Climate Change and Water Resources in South Asia - Chapter 11 (end) pdf

11
Using the Adaptation Policy Framework to
Assess Climate Risks and Response Measures in
South Asia: The Case of Floods and Droughts in
Bangladesh and India
11.1 INTRODUCTION
South Asia is noted for climate and hydrological extremes such as floods, droughts, heat
waves, and cyclones. The climate of South Asia is highly influenced by the Southwest
monsoon (see Chapter 1). More than three-quarters of the annual precipitation occurs in
the monsoon months (June-September). The onset and departure of the monsoon is
spatially highly variable, so the precipitation is also. The failure of the monsoon and high
summer temperatures leads to drought in many parts of Bangladesh, India and Pakistan. In
the Eastern Coast of India and in the coastal region of Bangladesh disastrous cyclones are
regular visitors. Glacier Lake Outburst Floods (GLOFs) in Bhutan, Nepal and Pakistan
cause disasters to life and property downstream, resulting in serious death tolls as well as
the destruction of valuable forests, farms and costly mountain infrastructure. In Nepal and
Bhutan, 44 glacier lakes have been identified as potentially dangerous and which may
result in GLOF (ICIMOD, 2001). In South Asia, particularly in the Himalayan region, the
frequency of the occurrence of GLOF events increased in the second half of the 20
th
century.
The 1990s was the warmest decade of the last century and several extreme climate
events occurred in the South Asia region. In July of 1993, the Tistung station in Nepal
registered 540 mm rain over a 24-hour period triggering a severe flood. Severe droughts
occurred over large regions in India and Pakistan in 2000. Bangladesh experienced the
worst flood in recent history in 1998 which engulfed about 70% of the country. It appears
that extreme climate events are increasing in frequency and magnitude, causing more deaths,
injury, disability and disease, economic and social impacts in the impoverished nations of
South Asia (Table 11.1). The Intergovernmental Panel on Climate Change (IPCC) (2001)
concluded that there would be likely increases in intense precipitation events, droughts,
tropical cyclone peak wind intensities and tropical cyclone mean and peak precipitation

intensities in the future due to climate change. Therefore, a dramatic increase in damage is
also expected.
M. MONIRUL QADER MIRZA
IAN BURTON
Copyright © 2005 Taylor & Francis Group plc, London, UK
Table 11.1 Extreme climate events and damages in selected countries in South Asia

Extreme Event Bangladesh India Nepal Pakistan

Flood * On average 21.5% is
inundated.
*Inundation may increase to
70%.
* During 1953-2000, 15,678
people died.
* On average 475 persons
died per event.
* Extreme flood can cause
economic damage of US$ 3 to
5 billion.
* 40 million ha area are
flood vulnerable.
* Average flooding
event affected 34
million people.
* 1,595 people/event.
* US$ 250 million
economic loss/event.
* During 1983-2000,
5,935 people died.

* On average 330
people died per
event.
* In July of 1993, one
single flood event
killed 1,336 people.
* Economic loss of
1993 flood was
Rs. 4,904 million
* Dig Tsho GLOF in
1985, destroyed the
Namche small
hydropower project
that built at the cost
of US$ 1.5 million.

* Floods affect
urban areas
severely.
280 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
Table 11.1 Continued

Extreme Event Bangladesh India Nepal Pakistan

Drought

* Past droughts typically
affected 47% of the country.
* Affected 53% of the

population.
* On average, loss crop in
drought is equal to a flood.
* In 1979, drought created
almost a famine like situation.

* 68% of the landmass
is vulnerable to
droughts.
* Severe droughts occur
once in every 9 years.
* The 1987 drought
affected 60% of the
cropped area and 285
million people.

* Droughts occur
occasionally but
effects are not
significant.

* Droughts affect
agriculture
severely.
* In 2000-2001,
due to drought
GDP growth rate
dropped to 2.6% as
against targeted
5%.

Cyclone

* Cyclones affect Bangladesh
severely.
* The 1970 cyclone killed
250,000 people.
* The 1991 cyclone killed
138,000 people.

* India is one of the
worst cyclone affected
countries.
* In 1999, the Orissa
super cyclone killed
10,000 people.


Not vulnerable

Not vulnerable
M. M. Q. MIRZA AND IAN BURTON 281
Copyright © 2005 Taylor & Francis Group plc, London, UK


0
500
1000
1500
2000

2500
3000
3500
1953 1958 1963 1968 1973 1978 1983 1988 1993 1998
Year
Damage (in million Rs. x 10)
5-y ear moving average
Fig. 11.1 Damage due to floods/heavy rains in India during 1953-2000. Figures for 1999 and 2000 are
tentative. Data source: Singh, 2001.
11.2 ADAPTATION POLICY FRAMEWORK
While substantial literature exists (Carter et al., 1994; IPCC, 2001; US Country Study
Program, 1996; and Feenstra et al., 1998) regarding climate change impacts, information on

Climate change poses a considerable risk to the future sustainable development of
countries in South Asia. How might the countries of the region best respond to these
risks? The diverse character and widespread nature of the risks is described as above.
From these we have selected urban floods in Dhaka, Bangladesh and droughts in Gujarat,
India as specific case studies.
We have selectively drawn upon some of the concepts and methods in the
Adaptation Policy Framework (APF) (Fig. 11.2) (UNDP, 2004) and applied them to the two
case studies. In doing so we recognize that present adaptation falls short of what is
necessary to prevent the further growth of vulnerability and damage potential. There is in
fact a current adaptation deficit in coping with climate variability and extremes even
without taking into account the added risk associated with climate change (Burton, 2004).
For this reason we recognize two types of adaptation. Type I Adaptation refers to current
adaptation strategy, policy, and measures without considering climate change. Most of
the adaptation measures are in practice belong to Type I. Type II Adaptation is the
additional adaptation that is required to cope with climate change.
Because climate change risks have still not been factored into many development
decisions, and because awareness of the need for adaptation has still not been well

