactor analysis, which could systematically evaluate the status and could simulate
the potential of the system for higher gains.
The twelve physical factors including economic factors used in the study were found
truly representing the current farming conditions. The criteria developed for selecting the
area for precision agriculture technology adoption could be used for any similar
condition including the small farmland-based agriculture of the developing countries.
Economic factor analysis showed a low investment in soya bean crop cultivation with
low current profit margin. The final suitability distribution revealed that 86 per cent of
the farmland could be suitable for PA technology adoption with very high potential. The
comparison of soya bean yield under five land classifications further justified the criteria
of suitability classification in the selected area for precision agriculture adoption.
Around 90 per cent of the farmland selected for precision agriculture adoption was found
capable of yielding higher profit margins over the farmlands practicing traditional
agriculture. The areas classified under permanently unsuitable could be diversified for
other crops, instead of soya bean befitting their suitability levels.

123
References
Aronoff, S. M., 1989. Geographic Information System: A Management Perspective. Canada: WDL Publication.
Bongiovanni, R., and J. Lowenberg-Deboer. 2004. Precision agriculture and sustainability. Precision
Agriculture, Vol. 5(4). Pp.359-387.
Colvin, T. S., and E. W. Kerkman. 1997. Precision agriculture requires precision tuning. In Proceedings of the
4th International Conference on Precision Agriculture. American Society of Agronomy. Madison.
Wisconsin. Pp.1083-1088.
Cook, S. E., R. O. Brien, R. J. Corner, and T. Oberthur. 2003. Is precision agriculture irrelevant to developing
countries? In Stafford J. and A. Werner edited Precision Agriculture. Wageningen Academic Publishers. The
Netherlands. Pp. 115-119.
Cruz, R. A. D. 1992. The Demonstration of Suitable Upland Agriculture Areas Using GIS Technology. Asian
Pacific Remote Sensing Journal, 5 (1). Pp. 35–56.
FAO. 1993. A Frame Work for Land Suitability Study for Irrigated Agriculture. Rome.
Gouvenian, R. D. 1995. Use of GIS for Sensitive Plant Species Conservation in Land Use Planning. The
Environmental Professional. USA. Pp. 17, 27-33.
Hardjono, W., S. L. Ranamukhaarachchi, and G. Singh. 2003. Factors affecting adoption of management
practices in smallholders’ oil palm plantation of Banten Province, Indonesia. Asia-Pacific Journal of Rural
Development, Vol. 13(1). Pp.24-42.
Hennessy T. 2003. Farming in EU: A Look at Competitors. http://www.fapri.miss
ouri.edu/Publications/2003Publications/RepFarms03/RepFrm03_ireland.pdf.
Hussain, I., F. Manikar, and W. Jehangir. 2002. Performance analysis of irrigated wheat farms in the lower
Indus basin. Asia-Pacific Journal of Rural Development, Vol. 12(2). Pp.82-105.
Jin, J., and C. Jiang. 2002. Spatial variability of soil nutrients and site-specific nutrient management in the P. R.
China. Computers and Electronics in Agriculture, 36(2002). Pp. 165-172.
Maohua, W. 2001. Possible adoption of precision agriculture for developing countries at the threshold of the
new millennium. Computer and Electronics in Agriculture, 30(2001). Pp. 45-50.
Rai, S. K. 1996. Land Use Suitability and Evaluation by Crop Modeling and GIS: A Case Study of Lopburi
Province, Thailand. AIT Thesis No. AE-96-38. Bangkok.
Ratanmart, S., and P. Kiratikasikorn. 1990. Soya bean soil and fertilizer research review paper. In Proceedings
of the 9th Thailand National Peanut and Soya bean Meeting, Nakhala Ratchisima, Thailand, 7-11 May 1990.
Runquist, S., N. Zhang, and R. K. Taylor. 2001. Development of a field level geographic information system.
Computer and Electronics in Agriculture, 31(2001). Pp. 201-209.
Setojima, M., T. M. Horiuchi, and K. Yoshinga. 1988. GIS for Agriculture Development by the Use of Remote
Sensing and GIS. Proceedings of Workshop on GIS, ESCAP/UNDP/National Research Council of Thailand,
Bangkok, 21-29 November 1998.
Singh, G. 1994. Energy input in production agriculture of India: During last four decades. Agricultural
Engineering Journal, Vol. 3 (1 & 2). Pp.1-20.
Srivastava, S. 2001. Space Inputs for Precision Agriculture: Scope for Proto-Type Experiments in the Diverse
Indian Agro-Ecosystems. ISRO. Bangalore, India.
Statistics on the Unites States of America. 2002. Cobra USA Statistics USA..
http://www.cobras.org/usastats.htm.
Swain, S. 2001. Geo-Information System for Land Suitability Assessment and Participatory Land Use Planning
and Management for a Micro-Watershed in Orissa, India. AIT Thesis No. RD-01-10. Asian Institute of
Technology. Bangkok, Thailand.
Vonlaya, K. 1983. Effect of Soil and Fertilizer-P and Lime on Soybean. AIT Thesis No. AE-83-8. Bangkok,
Thailand.

124
Vien, N. D. 1996. GIS-Assisted Land Suitability Analysis: A Case Study of Buonmathuot District, Vietnam.
AIT Thesis No. HS-96-9. Bangkok, Thailand.
Weaver, A. R., F. Chen, D. E. Kissel, L. T. West, W. Adkins, D. Rickman, and J. C. Luvall. 2004. Mapping soil
pH buffering capacity of selected fields in the coastal plain. Soil Science Society of America Journal, Vol.
68(2). Pp.662-687.
Yamamoto, H. 1988. Land Resources Evaluation for Agriculture Development by the Use of Remote Sensing
and GIS, Asia Pacific Remote Sensing J., 1(1). Pp. 37-53.
Zhang, N., M. Wang, and N. Wang. 2002. Precision Agriculture-a worldwide overview. Computers and
Electronics in Agriculture, 36(2002). Pp. 113-132.
Zhen, L., and R. K. Routray. 2002. Environmental Impact of groundwater exploitation in the North China
plain’s agriculture farmer’s perception and remedial measures. Asia-Pacific Journal of Rural Development,
Vol.12 (2). Pp.14-32.

125
 I N D E X

Asia-Pacific Journal of Rural Development

(Volume: I – XVI)

CONTENTS
Vol. I, July 1991, No. 1

1 Agriculture and Rural Development in the Asia-Pacific Region: Present

Scenario and Future Outlook
AZM Obaidullah Khan

7 South Asia: Those who are Left Behind - Issues and Perspectives
Nurul Islam

Gender and Rural Poverty in Asia: Implications for Agricultural Project Design
41
and Implementation
C. Safilios-Rothschild

Poverty and Gender Issues

62
Alexandra Stephens

75 Food Policy and Rural Poverty Alleviation in India

R. Thamarajakshi

84 Monitoring and Evaluation Mechanism of Agrarian Reform and Rural

Development in Selected Asian Countries
Durga Prasad Paudyal

100 Book Reviews

Vol. I, December 1991, No. 2

1 Strengthening the Linkage Between Ecological Security and Livelihood Security

in Rural Areas
M. S. Swaminathan

16 My Development Education
Akhtar Hameed Khan

Credit Programme for Rural Development and Poverty Alleviation in Malaysia

35 Chamhuri Siwar

126
 65 Labour Absorption in Bangladesh Agriculture: Problems and Policies
Muhammad Masum

82 Book Reviews

Vol. II, July 1992, No. 1

1 How Poor are Women in Rural India?

A. K. Rajuladevi

Participatory Planning for Rural Development and Disaster Management in

35
Bangladesh
Hiroyuki Nishimura

Managing Development through Institution Building

58
Salehuddin Ahmed

75 Mining Resettlement and Rural Development in Malaysia

Hassan Naziri Khalid

Farmers’ Use of Communication Media in Adopting Agricultural Technologies-

94 A Farm Level Study in Bangladesh
M. A. Kashem, A. Halim and M. Zulfikar Rahman

113 Book Reviews

Vol. II, December 1992, No. 2

Can and Should Social Science Disaster Research Knowledge and Findings from
1
Developed Societies be Applied in Developing Societies?
E. L. Quarantelli

15 Community-Based Resource Management as a Strategy for Sustainable

Development
Francisco P. Fellizar, Jr.

36 Subsistence Farming in Asia: Income and Resource Allocation

Salehuddin Ahmed and Shafiqur Rahman
Structural Adjustment, Trade Liberalization and Exchange Rate Policies in
77
South Asian Countries: Some Relevant Issues
Mustafa K. Mujeri
81 Book Reviews

Vol. III, July 1993, No. 1

Development of Rural Industries and Transformation of China’s Rural
1
Economy
Momtaz Uddin Ahmed

127
 A Framework for Analysis of Macro-Micro Transmission Mechanisms in
20 Bangladesh: Some Preliminary Considerations
Mustafa K. Mujeri, Quazi Shahabuddin and Salehuddin Ahmed