incorporated into the work of development agencies/ministries in the developing
countries and because adaptive capacity is lacking, not much Type II adaptation has taken
place. In this regard, the APF has been designed to help factor climate change risks in
to development decisions in order to reduce vulnerability and facilitate sustainable
development. The APF approach is briefly described in Section 11.2 and some of the major
concepts are described in Section 11.3. We then use the APF as a means of formulating an
analysis in the two case studies (Section 11.4). Finally, we discuss opportunities and
challenges associated with the APF with particular reference to the two case studies.
282 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
adaptation policies and strategies is limited. Burton et al. (2002) commented that effective
adaptation policy had to be responsive to a wide variety of economic, social, political, and
environmental circumstances. In order to do so, a common framework of concepts, linked
together in a flexible manner is required. Therefore the development of the Adaptation
Policy Framework (APF) has been motivated by the lack of practical guidance on
adaptation to climate change. The driving concern underlying the APF was that
discussions about climate change adaptation had not progressed significantly beyond
the identification of possible adaptation measures (UNDP, 2004). The goal of the APF is to
help narrow a wide range of policy options and measures into site-specific policies for
particular climate risks. Since the potential effects of climate change are pervasive,
adaptation can include a wide range of responses and policies in all economic sectors and
all regions (UNDP, 2001). The framework is intended to integrate short-, medium- and
long-term threats to national economic development planning, as well as the relevant
current policies and measures. In designing the APF, coping with present climate
variability is seen as an effective way to reduce long-term vulnerability to climate change.
Fig. 11.2 Outline of the Adaptation Policy Framework (APF) process (UNDP, 2004).
Countries in South Asia have now conducted some studies (ADB Country Study
Program, UNFCCC National Communications, US Country Study Program, etc.) under
Stage I Adaptation (Box 11.1). However, it is recognized that more work is needed to
progress to the next step and to prepare for Stage II Adaptation (Box 11.1), towards which

the APF is specifically directed. Over the long-term, this framework is critical for preparing
the ground for detailed analysis in Stage III Adaptation (Box 11.1).
The APF has five major steps (Fig. 11.2) compared to the seven steps of the “first
generation” of impact and vulnerability assessment method (Carter et al., 1994). The APF
(UNDP, 2004) is more robust and flexible and its “first generation” counterpart and
designed to fit present and future requirement in terms of climate variability and change.
The five-step analysis is supported by 9 Technical Papers (TPs) which are: APF Project
Scope and Design, Stakeholder Engagement in the Adaptation Process, Vulnerability
Assessment for Climate Adaptation, Vulnerability Assessment for Climate Adaptation,
M. M. Q. MIRZA AND IAN BURTON 283
TP

1

–

P
ROJECT
S
COPE
AND
D
ESIGN

TP

3

–

V
ULNERABILITY
A
SSESSMENT

TP

4

–

C
URRENT
C
LIMATE
R
ISKS

TP

5

–

F
UTURE
C
LIMATE
R
ISKS

TP

6

–

S
OCIO
-
ECONOMIC
C
ONDITIONS

TP

8

–

A
DAPTATION
S
TRATEGY

TP

9

–

C
ONTINUING
A
DAPTATION

TP

2

–

S
TAKEHOLDER
E
NGAGEMENT

Project scope and
design
Assessing current
vulnerability
Continuing the
adaptation process
Characterizing
future climate risks
Developing an
adaptation strategy
TP

7

–

M
EASURING AND
E
NHANCING
A
DAPTIVE
C
APACITY

Copyright © 2005 Taylor & Francis Group plc, London, UK
Assessing Current Climate Risks, Assessing Future Climate Risks, Socio-Economic
Conditions, Measuring and Enhancing Adaptive Capacity, Formulation of an Adaptive
Strategy and Continuing the Adaptation Process. The APF and TPs can be downloaded
from />11.3 VULNERABILITY AND ADAPTATION: A BRIEF SYNTHESIS
11.3.1 VULNERABILITY
The concept of vulnerability has gone through a comprehensive evolution process in the
last few decades. Generally it is defined from three perspectives: natural hazard, climate
change and variability and entitlement.
From a natural hazards perspective Blaikie et al. (1994) defined vulnerability as “…the
characteristics of a person or group in terms of their capacity to anticipate, cope with,
resist and recover from the impact of a natural hazard (p.57)”. It is focused on only human
systems and three temporal situations in terms of natural extreme events that cause
hazards are taken into account: pre-event and post-event and during the event. The
authors also argue that vulnerability “…is a measure of a person or group’s exposure to
the effects of a natural hazard, including the degree to which they can recover from the
impact of that event (p.57)”. The exposure refers to physical, economic and human
well-being and recovery is related to adaptive capacity and resiliency.

Kelly and Adger (2000) widened the definition of vulnerability as “…the ability or
inability of individuals or social groupings to respond to, in the sense of cope with, recover
from or adapt to, any external stress placed on their livelihoods and well-being (p.300).”
Their approach focuses on existing “wounds” (or prior damage), which might limit
capacity to respond to stresses and are independent of future threats.
The Intergovernmental Panel on Climate Change (IPCC) (2001) broadened natural
hazard perspective based definition by focusing on the future as well as incorporating
natural systems in addition to human system. It defines vulnerability as “…the degree to
which a system is susceptible to, or unable to cope with, adverse effects of climate change,
including climate variability and extremes. Vulnerability is a function of the character,
magnitude and rate of climate change and variation to which a system is exposed, its
sensitivity, and its adaptive capacity (p.18).”
Many authors (e.g., Liverman, 1994; Adger and Kelly, 1999) have argued for the use
of a political economy framework, often using the “entitlements approach” which begins


Box 11.1 Initial Guidance from the Conference of the Parties on Adaptation
(Decision 11/CP.1)

Stage I
: “Planning, which includes s tudies of pos s ible impacts of climate change
to identify particularly vulnerable countries or regions and policy options for
adaptation and appropriate capacity building”.

Stage II
: “Measures, including further capacity building which may be taken to
prepare for adaptation as envisaged in Article 4.1(e)”.

Stage III
: “Meas ures to facilitate adequate adaptation, including ins urance and

other adaptation measures as envisaged by articles 4.1(b) and 4.4”.
284 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
at household level (developed by Sen (1981, 1987), in analyses of vulnerability. A
household’s food entitlement consists of the food that the household can obtain through
production, exchange, or extra-legal legitimate conventions - such as reciprocal relations
or kinship obligations (Drèze and Sen, 1989). Ribot (1996) argues this approach
introduces a household perspective on vulnerability, one that replaces “eco-centric”
approaches to environmental change. The main contribution of this approach lies, perhaps,
in its focus on the vulnerability of individuals and social groups. Within this framework
vulnerability is understood as being determined by access to resources-specifically, by
individuals’ “entitlement” to call on these resources. Watts and Bohle (1993), using Drèze
and Sen’s (1989) analysis of entitlements, argue that vulnerability is configured by the
mutually constituted triad of entitlements, empowerment and political economy. Here
empowerment is the ability to shape the political economy that in turn shapes entitlement.
The Food and Agriculture Organization (FAO) of the United Nations (1999) defines
vulnerability from the food security perspective as “the presence of factors that place people
at risk of becoming food insecure or malnourished.” This definition focuses on causes of
food insecurity due to human interventions, such as political decisions, armed conflicts and
international economic embargo. Inappropriate political decisions often cause hunger in
Sub-Saharan Africa and Asia; armed conflicts either do not allow food distribution or
purchase of food due to diversion of resources for buying military hardware/software; and
international economic embargoes often lead to hunger by limiting a country or government’s
spending power or accumulation of economic resources.
11.3.2 ADAPTATION
A number of definitions of adaptation can be found in the literature. IPCC (2001) defined
adaptation as an adjustment in natural or human systems in response to actual or expected
climate stimuli and their effects or impacts, which moderates harm or exploits beneficial
opportunities. It refers to changes in processes, practices and structures to moderate
potential damages or to benefit from opportunities associated with climate change. Smithers

and Smit (1997) describe adaptation as involving “change in a system in response to some
force or perturbation”. Pielke (1998) refers adaptation “to adjustment in individual, group
and institutional behavior in order to reduce society’s vulnerabilities to climate. Adger
(2001) views adaptation as a dynamic social process and believes that the ability of a
society to act collectively determines its ability to adapt.
11.3.2.1 ADAPT TO WHAT?
Adaptation occurs in both natural and socio-economic systems (Burton et al., 1998).
People generally adapt and practice measures to adapt to the variability of natural climate
and extreme weather events. Human intervention modifies the threat of natural variability.
However, human action can cause irreversible damage to systems and their natural
resiliency may be lost. Burton et al. (1993) pointed out that human activities are not
always as well adapted to climate as they might be. The mounting losses from great natural
disasters are in substantial part associated with extreme weather events. Therefore, in a
situation where natural climate and hydrologic systems have been modified by human
intervention, even efficiently designed corrective measures might be proven to be either
partially effective or ineffective.
M. M. Q. MIRZA AND IAN BURTON 285
Copyright © 2005 Taylor & Francis Group plc, London, UK
11.3.2.2 ADAPTIVE CAPACITY
IPCC (2001) defined adaptive capacity as the potential, capability, or ability of a system to
adapt to climate change stimuli or their effects or impacts. Adaptive capacity depends on a
number of determinants that include: socio-economic wealth, governance, technology,
information and skills, infrastructure, institutions and equity. Among these determinants,
socio-economic factors are the most important determinants that help develop adaptive
capacity.
Socio-economic factors affect the ability of a system to absorb (robustness) or
respond to changes that occur to natural system due to natural causes or human
interventions (Smith et al., 1998). In South Asia, socio-economic conditions of various
economic groups, location and living conditions, inequality between rural and urban
population (including their intra inequality) and gender broadly defines exposure of these

groups to extreme weather events or human interventions.
11.3.2.3 ADAPTATION TYPES
Various types of adaptation include anticipatory and reactive adaptation, private and
public adaptation, and autonomous and planned adaptation. Salient features of various
types of adaptation are presented in Table 11.2.
11.3.2.4 ADAPTATION MEASURES
There are many potential adaptation measures that may be adopted in response to climate
change and variability. Burton et al. (1993) divided them into the following eight
categories depending on the individual’s choice of options. The choice typology has been
extended to include the role of community structures, institutional arrangements, and
public policies (also see Fig. 11.3). Table 11.3 summarizes the measures.
286 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
Fig. 11.3 Classification of adaptation options (Burton et al., 1993).
Table 11.3 Classification of adaptation measures

Classification Examples

Bear the cost Accept the cost because there is no other choice.
Share the losses Use insurance or government relief, or community or
family sharing.
Modify the events Modify the actual physical events themselves
(e.g. flood control, coastal surge protection).
Prevent the events or their
effects
Preventing drought by cloud seeding but very few
success stories.
Change use occurs when
an economic activity is
impossible or extremely

risky
Change human use activities (e.g. regulate floodplain
land-use; use drought-tolerant crops).
Change location Relocating major crops and farming regions, shifting
human settlement and livestock population.
Research Development of salt tolerant crops for coastal region,
rice varieties that can remain underwater for a longer
period, etc.
Education for behavioral
change
Saving water to reduce climate driven water demand;
conservation of energy to reduce cooling demand, etc.

11.4 PRESENT VULNERABILITY AND ADAPTATION MEASURES AND POLICIES
IN SOUTH ASIA: URBAN FLOODING IN DHAKA
11.4.1 URBAN FLOODS IN DHAKA, BANGLADESH
Bangladesh acts as the drainage outlet for the three large rivers: the Ganges, Brahmaputra
and Meghna (GBM). Huge rainfall in the basins during the monsoon, geographical
A dap ta tio n/
Resp ons e Opti ons
Legislativ e, Regulatory , Financial
Struc tur al, Technological
Insti tutional, Adminis trativ e
M arket Base d
Share the Loss
On-Site Operati ons
Bear the Loss
M odify the Ev ents
Prev ent the Effec ts
C hange U se

C hange Loc ation
Research
Educa tion, Behav ior al
1
2
3
4
5
6
7
8
M. M. Q. MIRZA AND IAN BURTON 287
Copyright © 2005 Taylor & Francis Group plc, London, UK
In the 1980s and 1990s, three extreme floods in 1987, 1988 and 1998 engulfed 36%,
63% and 69% of the country, respectively and caused human, environmental and
economic devastation in Bangladesh. During the flood of 1988, Dhaka City - the capital of
Bangladesh was severely affected. Again in 1998, a catastrophic flood engulfed the greater
Dhaka area in the months of August and September. Due to the flooding, about 56% of the
greater Dhaka was submerged, and affected about 1.9 million people (30% of the
population).
proximity and flat terrain of Bangladesh make it highly vulnerable to recurring flooding.
On average about 20.5% of Bangladesh gets inundated annually. In extreme flooding years,
the extent of inundation may be as much as 70%. Figure 11.4 shows year to year extent of
flooding in Bangladesh. Four types of floods commonly occur in Bangladesh: flash,
riverine, rainfall and storm-surge floods (Box 11.2).
Fig. 11.4 Extent of flooded area (%) in Bangladesh from 1954 to 2001. Source: Flood Forecasting and
Warning Center (FFWC), Dhaka.