Questions of Gender in Development Planning: Women’s Experiences in a New

44 Settlement of the Mahaweli Project in Sri Lanka
Joke Schrijvers

Transfer of Technology Systems in Bangladesh Agriculture

64
Indrajit Roy, M. A. Razzaque and A. K. Dey Sarker

71 Managing Project Sustainability

Adil Khan

75 Book Reviews

Vol. III, December 1993, No. 2

People’s Participation: Some Methods for Measuring Intensities Across

1
Development Sectors
Lokendra Prasad Poudyal and Karl E. Weber

19 Integrated Rural Development - Problems in Methodology and Institutional

Environment
Dirk Van Dusseldorp

41 The North-South Economic Interaction and the Environment

S. Mansoob Murshed

55 Book Reviews

Vol. IV, July & Dec. 1994, Nos. 1&2

Land Tenure and Production Organization of Agriculture in China, Vietnam

1 and Lao PDR
S. A. Subramani

52 Development from Within - A Rethinking on an Alternative Development Path

K. V. Sundaram

Participatory Development - The Community Information and Planning System

68
(CIPS): Management Experience of CIRDAP
S. Narayan

85 Local Orientation of Rural small Scale Industries : An Empirical Study from

Ciomas Subdistrict West Java, Indonesia
Tulus Tambunan

128
 118 Impact of the Integrated Rural Development Programme on the Rural Poor - A
Case Study of the Cuttack District in India
Krishna Chandra Rath and Jayant Kumar Routray
147 Book Reviews

Vol. V, July 1995, No. 1

Gender Aspects of Irrigation Management: the Chhattis Mauja Irrigation

1
System in Nepal
Margreet Zwartwveen & Nita Neupane
Gender in Agriculture : an Asian Perspective
27 B. Bhattacharya & G. Jhansi Rani

Women Farmers’ Worsening World : Can Gender Analysis Help?

49 Alexandra Stephens

59 Twenty Years of WID and Rural Women of Nepal

Meena Acharya

78 Rural Women, Poverty and Development in Pakistan

Shahnaz Kazi

93 Rural Women and the National Renovation Process in Vietnam

Le Thi
Is Women’s Loan Repayment Behaviour Different than Men’s? An Analysis of
103
Gender Differentials in Loan Repayment in Four CIRDAP Member Countries
Fahmeeda R. Wahab
Integrated Approach towards Small Family Norm, Income Generation and
119
Empowerment of Rural Women - A Case Study
G. N. Reddy, A. Rizwana and C. K. Gariyali

Vol. V, December 1995, No 2

Institutional Development for Managing a Rice Processing Complex in the
1 Philippines
Letecia N. Damole and Ebel Wickramanayake

The Applications of Geographical Information Systems in the Management of

19 Rubber Smallholdings : A Case Study of Malaysia
Ruslan Rainis and Abdul Halim Hashim

Participatory Process and Watershed Management - A Study of the Shiwalik

35 Foothill Villages in Northern India
Swarn Lata Arya and J. S. Samra
An Assessment of the Training and Visit Model of Agricultural Extension Work
58 in Bangladesh
Mohammad Hassanullah and Muzaffer Ahmad

129
 Some Recent Strategies on Women in Development and Poverty in the
84 Philippines
Florentina A. Tan

Effects of Irrigation on Household Income and Food Security in a Deeply

104
Flooded Area in Bangladesh
Mustafizur Rahman, S. N. Saha and Rezaul Karim

117 Book Reviews

Vol. VI, July 1996, No. 1

1 Sustaining Food Security for Rainfed Ecosystems in Asia

Mahabub Hossain

A Model for Analyzing Gender Relations in Two Tribal Communities in Orissa,

23 India
Smita Mishra and Reidar Dale

Village Livestock and Disease Control in Northern Thailand: A Survey

47 Examining Socio-economic Factors
T. Murphy and C.A. Tisdell

71 Rethinking Food Security in the Context of Development Paradigms

Kamal Nayan Kabra
83 Book Reviews

Vol. VI, December 1996, No. 2

1 Disaster Reduction and Preparedness

G.N. Ritchie

Disaster Mitigation through People’s Participation: Role of Local Self-

15
government Insitutions
A.V.S. Reddy, B.K. Thapliyal and K.R. Sastry

Politicizing Rural Development : Lessons from India’s West Bengal

35 Prabhat Datta

63 Stagnating Productivity in Crop Agriculture : The Quest for Sources

of Growth
Indrajit Roy

85 Research Notes - Loan Repayment Behaviour among the Rural Poor

- Lao PDR and Malaysia

103 Book Reviews

130
Vol. VII, July 1997, No. 1

1 Development Discourse and Post-Colonial Theory

Uma Kothari

11 Participation: Connotations and Content

K. R. Sastry

33 Micro-economic Reforms and Poverty Alleviation in Sri Lanka

Gamini Wickramasinghe and T.A. Dharmarathne

Practitioners’ Papers
Food Security in Lao People's Democratic Republic: An Assessment of the
57
Prevailing Situation and the Government's Food Security Regime

CIPS and PCM in the Implementation of the CIRDAP Action Research Project
77 on the Economic Impact of Peripheral Infrastructure on the rural Poor
G.N. Reddy

Self-help Groups - The Experiment of RASS

87 G. Muniratnam

Research Notes - Loan Repayment Behaviour among the Rural Poor

97
- Nepal, Pakistan, Philippines
123 Book Reviews

Vol. VII, December 1997, No. 2

Loan Contracts in Agrarian Credit Markets under Rationing Spillover

1
and Tied Situation
K.N. Selvaraj and N. Srinivasan

33 Decentralized Planning Framework for Public Interventions: Lessons from

Untied Funds Programme
K.R. Sastry and K.H. Rao
Poverty in Sri Lanka: Nature, Incidence and Measurement Issues
51
D.Gamage
Urbanization, Migration and Poverty: Some Recent Debates and Their Policy
81 Implications
K. N. Kabra
Practitioners' Papers
Utilization of Remote Sensing Techniques in Disaster Management - An
97 Administrative Experience
A.V.S. Reddy
Integrated Pest management: Institutional Constraints and Opportunities in the
101 Philippines
Marinela R. Castillo

131
 Farmers' Awareness on Environmental Population: A Study in Two Selected
107
Villages in Bangladesh
M.A. Hamid, A. Halim and S.M.A. Hossain

119 Research Notes

Vol. VIII, July 1998, No. 1

Spatial Patterns of Agricultural Development and Their Investment Priorities in

5 India
T. Haque, S.R. Hashim and Shanggen Fan

23 Malnutrition and Development: A Case Study of Sri Lanka

Mohottige U. Sedere
An Appraisal of the Adarsha Gram Project in Bangladesh
49 Syed Marghub Murshed and Syed Mansoob Murshed

71 Alternative Agriculture in Thailand and Japan

Sununtar Setboonsarng and Jonathan Gilman
Minor Irrigation, Input Response and Crop Output: The Study of a Block in a
87
North Bengal District, India
Sudip Chakranborty
Practitioner’s Paper
95 Micro-Credit and Income Generation in Rural Pakistan
Shabbir Hussain

109 Book Review

Vol. VIII, December 1998, No. 2

1 Access to Credit and the Effects of Credit on Resource Allocation

Decisions and Productivity in Indian Agriculture
K.N. Selvaraj, K. Chandran and K. R. Sundaravaradarajan

23 Rural Poverty and Importance of Small-Scale Industries:

The Case of Indonesia
Tulus Tambunan

Food Security and Sustainability Under Internationalization of Agriculture:

47
Some Reflections and Scope for Action with Reference to India
Sukhpal Singh

Farmers’ Organizations in Malaysia: Perspective, Role and Outlook

65 Abdul Aziz Abdul Rahman

83 Rural Small and Micro-Enterprises in Pakistan: Issues and Options

Rana Nasir Ali Khan

132
 93 Research Notes
Livestock in Asia: Action Research Experiences

101 The Rural Non-Farm Sector in Asia

Practitioners’ Paper
105 Vermiculture Biotechnology for Promoting Sustainable Agriculture
Chalasani Dutt, Siva Ram Krishna and A.V.S Reddy

Vol. IX, July 1999, No. 1

1 Development of Rainwater Harvesting for Domestic Water Use in Rural

Sri Lanka
R. de. S. Ariyabandu
Functioning of Informal Credit Market and its Linkages:
15
Evidences from Rural Credit Markets of South India
K.N. Selvaraj and K. R. Sundaravaradarajan
Participation of Rural Poor in Government and NGO Programmes:
37 A Comparative Study
Muhammad Samad
Corporate Participation in Rural Development: A New Model
57 Suggested
B. Yerram Raju
Impact of Irrigation Management Policy and Environment: Lessons
71
from Sri Lanka
M. M. Mohamed Aheeyar

Practitioner’s Paper
Microfinance in the Selected Asian Countries: An Overview
79 Shabbir Hussain

Vol. IX, December 1999, No. 2

Foodgrain Production in Bangladesh as influenced by Trends in Use of
1
Inputs and their Management at the Farm Level
Indrajit Roy

15 Production Structure and Technical Efficiency Analysis of Sericulture

in Pakistani Punjab
Munir Ahmad and Tanvir Khaliq Shami

33 Peoples’ Participation and Forest Management in India: Few Emerging

Issues
V. Reddappa Reddy

49 The Mystery of Mis-Targeting in Micro Credit

P. Subrahmanyam

133
 Measuring Women’s Empowerment: Some Methodological Issues
63 Tapash Kumar Biswas

93 Book Review

Vol. X, July 2000, No. 1

1 Assessment of Factors Influencing Agricultural Transformation Process:

Evidence from Midhills Region of Nepal
Surendra B. Thapa and Ganesh P. Shivakoti

27 Determinants of Household Earnings in Rural Economy of Thailand

Rajendra P. Shrestha and Apisit Eiumnoh

43 Economic Analysis of Peri-Urban Milk Production System Based on Findings

from a Survey of Small Dairy Herds Located around Islamabad City
Muzaffar Iqbal and R.H.Usmani
Institutions for Efficient and Equitable Use of Groundwater:
52 Irrigation Management Institutions and Water Markets in Gujarat,
Western India
M. Dinesh Kumar
Development Strategy of Rural Small-Scale Industries with a Cluster Approach: A
66
Case of Indonesia
Tulus Tambunan
Practitioner’s Paper
99 A Case Study Training as an Element of Post-1991 Flood
Disaster Recovery in Anhui and Jiangsu Provinces’ and in
Development of the China National Disaster Reduction Plan
Colonel G.N. Ritchie
Vol. X, December 2000, No. 2