Box 11.2 Bangladesh flood types

The Northern, Northeastern and Southeast parts of Bangladesh are vulnerable to
f
lash floods
. They usually occur due to a heavy rainfall in the neighboring hills and
mountains in India as well as in Bangladesh. The normal period of flash flooding
is late April to early May.
Riverine floods
are caused by over bank spillage of
monsoon flows in the major rivers and their distributaries. Riverine floods may
occur several times depending on timing and magnitude of rainfall in the basins
and may prolong for months in the monsoon (June-September).
Rainfall floods

occur when high local rainfall generates huge volume of runoff in the rivers and
streams exceeding the drainage capacity. Occurrences of such floods are common
when the three major rivers are at high stages.
Storm-surge floods
occur during
October-December and April-May in the low-lying coastal areas of Bangladesh.
Tropical cyclones generate storm-surges that bring tidal bores often 9 m high
(Ahmed and Mirza, 2000).
288 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
11.4.2 VULNERABILITY OF DHAKA TO FLOOD HAZARDS
Dhaka City is highly vulnerable to flooding and subsequent hazards (Fig. 11.5) for a
number of reasons. First, the city is surrounded by the distributaries of the two major
rivers, the Brahmaputra and Meghna. The surrounding rivers are Buriganga to the South,
Turag to the West, Tongi Khal to the North and Balu to the East. The combined area of

Dhaka East and Dhaka West known as Greater Dhaka covers an area approximately
275 km
2
(JICA, 1991). Dhaka is situated on a flat terrain which makes it vulnerable to
flooding. The elevation of Greater Dhaka is only 2 m-13 m above mean sea level (MSL)
and most of the urbanized areas are at 6 m-8 m above msl. About 62% area of Greater
Dhaka is below 6 m (JICA, 1987). This is consistent with the overall elevation of Bangladesh
where 80% of land area is below 12 m above MSL.
Second, Dhaka’s population has been growing at a very fast rate. Urbanization in
Bangladesh is poverty driven-caused by an unsustainable rural economy characterized
by extreme entitlement contraction among the majority of marginalized peasantry
(Barkat et al., 1997). Other causes include riverbank erosion, flooding, droughts and
cyclones. The present population of the Dhaka Metropolitan Area is more than 10 million.
The last decadal growth rate was about 70%, though the population growth rate was even
higher. In the decade 1981-1991, population doubled. Population statistics of Dhaka City
show that the annual growth rate was 2.9% (1951-1961), 10.2% (1961-1974) and 8.1%
(1974-1981) (Table 11.4).
Third, poverty is another important factor that makes the poor sections of Dhaka
more vulnerable to flooding. About 30% of Dhaka’s population is classified as a hardcore
poor (per capita monthly income ≤US$ 43) and 50% as poor (per capita monthly income
≤US$ 65). Altogether about 3 million or nearly one-third of Dhaka’s population live in
2,100 slums and squatter settlements (Rahman and Tariquzzaman, 2001). In particular,
Dhaka suffers from shortage of basic infrastructure and services such as water supply,
sanitation, solid waste disposal and transport. The problem is not only shortage, but also
unequal distribution of service, with much of the impact absorbed by low income and
poorer section of Dhaka.
Fourth, loss of internal water bodies increases the vulnerability to flooding. Dhaka
used to have a number of canals (Dhoali Khal, Begunbari Khal, etc.) connected to the
surrounding rivers and large water bodies. Most of these canals and water bodies have
disappeared over the last 3-4 decades mainly due to private and public encroachments.

Therefore, drainage congestion is a regular event and flooding from drainage overflow is a
severe problem even after a moderate shower. The water depth in some areas may be as
high as 40 cm-60 cm, which results in large infrastructure problems for the city, economic
losses in production, and damage to existing property and goods (Huq and Alam, 2003).
Fifth, loss of carrying capacity of surrounding rivers increases vulnerability. Dhaka is
surrounded by four rivers: Buriganga, Turag, Tongi Khal and Balu. Over the years,
the water carrying capacity of these rivers has been lost due to siltation and illegal
encroachments. Therefore, floodwater quickly overtops the bank and inundates the
surrounding urban area. Recession of floodwaters also takes longer time for the same
reason.
Sixth, rapid urbanization and built-up areas lead to shortening of the runoff
concentration time and an increase of the peak flow. In the last three decades rapid
urbanization has occurred in Dhaka. Therefore a substantial increase in development of
residential and commercial areas has taken place to accommodate rapid growth of
population at the initiatives of private land developers, real state business and public
M. M. Q. MIRZA AND IAN BURTON 289
Copyright © 2005 Taylor & Francis Group plc, London, UK
290 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
sector. These actions resulted in substantial growth of impervious areas, created
obstruction to natural drainage patterns, and reduced detention basins. Bari and Hasan
(2001) investigated the effects of urbanization on runoff concentration time and peak flow
with the aid of NAM
1
conceptual model and the Rational Formula
2
(Kuichling, 1889) and
found an increased volume of runoff with the increase of built-up areas in Dhaka City.
Table 11.4 Population of Dhaka in various decades
1

The NAM conceptual lumped hydrologic model was developed by the Technical University of
Denmark in 1973 and is widely applied in Bangladesh.
2
Kuichling (1889) first applied the Rational Method to estimate peak discharge. The formula is
Q
p
= FCIA, where F = unit conversion factor, C = runoff coefficient, I = intensity of rainfall (mm/hr)
and A = drainage area (km
2
).
3
In Bangladesh danger level at a river location is the level above which it is likely that the flood may
cause damages to nearby crops and homesteads. In a river having no embankment, danger level is
about annual average flood level. In an embanked river danger level is fixed slightly below design
flood level of the embankment.
Dhaka City was also severely affected by the 1998 flood (Fig. 11.6). The water levels in
the rivers surrounding Dhaka approached the respective danger levels in the second half
of July and crossed the danger levels in mid-August. Peak floods usually occur in the last
week of August and the first two weeks of September and floodwater recedes in the last
ten days of September. The floods of 1987 and 1988 followed this pattern. However,
during the flood of 1998, the water levels crossed the danger levels almost a month earlier
and stayed there until the last week of September (Faisal et al., 2003). Peak water levels,
return period and days above the danger level
3
of the Buriganga River are given in
Table 11.5. Seventy out of ninety, Dhaka City corporations were under water (Jahan, 2000).
The flood affected almost all aspects of human life including income, health and occupation.
People of various income and occupation suffered in varying degrees, and there were also
significant spatial variations in the impact of the flood (Jahan, 2000).
11.4.3 ADAPTATION AND COPING MECHANISMS

The location of the Dhaka City has made it particularly vulnerable to floods. It is
surrounded by the Buriganga to the South, Turag to the West, Tongi Khal to the North,
and Balu to the East. Dhaka City and the adjoining areas are composed of alluvial terraces
of the Southern part of the Madhupur tract and low-lying areas of doab of the rivers
Meghna and Lakhya. The city suffered from flooding mainly due to the spillage of the
Year Area (km
2
) Population
(million)
Source