1 Additionality, Fungibility, Convergence and Income and Substitution Effects of

Micro-credit: Findings of an Empirical Study in Andhra Pradesh, India
Dr. P. Dayachari, Dr. P. Subrahmanyam, and Prof. G. Raghava Reddy

19 Market Analysis and Development of Strategies: A Case Study from

Samal Municipality of the Philippines
Lin Zhen and Jayant Kumar Routray

37 What did they Need and What did we Bring? An Analysis of the Impact of
Giving Tenurial Rights to Encroached Land on Rural Development, An
Example of Sri Lanka
Sarachchandra Gamlath, and Piyadasa Edirisuriya
Community Management and Water Quality in Rural Water Supply
53 System in Nepal
Bhandari B.S. and Wickramanayake B.W
Institutional Opportunities and Constraints in the Performance of
67 Farmer-Managed Irrigation Systems in Nepal
Neeraj N. Joshi, Elinor Ostrom, Ganesh P. Shivakoti and Wai Fung Lam

134
 Practitioners’ Papers
93 Counter Disaster Staff Training
G.N. Ritchie

101 Public Information Kiosks for Inspiring Rural Change in India

K.A. Raju

Vol. XI, July 2001, No. 1

1 Non-Timber Forest Products (NTFPs) on Shifting Cultivated Lands (Khoriya),

A Means of Improving Livelihood of Chepang Rural Hill Tribe of Nepal
Bishnu Hari Pandit

Technical Efficiency Analysis of Pakistan Agriculture with Special Reference to

15
Chemical Fertilizer Use
Ghulam Mustafa Chaudhry

Analysis of Changes in the Cropping Pattern of Assam During 1965/66-1994/95:

26
An Econometric Study
Mercydi Maibangsa and S. Maibangsa

The Personal and Professional Problems of Field Workers of NGOs in

40 Bangladesh
Mokbul Morshed Ahmad

Agricultural Sustainability Through the Empowerment of Rubber Smallholders

65
in Thailand
Somboonsuke, B., Ganesh P. Shivakoti and H. Demaine

90 Contribution of Beneficiary Participation in Project Effectiveness in Watershed

Management Projects -A Case Study in Shivalik Foothill Region in Northern
India
Swarn Lata Arya, J. S. Samra and R.K. Aggarwal

Vol. XI, December 2001, No. 2

HRD Interventions for Rural Development: Analysis of Asian Experiences

1
P. Durgaprasad

15 Local Organizational Capability for Land Management: A SWOT Analysis of

Organizations in the Hills of Nepal
Giridhari Sharma Paudel and Gopal B. Thapa

35 Community Based Resource Management through Social Forestry Programme

of India: A Micro Level Study of Hindol Block in Orissa State
Dolagobinda Pradhan and Jayant K. Routray
Promotion of Integrated Agriculture-Aquaculture Farming Systems in
50
Northeast Thailand: Need for Judicious Considerations
Jharendu Pant, Harvey Demaine and Peter Edwards

135
 Regional Variations in Health Status: The Indian Case of Health Care and
64
Economic Reforms
K. Hanumantha Rao
Managing System within A Non-Systemic Vicious Circle: Institutional Linkage
86
Analysis to Identify the constraints of Technology Transfer and Adoption under
Crop Diversification Programme in Bangladesh
Jiban Ranjan Majumder and Ganesh P. Shivakoti

Vol. XII, July 2002, No. 1

Financial Viability of Drip-Irrigation System for Sugarcane and Grape

1 Cultivation in Maharashtra
AN Sarker and J S Hanamashetti
Exploring the Impacts of Structural Adjustment on the Provision of Selected
32
Economic and Social Services in Rural Sri Lanka
Joseph Mensah and Widanage Rupananda
The Rice Cultivation in Bangladesh: A Linear and Quadratic Programming
52
Approach
Sayed Hossain, Ming Yu Cheng and L.V.L.N Sarma

65 Arsenic Accumulation in Kangkong (Ipomoea reptans) and the Effects of

Phosphate Fertilizer in its Availability in Soil
M. M. Rahman, Ganesh P. Shivakoti, M. J. Uddin and S.M.A.T. Khandakar
Productivity, Technology, Infrastructure Growth and Investment Assessment
76
for Poverty Reduction in Dryland Agriculture
K.N.Selvaraj, C.Ramasamy, Anil Kuruvila and A.Rohini

Technological Problems in Small and Medium Fruit Processing Industries

89
(SMFPIs): A Case Study of East Java, Indonesia
I. B. Suryaningrat and V. M. Salokhe
Practitioners’ Paper
103 Participation in Natural Resource Management: A Case of Rural Energy Systems
B. Bhattacharya and S.K. Nanda