1951 - 0.34 Census of Pakistan, 1951
1961 28 0.55 Census of Pakistan, 1951
1971 40 1.5 Census of Bangladesh, 1974
1974 40 1.6 Census of Bangladesh, 1974
1981 155.4 3.44 Census of Bangladesh, 1981
1991 - 6.95 Census of Bangladesh, 1991
2001 1,530 10.6 Census of Bangladesh, 2001

M. M. Q. MIRZA AND IAN BURTON 291
Copyright © 2005 Taylor & Francis Group plc, London, UK
Fig. 11.6 Flooded area in Dhaka during 1998 floods. Source: Faisal et al., 2003. Reprinted with the
permission of Kluwer Academic Publishers, the Netherlands.
Table 11.5 Water levels in the Buriganga River and their return periods

River/Station Flood
Year
Danger
Level
Peak

Level
Return
Period
Days Above
Danger Level

1954 6.00 7.06 29.5 46
1955 7.09 31.5 31
1974 6.61 13.5 24
1987 6.64 11.7 17
1988 7.58 94.0 23
Buriganga/
Dhaka
1998 6.70 13.5 36

surrounding rivers. Local rainfall often complicates the flooding situation. Although the
city was periodically flooded, adaptation and coping mechanisms are not well documented
but some initiatives were taken in the wake of disastrous flooding of 1988 and 1998
(Huq and Alam, 2003; Hye, 1999; Faisal et al., 1999; and Jahan, 2000).
Dhaka Flood Protection Embankment: The first flood protection embankment was
the Buckland Flood Protection Embankment along the Buriganga River constructed
during the early period of the British rule. It was the first attempt to mitigate flood damage
292 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
in Dhaka City (Huq and Alam, 2003). In the wake of the 1988 floods, construction of an
embankment encircling Dhaka City was commenced under a “crash programme”. Under
Phase I of the project, a flood protection embankment for the Western part of the city was
completed. However, the Eastern part of the city remains unprotected (Faisal et al., 1999).
As a result, almost the entire Eastern block was inundated during floods in 1998. Although
the embankment provides some protection, it has increased vulnerability to flooding.

Unplanned urbanization is taking place in the low-lying areas adjacent to the Western part
of the embankment. In 1998, about 20% of the Western block was also inundated by
floods.
Non-Structural Measures: Important non-structural measures include flood
forecasting and warning, retention ponds, natural water bodies and drainage network,
land-use planning and relief and rehabilitation. Other practiced non-structural measures
are summarized in Table 11.6.
The FFWC of Bangladesh Water Development Board (BWDB) administers flood
forecasting and warning in collaboration with the Institute of Water Modeling (IWM),
Bangladesh Meteorological Department (BMD) and Space Research and Remote Sensing
Organization (SPARRSO). Three hydrologic forecasting techniques-MIKE 11 Simulation
Model, Muskingum/Cunge Flood Routing Method and Gauge-to-Gauge correlation used.
The FFWC provides flood and river forecasts for 16 locations for 24-h and 48-h periods.
In addition to this, it also provides daily river level and rainfall data for 50 rivers and
49 rainfall stations. Although these warnings are useful, a forecast in terms of inundation
area would be more useful in making people understand the danger of floods. Apart from
radio, TV and newspapers, there is no community-based mechanism to communicate flood
forecasting and warning to the city dwellers.
Dhaka City used to have many natural water bodies, which functioned as a buffer for
floodwaters. Over the years, the natural water bodied dwindled significantly due to public
encroachments for land development. Virtually no natural water bodies left in the old part
of the city. Encroachments are continuing even in the new upscale residential areas of
Gulshan, Banani and Baridhara. The minimum standard for a retention pond is 12% of the
urban area whereas the present area is estimated to be less than 4% (RAJUK, 1995). The
government has recently issued a decree banning the filling in of any wetland for urban
development (Huq and Alam, 2003).
Jahan (2000) investigated socio-economic coping mechanisms in Dhaka City during
floods in 1998. The poorer sections of the society were hard-hit as the duration of the
flood was more than two months. Many used up their savings and in addition, borrowed
money to survive (Fig. 11.7). More than 35% of the credit came from relatives followed by

shopkeepers and neighbors. Some people sold assets and mortgaged properties to buy
food and other daily necessities. Help and assistance also came from various public and
private organizations in terms of food, clothing, housing materials, medicine, water
purification tablets, money, etc. (Table 11.7). Among the respondents, 44% and 40% said
that they partly and completely recovered from the flood. The remaining 16% could not
recover at all.
11.4.4 DROUGHTS IN INDIA: CASE STUDY OF GUJARAT
High drought prone areas in India are located mainly in the Western part of the country
with arid and semi-arid climate (Fig. 11.8). However, occasionally other parts of the country
are also vulnerable to droughts. The Planning Commission of India identified 54 drought
prone districts distributed over 13 states (Kulshrestha, 1997). In India a drought is
M. M. Q. MIRZA AND IAN BURTON 293
Copyright © 2005 Taylor & Francis Group plc, London, UK
Table 11.6 Non-structural activities practiced by various groups
Activity Description
Remarks


Emergency services Medical care, potable water, food, candle, clothing,
temporary housing, shelter
Widely practiced but limited to accessible places
Flood proofing Raising plinth level of the house, building on tall pillars,
flood walls along properties, raising important roads and
some power stations above the 1988 flood level, special
embankment for the Zia International Airport
Minimum ground elevation proposed for houses for
the eastern part of the city
Flood fighting Temporary flood wall made by brick or sand bags, water
pumping, moving assets to upper floors, roof or elevated
high grounds such as roads and embankments

People fight with floods until unbearable
Flood shelter Community centers, educational institutions, public
buildings, roads, embankments
No specific urban flood shelters, other facilities used
as shelters are either not designed as such or have
insufficient capacity
Recovery and
reconstruction
Vulnerable group feeding, food for work, supply of
building materials, soft or interest free loans for business
and agriculture
Affected group specially the poor people has limited
access to such help. Misappropriation of relief and
rehabilitation material or fund is common

Note: Modified from Faisal et al. (1999).
294 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
Source: Jahan, 2000.
In 2000, Gujarat and some other Indian states were severely hit by a drought.
It affected 25 million people in 17 out of 25 districts. Most severely impacted districts
were: Kutch, Jamnagar, Junagadh, Rajkot, Amreli, Bhavnagar, Surendranagar, Mehsana,
Banaskantha, Sabarkantha, Panchmahals, Vadodara, Bharuch and Saurastra.
In 1987, another severe drought hit India. Rainfall in seven meteorological
subdivisions was significantly lower than normal: Saurastra, Kutch and Diu (-74%), West
Rajasthan (-67%), Haryana and Delhi (-67%), Punjab (-58%), Himachal Pradesh (-51%),
Plains of West Uttar Pradesh (-51%) and East Rajasthan (-50%). The drought was caused
considered to be “moderate” if the deficiency is between 26%-50% of the normal rainfall
and “severe” if the rainfall deficiency is greater than 50% of the normal (IMD, 1971; GOI,
1976). The largest number of droughts occurred in the first quarter of the last century. The

second quarters of both centuries experienced comparatively small number of droughts.