Vol. XII, No. 2, December 2002

1 Women’s Empowerment and Current use of Contraception

Tapash Kumar Biswas and Prof. Dr. M. Kabir

14 Environmental Impacts of Groundwater Exploitation in the North China Plain’s

Agriculture: Farmers’ Perceptions and Remedial Measures, A Case Study in
Ningjin County of Shandong Province, P.R. China
Lin Zhen and Jayant K. Routray
33 NGOs as Partners of Government in Resettlement – The Case of Rengali
Irrigation Project, India
Binaya Kumar Rout, and Reidar Dale

136
 47 Industrial Districts and Rural Development: A Case Study of Markazi Province
in Iran
Abdoreza R. Eftekhari, Ali Asgary and Mehdi Taherkhani

60 Local Participants’ Evaluation of Community Forestry in the Middle Hills of

Nepal
Ambika P. Gautam, Edward L. Webb and Ganesh P. Shivakoti

82 Performance Analysis of Irrigated Wheat Farms in the Lower Indus Basin

Intizar Hussain, Fuard Marikar and Waqar Jehangir

Practitioners’ Paper
Agricultural Credit and Rural Development
106
D Jha and S K Jha
115 Book Review

Vol. XIII, No. 1, July 2003

1 Monitoring and Evaluation of the Bangladesh Integrated Nutrition Project

Rezaul Karim and F. James Levinson

13 Employment Pattern among Tribals and non-Tribals in Eastern India: A

Retrogressive View
Sujeet K. Jha, O.N. Kunzru, Sabyasachi Das, G S Bisht, S.R.K. Singh and
S. K. Jha

24 Factors Affecting Adoption of Management Practices in Smallholder Oil Palm

Plantations of Banten Province, Indonesia
Widi Hardjono, S.L. Ranamukhaarachchi and Gajendra Singh

43 The Efficacy of Selected Predators against White-backed Planthopper (WBPH),

Sogatella furcifera (Horvath) (Homoptera: Delphacidae) in rice in Myanmar
Mu Mu Kyaw and S.L.Ranamukhaarachchi

60 Prominent Livelihood Asset Pentagon within the Analytical Framework of

Irrigation System Performance Assessment
Shiddi Ganesh Shrestha and Ganesh Prasad Shivakoti

Practitioners’ Paper
89 Does Leadership Matter? A Study of Self-Initiated Forest Management from
Central India
Rucha Ghate and Deepshikha Mehra

105 Book Review

137
Vol. XIII, No. 2, December 2003

1 Crop Diversification: An Empirical Analysis on Kangra Farms of Himachal

Pradesh, India
Mahajan Girish and Parkash Mehta

19 Assessment of Participatory Extension Approaches for Sustainable Agriculture

Development in Uplands of West Sumatra, Indonesia
Astia Dendi and Ganesh P. Shivakoti

44 Profit Gains by Collectors and Traders from Non-timber Forest Products

Trading in the Malekukhola Watershed in Nepal
Bishnu Hari Pandit and Gopal B. Thapa

56 Budgeting Process Improvement in Rural Infrastructure Develpment of Thai

Sub-District Local Government
N. Leungbootnak and C. Charoenngam

79 Management Information System for Microfinance Institutions

Chennapragada Ram Seshagiri Rao

Practitioners’ Papers
95 Perspectives on Emergence and Growth of Microfinance Sector
Naresh Singh

87 People’s Participation in Planning and Management of Rural Infrastructure: A

Study in Kerala State of India
Y. Gangi Reddy

109 Book Review

Vol. XIV, No. 1, July 2004

1 Empowerment of Women in Bangladesh: Does it Help Make Reproductive

Decision?
AKM Ahsan Ullah

10 Impacts of Rice-Prawn Gher Farming on Cropping Patterns, Land Tenant

System, and Household Income in Bangladesh – A Case Study of Khulna District
Basanta Kumar Barmon, Takumi Kondo, and Fumio Osanami
29 Constraints of Tilapia (Oreochromis niloticus) Production in Bangladesh:
Looking from Socio-economic Perspective
Ganesh P. Shivakoti and Jiban R. Majumder
56 Multiplication of Crossandra (Crossandra unduleafolia var: “Danica”) by
In-vitro Shoot Tip Culture
A. H. S. Wijesekara, D. C. Bandara and S.L. Ranamukhaarachchi

68 Explaining Intervention Outcome in Farmer Managed Irrigation System

Pradyumna Amatya, Wai Fung Lam, Nicole McCoy and Prashant Amatya

138
 Practitioner’s Paper
87 Agricultural Uses of Fishpond Sediment and Sustainability of Asian Integrated
Farming Systems
Md. Mizanur Rahman
95 Book Review

Vol. XIV, No. 2, December

1 Fruit Marketing Systems in Asia: Patterns, Problems and Policies

Deepak M. Pokhrel and Gopal B. Thapa
14 Women’s Empowerment under the Wavli System: A Study on Tribal Women in
South Gujarat, India
P. C. Sikligar
24 Linking Entrepreneurial Development and Natural Resources Management: An
Indian Experience
Chukka Kondaiah
39 Promotion of Farm Forestry in Laos Enhances Creation of Individual Land
Property
Dietrich Darr and Holm Uibrig
52 Initiation of Seawater Irrigation Co-Management for Marine Shrimp Farming in
Thailand
Pornpimon Chuaduangpui and Ganesh P. Shivakoti
Practitioners’ Papers
70 Community Participation in Land Reclamation Project: A Case Study
C. S. Singhal
89 Sustainable Development Needs and Options for Tripura (India)
Shashi Shekhar

Vol. XV, No. 1, July 2005

1 “State-in-Society” Approach and Implications for Rural Development Policy

Sony Pellissery
Balanced Scorecard-Based Performance Measurement for Rural Infrastructure
21
Development of Thai Sub-District Local Government
S. Posayanant and C. Charoenngam
Public Hearing in Rural Infrastructure Development Projects in Thailand: the
41 Case of the Ta Chin River Barrages
Ektewan Manowong and Stephen O. Ogunlana

55 Natural Resources Management for Sustainable Rural Development in Tribal

Areas: A Suggested Strategy for District Kinnaur in Himachal Pradesh, India- A
Case Study
Akbar S. Rawat

139
 Practitioners’ Papers
63 Soil Fertility and Land Productivity under Different Cropping Systems in
Highlands and Medium Highlands of Chandina Sub-District, Bangladesh
S. L. Ranamukhaarachchi, Md. Mizanur Rahman, Shamsun Nahar Begum
Female headed Households Resources and Constraints: A Study in Orissa, India
77
Trilochan Tripathy and Padmaja Mishra
Delineating the discipline of Natural Resource Management: Domain and
95
Approaches
Manoj T. Thomas
107 Book Review

Vol. XV, No. 2, December 2005

1 Institutional Framework for Promoting Rural Development in Bangladesh: An

Analytical Review
Momtaz Uddin Ahmed, Eva Benita A. Tuzon, and Shafiqur Rahman
33 Household Water Security through Demand Responsive Approaches in Sri Lanka
Rajindra de S. Ariyabandu and M. M. M. Aheeyar

51 Site-specific Fertiliser Management by GIS for Higher Yield Goals: A Precision

Agriculture Approach for Sustainability in Irrigated Rice Production of
Bangladesh
Abu Ahmed Mokammel Haque, H. P. W. Jayasuriya and V. M. Salokhe
Institutional Mechanism for Rural Development: The Role of NGOs with
65 Special Reference to the Bangladesh Rural Advancement Committee (BRAC)
Abdul-Muyeed Chowdhury and Mehnaz Rabbani
Quality of Life (QoL) Concept for the Evaluation of Societal Development
83
Rural Community in West Bengal, India
A. K. Pal and U. C. Kumar
95 An Analysis of Marketable Surplus of Potato in Bangladesh
A S M Anwarul Huq, Shamsul Alam and Shaheen Akter

Vol. XVI, No.1, July 2006

1 From the Peasant Charter to the ICARRD: An Overview of the Current Trends
and Emerging Issues in Rural Development in the Asia-Pacific Region
Durga P. Paudyal

61 Rural Infrastructure: A Critical Issue for Farm Productivity in Asia

B. Sudhakar Rao

79 Rural Development and Poverty Alleviation through Promotion of Rural Non-

farm Activities in the Asia-Pacific Region: Review of Evidence, Patterns and
Issues
Momtaz Uddin Ahmed

140
 Vol. XVI, No. 2, December 2006

1 Rural Poverty Assessment with Gender Dimension in Myanmar

Dolly Kyaw and Jayant K. Routray

The Role of Multi-Purpose Co-operatives in Rural Development in

31
Sri Lanka: A Case Study of Kurunegala District
M.B. Ranathilaka, Yoshiharu Shiratake and Piyadasa Edirisuriya

Community-led Mangrove Rehabilitation: Experiences from

53
Hua Khao Community, Songkhla, Thailand
Somsak Boromthanarat, Md. Zakir Hossain and Bussabong Chaijaroenwatana

69 Brooding Management Practices of Smallholder Turkey

Farmers in Imo State, Nigeria
I.C. Okoli, C. H. Nwodu, M. C. Uchegbu and O. M. Adesope

77 Communication Strategies for Disaster Preparedness in

Agriculture Sector in Bangladesh
M. A. Kashem

Carp Seed Traders in Bangladesh: Sources of Livelihood and

97
Vulnerability Resulting from Fish Seed Mortality
Mahmud Hasan and Amrit N. Bart

121 Fertiliser Requirements of Morning Glory for its

Higher Production in the Asian Region
Md. Mizanur Rahman, Amararatne Yakupitiyage and S. L. Ranamukhaarachchi