Types of Assistance Govt.
Agencies
Non-Govt.
Agencies
Voluntary
Agencies
Other

Food 29.80 23.08 43.27 11.54
Clothing 0.96 1.92 2.88 1.92
Housing materials 0.96 0 0.96 0
Medicine 6.73 10.57 15.38 4.81
Water purification tablets 6.73 8.65 13.46 2.88
Money 0 4.81 8.65 0.96
Other 0.96 1.92 0.96 1.92

Fig. 11.7 Distribution of households by sources of credit. Source: Jahan, 2000.
Table 11.7 Percentage distribution of households by types and sources of assistance
M. M. Q. MIRZA AND IAN BURTON 295
Copyright © 2005 Taylor & Francis Group plc, London, UK
11.5 VULNERABILITY OF GUJARAT TO DROUGHT HAZARD
11.5.1 ANNUAL RAINFALL
The main cause of the vulnerability of Gujarat to recurrent drought is low rainfall. Annual
rainfall in the state varies from 330 mm-1520 mm. The rainfall in the Southern highlands of
Saurashtra and the Gulf of Cambay is approximately 630 mm while the other parts of
Saurashtra have annual rainfall of less than 630 mm. The semi-arid area of Kutch has a very
low rainfall. Long-term departure of rainfall from the mean (= 431 mm for the period
1871-1994) is shown in Figure 11.9. In 1999, the India Meteorological Department (IMD)

noted that Saurastra and Kutch in Gujarat had received 58% deficient rainfall in the
monsoon (Mirza, 2000). Potential evapo-transpiration (PET) in the state is also high
(Table 11.8). In some areas, water deficit weeks exceeded the number of moist weeks.
by a record delay in the progression of the monsoon. Although the monsoon onset at
Kerala was on time (2 June) - it slowly reached Delhi on 26 July - a full four weeks delay.
The drought problem in 1987 was compounded by the weak monsoons in the two
preceding years of 1986 and 1985 (Kulshrestha, 1997). Food grain production of the Kharif
crop fell by 5.6 million tons.
Fig. 11.8 Mean annual rainfall in India. Source: Kulshrestha, 1997.
296 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
Figure 11.9 Mean annual precipitation in Saurashtra and Kutch in Gujarat. Data source: IITM,
Pune, India.
11.5.2 LOW RIVER RUNOFF
With the exception of the Narmada and Tapi Rivers, there are few year-round water
resources to sustain agricultural production in the region. Though Kutch has many rivers,
they are small and do not have much water. Those flowing North disappear in the desert,
while those flowing in other directions join the sea. Most of the rivers of Saurashtra and
Kutch dry up in the summer.
11.5.3 SENSITIVITY OF CROP AND VEGETATION TO AMOUNT OF RAINFALL
Crops and vegetation in the region are highly sensitive to the amount of rainfall. The Kutch
region, once covered with 1.5 m tall green grasses, has now been reduced to dusty plains.
Most of the cattle grazers are battling the drought for their livelihood (Nathan, 2001). The
-600
-400
-200
0
200
400
600

800
1871
1878
1885
1892
1899
1906
1913
1920
1927
1934
1941
1948
1955
1962
1969
1976
1983
1990
1997
M. M. Q. MIRZA AND IAN BURTON 297
Copyright © 2005 Taylor & Francis Group plc, London, UK
available water is inadequate to support agriculture. Table 11.9 shows the water
availability period for Gujarat and Saurashtra regions. In 1986 and 1987 in Gujarat, against
a target of 2 million tons, rabi production fell by 0.9 million tons (Nathan, 2001). In 2000,
food grain production declined by 30%.
11.5.4 FORECASTING
Forecasting and warning of monsoon in advance can save crops and reduce human misery
to a great extent. India Meteorological Department (IMD) has a 16-parameter model for
monsoon forecasting. The 16 parameters include regional and global scale temperature,

wind pressure and snow related meteorological variables. Based on favorable characters
of the parameters a “good” or “bad” monsoon is predicted. The model is not capable of
determining the “inter-spell duration” which is very important for crop agriculture. In the
past, the model successfully made predictions of monsoon but there have been instances of
failure too (for example, in the year 2002).
11.5.5 OVEREXPLOITATION OF GROUND WATER
Over the years, due to drought and many other socio-economic reasons, ground water in
the region has been significantly depleted. Ground water resources are overexploited in
the state, with the water table going down nearly 4 m per year, particularly in the
pre-monsoon season. In the drought years of 1999-2000, water levels began to dip
drastically. The immediate response of the people to this depletion was the deepening of
existing wells, drilling boreholes and drilling radial boreholes in the already deepened wells.
Overdraft of ground water has led to the problem of seawater ingress, particularly along
the Saurashtra Coast (Nathan, 2001).
11.5.6 GENDER INEQUALITY
Children and women are most vulnerable to the aftermath of natural hazards in South Asia.
As a nation, India is committed to a policy of increasing women rights and freedom, but
gender inequality is deeply rooted like many other similar global situations. During sudden
298 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
or prolonged disasters, the patterns of inequality are translated into lack of economic
assets (which include savings, credit, land, tools and training), personal safety and
nutrition, health care, social security, etc. In Gujarat, due to social class differences, which
cross-cut caste identities, women from 23 villages (62%) observed that low income women
were most hurt by intersecting effects of embedded disasters (Enarson, 2001).
11.5.7 ADAPTATION AND COPING MECHANISMS
Water scarcity was the most crucial problem in the drought-affected areas. Gujarat state
government and relief agencies spent Rs. 20,000 million for relief and rehabilitation
operation. About half of the resources went to supply drinking water by tankers and piped
water system. These coping mechanisms are ad-hoc and not sustainable in the long-run.