Practitioner’s Paper
133 Building Sustainable Rural Development Index in Asia and the Pacific:
Scoreboard on the Levels of Development of Rural Communities and
Country Performance in Rural Development
Eva Benita A. Tuzon

141
 Asia-Pacific Journal of Rural Development

NOTE TO CONTRIBUTORS

1.0 Aims

The Asia-Pacific Journal of Rural Development (APJORD) is an interdisciplinary journal

devoted to issues and discussions on rural development primarily in the Asia-Pacific
region. APJORD focuses on poverty issues and rural transformation, keeping in view the
four programme priorities of CIRDAP, namely (1) Agrarian Development; (2)
Institutional/ Infrastructural Development; (3) Resource Development including Human
Resources; and (4) Employment.

2.0 Submission of Articles

Articles are accepted for publication provided they are submitted to APJORD only. If
there is more than one author, a statement to the effect that the article has been approved
by all the authors must accompany it. It is the responsibility of the author(s) to obtain
permission for the use of previously published or unpublished materials including tables
and figures. All articles are subject to peer review and may be revised if necessary. The
editor's decision regarding articles submitted for publication is final. Copyright rests with
CIRDAP.
Authors are entitled to three complimentary copies of the issue in which their articles
appear. Articles written between 4,000-5,000 words are preferred. Both the hard copy and
soft copy of the write-up must have to be submitted to the editor. The soft copy either may
be given in floppy diskette/pen drive/CD or can be sent as an attachment to the e-mail
address: infocom@cirdap.org.
Typescripts should be submitted in duplicate, on A4 size paper, typewritten on one side, in
double space, with margins of at least 2.54 cm. (one inch). Statistical tables, illustrations
and charts should be submitted on separate sheets, and their positions should be indicated
in the text.

3.0 Style and Presentation

APJORD mostly follows the Chicago Manual of Style for preparing articles. Authors
should, as far as possible, adhere to the following basic requirements of style:

i. The name and institutional affiliation of the author should appear as a footnote on the
first page together with acknowledgement, if any.
ii. An abstract of approximately 100 words should precede the text on the first page. It
should be titled ‘Abstract’ indented, italicised and centred.

142
iii. British and ‘s’ spellings should be used. Numbers from zero to ten should be spelled
out, and numerals should be used for all other numbers.
iv. Where references are made without any comment, they should be inserted at the
appropriate places in the body of the text in parenthesis giving the surname of the
author(s) and year of publication of the work under reference, e.g. (Shama 1915: 41),
(Mayco and Santalines 1991: 18), (Jupitar 1985: 21; Razia 1999: 66).
v. Where references are made with detailed notes/comments, they should appear as
Notes, numbered sequentially throughout the article by superscripting in raised arabic
numbers and the corresponding notes appearing as footnotes or listed sequentially at
the end of the article.
vi. The reference list should appear after the Notes. For books give author’s surname,
other name(s) and/initial(s), year of publication, title of publication, place of
publication, and publisher, e.g.
Syed, Shireen Kamal. 1992. Women in Post-harvest Loss Prevention. Dhaka: Centre
on Integrated Rural Development for Asia and the Pacific.
Tamanga, A., N Shrestha., K. Sharma, and S. Kathi, 1994. Case Studies of Women
with Abortion Experiences. Kathmandu: IDS.
Acharya, Meena. 1994. Political Participation of Women in Nepal. In Nelson, Barbara
and Chowdhury, and Norma edited Women and Politics Worldwide. London: Yale
University Press.

For Journal articles, give the author’s surname, other name(s) and/or initial(s), year of
publication, title of article, volume and issue number, and inclusive pagination. e.g.
Quarantelli, E.L. 1992. Can and Should Social Science Disaster Research Knowledge
and Findings from Developed Societies be Applied in Developing Societies?
Asia-Pacific Journal of Rural Development, Vol. II, No. 2, December. Pp.1-14.

143
 ASIA-PACIFIC JOURNAL OF RURAL DEVELOPMENT

ISSN 1018 - 5291

ORDER FORM

Asia-Pacific Journal of Rural Development, Six-Monthly, First issue published in July 1991

Please enter my/our subscription to Asia-Pacific Journal of Rural Development beginning with

Vol. ______ No. ______

Name :

Designation :

Organisation :

Address :

City : Postcode :

Country : E-mail :

Signed Date

Annual Subscription Rates for two issues (including postage): US$20 for Asia-Pacific
countries excluding Bangladesh. Tk. 300.00 for Bangladesh. US$35 for all other countries

Please send Order Form to:

Editor
Asia-Pacific Journal of Rural Development (APJORD)
Chameli House, 17 Topkhana Road, GPO Box 2883, Dhaka-1000, Bangladesh
Fax : 880-2-9562035 E-mail: infocom@cirdap.org

Visit our website at

http://www.cirdap.org.sg

144