A number of options were suggested. First, policy-level recognition of traditional sources
of water such as talavs (lakes), virdas (shallow holes into which ground water seeps and is
collected for drinking) and vavs (stepped well) is required. In addition to this, check dams
and storage dams to for harvesting water at the village level will reduce the scarcity of
water (Vabadam, 2001). Second, decentralization of water management systems need to
be in place. It was argued that the people themselves can very well point out decentralized
solutions that might yield benefits much earlier and more cheaply than mega water project
such as the Narmada (Sangvai, 2000). Third, modifying the present structure of property
rights over ground water. Ground water is not presently considered to be a common
resource. According to the law, it belongs to the owners of the land in which it is located.
This law has resulted in landowners trying to withdraw as much ground water possible
regardless of the extent of their needs (Vabadam, 2001).
Agriculture was also hard hit due to lack of rain, irrigation water and soil dryness.
Somewhere between 190 mm-250 mm of rainfall fell in a span of 60 days to 80 days of
crop growth is the requirement of dry land agriculture. Crops fail if the amount of rainfall
is less than 115 mm-150 mm. Therefore, at least this amount of irrigation is required to
avoid crop failure. Canal irrigation in Gujarat is dependent on the water available in the
dams, which is also a function of rainfall. It was found that crop productivity is directly
proportional to the filling of dam with water. If a dam is less than 50% full, water is usually
not supplied for canal irrigation but is conserved for future need. Another alternative is
large-scale recharge and decentralized ways of water harvesting. Sangvai (2000) suggested
that the recharging of 200,000 wells would raise the ground water level throughout
Saurastra. In the past such a campaign was found to be successful. During 1995-1998,
farmers recharged thousands of wells. The endeavor does not involve big budget,
bureaucratic and unwieldy planning. Peasants can implement this speedily without
complicated technology at a cheaper cost.
Altering pricing policy for agriculture is an option to reduce water demand. The
pricing policy should be formulated in such a way that people are encouraged to grow
other crops. Groundnut, a water intensive crop, is widely cultivated in Saurashtra, is the
world’s largest supplier of this cash crop. Any attempt to make farmers switch to other

crops will be resisted by the powerful exporters and vested quarters.
11.6 STAKEHOLDERS’ PARTICIPATION
11.6.1 BANGLADESH
Stakeholders’ participation in flood control/mitigation measures and disaster management
M. M. Q. MIRZA AND IAN BURTON 299
Copyright © 2005 Taylor & Francis Group plc, London, UK
has undergone a long evolutionary process. Flood control/mitigation received major
attention after successive disastrous floods in 1954 and 1955. Stakeholders’ participation
in planning, design and implementation of the water sector in Bangladesh can be divided
into three major phases.
Phase I (1955-1990): After the successive floods in the mid-1950s, the then
government of Pakistan undertook a massive flood control program for East Pakistan
(now Bangladesh). The highly bureaucratic East Pakistan Water and Power Development
Authority (EPWAPDA) was created in 1959 and was administered mainly by the
engineers. Up to 1991, all public sector water projects were driven by a Master Plan
developed in 1964. The approach to development was centrally driven and planned.
All the administrators and technicians had been trained primarily in Pakistan and were not
able to adjust to the reality in Bangladesh. The orientation of the EPWAPDA was like a
military administration where information was controlled in a military way. For example,
maps were restricted and office of the Surveyor General of East Pakistan was under the
Ministry of Defense (Pittman, 1994).
This kind of management without the participation of various other levels of
stakeholders created conflicts between farmers, fishers, and tradesmen with different
interests in the project area. First, “public cuts” are one of such problems during a flood
when people inside and outside the project area cut an embankment to reduce the threat.
Second, unnecessary projects were implemented at the wish of politicians and engineers.
These projects created more problems than well-being. Third, operation and maintenance
are also affected. Projects are usually imposed from the top upon the landscape. Therefore
the structures quickly dry up, wash out, or silt up because of lack of local level
participation in their maintenance. Eventually the projects tend to run down and fail (Pittman,

1994). Fourth, lack of participation worsened environmental hazards. Many flood control
projects (e.g., the Chandpur Irrigation Project) created environmental hazards such as the
depletion of floodplain fisheries, employment, reduced supply of protein, water logging,
soil salinity and agriculture pollution (Mirza and Ericksen, 1996).
The top level bureaucracy was further expanded with the creation of National Water
Council (NWC) in 1986 headed by the head of the government. One positive aspect was
that several experts outside the government were rotationally chosen as members of the
council. After the disastrous floods of 1987 and 1988, the government decided to
re-examine the flood problem in Bangladesh. The Flood Action Plan (FAP) with
27 components supported by 15 donors was launched. In order to oversee the activities of
the FAP, the Flood Plan Coordination Organization (FPCO) was created. The FPCO was
under the Ministry of
Water Resources and was independent from the BWDB. But the
majority of the manpower of FPCO with ‘old school of thought’ was drawn from the
BWDB. Initially a broad-based stakeholder participation was not in the statute of the
FPCO.
Phase II (1991-1998): Under the pressure from the non-government organizations
(NGOs) and Civil Society, the plan was gradually changed from a structurally-oriented
plan in 1990 to a plan with more emphasis on the environment and public participation.
Special components for public participation were built into the FAP (Pittman, 1994). The
Compartmentalization Pilot Project (CPP), Tangail planned and executed for the first time
from a multi-disciplinary approach by taking into account the needs of fisheries, navigation
and agriculture. Guidelines for public participation were produced and the government
approved them. In the 2
nd
and 3
rd
national FAP conferences, many professionals outside of
government as well as grass roots representatives were allowed to attend and raise
questions. The FAP was completed in 1995 and the FPCO was renamed “Water Resources

300 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Copyright © 2005 Taylor & Francis Group plc, London, UK
Planning Organization (WARPO)”.
Phase III (1999-): A National Water Policy was formulated in 1999, recognizing that
the participation of stakeholders was an integral part of water resources management.
Special attention has been given to stakeholder participation intended to elicit direct input
from people at all levels of engagement. It stated that the “Guidelines for People’s
Participation (GPP) in Water Development Projects be adhered to as part of the project
planning by all institutions and agencies involved in public sector management of water
resources”. It also emphasized exploring opportunities and undertaking efforts to ensure
participation of the landless and other disadvantaged groups (MWR, 1999). During
preparation of the National Water Plan (1998-2002), the WARPO developed and
implemented a People’s Participation and Consultation Programme (PPCP) targeting a
wide cross-section of stakeholders. It was conducted through programmes of village, union,
thana and district meetings, national and regional workshops and special group meetings
with government agencies.
11.6.2 INDIA
In India two tiers of administrations the central and state governments, conduct water
resources management. Therefore, the participatory model of stakeholders is rather
complex. Figure 11.10 shows institutional arrangements in a top-down approach for the
water sector in India. The central Ministry of Water Resources is responsible for policy
guidelines and programs for the development and regulation of country’s water resources.
One of its main functions is to provide technical guidance, clearance and monitoring of the
irrigation, flood control and multipurpose projects (major/medium). The Ministry’s
other major function is the operation of the central network for flood forecasting and
warning on inter-state rivers. The state government has also responsibility of water
management. The central Ministry has 17 organizations, which are involved with water
resources research, development and management. The Ministry of Agriculture and
Department of Rural Development also have watershed development programs, but the
inter-ministerial and inter-departmental coordination is rather weak. Unlike Bangladesh,

there are no exclusive guidelines for stakeholders’ participation in water management
projects although the Indian National Water Policy 2002 states “Efforts should be made to
involve farmers progressively in various aspects of management of irrigation systems,
particularly in water distribution and collection of water rates. Assistance of voluntary
agencies should be enlisted in educating the farmers in efficient water use and water
management”(MWR, 2002).
Although water resources management including extreme events like floods and
droughts are dominated by top-down approach, some bottom-up approach is taking place.
Successive droughts in Gujarat and Andhra Pradesh (AP) in 2000 and 2001 compelled the
state governments to launch a local level participatory water conservation programme
involving NGOs who have very strong grass roots level networks. One such programme is
Sardar Patel Participatory Water Conservation Programme (SPPWCP) in Gujarat launched
in January 2000; and the AP government launched the Neeru Meeru (Water and You)
programme in May 2000. Regarding the SPPWCP, Down to Earth (2000) made two
important observations. First, through launching the SPPWCP, the Gujarat government
learned from its past mistakes and also from the successes of villages led by civil society.
Second, the SPPWCP was formulated in a way that reduces bureaucratic wrangling.
The people responded with enthusiasm and submitted proposals for more than 25,000
check dams.
M. M. Q. MIRZA AND IAN BURTON 301
Copyright © 2005 Taylor & Francis Group plc, London, UK
Fig. 11.10 Stakeholders arrangements in a top-down approach for water sector in India. Source:
Enarth, 2002.
A bottom-up stakeholders’ participatory model in water management may run into
trouble because of conflicts of interests of various groups. Enarth (2002) discussed such a
case in Thalota, Meshna, Gujarat, India. Thalota, a village located at the tail end of a
medium sized irrigation project served by a reservoir located about 110 km upstream. The
majority of the population belongs to Patels and Thakores and Muslims, Brahmins, Banias
(traders) and Harjans (untouchables) are a minority. Rich upstream farmers used most of
the water flowing through the irrigation canal. An NGO negotiated with the upstream

farmers and government department about the scarce water situation of Thalota. It also
convinced the water users at Thalota to share the maintenance cost of the irrigation canal.
An MOU was signed in 1996, and for the first time in more than 15 years, farmers saw
water flow from the canal not just into the village but to the last plot of the farmland along
the water course. However, the expansion of this Participatory Irrigation Management
(PIM) was eventually stalled due to fear of NGO control over resources, a feeling of
subordination of the concerned government officials of the irrigation department and a
local politician member of the legislative assembly (MLA). In 1999, under intense political
pressure the NGO had to abandon their community building process in many of these
villages.
302 ADAPTATION POLICY FRAMEWORK TO ASSESS CLIMATE RISKS
Government of
India
Ministries of
Agriculture
Water Resources
Rural Development
Forest & Envir.
Departments of
Agriculture
Water Resources
Rural Development
Forest & Envir.
District Advisory Committee (MPs/MLAs
/
Panchayat Leaders/Civil Servants)
State
Government
District Rural
Development

Agency
Forest
Department
Projects Projects Contractors
User Groups User Groups User Groups
Irrigation
Department
Land
Development
Corporation
Copyright © 2005 Taylor & Francis Group plc, London, UK
11.7 PRESENT ADAPTATION POLICIES
11.7.1 BANGLADESH
There are no exclusive ‘adaptation policies’ for the water sector in the countries of South
Asia. Control of extreme weather events (floods and droughts) primarily focused on
‘reduction of crop loss’ and ‘maximizing agriculture production’. In Bangladesh, for
three decades (1960-1990) the control of floods was the prioritized policy. This policy
focused on ‘structural solutions’ through the construction of flood control embankments
to protect mainly agriculture and human settlements from flooding. The planning and
design of the projects were based on 100-year and 20-year return period floods along the
main rivers (Ganges, Brahmaputra and Meghna) and the minor rivers, respectively. This
policy was implemented through the creation of a large engineering organization the
Bangladesh Water Development Board (BWDB) (formerly EPWAPDA). The success
of such structural solutions is questionable, as in many areas failure and over-topping of
embankments are regular phenomena during severe floods (for example, the floods of
1998 and 1988).
In the early years of flood control initiatives, less importance was given to
non-structural measures that include flood modeling, flood forecasting and warning,
evacuation, flood shelters, etc. Flood modeling received attention during the preparation
of the 2

nd
Water Master Plan during 1983-1986 when the Danish Hydrologic Model ‘NAM’
was introduced. Subsequently, more sophisticated MIKE 11 and MIKE 21 hydrodynamic
models were also introduced, calibrated and validated. The Surface Water Modeling
Center (SWMC) (now IWM) was created in the late 1980s. The BWDB created flood
forecasting and warning center (FFWC) in 1972. It received technical and financial
assistance from the United Nations during 1981-1986 and 1989-1992. The Center now
uses hourly rainfall, discharge and water level data collected from selected stations for
simulating floods with the aid of MIKE 11-GIS model, and disseminating the warnings in
electronic and print media and to several government departments. The forecasting and
warning information is also available on the internet (www.ffwc.net). During the 1998
flood, forecasting and warning was found to be effective in reducing loss of lives and
property (Chowdhury, 2000).
Drought forecasting and management policies are rather neglected although its not
a lesser ‘menace’ than floods. In the period 1973-1986, the average loss of crops due to
droughts was as same as for floods (Mirza, 2002). There is no effective mechanism in
place in Bangladesh for drought forecasting. Drought occurs for three main reasons: low
residual soil moisture as a result of inadequate rainfall in the monsoon; low or no rainfall in
summer together with high evapo-transpiration; and low summer flow in the rivers/streams
unable to meet irrigation water demand. In addition, in many areas, low monsoon recharge
can cause very high draw downs of water tables in summer leading to crop loss.
Departments responsible for agriculture development do not prepare projected soil
moisture maps. The BWDB do not forecast river flows for the summer months or estimate
the amount of recharge that occurred in the monsoon or prepare maps of vulnerable areas.
The meteorological department does not have any long-range weather forecasting model.
Although the Ministry of Agriculture drafted ‘drought codes’ in 1980, they are focused at
‘reactive’ rehabilitation rather than ‘anticipatory’. As the delivery and implementation of
the measures suggested in the codes are heavily top-down, by the time they reach at the
grass roots level, the damage is already done.
M. M. Q. MIRZA AND IAN BURTON 303

Copyright © 2005 Taylor & Francis Group plc, London, UK